MCP41HVXX Data Sheet

MCP41HVX1
7/8-Bit Single, +36V (±18V) Digital POT
with SPI Serial Interface and Volatile Memory
 2013-2015 Microchip Technology Inc.
MCP41HVX1 Single Potentiometer
TSSOP (ST)
VL
SCK
CS
SDI
SDO
WLAT
SHDN
14
13
12
11
10
9
8
1
2
3
4
5
6
7
V+
P0A
P0W
P0B
VDGND
NC (2)
NC (2)
NC (2)
V+
5 x 5 VQFN (MQ)
NC (2)
NC (2)
20 19 18 17 16
VL
1
SCK
CS
SDI
SDO
2
3
4
21 EP(1)
15 P0A
14 P0W
13 P0B
12 V11 DGND
5
6 7 8 9 10
NC (2)
• High-Voltage Analog Support:
- +36V Terminal Voltage Range (DGND = V-)
- ±18V Terminal Voltage Range
(DGND = V- + 18V)
• Wide Operating Voltage:
- Analog: 10V to 36V (specified performance)
- Digital: 2.7V to 5.5V
1.8V to 5.5V (DGND  V- + 0.9V)
• Single Resistor Network
• Potentiometer Configuration Options
• Resistor Network Resolution
- 7-bit: 127 resistors (128 Taps)
- 8-bit: 255 resistors (256 Taps)
• RAB Resistance Options:
- 5 k
10 k
- 50 k
100 k
• High Terminal/Wiper Current (IW) Support:
- 25 mA (for 5 k)
- 12.5 mA (for 10 k)
- 6.5 mA (for 50 k and 100 k)
• Zero-Scale to Full-Scale Wiper Operation
• Low Wiper Resistance: 75  (Typical)
• Low Temperature Coefficient:
- Absolute (Rheostat): 50 ppm typical
(0°C to +70°C)
- Ratiometric (Potentiometer): 15 ppm typical
• SPI Serial Interface
(10 MHz, Modes 0,0 and 1,1)
• Resistor Network Terminal Disconnect Via:
- Shutdown pin (SHDN)
- Terminal Control (TCON) register
• Write Latch (WLAT) Pin to Control Update of
Volatile Wiper Register (such as Zero Crossing)
• Power-on Reset/Brown-out Reset for Both:
- Digital supply (VL/DGND); 1.5V typical
- Analog supply (V+/V-); 3.5V typical
• Serial Interface Inactive Current (3 µA Typical)
• 500 kHz Typical Bandwidth (-3 dB) Operation
(5.0 k Device)
• Extended Temperature Range (-40°C to +125°C)
• Package Types: TSSOP-14 and VQFN-20 (5x5)
Package Types
WLAT
SHDN
NC (2)
NC (2)
Features
Note 1: Exposed Pad (EP)
2: NC = Not Internally Connected
Description
The MCP41HVX1 family of devices have dual power
rails (analog and digital). The analog power rail allows
high voltage on the resistor network terminal pins. The
analog voltage range is determined by the V+ and Vvoltages. The maximum analog voltage is +36V, while
the operating analog output minimum specifications are
specified from either 10V or 20V. As the analog supply
voltage becomes smaller, the analog switch resistances
increase, which affects certain performance specifications. The system can be implemented as dual rail
(±18V) relative to the digital logic ground (DGND).
The device also has a Write Latch (WLAT) function,
which will inhibit the volatile wiper register from being
updated (latched) with the received data until the WLAT
pin is low. This allows the application to specify a condition where the volatile wiper register is updated (such
as zero crossing).
DS20005207B-page 1
MCP41HVX1
Device Block Diagram
V+
VL
DGND
CS
SCK
SDI
SDO
Power-up/
Brown-out
Control
(Digital)
V–
Power-up/
Brown-out
Control
(Analog)
SPI Serial
Interface
Module and
Control
Logic
P0A
Resistor
Network 0
(Pot 0)
WLAT
SHDN
P0W
Wiper 0
and TCON
Register
Memory (2x8)
Wiper0 (V)
P0B
TCON
RAB Options Wiper (k)
RW ()
Specified Operating
Range
1 Potentiometer (1)
SPI
3Fh
5.0, 10.0,
50.0, 100.0
75
127 128
1.8V to
5.5V
10V(4) to 36V
MCP41HV51
1 Potentiometer (1)
SPI
7Fh
5.0, 10.0,
50.0, 100.0
75
255 256
1.8V to
5.5V
10V(4) to 36V
MCP45HV31(5) 1 Potentiometer(1) I2C™
3Fh
5.0, 10.0,
50.0, 100.0
75
127 128
1.8V to
5.5V
10V(4) to 36V
MCP45HV51(5) 1 Potentiometer (1)
7Fh
5.0, 10.0,
50.0, 100.0
75
255 256
1.8V to
5.5V
10V(4) to 36V
Note 1:
2:
3:
4:
5:
Wiper
Configuration
I2C
RS
MCP41HV31
Device
# of POTs
POR Wiper
Setting
Number
of:
Control
Interface
Resistance (Typical)
Taps
Device Features
VL(2)
V+(3)
Floating either terminal (A or B) allows the device to be used as a Rheostat (variable resistor).
This is relative to the DGND signal. There is a separate requirement for the V+/V- voltages: VL  V- + 2.7V.
Relative to V-, the VL and DGND signals must be between (inclusive) V- and V+.
Analog operation will continue while the V+ voltage is above the device’s analog Power-on Reset
(POR)/Brown-out Reset (BOR) voltage. Operational characteristics may exceed specified limits while the
V+ voltage is below the specified minimum voltage.
For additional information on these devices, refer to DS20005304.
DS20005207B-page 2
 2013-2015 Microchip Technology Inc.
MCP41HVX1
1.0
ELECTRICAL
CHARACTERISTICS
Absolute Maximum Ratings †
Voltage on V- with respect to DGND ......................................................................................... DGND + 0.6V to -40.0V
Voltage on V+ with respect to DGND ........................................................................................... DGND - 0.3V to 40.0V
Voltage on V+ with respect to V- .................................................................................................. DGND - 0.3V to 40.0V
Voltage on VL with respect to V+ ............................................................................................................ -0.6V to -40.0V
Voltage on VL with respect to V- ............................................................................................................. -0.6V to +40.0V
Voltage on VL with respect to DGND ....................................................................................................... -0.6V to +7.0V
Voltage on CS, SCK, SDI, WLAT, and SHDN with respect to DGND ................................................ -0.6V to VL + 0.6V
Voltage on all other pins (PxA, PxW, and PxB) with respect to V- ......................................................-0.3V to V+ + 0.3V
Input clamp current, IIK (VI < 0, VI > VL, VI > VPP on HV pins) ............................................................................ ±20 mA
Output clamp current, IOK (VO < 0 or VO > VL) ................................................................................................... ±20 mA
Maximum current out of DGND pin ...................................................................................................................... 100 mA
Maximum current into VL pin................................................................................................................................ 100 mA
Maximum current out of V- pin ............................................................................................................................. 100 mA
Maximum current into V+ pin ................................................................................................................................100 mA
Maximum current into PXA, PXW, & PXB pins (Continuous)
RAB = 5 k ............................................................................................................................. ±25 mA
RAB = 10 k ........................................................................................................................ ±12.5 mA
RAB = 50 k .......................................................................................................................... ±6.5 mA
RAB = 100 k ........................................................................................................................ ±6.5 mA
Maximum current into PXA, PXW, & PXB pins (Pulsed)
FPULSE > 10 kHz ........................................................................................................... (Max IContinuous)/(Duty Cycle)
FPULSE  10 kHz ........................................................................................................ (Max IContinuous)/(Duty Cycle)
Maximum output current sunk by any Output pin .................................................................................................. 25 mA
Maximum output current sourced by any Output pin ............................................................................................ 25 mA
Package Power Dissipation (TA = + 50°C, TJ = +150°C)
TSSOP-14 ............................................................................................................................................. 1000 mW
VQFN-20 (5x5) ...................................................................................................................................... 2800 mW
Soldering temperature of leads (10 seconds) ..................................................................................................... +300°C
ESD protection on all pins
Human Body Model (HBM) ......................................................................................................................  ±4 kV
Machine Model (MM) ..............................................................................................................................  ±400V
Charged Device Model (CDM) for TSSOP-14  ±1 kV
Maximum Junction Temperature (TJ) ..................................................................................................................... 150°C
Storage temperature ............................................................................................................................. -65°C to +150°C
Ambient temperature with power applied .............................................................................................. -40°C to +125°C
† Notice: Stresses above those listed under “Maximum Ratings” may cause permanent damage to the device. This
is a stress rating only and functional operation of the device at those or any other conditions above those indicated in
the operational listings of this specification is not implied. Exposure to maximum rating conditions for extended periods
may affect device reliability.
 2013-2015 Microchip Technology Inc.
DS20005207B-page 3
MCP41HVX1
AC/DC CHARACTERISTICS
Standard Operating Conditions (unless otherwise specified)
Operating Temperature -40°C ≤ TA ≤ +125°C (extended)
All parameters apply across the specified operating ranges unless noted.
V+ = 10V to 36V (referenced to V-);
V+ = +5V to +18V & V- = -5.0V to -18V (referenced to DGND ≥ ±5V to ±18V),
VL = +2.7V to 5.5V, 5 k, 10 k, 50 k, 100 k devices.
Typical specifications represent values for VL = 5.5V, TA = +25°C.
DC Characteristics
Parameters
Sym.
Min.
Typ.
Max.
Units
Digital Positive
Supply Voltage (VL)
VL
2.7
—
5.5
V
With respect to DGND(4)
1.8
—
5.5
V
DGND = V- + 0.9V (referenced to V-)(1,4)
—
—
0
V
With respect to V+
Analog Positive
Supply Voltage (V+)
V+
VL(16)
—
36.0
V
With respect to V-(4)
VDGND
V-
—
V+ - VL
V
With respect to V-(4,5)
V-
-36.0 + VL
—
0
V
With respect to DGND and VL = 1.8V
VRN
—
—
36V
V
Delta voltage between V+ and V-(4)
VL Start Voltage to
ensure Wiper Reset
VDPOR
—
—
1.8
V
With respect to DGND, V+ > 6.0V
RAM retention voltage (VRAM) < VDBOR
V+ Voltage to ensure
Wiper Reset
VAPOR
—
—
6.0
V
With respect to V-, VL = 0V
RAM retention voltage (VRAM) < VBOR
Digital to Analog
Level Shifter
Operational Voltage
VLS
—
—
2.3
V
VL to V- voltage.
DGND = V-
Power Rail Voltages
during Power-Up(1)
VLPOR
—
—
5.5
V
Digital Powers (VL/DGND) up 1st:
V+ and V- floating
or
as V+/V- powers up
(V+ must be  to DGND)(18)
V+POR
—
—
36
V
Analog Powers (V+/V-) up 1st:
VL and DGND floating
or
as VL/DGND powers up
(DGND must be between V- and V+)(18)
Digital Ground
Voltage (DGND)
Analog Negative
Supply Voltage (V-)
Resistor Network
Supply Voltage
VL Rise Rate to
ensure Power-on
Reset
VLRR
Note 6
V/ms
Conditions
With respect to DGND
Note 1
This specification is by design.
Note 4
V+ voltage is dependent on V- voltage. The maximum delta voltage between V+ and V- is 36V. The digital
logic DGND potential can be anywhere between V+ and V-. The VL potential must be ≥ DGND and ≤ V+.
Note 5
The minimum value determined by maximum V- to V+ potential equals 36V, and the minimum value for operation equals 1.8V. So, 36V - 1.8V = 34.2V.
Note 6
POR/BOR is not rate dependent.
Note 16 For specified analog performance, V+ must be 20V or greater (unless otherwise noted).
Note 18 During the power-up sequence, to ensure expected Analog POR operation, the two power systems (Analog
and Digital) should have a common reference to ensure that the driven DGND voltage is not at a higher
potential than the driven V+ voltage.
DS20005207B-page 4
 2013-2015 Microchip Technology Inc.
MCP41HVX1
AC/DC CHARACTERISTICS (CONTINUED)
Standard Operating Conditions (unless otherwise specified)
Operating Temperature -40°C ≤ TA ≤ +125°C (extended)
All parameters apply across the specified operating ranges unless noted.
V+ = 10V to 36V (referenced to V-);
V+ = +5V to +18V & V- = -5.0V to -18V (referenced to DGND ≥ ±5V to ±18V),
VL = +2.7V to 5.5V, 5 k, 10 k, 50 k, 100 k devices.
Typical specifications represent values for VL = 5.5V, TA = +25°C.
DC Characteristics
Parameters
Sym.
Min.
Typ.
Max.
Units
Delay after
device exits the
reset state (VL >
VBOR)
TBORD
—
10
20
µs
Supply Current(7)
IDDD
—
45
300
µA
Serial Interface Active,
Write all 0’s to Volatile Wiper 0 (address 0h)
VL = 5.5V, CS = VIL, FSCK = 5 MHz,
V- = DGND
—
—
7
µA
Serial Interface Inactive,
VL = 5.5V, SCK = VIH, CS = VIH, Wiper = 0,
V- = DGND
IDDA
—
—
5
µA
Current V+ to V-, PxA = PxB = PxW,
DGND = V- +(V+/2)
RAB
4.0
5
6.0
k
-502 devices, V+/V- = 10V to 36V
8.0
10
12.0
k
-103 devices, V+/V- = 10V to 36V
Resistance
(± 20%)(8)
RAB Current
Resolution
Step Resistance
(see Appendix
B.4)
IAB
40.0
50
60.0
k
-503 devices, V+/V- = 10V to 36V
80.0
100
120.0
k
-104 devices, V+/V- = 10V to 36V
—
—
9.00
mA
—
—
4.50
mA
-502 devices 36V / RAB(MIN),
(9)
-103 devices V- = -18V, V+ = +18V
—
—
0.90
mA
-503 devices
—
—
0.45
mA
-104 devices
256
Taps
8-bit
No Missing Codes
128
Taps
7-bit
No Missing Codes
N
RS
Conditions
—
RAB/(255)
—

8-bit
Note 1
—
RAB/(127)
—

7-bit
Note 1
Note 1
This specification is by design.
Note 7
Supply current (IDDD and IDDA) is independent of current through the resistor network.
Note 8
Resistance (RAB) is defined as the resistance between Terminal A to Terminal B.
Note 9
Guaranteed by the RAB specification and Ohms Law.
 2013-2015 Microchip Technology Inc.
DS20005207B-page 5
MCP41HVX1
AC/DC CHARACTERISTICS (CONTINUED)
Standard Operating Conditions (unless otherwise specified)
Operating Temperature -40°C ≤ TA ≤ +125°C (extended)
All parameters apply across the specified operating ranges unless noted.
V+ = 10V to 36V (referenced to V-);
V+ = +5V to +18V & V- = -5.0V to -18V (referenced to DGND ≥ ±5V to ±18V),
VL = +2.7V to 5.5V, 5 k, 10 k, 50 k, 100 k devices.
Typical specifications represent values for VL = 5.5V, TA = +25°C.
DC Characteristics
Parameters
Wiper Resistance
(see Appendix B.5)
Sym.
Min.
Typ.
Max.
Units
RW
—
75
170

IW = 1 mA
V+ = +18V, V- = -18V,
code = 00h, PxA = floating,
PxB = V-.
—
145
200

IW = 1 mA
V+ = +5.0V, V- = -5.0V,
code = 00h, PxA = floating,
PxB = V-(2)
Nominal Resistance
Temperature Coefficient
(see Appendix B.23)
RAB/T
Ratiometeric Tempco
(see Appendix B.22)
Conditions
—
50
—
ppm/°C TA = -40°C to +85°C
—
100
—
ppm/°C TA = -40°C to +125°C
VWB/T
—
15
—
ppm/°C Code = Mid-scale (80h or 40h)
Resistor Terminal Input VA,VW,VB
Voltage Range
(Terminals A, B and W)
V-
—
V+
V
Current through
Terminals
(A, B, and Wiper)(1)
—
—
25
mA
—
—
12.5
mA
—
—
6.5
mA
—
—
6.5
mA
-503 devices IBW(W ≠ ZS) and IAW(W ≠ FS)
-104 devices IBW(W ≠ ZS) and IAW(W ≠ FS)
—
—
36
mA
IBW(W = ZS), or IAW(W = FS)
—
5
—
nA
A = W = B = V-
Leakage current into A,
W or B
IT, IW
ITL
Note 1, Note 11
-502 devices IBW(W ≠ ZS) and IAW(W ≠ FS)
-103 devices IBW(W ≠ ZS) and IAW(W ≠ FS)
Note 1
This specification is by design.
Note 2
This parameter is not tested, but specified by characterization.
Note 11 Resistor terminals A, W and B’s polarity with respect to each other is not restricted.
DS20005207B-page 6
 2013-2015 Microchip Technology Inc.
MCP41HVX1
AC/DC CHARACTERISTICS (CONTINUED)
Standard Operating Conditions (unless otherwise specified)
Operating Temperature -40°C ≤ TA ≤ +125°C (extended)
All parameters apply across the specified operating ranges unless noted.
V+ = 10V to 36V (referenced to V-);
V+ = +5V to +18V & V- = -5.0V to -18V (referenced to DGND ≥ ±5V to ±18V),
VL = +2.7V to 5.5V, 5 k, 10 k, 50 k, 100 k devices.
Typical specifications represent values for VL = 5.5V, TA = +25°C.
DC Characteristics
Parameters
Full-Scale Error
(Potentiometer)
(8-bit code = FFh,
7-bit code =
7Fh)(10,17)
(VA = V+, VB = V-)
(see Appendix B.10)
Sym.
Min.
Typ.
Max.
Units
Conditions
VWFSE
-10.5
—
—
LSb
-8.5
—
—
LSb
-13.5
—
—
LSb
VAB = 10V to 36V
-5.5
—
—
LSb
VAB = 20V to 36V
-4.5
—
—
LSb
5 k
VAB = 20V to 36V
8-bit
7-bit
VAB = 20V to 36V
-40°C  TA  +85°C(2)
VAB = 20V to 36V
-40°C  TA  +85°C(2)
-7.0
—
—
LSb
-4.5
—
—
LSb
VAB = 10V to 36V
-6.0
—
—
LSb
-2.65
—
—
LSb
-2.25
—
—
LSb
-3.5
—
—
LSb
-1.0
—
—
LSb
-0.9
—
—
LSb
-1.4
—
—
LSb
-1.25
—
—
LSb
-0.95
—
—
LSb
VAB = 20V to 36V
-1.2
—
—
LSb
VAB = 10V to 36V
-1.1
—
—
LSb
-0.7
—
—
LSb
-0.95
—
—
LSb
-0.7
—
—
LSb
-0.85
—
—
LSb
-0.9
—
—
LSb
10 k
8-bit
VAB = 20V to 36V
VAB = 10V to 36V
VAB = 20V to 36V
7-bit
VAB = 20V to 36V
-40°C  TA  +85°C(2)
VAB = 10V to 36V
50 k
VAB = 20V to 36V
8-bit
VAB = 20V to 36V
-40°C  TA  +85°C(2)
VAB = 10V to 36V
VAB = 10V to 36V
-40°C  TA  +85°C(2)
7-bit
100 k
Note 2
This parameter is not tested, but specified by characterization.
Note 10
Measured at VW with VA = V+ and VB = V-.
VAB = 10V to 36V
-40°C  TA  +85°C(2)
VAB = 20V to 36V
8-bit
7-bit
VAB = 10V to 36V
VAB = 10V to 36V
-40°C  TA  +85°C(2)
VAB = 20V to 36V
VAB = 10V to 36V
Note 17 Analog switch leakage affects this specification. Higher temperatures increase the switch leakage.
 2013-2015 Microchip Technology Inc.
DS20005207B-page 7
MCP41HVX1
AC/DC CHARACTERISTICS (CONTINUED)
Standard Operating Conditions (unless otherwise specified)
Operating Temperature -40°C ≤ TA ≤ +125°C (extended)
All parameters apply across the specified operating ranges unless noted.
V+ = 10V to 36V (referenced to V-);
V+ = +5V to +18V & V- = -5.0V to -18V (referenced to DGND ≥ ±5V to ±18V),
VL = +2.7V to 5.5V, 5 k, 10 k, 50 k, 100 k devices.
Typical specifications represent values for VL = 5.5V, TA = +25°C.
DC Characteristics
Parameters
Sym.
Min.
Zero-Scale Error
VWZSE
—
—
(Potentiometer)
(8-bit code = 00h,
7-bit code =
00h)(10,17)
(VA = V+, VB = V- )
(see Appendix B.11)
Typ.
Max.
Units
—
+8.5
LSb
—
+13.5
LSb
—
—
+4.5
LSb
—
—
+7.0
LSb
—
—
+4.0
LSb
—
—
+6.5
LSb
—
—
+6.0
LSb
—
—
+2.0
LSb
—
—
+3.25
LSb
—
—
+3.0
LSb
—
—
+0.9
LSb
—
—
+0.8
LSb
—
—
+1.3
LSb
—
—
+1.2
LSb
—
—
+0.5
LSb
—
—
+0.7
LSb
—
—
+0.5
LSb
—
—
+0.95
LSb
—
—
+0.7
LSb
—
—
+0.25
LSb
—
—
+0.4
LSb
Conditions
5 k
8-bit
7-bit
10 k
VAB = 20V to 36V
VAB = 10V to 36V
VAB = 20V to 36V
VAB = 10V to 36V
VAB = 20V to 36V
8-bit
VAB = 10V to 36V
VAB = 10V to 36V
-40°C  TA  +85°C(2)
VAB = 20V to 36V
7-bit
50 k
VAB = 10V to 36V
VAB = 10V to 36V
-40°C  TA  +85°C(2)
VAB = 20V to 36V
8-bit
VAB = 20V to 36V
-40°C  TA  +85°C(2)
VAB = 10V to 36V
VAB = 10V to 36V
-40°C  TA  +85°C(2)
7-bit
100 k
VAB = 20V to 36V
VAB = 10V to 36V
VAB = 20V to 36V
8-bit
7-bit
VAB = 10V to 36V
VAB = 10V to 36V
-40°C  TA  +85°C(2)
VAB = 20V to 36V
VAB = 10V to 36V
Note 2
This parameter is not tested, but specified by characterization.
Note 10
Measured at VW with VA = V+ and VB = V-.
Note 17
Analog switch leakage affects this specification. Higher temperatures increase the switch leakage.
DS20005207B-page 8
 2013-2015 Microchip Technology Inc.
MCP41HVX1
AC/DC CHARACTERISTICS (CONTINUED)
Standard Operating Conditions (unless otherwise specified)
Operating Temperature -40°C ≤ TA ≤ +125°C (extended)
All parameters apply across the specified operating ranges unless noted.
V+ = 10V to 36V (referenced to V-);
V+ = +5V to +18V & V- = -5.0V to -18V (referenced to DGND ≥ ±5V to ±18V),
VL = +2.7V to 5.5V, 5 k, 10 k, 50 k, 100 k devices.
Typical specifications represent values for VL = 5.5V, TA = +25°C.
DC Characteristics
Parameters
Sym.
Potentiometer
Integral
Nonlinearity(10, 17)
(see Appendix
B.12)
P-INL
Potentiometer
Differential
Nonlinearity(10, 17)
(see Appendix
B.13)
Note 2
P-DNL
Min.
Typ.
Max.
Units
Conditions
8-bit
VAB = 10V to 36V
VAB = 10V to 36V
8-bit
VAB = 10V to 36V
7-bit
VAB = 10V to 36V
8-bit
VAB = 10V to 36V
±0.5
+1
LSb
±0.25
+0.5
LSb
-1
±0.5
+1
LSb
-0.5
±0.25
+0.5
LSb
-1.1
±0.5
+1.1
LSb
-1
±0.5
+1
LSb
VAB = 20V to 36V(2)
-1
±0.5
+1
LSb
VAB = 10V to 36V,
-40°C  TA  +85°C(2)
-0.6
±0.25
+0.6
LSb
-1.85
±0.5
+1.85
LSb
-1.2
±0.5
+1.2
LSb
VAB = 20V to 36V(2)
-1
±0.5
+1
LSb
VAB = 10V to 36V,
-40°C  TA  +85°C(2)
-1
±0.5
+1
LSb
-0.5
±0.25
+0.5
LSb
-0.25
±0.125
+0.25
LSb
-0.375
±0.125
+0.375
LSb
-0.125
±0.1
+0.125
LSb
-0.25
±0.125
+0.25
LSb
-0.125
±0.1
+0.125
LSb
-0.25
±0.125
+0.25
LSb
-0.125
-0.15
+0.125
LSb
5 k
7-bit
-1
-0.5
10 k
50 k
100 k
5 k
10 k
50 k
100 k
7-bit
VAB = 10V to 36V
8-bit
VAB = 10V to 36V
7-bit
VAB = 10V to 36V
8-bit
VAB = 10V to 36V
7-bit
VAB = 10V to 36V
8-bit
VAB = 10V to 36V
7-bit
VAB = 10V to 36V
8-bit
VAB = 10V to 36V
7-bit
VAB = 10V to 36V
8-bit
VAB = 10V to 36V
7-bit
VAB = 10V to 36V
This parameter is not tested, but specified by characterization.
Note 10 Measured at VW with VA = V+ and VB = V-.
Note 17 Analog switch leakage affects this specification. Higher temperatures increase the switch leakage.
 2013-2015 Microchip Technology Inc.
DS20005207B-page 9
MCP41HVX1
AC/DC CHARACTERISTICS (CONTINUED)
Standard Operating Conditions (unless otherwise specified)
Operating Temperature -40°C ≤ TA ≤ +125°C (extended)
All parameters apply across the specified operating ranges unless noted.
V+ = 10V to 36V (referenced to V-);
V+ = +5V to +18V & V- = -5.0V to -18V (referenced to DGND ≥ ±5V to ±18V),
VL = +2.7V to 5.5V, 5 k, 10 k, 50 k, 100 k devices.
Typical specifications represent values for VL = 5.5V, TA = +25°C.
DC Characteristics
Parameters
Bandwidth -3 dB
(load = 30 pF)
(see Appendix B.24)
VW Settling Time
(VA = 10V, VB = 0V,
±1LSb error band,
CL = 50 pF)
(see Appendix B.17)
DS20005207B-page 10
Sym.
BW
tS
Min.
Typ.
Max.
Units
—
480
—
kHz
—
480
—
kHz
—
240
—
kHz
—
240
—
kHz
—
48
—
kHz
—
48
—
kHz
Conditions
5 k
10 k
50 k
100 k
8-bit
Code = 7Fh
7-bit
Code = 3Fh
8-bit
Code = 7Fh
7-bit
Code = 3Fh
8-bit
Code = 7Fh
7-bit
Code = 3Fh
8-bit
Code = 7Fh
7-bit
Code = 3Fh
—
24
—
kHz
—
24
—
kHz
—
1
—
µs
5 k
Code = 00h → FFh (7Fh);
FFh (7Fh) → 00h
—
1
—
µs
10 k
Code = 00h → FFh (7Fh);
FFh (7Fh) → 00h
—
2.5
—
µs
50 k
Code = 00h → FFh (7Fh);
FFh (7Fh) → 00h
—
5
—
µs
100 k
Code = 00h → FFh (7Fh);
FFh (7Fh) → 00h
 2013-2015 Microchip Technology Inc.
MCP41HVX1
AC/DC CHARACTERISTICS (CONTINUED)
Standard Operating Conditions (unless otherwise specified)
Operating Temperature -40°C ≤ TA ≤ +125°C (extended)
All parameters apply across the specified operating ranges unless noted.
V+ = 10V to 36V (referenced to V-);
V+ = +5V to +18V & V- = -5.0V to -18V (referenced to DGND ≥ ±5V to ±18V),
VL = +2.7V to 5.5V, 5 k, 10 k, 50 k, 100 k devices.
Typical specifications represent values for VL = 5.5V, TA = +25°C.
DC Characteristics
Parameters
Sym.
Min.
Typ.
Max.
Units
Rheostat Integral
R-INL
Nonlinearity(12,13,14,17)
(see Appendix B.5)
-1.75
—
+1.75
LSb
-2.5
—
+2.5
LSb
Note 2
Conditions
5 k
8-bit
IW = 6.0 mA, (V+ - V-) = 36V(2)
IW = 3.3 mA, (V+ - V-) = 20V(2)
-4.0
—
+4.0
LSb
-1.0
—
+1.0
LSb
IW = 1.7 mA, (V+ - V-) = 10V
-1.5
—
+1.5
LSb
IW = 3.3 mA, (V+ - V-) = 20V(2)
-2.0
—
+2.0
LSb
IW = 1.7 mA, (V+ - V-) = 10V
-1.0
—
+1.0
LSb
-1.75
—
+1.75
LSb
7-bit
10 k
8-bit
IW = 6.0 mA, (V+ - V-) = 36V(2)
IW = 3.0 mA, (V+ - V-) = 36V(2)
IW = 1.7 mA, (V+ - V-) = 20V(2)
-2.0
—
+2.0
LSb
-0.6
—
+0.6
LSb
IW = 830 µA, (V+ - V-) = 10V
-0.8
—
+0.8
LSb
IW = 1.7 mA, (V+ - V-) = 20V(2)
-1.0
—
+1.0
LSb
IW = 830 µA, (V+ - V-) = 10V
-1.0
—
+1.0
LSb
-1.0
—
+1.0
LSb
7-bit
50 k
8-bit
IW = 3.0 mA, (V+ - V-) = 36V(2)
IW = 600 µA, (V+ - V-) = 36V(2)
IW = 330 µA, (V+ - V-) = 20V(2)
-1.2
—
+1.2
LSb
-0.5
—
+0.5
LSb
IW = 170 µA, (V+ - V-) = 10V
-0.5
—
+0.5
LSb
IW = 330 µA, (V+ - V-) = 20V(2)
-0.6
—
+0.6
LSb
IW = 170 µA, (V+ - V-) = 10V
-1.0
—
+1.0
LSb
-1.0
—
+1.0
LSb
7-bit
100 k 8-bit
IW = 600 µA, (V+ - V-) = 36V(2)
IW = 300 µA, (V+ - V-) = 36V(2)
IW = 170 µA, (V+ - V-) = 20V(2)
-1.2
—
+1.2
LSb
-0.5
—
+0.5
LSb
IW = 83 µA, (V+ - V-) = 10V
-0.5
—
+0.5
LSb
IW = 170 µA, (V+ - V-) = 20V(2)
-0.6
—
+0.6
LSb
IW = 83 µA, (V+ - V-) = 10V
7-bit
IW = 300 µA, (V+ - V-) = 36V(2)
This parameter is not tested, but specified by characterization.
Note 12 Nonlinearity is affected by wiper resistance (RW), which changes significantly over voltage and temperature.
Note 13 Externally connected to a Rheostat configuration (RBW), and then tested.
Note 14 Wiper current (IW) condition determined by RAB(max) and Voltage Condition, the delta voltage between V+
and V- (voltages are 36V, 20V, and 10V).
Note 17 Analog switch leakage affects this specification. Higher temperatures increase the switch leakage.
 2013-2015 Microchip Technology Inc.
DS20005207B-page 11
MCP41HVX1
AC/DC CHARACTERISTICS (CONTINUED)
Standard Operating Conditions (unless otherwise specified)
Operating Temperature -40°C ≤ TA ≤ +125°C (extended)
All parameters apply across the specified operating ranges unless noted.
V+ = 10V to 36V (referenced to V-);
V+ = +5V to +18V & V- = -5.0V to -18V (referenced to DGND ≥ ±5V to ±18V),
VL = +2.7V to 5.5V, 5 k, 10 k, 50 k, 100 k devices.
Typical specifications represent values for VL = 5.5V, TA = +25°C.
DC Characteristics
Parameters
Rheostat
Differential
Nonlinearity
(12,13,14,17)
Sym.
Min.
Typ.
Max.
Units
Conditions
R-DNL
-0.5
—
+0.5
LSb
-0.5
—
+0.5
LSb
IW = 3.3 mA, (V+ - V-) = 20V(2)
-0.8
—
+0.8
LSb
IW = 1.7 mA, (V+ - V-) = 10V
-0.6
—
+0.6
LSb
IW = 1.7 mA, (V+ - V-) = 10V
-40°C  TA  +85°C(2)
-0.25
—
+0.25
LSb
-0.25
—
+0.25
LSb
IW = 3.3 mA, (V+ - V-) = 20V(2)
-0.3
—
+0.3
LSb
IW = 1.7 mA, (V+ - V-) = 10V
-0.5
—
+0.5
LSb
-0.5
—
+0.5
LSb
5 k
8-bit
IW = 6.0 mA, (V+ - V-) = 36V(2)
(see Appendix
B.5)
Note 2
7-bit
10 k 8-bit
IW = 6.0 mA, (V+ - V-) = 36V(2)
IW = 3.0 mA, (V+ - V-) = 36V(2)
IW = 1.7 mA, (V+ - V-) = 20V(2)
-0.5
—
+0.5
LSb
-0.25
—
+0.25
LSb
IW = 830 µA, (V+ - V-) = 10V
-0.25
—
+0.25
LSb
IW = 1.7 mA, (V+ - V-) = 20V(2)
-0.25
—
+0.25
LSb
IW = 830 µA, (V+ - V-) = 10V
-0.5
—
+0.5
LSb
-0.5
—
+0.5
LSb
7-bit
50 k 8-bit
IW = 3.0 mA, (V+ - V-) = 36V(2)
IW = 600 µA, (V+ - V-) = 36V(2)
IW = 330 µA, (V+ - V-) = 20V(2)
-0.5
—
+0.5
LSb
-0.25
—
+0.25
LSb
IW = 170 µA, (V+ - V-) = 10V
-0.25
—
+0.25
LSb
IW = 330 µA, (V+ - V-) = 20V(2)
-0.25
—
+0.25
LSb
IW = 170 µA, (V+ - V-) = 10V
-0.5
—
+0.5
LSb
-0.5
—
+0.5
LSb
7-bit
100 k 8-bit
IW = 600 µA, (V+ - V-) = 36V(2)
IW = 300 µA, (V+ - V-) = 36V(2)
IW = 170 µA, (V+ - V-) = 20V(2)
-0.5
—
+0.5
LSb
-0.25
—
+0.25
LSb
IW = 83 µA, (V+ - V-) = 10V
-0.25
—
+0.25
LSb
IW = 170 µA, (V+ - V-) = 20V(2)
-0.25
—
+0.25
LSb
IW = 83 µA, (V+ - V-) = 10V
7-bit
IW = 300 µA, (V+ - V-) = 36V(2)
This parameter is not tested, but specified by characterization.
Note 12 Nonlinearity is affected by wiper resistance (RW), which changes significantly over voltage and temperature.
Note 13 Externally connected to a Rheostat configuration (RBW), and then tested.
Note 14 Wiper current (IW) condition determined by RAB(max) and Voltage Condition, the delta voltage between V+
and V- (voltages are 36V, 20V, and 10V).
Note 17 Analog switch leakage affects this specification. Higher temperatures increase the switch leakage.
DS20005207B-page 12
 2013-2015 Microchip Technology Inc.
MCP41HVX1
AC/DC CHARACTERISTICS (CONTINUED)
Standard Operating Conditions (unless otherwise specified)
Operating Temperature -40°C ≤ TA ≤ +125°C (extended)
All parameters apply across the specified operating ranges unless noted.
V+ = 10V to 36V (referenced to V-);
V+ = +5V to +18V & V- = -5.0V to -18V (referenced to DGND ≥ ±5V to ±18V),
VL = +2.7V to 5.5V, 5 k, 10 k, 50 k, 100 k devices.
Typical specifications represent values for VL = 5.5V, TA = +25°C.
DC Characteristics
Parameters
Sym.
Min.
Typ.
Max.
Units
Capacitance (PA)
CA
—
75
—
pF
Measured to V-, f =1 MHz,
Wiper code = Mid-Scale
Capacitance (Pw)
CW
—
120
—
pF
Measured to V-, f =1 MHz,
Wiper code = Mid-Scale
Capacitance (PB)
CB
—
75
—
pF
Measured to V-, f =1 MHz,
Wiper code = Mid-Scale
Common-Mode
Leakage
ICM
—
5
—
nA
VA = VB = VW
CIN,
COUT
—
10
—
pF
fC = 400 kHz
Digital Interface Pin
Capacitance
Conditions
Digital Inputs/Outputs (CS, SDI, SDO, SCK, SHDN, WLAT)
Schmitt Trigger HighInput Threshold
VIH
0.45 VL
—
VL + 0.3V
V
2.7V  VL  5.5V
0.5 VL
—
VL + 0.3V
V
1.8V  VL  2.7V
Schmitt Trigger
Low-Input Threshold
VIL
DGND - 0.5V
—
0.2 VL
V
Hysteresis of Schmitt
Trigger Inputs
VHYS
—
0.1 VL
—
V
Output Low
Voltage (SDO)
VOL
Output High
Voltage (SDO)
VOH
Input Leakage
Current
IIL
 2013-2015 Microchip Technology Inc.
DGND
—
0.2 VL
V
VL = 5.5V, IOL = 5 mA
DGND
—
0.2 VL
V
VL = 1.8V, IOL = 800 µA
0.8 VL
—
VL
V
VL = 5.5V, IOH = -2.5 mA
0.8 VL
—
VL
V
VL = 1.8V, IOL = -800 µA
1
uA
VIN = VL and VIN = DGND
-1
DS20005207B-page 13
MCP41HVX1
AC/DC CHARACTERISTICS (CONTINUED)
Standard Operating Conditions (unless otherwise specified)
Operating Temperature -40°C ≤ TA ≤ +125°C (extended)
All parameters apply across the specified operating ranges unless noted.
V+ = 10V to 36V (referenced to V-);
V+ = +5V to +18V & V- = -5.0V to -18V (referenced to DGND ≥ ±5V to ±18V),
VL = +2.7V to 5.5V, 5 k, 10 k, 50 k, 100 k devices.
Typical specifications represent values for VL = 5.5V, TA = +25°C.
DC Characteristics
Parameters
Sym.
Min.
Typ.
Max
Units
Conditions
N
0h
—
FFh
hex
8-bit
—
7Fh
hex
7-bit
hex
8-bit
hex
7-bit
RAM (Wiper, TCON) Value
Wiper Value Range
0h
Wiper POR/BOR Value
7Fh
NPOR/BOR
3Fh
TCON Value Range
TCON POR/BOR Value
N
0h
—
FFh
FF
NTCON
hex
hex
All Terminals connected
Power Requirements
Power Supply
Sensitivity
(see Appendix B.20)
Power Dissipation
PSS
PDISS
—
0.0015
0.0035
%/%
8-bit
VL = 2.7V to 5.5V,
V+ = 18V, V- = -18V,
Code = 7Fh
—
0.0015
0.0035
%/%
7-bit
VL = 2.7V to 5.5V,
V+ = 18V, V- = -18V,
Code = 3Fh
VL = 5.5V, V+ = 18V, V- =
-18V(15)
—
260
—
mW
5 k
—
130
—
mW
10 k
—
26
—
mW
50 k
—
13
—
mW
100 k
Note 15 PDISS = I  V, or ((IDDD  5.5V) + (IDDA  36V) + (IAB  36V)).
DS20005207B-page 14
 2013-2015 Microchip Technology Inc.
MCP41HVX1
AC/DC Notes:
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
This specification is by design.
This parameter is not tested, but specified by characterization.
See Absolute Maximum Ratings.
V+ voltage is dependent on V- voltage. The maximum delta voltage between V+ and V- is 36V. The digital logic
DGND potential can be anywhere between V+ and V-. The VL potential must be ≥ DGND and ≤ V+.
The minimum value determined by maximum V- to V+ potential equals 36V, and the minimum value for operation
equals 1.8V. So, 36V - 1.8V = 34.2V.
POR/BOR is not rate dependent.
Supply current (IDDD and IDDA) is independent of current through the resistor network.
Resistance (RAB) is defined as the resistance between Terminal A to Terminal B.
Guaranteed by the RAB specification and Ohms Law.
Measured at VW with VA = V+ and VB = V-.
Resistor terminals A, W and B’s polarity with respect to each other is not restricted.
Nonlinearity is affected by wiper resistance (RW), which changes significantly over voltage and temperature.
Externally connected to a Rheostat configuration (RBW), and then tested.
Wiper current (IW) condition determined by RAB(max) and Voltage Condition, the delta voltage between V+ and V(voltages are 36V, 20V, and 10V).
PDISS = I  V, or ((IDDD  5.5V) + (IDDA  36V) + (IAB  36V)).
For specified analog performance, V+ must be 20V or greater (unless otherwise noted).
Analog switch leakage affects this specification. Higher temperatures increase the switch leakage.
During the power-up sequence, to ensure expected Analog POR operation, the two power systems (Analog and
Digital) should have a common reference to ensure that the driven DGND voltage is not at a higher potential than
the driven V+ voltage.
 2013-2015 Microchip Technology Inc.
DS20005207B-page 15
MCP41HVX1
1.1
SPI Mode Timing Waveforms and Requirements
± 1 LSb
W
Old Value
FIGURE 1-1:
TABLE 1-1:
New Value
Settling Time Waveforms.
WIPER SETTLING TIMING
Standard Operating Conditions (unless otherwise specified)
Operating Temperature
-40°C ≤ TA ≤ +125°C (extended)
All parameters apply across the specified operating ranges unless noted.
V+ = 10V to 36V (referenced to V-);
V+ = +5V to +18V & V- = -5.0V to -18V (referenced to DGND ≥ ±5V to ±18V),
VL = +2.7V to 5.5V, 5 k, 10 k, 50 k, 100 k devices.
Typical specifications represent values for VL = 5.5V, TA = +25°C.
Timing Characteristics
Parameters
VW Settling Time
(VA = 10V, VB = 0V,
±1LSb error band,
CL = 50 pF)
(see Appendix B.17)
DS20005207B-page 16
Sym.
Min.
Typ.
Max.
Units
Conditions
tS
—
1
—
µs
5 k
Code = 00h ≥ FFh (7Fh);
FFh (7Fh) ≥ 00h
—
1
—
µs
10 k
Code = 00h ≥ FFh (7Fh);
FFh (7Fh) ≥ 00h
—
2.5
—
µs
50 k
Code = 00h ≥ FFh (7Fh);
FFh (7Fh) ≥ 00h
—
5
—
µs
100 k Code = 00h ≥ FFh (7Fh);
FFh (7Fh) ≥ 00h
 2013-2015 Microchip Technology Inc.
MCP41HVX1
CS
84
“1”
WLAT
“1”
85
“0”
“0”
70b
70a
71
83b
SCK
83a
72
80
MSb
SDO
LSb
BIT6 - - - - - -1
77
SDI
MSb IN
73
FIGURE 1-2:
TABLE 1-2:
#
BIT6 - - - -1
LSb IN
74
SPI Timing Waveform (Mode = 11).
SPI REQUIREMENTS (MODE = 11)
Characteristic
SCK Input Frequency
Symbol
FSCK
Min.
Max. Units
—
10
MHz VL = 2.7V to 5.5V
—
1
MHz VL = 1.8V to 2.7V
70a CS Active (VIL) to SCK input
TcsA2scH
25
—
ns
70b WLAT Active (VIL) to eighth (or sixteenth) SCK
of the Serial Command to ensure previous data is
latched (set-up time)
TwlA2scH
20
—
ns
35
—
ns
71
SCK input high time
TscH
72
SCK input low time
TscL
73
Set-up time of SDI input to SCK edge
—
ns
VL = 1.8V to 2.7V
35
—
ns
VL = 2.7V to 5.5V
VL = 1.8V to 2.7V
120
—
ns
TDIV2scH
10
—
ns
74
Hold time of SDI input from SCK edge
TscH2DIL
20
—
ns
CS Inactive (VIH) to SDO output high-impedance
TcsH2DOZ
—
50
ns
80
SDO data output valid after SCK edge
TscL2DOV
—
83a CS Inactive (VIH) after SCK edge
84
Hold time of CS (or WLAT) Inactive (VIH) to
CS (or WLAT) Active (VIL)
85
WLAT input low time
Note 1:
VL = 2.7V to 5.5V
120
77
83b WLAT Inactive (VIH) after eighth (or sixteenth)
SCK edge (hold time)
Conditions
Note 1
55
ns
VL = 2.7V to 5.5V
90
ns
VL = 1.8V to 2.7V
TscH2csI
100
—
ns
TscH2wlatI
50
—
ns
TcsA2csI
20
—
ns
TWLATL
25
—
ns
This specification is by design.
 2013-2015 Microchip Technology Inc.
DS20005207B-page 17
MCP41HVX1
82
CS
84
“1”
“1”
WLAT
“0”
“0”
70b
83a
83b
70a
SCK
71
MSb
SDO
BIT6 - - - - - -1
LSb
75, 76
73
SDI
80
72
MSb IN
77
BIT6 - - - -1
LSb IN
74
FIGURE 1-3:
TABLE 1-3:
SPI Timing Waveform (Mode = 00).
SPI REQUIREMENTS (MODE = 00)
#
Characteristic
SCK Input Frequency
Symbol
FSCK
Min.
Max. Units
—
10
MHz VL = 2.7V to 5.5V
—
1
MHz VL = 1.8V to 2.7V
ns
70a CS Active (VIL) to SCK input
TcsA2scH
25
—
70b WLAT Active (VIL) to eighth (or sixteenth) SCK
of the Serial Command to ensure previous data
is latched (setup time)
TwlA2scH
20
—
ns
35
—
ns
71
SCK input high time
TscH
72
SCK input low time
TscL
73
Set-up time of SDI input to SCK edge
—
ns
VL = 1.8V to 2.7V
35
—
ns
VL = 2.7V to 5.5V
VL = 1.8V to 2.7V
120
—
ns
TDIV2scH
10
—
ns
74
Hold time of SDI input from SCK edge
TscH2DIL
20
—
ns
CS Inactive (VIH) to SDO output high-impedance
TcsH2DOZ
—
50
ns
80
SDO data output valid after SCK edge
TscL2DOV
—
82
SDO data output valid after CS Active (VIL)
83b WLAT Inactive (VIH) after SCK edge
84
Hold time of CS (or WLAT) Inactive (VIH) to
CS (or WLAT) Active (VIL)
85
WLAT input low time
Note 1:
VL = 2.7V to 5.5V
120
77
83a CS Inactive (VIH) after SCK edge
Conditions
Note 1
55
ns
VL = 2.7V to 5.5V
90
ns
VL = 1.8V to 2.7V
TscL2DOV
—
70
ns
TscL2csI
100
—
ns
TscL2wlatI
50
—
ns
TcsA2csI
20
—
ns
TWLATL
25
—
ns
This specification is by design.
DS20005207B-page 18
 2013-2015 Microchip Technology Inc.
MCP41HVX1
TEMPERATURE CHARACTERISTICS
Electrical Specifications: Unless otherwise indicated, VDD = +2.7V to +5.5V, VSS = GND.
Parameters
Sym.
Min.
Typ.
Max.
Units
Conditions
Temperature Ranges
Specified Temperature Range
TA
-40
—
+125
°C
Operating Temperature Range
TA
-40
—
+125
°C
Storage Temperature Range
TA
-65
—
+150
°C
Thermal Resistance, 14L-TSSOP (ST)
JA
—
100
—
°C/W
Thermal Resistance, 20L-VQFN (MQ)
JA
—
38.3
—
°C/W
Thermal Package Resistances
 2013-2015 Microchip Technology Inc.
DS20005207B-page 19
MCP41HVX1
2.0
Note:
TYPICAL PERFORMANCE CURVES
The device Performance Curves are available in a separate document. This is done to keep the file size of
this PDF document less than the 10 MB file attachment limit of many mail servers.
The MCP41HVX1 Performance Curves document is literature number DS20005209, and can be found on
the Microchip website. Look at the MCP41HVX1 Product Page under Documentation and Software, in the
Data Sheets category.
DS20005207B-page 20
 2013-2015 Microchip Technology Inc.
MCP41HVX1
3.0
PIN DESCRIPTIONS
The descriptions of the pins are listed in Table 3-1.
Additional descriptions of the device pins follows.
TABLE 3-1:
PINOUT DESCRIPTION FOR THE MCP41HVX1
Pin
TSSOP
VQFN
14L
20L
1
2
3
4
5
6
Buffer
Type
Function
Symbol
Type
1
VL
P
—
Positive Digital Power Supply Input
2
SCK
I
ST
SPI Serial Clock pin
3
CS
I
ST
Chip Select
4
SDI
I
ST
SPI Serial Data In pin
5
SDO
O
—
SPI Serial Data Out
6
WLAT
I
ST
Wiper Latch Enable
0 = Received SPI Shift Register Buffer (SPIBUF) value is
transferred to Wiper register
1 = Received SPI data value is held in SPI Shift Register
Buffer (SPIBUF)
7
7
SHDN
I
ST
Shutdown
8
11
DGND
P
—
Ground
9
8, 9, 10, 17,
18, 19, 20
NC
—
—
Pin not internally connected to die. To reduce noise
coupling, connect pin either to DGND or VL.
10
12
V-
P
—
Analog Negative Potential Supply
11
13
P0B
I/O
A
Potentiometer 0 Terminal B
12
14
P0W
I/O
A
Potentiometer 0 Wiper Terminal
13
15
P0A
I/O
A
Potentiometer 0 Terminal A
14
16
V+
P
—
Analog Positive Potential Supply
—
21
EP
P
—
Exposed Pad, connect to V- signal or Not Connected
(floating)(1)
Legend: A = Analog, ST = Schmitt Trigger, I = Input, O = Output, I/O = Input/Output, P = Power
Note 1:
The VQFN package has a contact on the bottom of the package. This contact is conductively connected to the
die substrate, and therefore should be unconnected or connected to the same ground as the device’s V- pin.
 2013-2015 Microchip Technology Inc.
DS20005207B-page 21
MCP41HVX1
3.1
Positive Power Supply Input (VL)
The VL pin is the device’s positive power supply input.
The input power supply is relative to DGND and can
range from 1.8V to 5.5V. A decoupling capacitor on VL
(to DGND) is recommended to achieve maximum
performance.
While the device’s VL < Vmin (2.7V), the electrical
performance of the device may not meet the data sheet
specifications.
3.2
Serial Clock (SCK)
The SCK pin is the serial interface's Serial Clock pin.
This pin is connected to the host controllers’ SCK pin.
The MCP41HVX1 is an SPI slave device, so its SCK
pin is an input-only pin.
3.3
Chip Select (CS)
The CS pin is the serial interface’s chip select input.
Forcing the CS pin to VIL enables the serial commands.
3.4
Serial Data In (SDI)
The SDI pin is the serial interface’s Serial Data In pin.
This pin is connected to the host controller’s SDO pin.
3.5
Serial Data Out (SDO)
The SDO pin is the serial interface’s Serial Data Out
pin. This pin is connected to the host controller’s SDI
pin. This pin allows the host controller to read the digital
potentiometer registers (Wiper and TCON), or monitor
the state of the command error bit.
3.6
Wiper Latch (WLAT)
The WLAT pin is used to delay the transfer of the
received wiper value (in the shift register) to the wiper
register. This allows this transfer to be synchronized to
an external event (such as zero crossing). See
Section 4.3.2 “Wiper Latch”.
3.7
Shutdown (SHDN)
The SHDN pin is used to force the resistor network
terminals into the hardware shutdown state. See
Section 4.3.1 “Shutdown”.
3.8
Digital Ground (DGND)
3.11
Potentiometer Terminal B
The Terminal B pin is connected to the internal
potentiometer’s terminal B.
The potentiometer’s terminal B is the fixed connection
to the zero-scale wiper value of the digital
potentiometer. This corresponds to a wiper value of
0x00 for both 7-bit and 8-bit devices.
The Terminal B pin does not have a polarity relative to
the Terminal W or A pins. The Terminal B pin can
support both positive and negative current. The voltage
on Terminal B must be between V+ and V-.
3.12
Potentiometer Wiper (W) Terminal
The Terminal W pin is connected to the internal
potentiometer’s Terminal W (the Wiper). The wiper
terminal is the adjustable terminal of the digital
potentiometer. The Terminal W pin does not have a
polarity relative to terminal’s A or B pins. The Terminal
W pin can support both positive and negative current.
The voltage on Terminal W must be between V+ and V-.
If the V+ voltage powers-up before the VL voltage, the
wiper is forced to mid-scale once the Analog POR
voltage is crossed.
If the V+ voltage powers-up after the VL voltage is
greater than the Digital POR voltage, the wiper is
forced to the value in the wiper register once the
Analog POR voltage is crossed.
3.13
Potentiometer Terminal A
The Terminal A pin is connected to the internal
potentiometer’s Terminal A.
The potentiometer’s Terminal A is the fixed connection
to the full-scale wiper value of the digital potentiometer.
This corresponds to a wiper value of 0xFF for 8-bit
devices or 0x7F for 7-bit devices.
The Terminal A pin does not have a polarity relative to
the Terminal W or B pins. The Terminal A pin can
support both positive and negative current. The voltage
on Terminal A must be between V+ and V-.
3.14
Analog Positive Voltage (V+)
The analog circuitry’s positive supply voltage. The V+
pin must have a higher potential then the V- pin.
The DGND pin is the device’s digital ground reference.
3.15
3.9
This pad is only on the bottom of the VQFN packages.
This pad is conductively connected to the device
substrate. The EP pin must be connected to the Vsignal or left floating. This pad could be connected to a
Printed Circuit Board (PCB) heat sink to assist as a
heat sink for the device.
Not Connected (NC)
This pin is not internally connected to the die. To reduce
noise coupling, these pins should be connected to
either VL or DGND.
3.10
Analog Negative Voltage (V-)
Exposed Pad (EP)
Analog circuitry negative supply voltage. Must not
have a higher potential then the DGND pin.
DS20005207B-page 22
 2013-2015 Microchip Technology Inc.
MCP41HVX1
4.0
FUNCTIONAL OVERVIEW
This data sheet covers a family of two volatile digital
potentiometer devices that will be referred to as
MCP41HVX1.
As the Device Block Diagram shows, there are six
main functional blocks. These are:
•
•
•
•
•
•
Operating Voltage Range
POR/BOR Operation
Memory Map
Control Module
Resistor Network
Serial Interface (SPI)
4.1
Operating Voltage Range
The MCP41HVX1 devices have four voltage signals.
These are:
•
•
•
•
V+
VL
DGND
V-
- Analog power
- Digital power
- Digital ground
- Analog ground
Figure 4-1 shows the two possible power-up
sequences: analog power rails power-up first, or digital
power rails power-up first. The device has been
designed so that either power rail may power-up first.
The device has a POR circuit for both digital power
circuitry and analog power circuitry.
The POR/BOR operation and the Memory Map are
discussed in this section, and the Resistor Network and
SPI operation are described in their own sections. The
Device Commands are discussed in Section 7.0
“Device Commands”.
If the V+ voltage powers-up before the VL voltage, the
wiper is forced to mid-scale once the analog POR
voltage is crossed.
If the V+ voltage powers-up after the VL voltage is
greater than the digital POR voltage, the wiper is forced
to the value in the wiper register once the analog POR
voltage is crossed.
Figure 4-2 shows the three cases of the digital power
signals (VL/DGND) with respect to the analog power
signals (V+/V-). The device implements level shifts
between the digital and analog power systems, which
allows the digital interface voltage to be anywhere in
the V+/V- voltage window.
Analog Voltage Powers-Up First
Referenced to V-
Referenced to DGND
FIGURE 4-1:
V+
Digital Voltage Powers-Up First
Referenced to V-
V+
VL
VL
DGND
V-
DGND
V-
V+
Referenced to DGND
V+
VL
VL
DGND
DGND
V-
V-
Power-On Sequences.
 2013-2015 Microchip Technology Inc.
DS20005207B-page 23
MCP41HVX1
V+
Case 1
HighVoltage
Range
HighVoltage
Range
V+
Case 2
Anywhere
between
V+ and V(VL  DGND)
VL
DGND
V+ and VL
Case 3
DGND
HighVoltage
Range
VL
FIGURE 4-2:
V- and DGND
Voltage Ranges.
DS20005207B-page 24
V-
V-
 2013-2015 Microchip Technology Inc.
MCP41HVX1
4.2
POR/BOR Operation
4.2.1.1
Digital Circuitry
The resistor network’s devices are powered by the analog
power signals (V+/V-), but the digital logic (including the
wiper registers) is powered by the digital power signals
(VL/DGND). So, both the digital circuitry and analog
circuitry have independent POR/BOR circuits.
A Digital Power-on Reset (DPOR) occurs when the
device’s VL signal has power applied (referenced from
DGND) and the voltage rises above the trip point. A
Brown-out Reset (BOR) occurs when a device has power
applied to it, and the voltage drops below the trip point.
The wiper position will be forced to the default state
when the V+ voltage (relative to V-) is above the analog
POR/BOR trip point. The wiper register will be in the
default state when the VL voltage (relative to DGND) is
above the digital POR/BOR trip point.
The device’s RAM retention voltage (VRAM) is lower
than the POR/BOR voltage trip point (VPOR/VBOR). The
maximum VPOR/VBOR voltage is less then 1.8V.
4.2.1
POWER-ON RESET
Each power system has its own independent Power-on
Reset circuitry. This is done so that regardless of the
power-up sequencing of the analog and digital power
rails, the wiper output will be forced to a default value
after minimum conditions are met for either power supply.
Table 4-1 shows the interaction between the analog
and digital PORs for the V+ and VL voltages on the
wiper pin state.
TABLE 4-1:
WIPER PIN STATE BASED
ON POR CONDITIONS
V+ Voltage
VL Voltage
VL < VDPOR
V+ <
VAPOR
Note 1:
Comments
Unknown Mid-Scale
Unknown
VL ≥ VDPOR
V+ ≥
VAPOR
Wiper
Register
Value(1)
Wiper Register
can be updated
The default POR state of the wiper
register value is the mid-scale value.
 2013-2015 Microchip Technology Inc.
• The volatile wiper registers are loaded with the
POR/BOR value
• The TCON registers are loaded with the default
values
• The device is capable of digital operation
Table 4-2 shows the default POR/BOR wiper register
setting selection.
the
electrical
When
VPOR/VBOR < VDD < 2.7V,
performance may not meet the data sheet specifications.
In this region, the device is capable of incrementing,
decrementing, reading and writing to its volatile memory
if the proper serial command is executed.
TABLE 4-2:
DEFAULT POR/BOR WIPER
REGISTER SETTING
(DIGITAL)
Default
POR Wiper
Device
Wiper
Register Resolution Code
Setting
Typical
RAB
Value
Package
Code
The digital-signal-to-analog-signal voltage level shifters
require a minimum voltage between the VL and Vsignals. This voltage requirement is below the
operating supply voltage specifications. The wiper
output may fluctuate while the VL voltage is less than
the level shifter operating voltage, since the analog
values may not reflect the digital value. Output issues
may be reduced by powering-up the digital supply
voltages to their operating voltage before powering the
analog supply voltage.
When the device powers-up, the device VL will cross
the VPOR/VBOR voltage. Once the VL voltage crosses
the VPOR/VBOR voltage, the following happens:
5.0 k
-502
Mid-Scale
10.0 k
-103
Mid-Scale
50.0 k
-503
Mid-Scale
100.0 k -104
Mid-Scale
Note 1:
8-bit
7Fh
7-bit
3Fh
8-bit
7Fh
7-bit
3Fh
8-bit
7Fh
7-bit
3Fh
8-bit
7Fh
7-bit
3Fh
Register setting independent of analog
power voltage.
DS20005207B-page 25
MCP41HVX1
Analog Circuitry
TABLE 4-3:
An Analog Power-on Reset (APOR) occurs when the
device’s V+ pin voltage has power applied (referenced
from V-) and the V+ pin voltage rises above the trip point.
Once the VL pin voltage exceeds the digital POR trip point
voltage, the wiper register will control the wiper setting.
Table 4-3 shows the default POR/BOR Wiper Setting
for when the VL pin is not powered (< digital POR trip
point).
DEFAULT POR/BOR WIPER
SETTING (ANALOG)
Typical
RAB Value
Package
Code
4.2.1.2
Default
POR Wiper
Setting
5.0 k
-502
Mid-Scale
10.0 k
-103
Mid-Scale
50.0 k
-503
Mid-Scale
100.0 k
-104
Mid-Scale
Note 1:
Device
Resolution
8-bit
7-bit
8-bit
7-bit
8-bit
7-bit
8-bit
7-bit
Wiper setting is dependent on the wiper
register value if the VL voltage is greater
than the digital POR voltage.
Referenced to DGND
V+
VL
VPOR/VBOR
DGND
V-
Digital logic has been
reset (POR). This
includes the wiper register.
Note:
Brown-out
condition,
Wiper value Analog Power
unknown
is recovering (still low) and VL
rail/pin no longer sources current
Analog Power
to V+
is Low
Digital logic has been
reset (POR). This
includes the wiper register.
Brown-out condition,
Wiper value unknown
Digital logic has been
reset (POR). This
includes the wiper register.
When VL is above V+ (floating, the VL pin ESD clamping diode will cause the V+ level to be pulled up.
FIGURE 4-3:
DS20005207B-page 26
DGND, VL, V+, and V- Signal Waveform Examples.
 2013-2015 Microchip Technology Inc.
MCP41HVX1
4.2.2
BROWN-OUT RESET
Each power system has its own independent
Brown-out Reset circuitry. This is done so that regardless of the power-down sequencing of the analog and
digital power rails, the wiper output will be forced to a
default value after the low-voltage conditions are met
for either power supply.
Table 4-4 shows the interaction between the analog
and digital BORs for the V+ and VL voltages on the
wiper pin state.
TABLE 4-4:
WIPER PIN STATE BASED ON
BOR CONDITIONS
V+ Voltage
VL Voltage
VL < VDBOR
V+ <
VABOR
Note 1:
4.2.2.1
Comments
When 1.8V ≤ VL, the device is capable of digital
operation.
Table 4-5 shows the digital potentiometer’s level of
functionality across the entire VL range, while Figure 4-4
illustrates the Power-up and Brown-out functionality.
4.2.2.2
Analog Circuitry
An Analog Brown-out Reset (ABOR) occurs when the
device’s V+ pin has power applied (referenced from V-)
and the V+ pin voltage drops below the trip point. In this
case, the resistor network terminal pins can become an
unknown state.
Unknown Mid-Scale
Unknown
VL ≥ VDBOR
V+ ≥
VABOR
Whenever VL transitions from VL < VDBOR to VL >
VDBOR (a POR event), the wiper’s POR/BOR value is
latched into the wiper register and the volatile TCON
register is forced to the POR/BOR state.
Wiper
register
value (1)
Wiper register
can be updated
The default POR state of the wiper
register value is the mid-scale value.
Digital Circuitry
When the device’s digital power supply powers-down,
the device’s VL pin voltage will cross the digital
VDPOR/VDBOR voltage.
Once the VL voltage decreases below
VDPOR/VDBOR voltage, the following happens:
the
• Serial Interface is disabled
If the VL voltage decreases below the VRAM voltage,
the following happens:
• Volatile wiper registers may become corrupted
• TCON registers may become corrupted
Section 4.2.1 “Power-on Reset” describes what
occurs as the voltage recovers above the
VDPOR/VDBOR voltage.
Serial commands not completed due to a brown-out
condition may cause the memory location to become
corrupted.
The brown-out circuit establishes a minimum VDBOR
threshold for operation (VDBOR < 1.8V). The digital
BOR voltage (VDBOR) is higher than the RAM retention
voltage (VRAM) so that as the device voltage crosses
the digital BOR threshold, the value that is loaded into
the volatile wiper register is not corrupted due to RAM
retention issues.
When VL < VDBOR, all communications are ignored
and the potentiometer terminals are forced to the
analog BOR state.
 2013-2015 Microchip Technology Inc.
DS20005207B-page 27
MCP41HVX1
TABLE 4-5:
DEVICE FUNCTIONALITY AT EACH VL REGION
VL Level
VL < VDBOR < 1.8V
VDBOR ≤ VL < 1.8V
1.8V ≤ VL ≤ 5.5V
Note 1:
2:
V+ / V- Level
Serial
Interface
Potentiometer
Terminals(2)
Wiper
Register
Setting
Output(2)
Valid Range
Ignored
“unknown”
Unknown
Invalid
Invalid Range
Ignored
“unknown”
Unknown
Invalid
Valid Range
“Unknown”
connected
Invalid Range “Unknown”
connected
Volatile wiper
Register
initialized
Invalid
Valid Range
Accepted
connected
Invalid Range
Accepted
connected
Volatile wiper
Register
determines
Wiper Setting
Valid
Comment
The volatile registers
are forced to the
POR/BOR state when
VL transitions above
the VDPOR trip point
Valid
Invalid
For system voltages below the minimum operating voltage, it is recommended to use a voltage supervisor
to hold the system in reset. This ensures that MCP41HVX1 commands are not attempted out of the operating range of the device.
Assumes that V+ > VAPOR.
Normal Operation Range
VL
Outside Specified
AC/DC Range
Normal Operation Range
1.8V
VPOR/BOR
VRAM
DGND
Device’s Serial
Interface is
“Not Specified
FIGURE 4-4:
DS20005207B-page 28
Device’s Serial
VBOR Delay
Interface is
“Not Operational” Wiper Forced to Default POR/BOR setting
Power-up and Brown-out - V+/V- at Normal Operating Voltage.
 2013-2015 Microchip Technology Inc.
MCP41HVX1
4.3
Control Module
4.3.1.2
The control module controls the following functionalities:
• Shutdown
• Wiper Latch
4.3.1
SHUTDOWN
The MCP41HVX1 has two methods to disconnect the
terminal’s pins (P0A, P0W, and P0B) from the resistor
network. These are:
• Hardware Shutdown pin (SHDN)
• Terminal Control Register (TCON)
4.3.1.1
Hardware Shutdown Pin Operation
The SHDN pin has the same functionality as
Microchip’s family of standard-voltage devices. When
the SHDN pin is low, the P0A terminal will disconnect
(become open) while the P0W terminal simultaneously
connects to the P0B terminal (see Figure 4-5).
Note:
When the SHDN pin is Active (VIL), the
state of the TCON register bits is
overridden (ignored). When the state of
the SHDN pin returns to the Inactive state
(VIH), the TCON register bits return to
controlling the terminal connection state.
This ensures the value in the TCON
register is not corrupted
The Hardware Shutdown pin mode does not corrupt
the volatile wiper register. When Shutdown is exited,
the device returns to the wiper setting specified by the
volatile wiper value. See Section 5.7 for additional
description details.
Note:
When the SHDN pin is active, the Serial
Interface is not disabled and serial
interface activity is executed.
Terminal Control Register
The Terminal Control (TCON) register allows the
device’s terminal pins to be independently removed
from the application circuit. These terminal control
settings do not modify the wiper setting values. This
has no effect on the serial interface, and the
memory/wipers are still under full user control.
The resistor network has four TCON bits associated
with it: one bit for each terminal (A, W, and B) and one
to have a software configuration that matches the
configuration of the SHDN pin. These bits are named
R0A, R0W, R0B and R0HW. Register 4-1 describes the
operation of the R0HW, R0A, R0B, and R0W bits.
Note:
When the R0HW bit forces the resistor
network into the hardware SHDN state,
the state of the TCON register R0A, R0W,
and R0B bits is overridden (ignored).
When the state of the R0HW bit no longer
forces the resistor network into the
hardware SHDN state, the TCON register
R0A, R0W, and R0B bits return to
controlling the terminal connection state.
That is, the R0HW bit does not corrupt the
state of the R0A, R0W and R0B bits.
Figure 4-6 shows how the SHDN pin signal and the
R0HW bit signal interact to control the hardware
shutdown of each resistor network (independently).
SHDN (from pin)
R0HW
(from TCON register)
FIGURE 4-6:
Interaction.
To Pot 0 Hardware
Shutdown Control
R0HW Bit and SHDN Pin
Resistor Network
A
W
B
FIGURE 4-5:
Hardware Shutdown
Resistor Network Configuration.
 2013-2015 Microchip Technology Inc.
DS20005207B-page 29
MCP41HVX1
4.3.2
WIPER LATCH
The wiper latch pin is used to control when the new
wiper value in the wiper register is transferred to the
wiper. This is useful for applications that need to
synchronize the wiper updates. This may be for
synchronization to an external event, such as zero
crossing, or to synchronize the update of multiple
digital potentiometers.
Note 1: This feature only inhibits the data transfer
from the wiper register to the wiper.
2: When the WLAT pin becomes active, data
transferred to the wiper will not be corrupted due to the wiper register buffer getting loaded from an active SPI command.
4.3.3
DEVICE CURRENT MODES
When the WLAT pin is high, transfers from the wiper
register to the wiper are inhibited. When the WLAT pin
is low, transfers may occur from the Wiper register to
the wiper. Figure 4-7 shows the interaction of the WLAT
pin and the loading of the wiper.
There are two current modes for Volatile devices.
These are:
If the external event crossing time is long, then the
wiper could be updated the entire time that the WLAT
signal is low. Once the WLAT signal goes high, the
transfer from the wiper register is disabled. The wiper
register can continue to be updated. Only the CS pin is
used to enable/disable serial commands.
For the SPI interface, Static Operation occurs when
the CS pin is at the VIH voltage and the SCK pin is
static (high or low).
• Serial Interface Inactive (Static Operation)
• Serial Interface Active
If the application does not require synchronized wiper
register updates, then the WLAT pin should be tied low.
VIH
CS
VIL
VIH
WLAT
VIL
16 SCK
SCK
VIL
16 SCK
16 SCK
16 SCK
Wiper
Register
Loaded
Wiper
Register
Transferred
to Wiper
When WLAT goes low during an SPI active transfer, When WLAT goes high during an SPI active transfer,
the previously loaded Wiper Register value is
the wiper register value will be updated with
transferred to the wiper. (1)
the new value from this serial command when the
command completes. The wiper will retain the
value that was last transferred from the wiper
register before the WLAT pin went high.
Note 1: The wiper register may be updated on 16 SCK cycles for a Write command, or on 8 SCK cycles with
and Increment or Decrement command.
2: The WLAT pin should not be brought high during the falling edge of the 8th clock cycle of an Increment
or Decrement command or the 16th clock cycle of a Write command.
FIGURE 4-7:
DS20005207B-page 30
WLAT Interaction with Wiper During Serial Communication – (SPI Mode 1,1).
 2013-2015 Microchip Technology Inc.
MCP41HVX1
4.4
Memory Map
TABLE 4-6:
The device memory supports 16 locations that are
eight bits wide (16 x 8 bits). This memory space
contains only volatile locations (see Table 4-7).
4.4.1
VOLATILE MEMORY (RAM)
WIPER POR STANDARD
SETTINGS
Resistance
Typical
Code
RAB Value
Wiper
Default
Code
POR Wiper
Setting
8-bit 7-bit
There are two volatile memory locations. These are:
-502
5.0 k
Mid-Scale
7Fh
3Fh
• Volatile Wiper 0
• Terminal Control (TCON0) Register 0
-103
10.0 k
Mid-Scale
7Fh
3Fh
-503
50.0 k
Mid-Scale
7Fh
3Fh
The volatile memory starts functioning at the RAM
retention voltage (VRAM). The POR/BOR wiper code is
shown in Table 4-6.
-104
100.0 k
Mid-Scale
7Fh
3Fh
Table 4-7 shows this memory map and which serial
commands operate (and don’t) on each of these
locations.
Accessing an “invalid” address (for that device) or an
invalid command for that address will cause an error
condition (CMDERR) on the serial interface.
TABLE 4-7:
4.4.1.1
Write to Invalid (Reserved)
Addresses
Any write to a reserved address will be ignored and will
generate an error condition. To exit the error condition,
the user must take the CS pin to the VIH level and then
back to the active state (VIL).
MEMORY MAP AND THE SUPPORTED COMMANDS
Address
Function
Allowed Commands
Disallowed Commands (1)
Memory Type
00h
Volatile Wiper 0
Read, Write,
Increment, Decrement
—
RAM
01h - 03h
Reserved
none
Read, Write,
Increment, Decrement
—
04h
Volatile
TCON Register
Read, Write
Increment, Decrement
RAM
05h - 0Fh
Reserved
none
Read, Write,
Increment, Decrement
—
Note 1: This command on this address will generate an error condition. To exit the error condition, the user must
take the CS pin to the VIH level and then back to the active state (VIL).
 2013-2015 Microchip Technology Inc.
DS20005207B-page 31
MCP41HVX1
4.4.1.2
Terminal Control (TCON) Registers
The value that is written to this register will appear on
the resistor network terminals when the serial
command has completed.
The Terminal Control (TCON) register contains four
control bits for Wiper 0. Register 4-1 describes each bit
of the TCON register.
On a POR/BOR, these registers are loaded with FFh
for all terminals connected. The host controller needs
to detect the POR/BOR event and then update the
volatile TCON register values.
The state of each resistor network terminal connection
is individually controlled. That is, each terminal
connection (A, B and W) can be individually connected/disconnected from the resistor network. This
allows the system to minimize the currents through the
digital potentiometer.
TCON0 BITS(1)
REGISTER 4-1:
R-1
R-1
R-1
R-1
R/W-1
R/W-1
R/W-1
R/W-1
D7
D6
D5
D4
R0HW
R0A
R0W
R0B
bit 7
bit 0
Legend:
R = Readable bit
W = Writable bit
U = Unimplemented bit, read as ‘0’
-n = Value at POR
‘1’ = Bit is set
‘0’ = Bit is cleared
x = Bit is unknown
bit 7-4
D7-D4: Reserved. Forced to “1”
bit 3
R0HW: Resistor 0 Hardware Configuration Control bit
This bit forces Resistor 0 into the “shutdown” configuration of the Hardware pin
1 = Resistor 0 is not forced to the hardware pin “shutdown” configuration
0 = Resistor 0 is forced to the hardware pin “shutdown” configuration
bit 2
R0A: Resistor 0 Terminal A (P0A pin) Connect Control bit
This bit connects/disconnects the Resistor 0 Terminal A to the Resistor 0 Network
1 = P0A pin is connected to the Resistor 0 Network
0 = P0A pin is disconnected from the Resistor 0 Network
bit 1
R0W: Resistor 0 Wiper (P0W pin) Connect Control bit
This bit connects/disconnects the Resistor 0 Wiper to the Resistor 0 Network
1 = P0W pin is connected to the Resistor 0 Network
0 = P0W pin is disconnected from the Resistor 0 Network
bit 0
R0B: Resistor 0 Terminal B (P0B pin) Connect Control bit
This bit connects/disconnects the Resistor 0 Terminal B to the Resistor 0 Network
1 = P0B pin is connected to the Resistor 0 Network
0 = P0B pin is disconnected from the Resistor 0 Network
Note 1:
2:
These bits do not affect the wiper register values.
The hardware SHDN pin (when active) overrides the state of these bits. When the SHDN pin returns to the
inactive state, the TCON register will control the state of the terminals. The SHDN pin does not modify the
state of the TCON bits.
DS20005207B-page 32
 2013-2015 Microchip Technology Inc.
MCP41HVX1
5.0
RESISTOR NETWORK
5.1
The resistor network has either 7-bit or 8-bit resolution.
Each resistor network allows zero-scale to full-scale
connections. Figure 5-1 shows a block diagram for the
resistive network of a device. The resistor network has up
to three external connections. These are referred to as
Terminal A, Terminal B, and the wiper (or Terminal W).
The resistor network is made up of several parts. These
include:
• Resistor Ladder Module
• Wiper
• Shutdown Control (Terminal Connections)
Terminals A and B as well as the wiper W do not have
a polarity. These terminals can support both positive
and negative current.
RW
RS
RW
RS
RW
R
RAB S
8-Bit
N=
255
(1) (FFh)
7-Bit
N=
127
(7Fh)
254
(1) (FEh)
126
(7Eh)
253
(FDh)
125
(7Dh)
(1)
W
RW (1)
RS
RZS
RW
(1)
1
(01h)
1
(01h)
0
(00h)
0
(00h)
Analog MUX
B
Note 1: The wiper resistance is dependent on
several factors, including wiper code,
device V+ voltage, terminal voltages (on
A, B and W) and temperature.
Also, for the same conditions, each tap
selection resistance has a small variation.
This RW variation has a greater effect on
some specifications (such as INL) for the
smaller resistance devices (5.0 k)
compared to larger resistance devices
(100.0 k).
FIGURE 5-1:
Resistor Block Diagram.
The RAB resistor ladder is composed of the series of
equal value Step resistors (RS) and the Full-Scale
(RFS) and Zero-Scale (RZS) resistances:
RAB = RZS + n × RS + RFS
Where “n” is determined by the resolution of the device.
The RFS and RZS resistances are discussed in
Section 5.1.3 “RFS and RZS Resistors”.
There is a connection point (tap) between each RS
resistor. Each tap point is a connection point for an
analog switch. The opposite side of the analog switch
is connected to a common signal which is connected to
the Terminal W (Wiper) pin (see Section 5.2 “Wiper”).
Figure 5-1 shows a block diagram of the Resistor
Network. The RAB (and RS) resistance has small
variations over voltage and temperature.
The end points of the resistor ladder are connected to
analog switches, which are connected to the device
Terminal A and Terminal B pins. In the ideal case, these
switches would have 0 of resistance, that is
RFS = RZS = 0. This will also be referred as the
Simplified model.
A
RFS
Resistor Ladder Module
For an 8-bit device, there are 255 resistors in a string
between Terminal A and Terminal B. The wiper can be
set to tap onto any of these 255 resistors, thus providing 256 possible settings (including Terminal A and
Terminal B). A wiper setting of 00h connects Terminal
W (wiper) to Terminal B (Zero-Scale). A wiper setting of
7Fh is the Mid-Scale setting. A wiper setting of FFh
connects Terminal W (wiper) to Terminal A (Full-Scale).
Table 5-2 illustrates the full wiper setting map.
For a 7-bit device, there are 127 resistors in a string
between Terminal A and Terminal B. The wiper can be
set to tap onto any of these 127 resistors, thus providing 128 possible settings (including Terminal A and
Terminal B). A wiper setting of 00h connects Terminal
W (wiper) to Terminal B (Zero-Scale). A wiper setting of
3Fh is the Mid-scale setting. A wiper setting of 7Fh connects the wiper to Terminal A (Full-Scale). Table 5-2
illustrates the full wiper setting map.
5.1.1
RAB CURRENT (IRAB)
The current through the RAB resistor (A pin to B pin) is
dependent on the voltage on the VA and VB pins and
the RAB resistance, as shown in Equation 5-1.
EQUATION 5-1:
RAB
 V A – VB 
RAB = RZS +  n  R S  + R FS = --------------------------- I RAB 
Where:
VA = the voltage on the VA pin
VB = the voltage on the VB pin
IRAB = the current into the VREF pin
 2013-2015 Microchip Technology Inc.
DS20005207B-page 33
MCP41HVX1
5.1.2
STEP RESISTANCE (RS)
Step resistance (RS) is the resistance from one tap setting to the next. This value will be dependent on the
RAB value that has been selected (and the full-scale
and zero-scale resistances). The RS resistors are
manufactured so that they should be very consistent
with each other and track each other’s values as
voltage and/or temperature change.
Equation 5-2 shows the simplified and detailed equations for calculating the RS value. The simplified equation assumes RFS = RZS = 0. Table 5-1 shows
example step resistance calculations for each device,
and the variation of the detailed model (RFS  0;
RZS  0) from the simplified model (RFS = RZS = 0).
As the RAB resistance option increases, the effects of
the RZS and RFS resistances decrease.
The total resistance of the device has minimal variation
due to operating voltage (see device characterization
graphs).
Equation 5-2
resistance.
shows
calculations
for
the
step
Simplified Model (assumes RFS = RZS = 0)
R AB =  n  RS 
8-bit
R AB
RS = ----------255
RAB
RS = ----------n
7-bit
R AB
R S = ----------127
Detailed Model
R AB = R FS +  n  R S  + R ZS
RAB – R FS – R ZS
R S = --------------------------------------------n
or
 VFS – VZS 
-------------------------------n
RS = -------------------------------I AB
Where:
“n” = 255 (8-bit) or 127 (7-bit)
VFS = Wiper voltage at Full-Scale code
VZS = Wiper voltage at Zero-Scale code
IAB = Current between Terminal A and Terminal B
EQUATION 5-2:
TABLE 5-1:
RS CALCULATION
EXAMPLE STEP RESISTANCES (RS) CALCULATIONS
Example Resistance ()
RAB
5,000
10,000
50,000
100,000
Note 1:
2:
3:
RZS(3)
RFS(3)
0
Variation%(1) Resolution
RS
Equation
Value
0
5,000/127
39.37
0
80
60
4,860/127
38.27
-2.80
0
0
5,000/255
19.61
0
80
60
4,860/255
19.06
-2.80
0
0
10,000/127
78.74
0
80
60
9,860/127
77.64
-1.40
0
0
10,000/255
39.22
0
80
60
9,860/255
38.67
-1.40
0
0
50,000/127
393.70
0
80
60
49,860/127
392.60
-0.28
0
0
50,000/255
196.08
0
80
60
49,860/255
195.53
-0.28
0
0
100,000/127
787.40
0
80
60
99,860/127
786.30
-0.14
0
0
100,000/255
392.16
0
80
60
99,860/255
391.61
-0.14
Comment
7-bit (127 RS) Simplified Model(2)
8-bit (255 RS) Simplified Model(2)
7-bit (127 RS) Simplified Model(2)
8-bit (255 RS) Simplified Model (2)
7-bit (127 RS) Simplified Model(2)
8-bit (255 RS) Simplified Model(2)
7-bit (127 RS) Simplified Model(2)
8-bit (255 RS) Simplified Model(2)
Delta % from Simplified Model RS calculation value:
Assumes RFS = RZS = 0.
Zero-Scale (RZS) and Full-Scale (RFS) resistances are dependent on many operational characteristics of
the device, including the V+ / V- voltage, the voltages on the A, B and W terminals, the wiper code
selected, the RAB resistance and the temperature of the device.
DS20005207B-page 34
 2013-2015 Microchip Technology Inc.
MCP41HVX1
5.1.3
RFS AND RZS RESISTORS
The RFS and RZS resistances are artifacts of the RAB
resistor network implementation. In the ideal model, the
RFS and RZS resistances would be 0. These resistors
are included in the block diagram to help better model
the actual device operation. Equation 5-3 shows how to
estimate the RS, RFS, and RZS resistances based on
the measured voltages of VREF, VFS, VZS and the
measured current IVREF.
EQUATION 5-3:
ESTIMATING RS, RFS
AND RZS
 VA – VFS 
R FS = ------------------------------- I RAB 
 VZS – VB 
RZS = ------------------------------ I RAB 
Where:
VS
R S = ----------------- I RAB 
 V FS – V ZS 
VS = -------------------------------- (8-bit device)
255
 V FS – V ZS 
VS = -------------------------------- (7-bit device)
127
VFS = VW voltage when the wiper code is at
full-scale
VZS = VW voltage when the wiper code is at
zero-scale
5.2
Wiper
The wiper terminal is connected to an analog switch
MUX, where one side of all the analog switches are
connected together via the W terminal. The other side
of each analog switch is connected to one of the taps
of the RAB resistor string (see Figure 5-1).
The value in the volatile wiper register selects which
analog switch to close, connecting the W terminal to
the selected node of the resistor ladder. The wiper
register is eight bits wide, and Table 5-2 shows the
wiper value state for both 7-bit and 8-bit devices.
The wiper resistance (RW) is the resistance of the
selected analog switch in the analog MUX. This
resistance is dependent on many operational
characteristics of the device, including the V+/V- voltage, the voltages on the A, B and W terminals, the
wiper code selected, the RAB resistance and the
temperature of the device.
When the wiper value is at zero-scale (00h), the wiper
is connected closest to the B terminal. When the wiper
value is at full-scale (FFh for 8-bit, 7Fh for 7-bit), the
wiper is connected closest to the A terminal.
A zero-scale wiper value connects the W terminal
(wiper) to the B terminal (wiper = 00h). A full-scale
wiper value connects the W terminal (wiper) to the A
terminal (wiper = FFh (8-bit), or wiper = 7Fh (7-bit)). In
these configurations, the only resistance between
Terminal W and the other terminal (A or B) is that of the
analog switches.
TABLE 5-2:
VOLATILE WIPER VALUE VS.
WIPER POSITION
Wiper Setting
 2013-2015 Microchip Technology Inc.
7-bit
8-bit
7Fh
FFh
7Eh 40h
FEh 80h
3Fh
7Fh
3Eh 01h
7Eh 01h
00h
00h
Properties
Full-Scale (W = A),
Increment commands ignored
W=N
W = N (Mid-Scale)
W=N
Zero-Scale (W = B)
Decrement command ignored
DS20005207B-page 35
MCP41HVX1
5.2.1
WIPER RESISTANCE (RW)
5.2.2
Wiper resistance is significantly dependent on:
In a potentiometer configuration, the wiper resistance
variation does not affect the output voltage seen on the
W pin, and therefore is not a significant source of error.
• The resistor network’s supply voltage (VRN)
• The resistor network’s terminal (A, B, and W)
voltages
• Switch leakage (occurs at higher temperatures)
• IW current
5.2.3
Figure 5-2 shows the wiper resistance characterization
data for all four RAB resistances and temperatures. Each
RAB resistance determined the maximum wiper current
based on worst-case conditions RAB = RAB maximum
and at full-scale code, VBW ~= V+ (but not exceeding
V+). The V+ targets were 10V, 20V, and 36V. What this
graph shows is that at higher RAB resistances (50 k
and 100 k) and at the highest temperature (+125°C),
the analog switch leakage causes an increase in the
measured result of RW, where RW is measured in a
rheostat configuration with RW = (VBW - VBA)/IBW.
2400
Ͳ40C5kIW=1.7mA
Ͳ40C5kIW=3.3mA
Ͳ40C5kIW=6.0mA
Ͳ40C10kIW=830uA
Ͳ40C10kIW=1.7mA
Ͳ40C10kIW=3.0mA
Ͳ40C50kIW=170uA
Ͳ40C50kIW=330uA
Ͳ40C50kIW=600uA
Ͳ40C100kIW=83uA
Ͳ40C100kIW=170uA
Ͳ40C100kIW=300uA
2200
2000
Wiper Re
esistance RW (:)
1800
1600
1400
+25C5kIW=1.7mA
+25C5kIW=3.3mA
+25C5kIW=6.0mA
+25C10kIW=830uA
+25C10kIW=1.7mA
+25C10kIW=3.0mA
+25C50kIW=170uA
+25C50kIW=330uA
+25C50kIW=600uA
+25C100kIW=83uA
+25C100kIW=170uA
+25C100kIW=300uA
+85C5kIW=1.7mA
+85C5kIW=3.3mA
+85C5kIW=6.0mA
+85C10kIW=830uA
+85C10kIW=1.7mA
+85C10kIW=3.0mA
+85C50kIW=170uA
+85C50kIW=330uA
+85C50kIW=600uA
+85C100kIW=83uA
+85C100kIW=170uA
+85C100kIW=300uA
POTENTIOMETER
CONFIGURATION
+125C5kIW=1.7mA
+125C5kIW=3.3mA
+125C5kIW=6.0mA
+125C10kIW=830uA
+125C10kIW=1.7mA
+125C10kIW=3.0mA
+125C50kIW=170uA
+125C50kIW=330uA
+125C50kIW=600uA
+125C100kIW=83uA
+125C100kIW=170uA
+125C100kIW=300uA
RHEOSTAT CONFIGURATION
In a rheostat configuration, the wiper resistance variation creates nonlinearity in the RBW (or RAW) value. The
lower the nominal resistance (RAB), the greater the
possible relative error. Also, a change in voltage needs
to be taken into account. For the 5.0 k device, the
maximum wiper resistance at 5.5V is approximately 6%
of the total resistance, while at 2.7V it is approximately
6.5% of the total resistance.
5.2.4
LEVEL SHIFTERS
(DIGITAL-TO-ANALOG)
Since the digital logic may operate anywhere within the
analog power range, level shifters are present so that the
digital signals control the analog circuitry. This level shifter
logic is relative to the V- and VL voltages. A delta voltage
of 2.7V between VL and V- is required for the serial
interface to operate at the maximum specified frequency.
1200
IW =83uA,+125C(100k:)
1000
800
IW =170uA,+125C(100k:)
IW =170uA,+125C(50k:)
600
Increasedwiperresistance(RW)occurs
duetoincreasedanalog switchleakage at
highertemperatures(suchas+125C)and
larger RAB resistances.
largerR
resistances
IW =300uA,+125C(100k:)
400
200
0
0
32
64
96
128
160
DAC Wiper Code
192
224
256
FIGURE 5-2:
RW Resistance Vs. RAB,
Wiper Current (IW), Temperature and Wiper Code.
Since there is minimal variation of the total device
resistance (RAB) over voltage, at a constant temperature (see device characterization graphs), the change
in wiper resistance over voltage can have a significant
impact on the RINL and RDNL errors.
DS20005207B-page 36
 2013-2015 Microchip Technology Inc.
MCP41HVX1
Terminal Currents
values without violating the maximum terminal current
specification. Table 5-3 shows resistance and current
calculations based on the RAB resistance (RS resistance) for a system that supports ± 18V ( 36V). In
Rheostat configuration, the minimum wiper-code value
is shown (for VBW = 36V). As the VBW voltage
decreases, the minimum wiper-code value also
decreases. Using a wiper code less then this value will
cause the maximum terminal current (IT) specification
to be violated.
The terminal currents are limited by several factors,
including the RAB resistance (RS resistance). The
maximum current occurs when the wiper is at either the
zero-scale (IBW) or full-scale (IAW) code. In this case,
the current is only going through the analog switches
(see IT specification in Section 1.0 “Electrical
Characteristics”). When the current passes through
at least one RS resistive element, then the maximum
terminal current (IT) has a different limit. The current
through the RAB resistor is limited by the RAB
resistance. The worst case (max current) occurs when
the resistance is at the minimum RAB value.
Note:
Higher current capabilities allow a greater delta voltage
between the desired terminals for a given resistance.
This also allows a more usable range of wiper code
RBW ()
(= 36V/IT(MAX)) (2)
Typical
IT (A, B, or W (IW)) (mA)
(IBW(W = ZS), IAW(W = FS) (1)
RAB Resistance ()
9.00
25.0
1,440
62.992
4.50
12.5
2,880
91
45
0.392
0.787
314.961
0.90
6.5
5539
35
17
1.020
2.047
0.45
6.5
5539
17
8
2.039
4.094
RS(MIN) ()
Min.
Max.
8-bit
7-bit
4,000
6,000
15.686
31.496
10,000
8,000
12,000
31.373
50,000
40,000
60,000
156.863
100,000
80,000
120,000
313.725
629.9
5,000
Rheostat
VBW(MAX) When
Wiper = 01h (V)
(= IT(MAX) * RS(MIN))
TERMINAL (WIPER) CURRENT AND WIPER SETTINGS (RW = RFS = RZS = 0)
IAB(MAX) (mA)
(= 36V/RAB(MIN)) (1)
TABLE 5-3:
For high terminal-current applications, it is
recommended that proper PCB layout
techniques be used to address the
thermal implications of this high current.
The VQFN package has better thermal
properties than the TSSOP package.
Rheostat
Min ‘N’
when VBW = 36V
N * RS(MIN) * 36V
 IT (mA) (3)
5.3
8-bit
7-bit
8-bit
7-bit
91
45
0.392
0.787
Note 1: IBW or IAW currents can be much higher than this depending on the voltage differential between Terminal
B and Terminal W or Terminal A and Terminal W.
2: Any RBW resistance greater than this limits the current.
3: If VBW = 36V, then the wiper code value must be greater than or equal to Min ‘N’. Wiper codes less than
Min ‘N’ will cause the wiper current (IW) to exceed the specification. Wiper codes greater than Min ‘N’ will
cause the wiper current to be less than the maximum. The Min ‘N’ number has been rounded up from the
calculated number to ensure that the wiper current does not exceed the maximum specification.
 2013-2015 Microchip Technology Inc.
DS20005207B-page 37
MCP41HVX1
Figures 5-3 through 5-6 show graphs of the calculated
currents (minimum, typical, and maximum) for each
resistor option. These graphs are based on 25 mA
(5 k), 12.5 mA (10 k), and 6.5 mA (50 k and
100 k) specifications.
RAB = 5k:
30.0E-3
RAB(TYP)
4.0E-3
3.0E-3
2.0E-3
2 0E 3
RAB(MAX)
1.0E-3
000.0E+0
0
32
64
FIGURE 5-5:
Code – 50 k.
96
128 160
Wiper Code
192
224
256
Maximum IBW Vs. Wiper
RAB = 100k:
7.0E-3
6.0E-3
RAB(MIN)
5.0E-3
5 0E 3
RAB(TYP)
4.0E-3
3.0E-3
2.0E-3
2 0E 3
25.0E-3
1.0E-3
RAB(TYP)
IBW(MAX) (A)
RAB(MIN)
5.0E-3
5 0E 3
IBW(MAX)
AX) (A)
Looking at the 50 k device, the maximum terminal
current is 6.5 mA. That means that any wiper code
value greater than 36 ensures that the terminal current
is less than 6.5 mA. This is ~14% of the full-scale value.
If the application could change to the 100 k device,
which has the same maximum terminal current specification, any wiper-code value greater than 18 ensures
that the terminal current is less than 6.5 mA. This is
~7% of the full-scale value. Supporting higher terminal
current allows a greater wiper code range for a given
VBW voltage.
6.0E-3
IBW(MAX)
AX) (A)
To ensure no damage to the resistor network (including
long-term reliability) the maximum terminal current
must not be exceeded. This means that the application
must assume that the RAB resistance is the minimum
RAB value (RAB(MIN), see blue lines in graphs).
RAB = 50k:
7.0E-3
20.0E-3
RAB(MAX)
000.0E+0
RAB(MIN)
0
15.0E-3
32
64
96
128 160
Wiper Code
RAB(MAX)
10.0E-3
FIGURE 5-6:
Code – 100 k.
5.0E-3
000.0E+0
0
32
64
96
128
160
192
224
256
Wiper Code
FIGURE 5-3:
Code – 5 k.
Maximum IBW Vs. Wiper
RAB = 10k:
192
224
256
Maximum IBW Vs. Wiper
Figure 5-7 shows a graph of the maximum VBW voltage
versus wiper code (for 5 k and 10 k devices). To
ensure that no damage is done to the resistor network,
the RAB(MIN) resistance (blue line) should be used to
determine VBW voltages for the circuit. Devices where
the RAB resistance is greater than the RAB(MIN) resistance will naturally support a higher voltage limit.
14.0E-3
40.0
12.0E-3
RAB(TYP)
35.0
RAB(MIN)
8.0E-3
30.0
RAB(MAX)
6.0E-3
VBW(MAX) (V)
IBW(MAX) (A)
10.0E-3
4.0E-3
2.0E-3
RAB(MAX)
RAB(TYP)
25.0
20.0
RAB(MIN)
15.0
10.0
000.0E+0
0
32
64
96
128
160
192
224
256
Wiper Code
FIGURE 5-4:
Code – 10 k.
DS20005207B-page 38
Maximum IBW Vs. Wiper
5.0
0.0
0
32
64
96
128
160
192
224
256
Wiper Code
FIGURE 5-7:
Maximum VBW Vs. Wiper
Code (5 k and 10 k devices).
 2013-2015 Microchip Technology Inc.
MCP41HVX1
Table 5-4 shows the maximum VBW voltage that can be
applied across the Terminal B to Terminal W pins for a
given wiper-code value (for the 5 k and 10 k
devices). These calculations assume the ideal model
(RW = RFS = RZS = 0) and show the calculations
based on RS(MIN) and RS(MAX). Table 5-5 shows the
same calculations for the 50 k devices, and Table 5-6
shows the calculations for the 100 k devices. These
tables are supplied as a quick reference.
TABLE 5-4:
Code
Hex.
MAX VBW AT EACH WIPER CODE (RW = RFS = RZS = 0) FOR V+ – V- = 36V,
5 K AND 10 K DEVICES
VBW(MAX)
Dec.
RS(MIN)
RS(MAX)
00h
0
0.000
0.000
01h
1
0.392
0.588
02h
2
0.784
03h
3
1.176
04h
4
05h
Code
Hex.
VBW(MAX)
Code
Hex.
VBW(MAX)
Dec.
RS(MIN)
RS(MAX)
Dec.
RS(MIN)
20h
32
12.549
18.824
21h
33
12.941
19.412
40h
64
25.098
41h
65
25.490
1.176
22h
34
13.333
1.765
23h
35
13.725
20.000
42h
66
25.882
20.588
43h
67
1.569
2.353
24h
36
25.275
14.118
21.176
44h
68
5
1.961
2.941
25h
37
26.667
14.510
21.765
45h
69
27.059
06h
6
2.353
3.529
26h
38
14.902
22.353
46h
70
27.451
07h
7
2.745
4.118
27h
39
15.294
22.941
47h
71
27.843
08h
8
3.137
4.706
28h
40
15.686
23.529
48h
72
28.235
09h
9
3.529
5.294
29h
41
16.078
24.118
49h
73
28.627
0Ah
10
3.922
5.882
2Ah
42
16.471
24.706
4Ah
74
29.020
0Bh
11
4.314
6.471
2Bh
43
16.863
25.294
4Bh
75
29.412
0Ch
12
4.706
7.059
2Ch
44
17.255
25.882
4Ch
76
29.804
0Dh
13
5.098
7.647
2Dh
45
17.647
26.471
4Dh
77
30.196
0Eh
14
5.490
8.235
2Eh
46
18.039
27.059
4Eh
78
30.588
0Fh
15
5.882
8.824
2Fh
47
18.431
27.647
4Fh
79
30.980
10h
16
5.275
9.412
30h
48
18.824
28.235
50h
80
31.373
11h
17
6.667
10.000
31h
49
19.216
28.824
51h
81
31.765
12h
18
7.059
10.588
32h
50
19.608
29.412
52h
82
32.157
13h
19
7.451
11.176
33h
51
20.000
30.000
53h
83
32.549
14h
20
7.843
11.765
34h
52
20.392
30.588
54h
84
32.941
15h
21
8.235
12.353
35h
53
20.784
31.176
55h
85
33.333
16h
22
8.627
12.941
36h
54
21.176
31.765
56h
86
33.725
17h
23
9.020
13.529
37h
55
21.569
32.353
57h
87
34.118
18h
24
9.412
14.118
38h
56
21.961
32.941
58h
88
34.510
19h
25
9.804
14.706
39h
57
22.353
33.529
59h
89
34.902
1Ah
26
10.196
15.294
3Ah
58
22.745
34.118
5Ah
90
35.294
1Bh
27
10.588
15.882
3Bh
59
23.137
34.706
5Bh
91
35.686
1Ch
28
10.980
16.471
3Ch
60
23.529
35.294
5Ch
92 - 255
36.0 (1, 2)
1Dh
29
11.373
17.059
3Dh
61
23.922
35.882
1Eh
30
11.765
17.647
3Eh
62
24.314
36.0 (1, 2)
1Fh
31
12.157
18.235
3Fh
63
24.706
Note 1:
Calculated RBW voltage is greater than 36V (highlighted in color), must be limited to 36V (V+ - V-).
2:
RS(MAX)
This wiper code and greater will limit the IBW current to less than the maximum supported terminal current (IT).
 2013-2015 Microchip Technology Inc.
DS20005207B-page 39
MCP41HVX1
TABLE 5-5:
MAX VBW AT EACH WIPER CODE (RW = RFS = RZS = 0) FOR V+ - V- = 36V,
50 K DEVICES
Code
VBW(MAX)
Code
VBW(MAX)
Code
Hex.
Dec.
RS(MIN)
RS(MAX)
Hex.
Dec.
RS(MIN)
RS(MAX)
00h
0
0.000
0.000
10h
16
16.314
24.471
20h
32
32.627
01h
1
1.020
1.529
11h
17
17.333
26.000
21h
33
33.647
02h
2
2.039
3.059
12h
18
18.353
27.529
22h
34
34.667
03h
3
3.059
4.588
13h
19
19.373
29.059
23h
35
04h
4
4.078
6.118
14h
20
20.392
30.588
05h
5
5.098
7.647
15h
21
21.412
32.118
06h
6
6.118
9.176
16h
22
22.431
33.647
07h
7
7.137
10.706
17h
23
23.451
35.176
08h
8
8.157
12.235
18h
24
24.471
36.0(1, 2)
09h
9
9.176
13.765
19h
25
25.490
0Ah
10
10.196
15.294
1Ah
26
26.510
0Bh
11
11.216
16.824
1Bh
27
27.529
0Ch
12
12.235
18.353
1Ch
28
28.549
0Dh
13
13.255
19.882
1Dh
29
29.569
0Eh
14
14.275
21.412
1Eh
30
30.588
15
15.294
22.941
1Fh
31
31.608
0Fh
Note 1:
2:
Hex.
VBW(MAX)
Dec.
24h - FFh 36 - 255
RS(MIN)
35.686
36.0(1, 2)
Calculated RBW voltage is greater than 36V (highlighted in color), must be limited to 36V (V+ - V-).
This wiper code and greater will limit the IBW current to less than the maximum supported terminal current (IT).
TABLE 5-6:
MAX VBW AT EACH WIPER CODE (RW = RFS = RZS = 0) FOR V+ - V- = 36V,
100 K DEVICES
Code
VBW(MAX)
Code
VBW(MAX)
Hex.
Dec.
RS(MIN)
RS(MAX)
Hex.
Dec.
RS(MIN)
00h
0
0.000
0.000
10h
16
32.627
01h
1
2.039
3.059
11h
17
34.667
02h
2
4.078
6.118
12h - FFh
18 - 255
36.0(1, 2)
03h
3
6.118
9.176
04h
4
8.157
12.235
05h
5
10.196
15.294
06h
6
12.235
18.353
07h
7
14.275
21.412
08h
8
16.314
24.471
09h
9
18.353
27.529
0Ah
10
20.392
30.588
0Bh
11
22.431
33.647
0Ch
12
24.471
36.0(1, 2)
0Dh
13
26.510
0Eh
14
28.549
15
30.588
0Fh
Note 1:
2:
RS(MAX)
RS(MAX)
Calculated RBW voltage is greater than 36V (highlighted in color), must be limited to 36V (V+ - V-).
This wiper code and greater will limit the IBW current to less than the maximum supported terminal current (IT).
DS20005207B-page 40
 2013-2015 Microchip Technology Inc.
MCP41HVX1
Variable Resistor (Rheostat)
5.5
A variable resistor is created using Terminal W and either
Terminal A or Terminal B. Since the wiper-code value of
0 connects the wiper to Terminal B, the RBW resistance
increases with increasing wiper-code value. Conversely,
the RAW resistance will decrease with increasing
wiper-code value. Figure 5-8 shows the connections
from a potentiometer to create a rheostat configuration.
A
RAW
RAW or
W
B
RBW
RBW
Resistor
FIGURE 5-8:
Rheostat Configuration.
Equation 5-4 shows the RBW and RAW calculations.
The RBW calculation is for the resistance between the
wiper and Terminal B. The RAW calculation is for the
resistance between the wiper and Terminal A.
EQUATION 5-4:
RBW AND RAW
CALCULATION
Simplified Model (assumes RFS = RZS = 0)
Analog Circuitry Power
Requirements
This device has two power supplies. One is for the
digital interface (VL and DGND) and the other is for the
high-voltage analog circuitry (V+ and V-). The
maximum delta voltage between V+ and V- is 36V. The
digital power signals must be between V+ and V-.
If the digital ground (DGND) pin is at half the potential
of V+ (relative to V-), then the terminal pins’ potentials
can be ±(V+/2) relative to DGND.
Figure 5-9 shows the relationship of the four power signals. This shows that the V+/V- signals do not need to
be symmetric around the DGND signal.
To ensure that the wiper register has been properly
loaded with the POR/BOR value, the VL voltage must
be at the minimum specified operating voltage (referenced to DGND).
Voltages Relative to DGND
5.4
V+
VL
DGND
V—
This can be anywhere
between V- and V+.
R BW =  n  RS 
R AW =   FSV – n   R S 
Where:
R AB
RS = ---------------------------Resolution
8-bit
R AB
R S = ----------255
 V+ – V- Voltage
+36V max.
+10V min.
7-bit
RAB
RS = ----------127
n = Wiper code
FSV = Full-scale value
(255 for 8-bit or 127 for 7-bit)
Detailed Model
FIGURE 5-9:
Ranges.
5.6
5.6.1
Analog Circuitry Voltage
Resistor Characteristics
V+/V- LOW-VOLTAGE OPERATION
The resistor network is specified from 20V to 36V. At
voltages below 20V, the resistor network will function,
but the operational characteristics may be outside the
specified limits. Please refer to Section 2.0 “Typical
Performance Curves” for additional information.
R BW = R ZS +  n  RS 
5.6.2
R AW = R FS +   FSV – n   R S 
Biasing the ends (Terminal A and Terminal B) near
mid-supply ((V+ - |V-|)/2) will give the worst switch
resistance temperature coefficient.
Where
RESISTOR TEMPCO
n = Wiper code
FSV = The full-scale value
(255 for 8-bit or 127 for 7-bit)
 2013-2015 Microchip Technology Inc.
DS20005207B-page 41
MCP41HVX1
5.7
Shutdown Control
Note:
Shutdown is used to minimize the device’s current
consumption. The MCP41HVX1 has two methods to
achieve this:
• Hardware Shutdown Pin (SHDN)
• Terminal Control Register (TCON)
The Hardware Shutdown pin is backwards compatible
with the MCP42X1 devices.
5.7.1
HARDWARE SHUTDOWN PIN
(SHDN)
The SHDN pin is available on the potentiometer
devices. When the SHDN pin is forced active (VIL):
• The P0A terminal is disconnected
• The P0W terminal is connected to the P0B terminal (see Figure 4-5)
• The Serial Interface is NOT disabled, and all
Serial Interface activity is executed
The R0HW bit does NOT corrupt the values in the
Volatile Wiper registers nor the TCON register. When
the Shutdown mode is exited (R0HW bit = 1):
• The device returns to the wiper setting specified
by the volatile wiper value
• The TCON register bits return to controlling the
terminal connection state
The Hardware Shutdown pin mode does not corrupt
the values in the Volatile Wiper Registers nor the
TCON register. When the Shutdown mode is exited
(SHDN pin is inactive (VIH)):
Resistor Network
A
• The device returns to the wiper setting specified
by the volatile wiper value
• The TCON register bits return to controlling the
terminal connection state
Resistor Network
W
B
FIGURE 5-11:
Resistor Network Shutdown
State (R0HW = 0).
A
W
B
FIGURE 5-10:
Hardware Shutdown
Resistor Network Configuration.
5.7.2
When the R0HW bit forces the resistor
network into the hardware SHDN state,
the state of the TCON0 register’s R0A,
R0W and R0B bits is overridden (ignored).
When the state of the R0HW bit no longer
forces the resistor network into the
hardware SHDN state, the TCON0
register’s R0A, R0W and R0B bits return
to controlling the terminal connection
state. In other words, the R0HW bit does
not corrupt the state of the R0A, R0W and
R0B bits.
TERMINAL CONTROL REGISTER
(TCON)
The Terminal Control (TCON) register is a volatile
register used to configure the connection of each
resistor network terminal pin (A, B and W) to the
resistor network. This register is shown in Register 4-1.
5.7.3
INTERACTION OF SHDN PIN AND
TCON REGISTER
Figure 4-6 shows how the SHDN pin signal and the
R0HW bit signal interact to control the hardware
shutdown of the resistor network.
SHDN (from pin)
R0HW
(from TCON register)
FIGURE 5-12:
Interaction.
To Pot 0 Hardware
Shutdown Control
R0HW bit and SHDN pin
The R0HW bit forces the selected resistor network into
the same state as the SHDN pin. Alternate low-power
configurations may be achieved with the R0A, R0W
and R0B bits.
When the R0HW bit is ‘0’:
• The P0A terminal is disconnected
• The P0W terminal is simultaneously connected to
the P0B terminal (see Figure 5-11)
DS20005207B-page 42
 2013-2015 Microchip Technology Inc.
MCP41HVX1
6.0
SERIAL INTERFACE (SPI)
The MCP41HVX1 devices support the SPI serial
protocol. This SPI operates in the Slave mode (does
not generate the serial clock). The device’s SPI command format operates on multiples of eight bits.
The SPI interface uses up to four pins. These are:
•
•
•
•
CS – Chip Select
SCK – Serial Clock
SDI – Serial Data In
SDO – Serial Data Out
A typical SPI interface is shown in Figure 6-1. In the
SPI interface, the Master’s Output pin is connected to
the Slave’s Input pin, and the Master’s Input pin is
connected to the Slave’s Output pin.
The MCP41HVX1 SPI module supports two (of the
four) standard SPI modes. These are Mode 0,0 and
1,1. The SPI mode is determined by the state of the
SCK pin (VIH or VIL) when the CS pin transitions from
inactive (VIH) to active (VIL).
Note:
Some Host Controller SPI modules only
operate with 16-bit transfers. For these
Host Controllers, only the Read and Write
Commands or the Continuous Increment
or Decrement Commands that are an
even multiple of Increment or Decrement
commands may be used.
Typical SPI Interface Connections
Host
Controller
FIGURE 6-1:
MCP41HVX1
SDO
(Master Out - Slave In (MOSI))
SDI
SDI
(Master In - Slave Out (MISO))
SDO
SCK
SCK
I/O
CS
I/O
WLAT
I/O
SHDN
Typical SPI Interface Block Diagram.
 2013-2015 Microchip Technology Inc.
DS20005207B-page 43
MCP41HVX1
6.1
SDI, SDO, SCK, and CS Operation
The operation of the four SPI interface pins are
discussed in this section. These pins are:
•
•
•
•
Serial Data In (SDI)
Serial Data Out (SDO)
Serial Clock (SCK)
The Chip Select Signal (CS)
SERIAL DATA IN (SDI)
The Serial Data In (SDI) signal is the data signal into
the device. The value on this pin is latched on the rising
edge of the SCK signal.
6.1.2
THE CHIP SELECT SIGNAL (CS)
The Chip Select (CS) signal is used to select the device
and frame a command sequence. To start a command,
or sequence of commands, the CS signal must transition
from the inactive state (VIH) to an active state (VIL).
After the CS signal has gone active, the SDO pin is
driven and the clock bit counter is reset.
The serial interface works on either 8-bit or 16-bit
boundaries depending on the selected command. The
Chip Select (CS) pin frames the SPI commands.
6.1.1
6.1.4
SERIAL DATA OUT (SDO)
The Serial Data Out (SDO) signal is the data signal out
of the device. The value on this pin is driven on the
falling edge of the SCK signal.
Note:
There is a required delay after the CS pin
goes active to the 1st edge of the SCK pin.
If an error condition occurs for an SPI command, then
the command byte’s Command Error (CMDERR) bit (on
the SDO pin) will be driven low (VIL). To exit the error
condition, the user must take the CS pin to the VIH level.
When the CS pin returns to the inactive state (VIH), the
SPI module resets (including the Address Pointer).
While the CS pin is in the inactive state (VIH), the serial
interface is ignored. This allows the host controller to
interface to other SPI devices using the same SDI,
SDO and SCK signals.
Once the CS pin is forced to the active level (VIL), the
SDO pin will be driven. The state of the SDO pin is
determined by the serial bit’s position in the command,
the command selected, and if there is a command error
state (CMDERR).
6.1.5
6.1.3
At 1.8V VL operation, the DGND signal must be 0.9V or
greater above the V- signal. If VL is 2.0V or greater,
then the DGND signal can be tied to the V- signal (see
Table 6-1).
SERIAL CLOCK (SCK)
The Serial Clock (SCK) signal is the clock signal of the
SPI module. The frequency of the SCK pin determines
the SPI frequency of operation.
The SPI interface is specified to operate up to 10 MHz.
The actual clock rate depends on the configuration of
the system and the serial command used. Table 6-1
shows the SCK frequency.
TABLE 6-1:
SCK FREQUENCY
Command
VL
Voltage
Read
Write,
Increment,
Decrement
2.7V
10 MHz
10 MHz
1.8V
1 MHz
1 MHz
DGND = V- + 0.9V
2.0V
1 MHz
1 MHz
DGND = V-
Comment
LOW-VOLTAGE SUPPORT
The Serial Interface is designed to also support 1.8V
operation (at reduced specifications – frequency,
thresholds, etc.). This allows the MCP41HVX1 device
to interface to low-voltage host controllers.
6.1.6
SPLIT RAIL SUPPORT
The Serial Interface is designed to support split rail
systems. In a split rail system, the microcontroller can
operate at a lower voltage than the MCP41HXX1
device. This is achieved with the VIH specification.
For VL  2.7V, the minimum VIH = 0.45  VL. So if the
microcontroller VOH at 1.8V is 0.8  VDD, then VL can
be a maximum of 3.2V (see Equation 6-1).
See Section 8.1 “Split Rail Applications” for additional discussion on split rail support.
EQUATION 6-1:
CALCULATING MAX VL
FOR MICROCONTROLLER
AT 1.8V
If VOH = 0.8 × VDD = 0.8 × 1.8V = 1.44V
Then: VIH(MIN) = 1.44V
With VIH = 0.45 × VL
Then: VL = 1.44V/0.45 = 3.2V
DS20005207B-page 44
 2013-2015 Microchip Technology Inc.
MCP41HVX1
6.2
The SPI Modes
6.3
SPI Waveforms
The SPI module supports two (of the four) standard SPI
modes. These are Mode 0,0 and 1,1. The mode is
determined by the state of the SDI pin on the rising
edge of the first clock bit (of the 8-bit byte).
Figures 6-2 through 6-5 show the different SPI command waveforms. Figure 6-2 and Figure 6-3 are read
and write commands. Figure 6-4 and Figure 6-5 are
Increment and Decrement commands.
6.2.1
6.4
MODE 0,0
In Mode 0,0: SCK Idle state = low (VIL), data is clocked
in on the SDI pin on the rising edge of SCK and clocked
out on the SDO pin on the falling edge of SCK.
6.2.2
Daisy Chaining
This SPI Interface does NOT support daisy chaining.
MODE 1,1
In Mode 1,1: SCK Idle state = high (VIH), data is
clocked in on the SDI pin on the rising edge of SCK and
clocked out on the SDO pin on the falling edge of SCK.
VIH
CS
VIL
SCK
PIC Writes
to SSPBUF
CMDERR bit
SDO
bit15 bit14 bit13 bit12 bit11
AD3 AD2 AD1 AD0
bit15 bit14 bit13 bit12
SDI
C1
bit10 bit9
bit8
bit7
bit6
bit5
bit4
bit3
bit2
bit1
bit0
X
bit9
D8
bit8
D7
bit7
D6
bit6
D5
bit5
D4
bit4
D3
bit3
D2
D1
bit2 bit1
D0
bit0
C0
Input
Sample
FIGURE 6-2:
16-Bit Commands (Write, Read) – SPI Waveform (Mode 1,1).
VIH
CS
VIL
SCK
PIC Writes
to SSPBUF
SDO
SDI
CMDERR bit
bit15
bit14 bit13 bit12 bit11
AD3 AD2 AD1 AD0
bit15 bit14 bit13 bit12
C1
bit10 bit9
bit8
bit7
bit6
bit5
bit4
bit3
bit2
X
bit9
D8
bit8
D7
bit7
D6
bit6
D5
bit5
D4
bit4
D3
bit3
D2
D1
bit2 bit1
C0
bit1
bit0
D0
bit0
Input
Sample
FIGURE 6-3:
16-Bit Commands (Write, Read) – SPI Waveform (Mode 0,0).
 2013-2015 Microchip Technology Inc.
DS20005207B-page 45
MCP41HVX1
VIH
VIL
CS
SCK
PIC Writes
to SSPBUF
CMDERR bit
“1” = Valid Command
“0” = Invalid Command
SDO
bit7
SDI
AD3
bit6
AD2
bit5
AD1
bit4
AD0
bit3
C1
bit2
C0
bit1
X
bit0
X
bit0
bit7
Input
Sample
FIGURE 6-4:
8-Bit Commands (Increment, Decrement) – SPI Waveform with PIC MCU (Mode 1,1).
VIH
CS
VIL
SCK
PIC Writes
to SSPBUF
CMDERR bit
“1” = Valid Command
“0” = Invalid Command
SDO
bit7
SDI
AD3
bit7
bit6
AD2
bit5
AD1
bit4
AD0
bit3
C1
bit2
C0
bit1
X
bit0
X
bit0
Input
Sample
FIGURE 6-5:
DS20005207B-page 46
8-Bit Commands (Increment, Decrement) – SPI Waveform with PIC MCU (Mode 0,0).
 2013-2015 Microchip Technology Inc.
MCP41HVX1
7.0
DEVICE COMMANDS
7.1
All commands have a Command Byte which specifies
the register address and the command. Commands
which require data (write and read commands) also
have the Data Byte.
The MCP41HVX1’s SPI command format supports
sixteen memory address locations and four commands. These commands are shown in Table 7-1.
Commands may be sent when the CS pin is driven to VIL.
The 8-bit commands (Increment Wiper and Decrement
Wiper commands) contain a command byte, while 16-bit
commands (Read Data and Write Data commands)
contain a command byte and a data byte. The command
byte contains two data bits (see Figure 7-1).
7.1.1
Command Name
# of
Bits
11
Read Data
16-Bits
00
Write Data
16-Bits
01
Increment Wiper
8-Bits
10
Decrement Wiper
8-Bits
Command Byte
A A A A C C D D
D D D D 1 0 9 8
3 2 1 0
Data
Bits
Command
Bits
As the command byte is being loaded into the device
on the SDI pin, the device’s SDO pin is driving. The
SDO pin will output high bits for the first six bits of that
command. On the 7th bit, the SDO pin will output the
CMDERR bit state (see Section 7.1.1.1 “Error
Condition”). The 8th bit state depends on the
command selected.
16-bit Command
8-bit Command
Memory
Address
The device memory is accessed when the master
sends a proper command byte to select the desired
operation. The memory location to be accessed is contained in the command byte’s AD3:AD0 bits. The action
desired is contained in the command byte’s C1:C0 bits
(see Table 7-1). C1:C0 determines if the desired memory location will be read, written, incremented (wiper
setting +1) or decremented (wiper setting -1). The
Increment and Decrement commands are only valid on
the volatile wiper registers.
COMMANDS
C1:C0
Bit
States
COMMAND BYTE
The command byte has three fields: the address, the
command, and two data bits (see Figure 7-1).
Currently, only one of the data bits is defined (D8). This
is for the Write command.
Table 7-2 shows the supported commands for each
memory location and the corresponding values on the
SDI and SDO pins.
TABLE 7-1:
Command Format
Command Byte
Data Byte
A A A A C C D D D D D D D D D D
D D D D 1 0 9 8 7 6 5 4 3 2 1 0
3 2 1 0
Data
Bits
Memory
Address
Command
Bits
Command
Bits
CC
1 0
0 0 = Write Data
0 1 = INCR
1 0 = DECR
1 1 = Read Data
D9
This bit is only used as the CMDERR bit.
D8
This bit is not used. Maintained for code compatibility with MCP41XX, MCP42XX and MCP43XX devices.
FIGURE 7-1:
General SPI Command Formats.
 2013-2015 Microchip Technology Inc.
DS20005207B-page 47
MCP41HVX1
TABLE 7-2:
MEMORY MAP AND THE SUPPORTED COMMANDS
Address
Value
00h
Command
Function
MISO (SDO pin)(2)
nn nnnn nnnn
0000 00nn nnnn nnnn 1111 1111 1111 1111
Read Data
nn nnnn nnnn
0000 11nn nnnn nnnn 1111 111n nnnn nnnn
Increment Wiper
—
0000 0100
1111 1111
Decrement Wiper
—
0000 1000
1111 1111
—
—
—
—
Reserved
04h(3)
Volatile
Write Data
TCON Register Read Data
05h –
0Fh(4)
Reserved
3:
4:
MOSI (SDI pin)
Volatile Wiper 0 Write Data
01h –
03h(4)
Note 1:
2:
SPI String (Binary)
Data
(10-bits)(1)
—
nn nnnn nnnn
nn nnnn nnnn
—
0100 00nn nnnn nnnn 1111 1111 1111 1111
0100 11nn nnnn nnnn 1111 111n nnnn nnnn
—
—
The data memory is eight bits wide, so the two MSbs (D9:D8) are ignored by the device.
All these address/command combinations are valid, so the CMDERR bit is set. Any other
address/command combination is a command error state and the CMDERR bit will be clear.
Increment or Decrement commands are invalid for these addresses.
Reserved addresses: Any command is invalid for these addresses.
DS20005207B-page 48
 2013-2015 Microchip Technology Inc.
MCP41HVX1
7.1.1.1
Error Condition
The CMDERR bit indicates if the four address bits
received (AD3:AD0) and the two command bits
received (C1:C0) are a valid combination. The
CMDERR bit is high if the combination is valid and low
if the combination is invalid (see Table 7-3).
The command error bit will also be low if a write to a
Reserved Address has been specified. SPI commands
that do not have a multiple of eight clocks are ignored.
Once an error condition has occurred, any following
commands are ignored. All following SDO bits will be
low until the CMDERR condition is cleared by forcing
the CS pin to the inactive state (VIH).
TABLE 7-3:
CMDERR
Bit States
COMMAND ERROR BIT
Description
7.1.2
DATA BYTE
Only the Read command and the Write command use
the data byte (see Figure 7-1). These commands
concatenate the eight bits of the data byte with the one
data bit (D8) contained in the command byte to form
nine bits of data (D8:D0). The command byte format
supports up to nine bits of data, but the MCP41HVX1
only uses the lower eight bits. That means that the
full-scale code of the 8-bit resistor network is FFh.
When at full-scale, the wiper connects to Terminal A.
The D8 bit is maintained for code compatibility with the
MCP41XX, MCP42XX, and MCP43XX devices.
The D9 bit is currently unused, and corresponds to the
position on the SDO data of the CMDERR bit.
7.1.3
CONTINUOUS COMMANDS
1
“Valid” Command/Address combination
The device supports the ability to execute commands
continuously while the CS pin is in the active state (VIL).
Any sequence of valid commands may be received.
0
“Invalid” Command/Address combination
The following example is a valid sequence of events:
Aborting a Transmission
All SPI transmissions must have the correct number of
SCK pulses to be executed. The command is not
executed until the complete number of clocks have
been received. Some commands also require the CS
pin to be forced inactive (VIH). If the CS pin is forced to
the inactive state (VIH), the serial interface is reset.
Partial commands are not executed.
SPI is more susceptible to noise than other bus
protocols. The most likely case is that this noise
corrupts the value of the data being clocked into the
MCP41HVX1 or the SCK pin is injected with extra clock
pulses. This may cause data to be corrupted in the
device or cause a command error to occur, since the
address and command bits were not a valid combination. The extra SCK pulse will also cause the SPI data
(SDI) and clock (SCK) to be out of sync. Forcing the CS
pin to the inactive state (VIH) resets the serial interface.
The SPI interface will ignore activity on the SDI and
SCK pins until the CS pin transition to the active state
is detected (VIH to VIL).
1.
2.
3.
4.
5.
6.
7.
8.
CS pin driven active (VIL).
Read Command.
Increment Command (Wiper 0).
Increment Command (Wiper 0).
Decrement Command (Wiper 0).
Write Command.
Read Command.
CS pin driven inactive (VIH).
Note 1: It is recommended that while the CS pin
is active, only one type of command
should be issued. When changing commands, it is recommended to take the CS
pin inactive, then force it back to the
active state.
2: It is also recommended that long
command strings should be broken down
into shorter command strings. This
reduces the probability of noise on the
SCK pin corrupting the desired SPI
command string.
Note 1: When data is not being received by the
MCP41HVX1, it is recommended that the
CS pin be forced to the inactive level (VIL)
2: It is also recommended that long
continuous command strings should be
broken down into single commands or
shorter continuous command strings. This
reduces the probability of noise on the SCK
pin corrupting the desired SPI commands.
 2013-2015 Microchip Technology Inc.
DS20005207B-page 49
MCP41HVX1
7.2
Write Data
7.2.1
The Write command is a 16-bit command. The format
of the command is shown in Figure 7-2.
SINGLE WRITE TO VOLATILE
MEMORY
The write operation requires that the CS pin be in the
active state (VIL). Typically, the CS pin will be in the
inactive state (VIH) and is driven to the active state
(VIL). The 16-bit Write command (command byte and
data byte) is then clocked (SCK pin) in on the SDI pin.
Once all 16 bits have been received, the specified
volatile address is updated. A write will not occur if the
write command isn’t exactly 16 clocks pulses.
A Write command to a volatile memory location
changes that location after a properly formatted Write
command (16-clock) has been received.
Figures 6-2 and 6-3 show possible waveforms for a
single write.
DATA BYTE
COMMAND BYTE
A
D
SDI
3
1
SDO 1
A
D
2
1
1
A
D
1
1
1
A
D
0
1
1
0
0
D
9
D
8
D
7
D
6
D
5
D
4
D
3
D
2
D
1
D
0
1
1
1
1
1
0
1
0
1
0
1
0
1
0
1
0
1
0
1
0
1
0
1 Valid Address/Command combination
0 Invalid Address/Command combination (1)
Note 1: If an Error Condition occurs (CMDERR = L), all following SDO bits will be low until the CMDERR
condition is cleared (the CS pin is forced to the inactive state).
FIGURE 7-2:
DS20005207B-page 50
Write Command – SDI and SDO States.
 2013-2015 Microchip Technology Inc.
MCP41HVX1
7.2.2
CONTINUOUS WRITES TO
VOLATILE MEMORY
Continuous writes are possible only when writing to the
volatile memory registers (address 00h and 04h).
Figure 7-3 shows the sequence for three continuous
writes. The writes do not need to be to the same volatile
memory address.
COMMAND BYTE
SDI
SDO
A
D
3
1
A
D
2
1
A
D
1
1
A
D
0
1
A
D
3
1
A
D
2
1
A
D
1
1
A
D
0
1
A
D
3
1
A
D
2
1
A
D
1
1
A
D
0
1
DATA BYTE
0
0
D
9
D
8
D
7
D
6
D
5
D
4
D
3
D
2
D
1
D
0
1
1
1*
1
1
1
1
1
1
1
1
1
0
0
D
9
D
8
D
7
D
6
D
5
D
4
D
3
D
2
D
1
D
0
1
1
1*
1
1
1
1
1
1
1
1
1
0
0
D
9
D
8
D
7
D
6
D
5
D
4
D
3
D
2
D
1
D
0
1
1
1*
1
1
1
1
1
1
1
1
1
Note 1: If a Command Error (CMDERR) occurs at this bit location (*), then all following SDO bits will be
driven low until the CS pin is driven inactive (VIH).
FIGURE 7-3:
Continuous Write Sequence.
 2013-2015 Microchip Technology Inc.
DS20005207B-page 51
MCP41HVX1
7.3
Read Data
7.3.1
The read operation requires that the CS pin be in the
active state (VIL). Typically, the CS pin will be in the
inactive state (VIH) and is driven to the active state
(VIL). The 16-bit Read command (command byte and
data byte) is then clocked (SCK pin) in on the SDI pin.
The SDO pin starts driving data on the 7th bit
(CMDERR bit) and the addressed data comes out on
the 8th through 16th clocks. Figures 6-2 through 6-3
show possible waveforms for a single read.
The Read command is a 16-bit command. The format
of the command is shown in Figure 7-4.
The first six bits of the Read command determine the
address and the command. The 7th clock will output
the CMDERR bit on the SDO pin. The 8th clock will be
fixed at 1, and with the remaining eight clocks, the
device will transmit the eight data bits (D7:D0) of the
specified address (AD3:AD0).
Figure 7-4 shows the SDI and SDO information for a
Read command.
COMMAND BYTE
SDI
SDO
SINGLE READ
DATA BYTE
A
D
3
1
A
D
2
1
A
D
1
1
A
D
0
1
1
1
X
X
X
X
X
X
X
X
X
X
1
1
1
1
1
1
1
1
1
1
0
0
D
7
0
D
6
0
D
5
0
D
4
0
D
3
0
D
2
0
D
1
0
D Valid Address/Command combination
0
0 Attempted Memory Read of Reserved
Memory location
READ DATA
FIGURE 7-4:
DS20005207B-page 52
Read Command – SDI and SDO States.
 2013-2015 Microchip Technology Inc.
MCP41HVX1
7.3.2
CONTINUOUS READS
Figure 7-5 shows the sequence for three continuous
reads. The reads do not need to be to the same
memory address.
Continuous reads allow the device’s memory to be
read quickly. Continuous reads are possible to all
memory locations.
COMMAND BYTE
SDI
SDO
A
D
3
1
A
D
2
1
A
D
1
1
A
D
0
1
A
D
3
1
A
D
2
1
A
D
1
1
A
D
0
1
A
D
3
1
A
D
2
1
A
D
1
1
A
D
0
1
X
DATA BYTE
1
1
X
X
X
X
X
X
X
X
X
1
1 1* 1
D
7
D
6
D
5
D
4
D
3
D
2
D
1
D
0
1
1
X
X
X
X
X
X
X
X
X
1
1 1* 1
D
7
D
6
D
5
D
4
D
3
D
2
D
1
D
0
1
1
X
X
X
X
X
X
X
X
X
1
1 1* 1
D
7
D
6
D
5
D
4
D
3
D
2
D
1
D
0
X
X
Note 1: If a Command Error (CMDERR) occurs at this bit location (*), then all following SDO bits will be
driven low until the CS pin is driven inactive (VIH).
FIGURE 7-5:
Continuous Read Sequence.
 2013-2015 Microchip Technology Inc.
DS20005207B-page 53
MCP41HVX1
7.4
Increment Wiper
7.4.1
The Increment command is an 8-bit command. The
Increment command can only be issued to specific
volatile memory locations (the wiper register). The format of the command is shown in Figure 7-6.
An Increment command to the volatile memory location
changes that location after a properly formatted
command (eight clocks) has been received.
Increment commands provide a quick and easy
method to modify the value of the volatile wiper location
by +1 with minimal overhead.
COMMAND BYTE
(INCR COMMAND (n+1))
A
D
3
1
SDO 1
SDI
A
D
2
1
1
A
D
1
1
1
A
D
0
1
1
0
1
X
X
1
1
1 1* 1 Note 1, 2
1 0 0 Note 1, 3
Note 1: Only functions when writing the volatile
wiper register (AD3:AD0 = 0h).
2: Valid Address/Command combination.
3: Invalid Address/Command combination
all following SDO bits will be low until the
CMDERR condition is cleared (the CS
pin is forced to the inactive state).
FIGURE 7-6:
Increment Command –
SDI and SDO States.
Note:
Table 7-2 shows the valid addresses for
the Increment Wiper command. Other
addresses are invalid.
DS20005207B-page 54
SINGLE INCREMENT
Typically, the CS pin starts at the inactive state (VIH),
but may already be in the active state due to the
completion of another command.
Figures 6-4 through 6-5 show possible waveforms for a
single increment. The increment operation requires
that the CS pin be in the active state (VIL). Typically, the
CS pin will be in the inactive state (VIH) and is driven to
the active state (VIL). The 8-bit Increment command
(command byte) is then clocked in on the SDI pin by the
SCK pins. The SDO pin drives the CMDERR bit on the
7th clock.
The wiper value will increment up to FFh on 8-bit
devices and 7Fh on 7-bit devices. After the wiper value
has reached full-scale (8-bit = FFh, 7-bit = 7Fh), the
wiper value will not be incremented further. See
Table 7-4 for additional information on the Increment
command versus the current volatile wiper value.
The increment operations only require the Increment
command byte while the CS pin is active (VIL) for a
single increment.
After the wiper is incremented to the desired position,
the CS pin should be forced to VIH to ensure that
unexpected transitions on the SCK pin do not cause
the wiper setting to change. Driving the CS pin to VIH
should occur as soon as possible (within device
specifications) after the last desired increment occurs.
TABLE 7-4:
Current Wiper
Setting
7-bit
Pot
8-bit
Pot
INCREMENT OPERATION VS.
VOLATILE WIPER VALUE
Wiper (W)
Properties
7Fh
FFh
Full-Scale (W = A)
7Eh
40h
FEh
80h
W=N
3Fh
7Fh
W = N (Mid-Scale)
3Eh
01h
7Eh
01h
W=N
00h
00h
Zero-Scale (W = B)
Increment
Command
Operates?
No
Yes
Yes
 2013-2015 Microchip Technology Inc.
MCP41HVX1
7.4.2
CONTINUOUS INCREMENTS
When executing a continuous command string, the
Increment command can be followed by any other valid
command.
Continuous increments are possible only when writing
to the volatile wiper registers (address 00h).
The wiper terminal will move after the command has
been received (8th clock).
Figure 7-7 shows a continuous increment sequence.
When executing a continuous Increment command,
the selected wiper will be altered from n to n+1 for each
Increment command received. The wiper value will
increment up to FFh on 8-bit devices and 7Fh on 7-bit
devices. After the wiper value has reached full-scale
(8-bit = FFh, 7-bit = 7Fh), the wiper value will not be
incremented further.
After the wiper is incremented to the desired position,
the CS pin should be forced to VIH to ensure that
unexpected transitions on the SCK pin do not cause
the wiper setting to change. Driving the CS pin to VIH
should occur as soon as possible (within device
specifications) after the last desired increment occurs.
Increment commands can be sent repeatedly without
raising CS until a desired condition is met.
(INCR COMMAND (n+1))
A
D
3
1
1
SDO 1
1
SDI
A
D
2
1
1
1
1
A
D
1
1
1
1
1
A
D
0
1
1
1
1
X
COMMAND BYTE
COMMAND BYTE
COMMAND BYTE
(INCR COMMAND (n+2))
0
1
X
1
1
1
1
1 1* 1
1 0 0
1 1 1
1 1 1
A
D
3
1
0
1
1
A
D
2
1
0
1
1
A
D
1
1
0
1
1
A
D
0
1
0
1
1
X
(INCR COMMAND (n+3))
0
1
X
1
0
1
1
1 1* 1
0 0 0
1 0 0
1 1 1
A
D
3
1
0
0
1
A
D
2
1
0
0
1
A
D
1
1
0
0
1
A
D
0
1
0
0
1
0
1
X
X
1
0
0
1
1 1* 1 Note 1,
0 0 0 Note 3,
0 0 0 Note 3,
1 0 0 Note 3,
2
4
4
4
Note 1: Only functions when writing the volatile wiper register (AD3:AD0 = 0h).
2: Valid Address/Command combination.
3: Invalid Address/Command combination.
4: If an error condition occurs (CMDERR = L), all following SDO bits will be low until the CMDERR
condition is cleared (the CS pin is forced to the inactive state).
FIGURE 7-7:
Continuous Increment Command – SDI and SDO States.
 2013-2015 Microchip Technology Inc.
DS20005207B-page 55
MCP41HVX1
7.5
Decrement Wiper
7.5.1
The Decrement command is an 8-bit command. The
Decrement command can only be issued to volatile
wiper locations. The format of the command is shown
in Figure 7-8.
A Decrement command to the volatile wiper location
changes that location after a properly formatted
command (eight clocks) has been received.
Decrement commands provide a quick and easy
method to modify the value of the volatile wiper location
by -1 with minimal overhead.
COMMAND BYTE
(DECR COMMAND (n+1))
A
D
3
1
SDO 1
SDI
A
D
2
1
1
A
D
1
1
1
A
D
0
1
1
1
0
X
X
1
1
1 1* 1 Note 1, 2
1 0 0 Note 1, 3
Note 1: Only functions when writing the volatile
wiper registers (AD3:AD0 = 0h).
2: Valid Address/Command combination.
3: Invalid Address/Command combination,
all following SDO bits will be low until the
CMDERR condition is cleared.
(the CS pin is forced to the inactive
state).
FIGURE 7-8:
Decrement Command –
SDI and SDO States.
Note:
Table 7-2 shows the valid addresses for
the Decrement Wiper command. Other
addresses are invalid.
DS20005207B-page 56
SINGLE DECREMENT
Typically, the CS pin starts at the inactive state (VIH),
but may already be in the active state due to the
completion of another command.
Figures 6-4 through 6-5 show possible waveforms for a
single decrement. The decrement operation requires
that the CS pin be in the active state (VIL). Typically, the
CS pin will be in the inactive state (VIH) and is driven to
the active state (VIL). Then the 8-bit Decrement
command (command byte) is clocked in on the SDI pin
by the SCK pin. The SDO pin drives the CMDERR bit
on the 7th clock.
The wiper value will decrement from the wiper’s
full-scale value (FFh on 8-bit devices and 7Fh on 7-bit
devices). If the wiper register has a zero-scale value
(00h), then the wiper value will not decrement. See
Table 7-5 for additional information on the Decrement
command vs. the current volatile wiper value.
The Decrement commands only require the Decrement
command byte while the CS pin is active (VIL) for a
single decrement.
After the wiper is decremented to the desired position,
the CS pin should be forced to VIH to ensure that
unexpected transitions on the SCK pin do not cause
the wiper setting to change. Driving the CS pin to VIH
should occur as soon as possible (within device
specifications) after the last desired decrement occurs.
TABLE 7-5:
Current Wiper
Setting
7-bit
Pot
8-bit
Pot
DECREMENT OPERATION VS.
VOLATILE WIPER VALUE
Wiper (W)
Properties
7Fh
FFh
Full-Scale (W = A)
7Eh
40h
FEh
80h
W=N
3Fh
7Fh
W = N (Mid-Scale)
3Eh
01h
7Eh
01h
W=N
00h
00h
Zero-Scale (W = B)
Decrement
Command
Operates?
Yes
Yes
No
 2013-2015 Microchip Technology Inc.
MCP41HVX1
7.5.2
CONTINUOUS DECREMENTS
When executing a continuous command string, the
Decrement command can be followed by any other
valid command.
Continuous decrements are possible only when writing
to the volatile wiper register (address 00h).
The wiper terminal will move after the command has
been received (8th clock).
Figure 7-9 shows a continuous decrement sequence.
When executing continuous Decrement commands,
the selected wiper will be altered from n to n-1 for each
Decrement command received. The wiper value will
decrement from the wiper’s full-scale value (FFh on
8-bit devices and 7Fh on 7-bit devices). If the Wiper
register has a zero-scale value (00h), then the wiper
value will not decrement. See Table 7-5 for additional
information on the Decrement command vs. the current
volatile wiper value.
After the wiper is decremented to the desired position,
the CS pin should be forced to VIH to ensure that
“unexpected” transitions on the SCK pin do not cause
the wiper setting to change. Driving the CS pin to VIH
should occur as soon as possible (within device
specifications) after the last desired decrement occurs.
Decrement commands can be sent repeatedly without
raising CS until a desired condition is met.
(DECR COMMAND (n-1))
A
D
3
1
1
SDO 1
1
SDI
A
D
2
1
1
1
1
A
D
1
1
1
1
1
A
D
0
1
1
1
1
X
COMMAND BYTE
COMMAND BYTE
COMMAND BYTE
(DECR COMMAND (n-1))
1
0
X
1
1
1
1
1 1* 1
1 0 0
1 1 1
1 1 1
A
D
3
1
0
1
1
A
D
2
1
0
1
1
A
D
1
1
0
1
1
A
D
0
1
0
1
1
X
(DECR COMMAND (n-1))
1
0
X
1
0
1
1
1 1* 1
0 0 0
1 0 0
1 1 1
A
D
3
1
0
0
1
A
D
2
1
0
0
1
A
D
1
1
0
0
1
A
D
0
1
0
0
1
1
0
X
X
1
0
0
1
1 1* 1 Note 1,
0 0 0 Note 3,
0 0 0 Note 3,
1 0 0 Note 3,
2
4
4
4
Note 1: Only functions when writing the volatile wiper registers (AD3:AD0 = 0h).
2: Valid Address/Command combination.
3: Invalid Address/Command combination.
4: If an Error Condition occurs (CMDERR = L), all following SDO bits will be low until the CMDERR
condition is cleared (the CS pin is forced to the inactive state).
FIGURE 7-9:
Continuous Decrement Command – SDI and SDO States.
 2013-2015 Microchip Technology Inc.
DS20005207B-page 57
MCP41HVX1
NOTES:
DS20005207B-page 58
 2013-2015 Microchip Technology Inc.
MCP41HVX1
8.0
APPLICATIONS EXAMPLES
8.1
PIC® MCU
SDO
CS
SCK
SDI
FIGURE 8-1:
To ensure that communication properly occurs
between the devices, care must be done to verify the
compatibility of the VIL, VIH, VOL and VOH levels of the
interface signals between the devices. These interface
signals are:
VDD
(minimum)
CS
SCK
SDI
SDO
SHDN
WLAT
TABLE 8-1:
2.7V
Note 1:
Table 8-1
shows
the
calculated
maximum
MCP41HVX1 VL based on the microcontroller’s
minimum VOH.
Note:
VOH specifications typically have a current
load specified. This is due to the pin
expected to drive externally circuitry. If the
pin is unloaded (or lightly loaded), then the
VOH of the pin could approach the device
VDD (this is dependent on the implementation of the output driver circuit). For VOL,
unloaded (or lightly loaded) pins could
approach the device VSS.
For VOH and VOL characterization graphs
from an example microcontroller, see the
PIC16F1934 data sheet (DS41364), Figure 31-15 and Figure 31-16.
WLAT
SHDN
Example Split Rail System.
MCP41HVX1 VL VOLTAGE
BASED ON
MICROCONTROLLER VOH
PIC® MCU
1.8V
When the microcontroller is at a lower-voltage rail, the
VOH of the microcontroller needs to be greater than the
VIH of the MCP41HVX1, and the VIL of the microcontroller needs to be greater than the VOL of the
MCP41HVX1.
MCP41HVX1
SDI
CS
SCK
SDO
I/O
I/O
Split Rail Applications
Split rail applications are when one device operates
from one voltage level (rail) and the second device
operates from a second voltage level (rail). The typical
scenario will be when the microcontroller is operating at
a lower voltage level (for power savings, etc) and the
MCP41HVX1 is operating at a higher voltage level to
maximize operational performance. This configuration
is shown in Figure 8-1.
•
•
•
•
•
•
3.0V
Voltage
Regulator
2.0V (1.8V min)
Digital potentiometers have a multitude of practical
uses in modern electronic circuits. The most popular
uses include precision calibration of set point
thresholds, sensor trimming, LCD bias trimming, audio
attenuation, adjustable power supplies, motor control
overcurrent trip setting, adjustable gain amplifiers and
offset trimming.
2:
VOH (minimum)(1)
Formula
Calculated
(with load)
MCP41HVX1
Max VL
0.7 × VDD
1.26V
2.8V
0.8 × VDD
1.44V
3.2V
0.85 × VDD
1.53V
3.4V
0.9 × VDD
1.62V
3.6V
VDD
1.8V
4.0V
VDD - 0.7V
1.1V
2.44V
0.7 × VDD
1.89V
4.2V
0.8 × VDD
2.16V
4.8V
0.9 × VDD
2.43V
5.4V
VDD
2.7V
5.5V
The VOH minimum voltage is determined
by the load on the pin. If the load is small,
a typical output’s voltage should approach
the device’s VDD voltage. This is dependent on the device’s output driver design.
Split Rail voltages are dependent on VIL,
VIH, VOL, and VOH of the microcontroller
and the MCP41HVX1 devices.
FIGURE 8-2:
Example PIC®
Microcontroller VOH Characterization Graph
(VDD = 1.8V).
 2013-2015 Microchip Technology Inc.
DS20005207B-page 59
MCP41HVX1
8.2
Using Shutdown Modes
Figure 8-3 shows a possible application circuit where
the independent terminals could be used.
Disconnecting the wiper allows the transistor input to
be taken to the bias voltage level (disconnecting A
and/or B may be desired to reduce system current).
Disconnecting Terminal A modifies the transistor input
by the RBW rheostat value to the Common B.
Disconnecting Terminal B modifies the transistor input
by the RAW rheostat value to the Common A. The
Common A and Common B connections could be
connected to V+ and V-.
8.3
High-Voltage DAC
A high-voltage DAC can be implemented using the
MCP41HVXX, with voltages as high as 36V. The circuit
is shown in Figure 8-4. The equation to calculate the
voltage output is shown in Equation 8-1.
V+
High Voltage DAC
VD
D1
V+
+
OPA170
-
Common A
R2
MCP41HVXX
A
B
R1
Input
V+
+
OPA170
-
A
To base
of Transistor
(or Amplifier)
W
VOUT
FIGURE 8-4:
High-Voltage DAC.
EQUATION 8-1:
DAC OUTPUT VOLTAGE
CALCULATION
8-bit
B
N
R1
V OUT  N  = ---------   V D   1 + ------- 

255 
R2 
Input
N = 0 to 255 (decimal)
Common B
7-bit
Balance
Bias
FIGURE 8-3:
Example Application Circuit
using Terminal Disconnects.
DS20005207B-page 60
N
R1
V OUT  N  = ---------   V D   1 + ------- 
127
R2
N = 0 to 127 (decimal)
 2013-2015 Microchip Technology Inc.
MCP41HVX1
8.4
Variable Gain Instrumentation
Amplifier
A variable gain instrumentation amplifier can be
implemented using the MCP41HVXX along with a
high-voltage dual analog switch and a high-voltage
instrumentation amplifier. An example circuit is shown
in Figure 8-5. The equation to calculate the voltage
output is shown in Equation 8-2.
S8A
S1B
DB
B
W
AD8221
A
VOUT
S8B
FIGURE 8-5:
Variable Gain
Instrumentation Amplifier for Data Acquisition
System.
EQUATION 8-2:
Audio Volume Control
A digital volume control can be implemented with the
MCP41HVXX. Figure 8-6 shows a simple audio
volume control implementation.
Figure 8-7 shows a circuit-referenced voltage crossing
detect circuit. The output of this circuit could be used to
control the wiper latch of the MCP41HVXX device in
the audio volume control circuit to reduce zipper noise
or to update the different channels at the same time.
The op amp (U1) could be an MCP6001, while the general purpose comparators (U2 and U3) could be an
MCP6541. U4 is a simple AND gate.
V+
MCP41HVxx
ADG1207
S1A
DA
8.5
U1 establishes the signal zero reference. The upper
limit of the comparator is set above its offset. The WLAT
pin is forced high whenever the voltage falls between
2.502V and 2.497V (a 0.005V window).
The capacitor C1 AC couples the VIN signal into the circuit before feeding into the windowed comparator (and
MCP41HVXX Terminal A pin).
V+
DAC OUTPUT VOLTAGE
CALCULATION
8-bit
49.4 k 
Gain  N  = 1 + --------------------------N
---------  R AB
255
MCP41HVXX
A
VIN
49.4 k 
Gain  N  = 1 + --------------------------N
---------  R AB
127
+
SDI
SCK B
WLAT
N = 0 to 255 (decimal)
7-bit
V+
VL
GND
VOUT
V-
V-
FIGURE 8-6:
Audio Volume Control.
N = 0 to 127 (decimal)
+5V
VIN
R3
100 k
C1
1 µF
+5V
+
R4
R1 200 k
90 k
U2
U4
+5V
R2
10 k
+5V
U1
+
-
FIGURE 8-7:
Crossing Detect.
 2013-2015 Microchip Technology Inc.
WLAT
+
U3
R5
100 k
Referenced Voltage
DS20005207B-page 61
MCP41HVX1
Programmable Power Supply
8.7
The ADP1611 is a step-up DC-to-DC switching converter. Using the MCP41HVXX device allows the power
supply to be programmable up to 20V. Figure 8-8
shows a programmable power supply implementation.
Equation 8-3 shows the equation to calculate the
output voltage of the programmable power supply. This
output is derived from the RBW resistance of the
MCP41HVXX device and the R2 resistor. The
ADP1611 will adjust its output voltage to maintain
1.23V on the FB pin.
When power is connected, L1 acts as a short, and
VOUT is a diode drop below the +5V voltage. The VOUT
voltage will ramp to the programmed value.
MCP41HVXX
(100 k)
V+
A
W
IN
RT
FB
B
C3
22 nF
LOAD CURRENT (IL)
 R 2A + R3A 
IL = ---------------------------------  V W
R 1A  R3A
C2
10 µF
R1
8.5 k
R1B
R2B
150 k
15 k
C2
L1
4.7 µF
10 pF
D1
SS
COMP
R2
220 k
U2
VOUT
+
-15V
C5
10 µF
FIGURE 8-8:
Supply.
Programmable Power
EQUATION 8-3:
POWER SUPPLY OUTPUT
VOLTAGE CALCULATION
N  R AB
 --------------------

 255  

VOUT  N  = 1.23V   1 +  ---------------------- 
 R2  




R3B
50 k
+15V
SW
C4
150 pF
8-bit
EQUATION 8-4:
+5V
ADP1611
C1
0.1 µF
Programmable Bidirectional
Current Source
A programmable bidirectional current source can be
implemented with the MCP41HVXX. Figure 8-9 shows
an implementation where U1 and U2 work together to
deliver the desired current (dependent on selected
device) in both directions. The circuit is symmetrical
(R1A = R1B, R2A = R2B, R3A = R3B) in order to improve
stability. If the resistors are matched, the load current
(IL) calculation is shown below:
C1
V+
MCP41HVXX
8.6
A
W
B
+15V
+
U1
-15V
V-
FIGURE 8-9:
Current Source.
10 pF
R1A
R3A
50 k
R2A
150 k 14.95 k
R4
500
VL
IL
Programmable Bidirectional
N = 0 to 255 (decimal)
7-bit
N  R AB
 --------------------

 127  

VOUT  N  = 1.23V   1 +  ---------------------- 
 R2  




N = 0 to 127 (decimal)
DS20005207B-page 62
 2013-2015 Microchip Technology Inc.
MCP41HVX1
8.8
LCD Contrast Control
8.9
The MCP41HVXX can be used for LCD contrast
control. Figure 8-10 shows a simple programmable
LCD contrast control implementation.
Table 8-2 shows the time for each SPI serial interface
command as well as the effective data update rate that
can be supported by the digital interface (based on the
two SPI serial interface frequencies). So, the Serial
Interface performance, along with the wiper response
time, would be used to determine your application’s
volatile Wiper register update rate.
Some LCD panels support a fixed power supply of up
to 28V. The high voltage digital potentiometer's wiper
can support contrast adjustments through the entire
voltage range.
D1
VOUT (LCD Bias)
MCP41HVXX
C1
10 µF
uController
SDO
SCK
CS
FIGURE 8-10:
Control.
TABLE 8-2:
Serial Interface Communication
Times
LCD Panel
Fixed
(up to +28V)
A
W
+16V to +26V
Contrast Adj.
B
Programmable Contrast
SERIAL INTERFACE TIMES/FREQUENCIES(1)
Command
# of Serial
Interface bits
Example
# Bytes
Transferred
Command
Time (μs)
# of Serial
Interface bits
1 MHz
10 MHz
Effective Data
Update
Frequency (kHz)(2)
1 MHz
10 MHz
Write Single Byte
16
1
16
16
1.6
62,500
625,000
Write Continuous
Bytes
N × 16
5
80
80
8
12,500
125,000
Read Byte
16
1
16
16
1.6
62,500
625,000
Read Continuous
Bytes
N × 16
5
80
80
8
12,500
125,000
Increment Wiper
8
1
8
8
0.8
125,000
1,250,000
N×8
5
40
40
4
25,000
250,000
8
1
8
8
0.8
125,000
1,250,000
N×8
5
40
40
4
25,000
250,000
Continuous
Increments
Decrement Wiper
Continuous
Decrements
Note 1:
2:
Includes the Start or Stop bits.
This is the command frequency multiplied by the number of bytes transferred.
 2013-2015 Microchip Technology Inc.
DS20005207B-page 63
MCP41HVX1
8.10
Implementing Log Steps with a
Linear Digital Potentiometer
In audio volume control applications, the use of
logarithmic steps is desirable since the human ear
hears in a logarithmic manner. The use of a linear
potentiometer can approximate a log potentiometer,
but with fewer steps. An 8-bit potentiometer can
achieve fourteen 3 dB log steps plus a 100% (0 dB)
and a mute setting.
Figure 8-11 shows a block diagram of one of the
MCP45HVx1 resistor networks being used to attenuate
an input signal. In this case, the attenuation will be
ground referenced. Terminal B can be connected to a
common-mode voltage, but the voltages on the A, B
and wiper terminals must not exceed the
MCP45HVX1’s V+/V- voltage limits.
MCP45HVX1
P0A
P0W
P0B
FIGURE 8-11:
Signal Attenuation Block
Diagram – Ground Referenced.
Equation 8-5 shows the equation to calculate voltage
dB gain ratios for the digital potentiometer, while
Equation 8-6 shows the equation to calculate
resistance dB gain ratios. These two equations assume
that the B terminal is connected to ground.
If Terminal B is not directly resistively connected to
ground, then this Terminal B to ground resistance
(RB2GND) must be included into the calculation.
Equation 8-7 shows this equation.
DS20005207B-page 64
EQUATION 8-5:
dB CALCULATIONS
(VOLTAGE)
L = 20 × log10 (VOUT/VIN)
dB
-3
-2
-1
EQUATION 8-6:
VOUT/VIN Ratio
0.70795
0.79433
0.89125
dB CALCULATIONS
(RESISTANCE) – CASE 1
Terminal B connected to Ground (see Figure 8-11)
L = 20 × log10 (RBW/RAB)
EQUATION 8-7:
dB CALCULATIONS
(RESISTANCE) – CASE 2
Terminal B through RB2GND to Ground
L = 20 × log10 ( (RBW + RB2GND)/(RAB + RB2GND) )
Table 5-3 shows the codes that can be used for 8-bit
digital potentiometers to implement the log attenuation.
The table shows the wiper codes for -3 dB, -2 dB, and
-1 dB attenuation steps. This table also shows the
calculated attenuation based on the wiper code’s linear
step. Calculated attenuation values less than the
desired attenuation are shown with red text. At lower
wiper code values, the attenuation may skip a step. If
this occurs, the next attenuation value is colored
magenta to highlight that a skip occurred. For example,
in the -3 dB column the -48 dB value is highlighted
since the -45 dB step could not be implemented (there
are no wiper codes between 2 and 1).
 2013-2015 Microchip Technology Inc.
MCP41HVX1
TABLE 8-3:
LINEAR TO LOG ATTENUATION FOR 8-BIT DIGITAL POTENTIOMETERS
-3 dB Steps
# of
Steps
Desired
Attenuation
Wiper
Code
-2 dB Steps
Calculated
Attenuation
(1)
Desired
Attenuation
Wiper
Code
-1 dB Steps
Calculated
Attenuation
(1)
Desired
Attenuation
Wiper
Code
Calculated
Attenuation
(1)
0
0 dB
255
0 dB
0 dB
255
0 dB
0 dB
255
0 dB
1
-3 dB
180
-3.025 dB
-2 dB
203
-1.981 dB
-1 dB
227
-1.010 dB
2
-6 dB
128
-5.987 dB
-4 dB
161
-3.994 dB
-2 dB
203
-1.981 dB
3
-9dB
90
-9.046 dB
-6 dB
128
-5.987 dB
-3 dB
180
-3.025 dB
4
-12 dB
64
-12.007 dB
-8 dB
101
-8.044 dB
-4 dB
161
-3.994 dB
-5.024 dB
5
-15 dB
45
-15.067 dB
-10 dB
81
-9.961 dB
-5 dB
143
6
-18 dB
32
-18.028 dB
-12 dB
64
-12.007 dB
-6 dB
128
-5.987 dB
7
-21 dB
23
-20.896 dB
-14 dB
51
-13.979 dB
-7 dB
114
-6.993 dB
8
-24 dB
16
-24.048 dB
-16 dB
40
-16.090 dB
-8 dB
101
-8.044 dB
9
-27 dB
11
-27.303 dB
-18 dB
32
-18.028 dB
-9 dB
90
-9.046 dB
10
-30 dB
8
-30.069 dB
-20 dB
25
-20.172 dB
-10 dB
81
-9.961 dB
11
-33 dB
6
-32.568 dB
-22 dB
20
-22.110 dB
-11 dB
72
-10.984 dB
12
-36 dB
4
-36.090 dB
-24 dB
16
-24.048 dB
-12 dB
64
-12.007 dB
-13.013 dB
13
-39 dB
3
-38.588 dB
-26 dB
13
-25.852 dB
-13 dB
57
14
-42 dB
2
-42.110 dB
-28 dB
10
-28.131 dB
-14 dB
51
-13.979 dB
15
-48 dB
1
-48.131 dB
-30 dB
8
-30.069 dB
-15 dB
45
-15.067 dB
16
Mute
0
Mute
-32 dB
6
-32.602 dB
-16 dB
40
-16.090 dB
17
-34 dB
5
-34.151 dB
-17 dB
36
-17.005 dB
18
-36 dB
4
-36.090 dB
-18 dB
32
-18.028 dB
19
-38 dB
3
-38.588 dB
-19 dB
29
-18.883 dB
20
-42 dB
2
-42.110 dB
-20 dB
25
-20.172 dB
21
-48 dB
1
-48.131 dB
-21 dB
23
-20.896 dB
22
Mute
0
Mute
-22 dB
20
-22.110 dB
23
-23 dB
18
-23.025 dB
24
-24 dB
16
-24.048 dB
25
-25 dB
14
-25.208 dB
26
-26 dB
13
-25.852 dB
27
-27dB
11
-27.303 dB
28
-28 dB
10
-28.131 dB
29
-29 dB
9
-29.046 dB
30
-30 dB
8
-30.069 dB
31
-31 dB
7
-31.229 dB
32
-33 dB
6
-32.568 dB
33
-34 dB
5
-34.151 dB
34
-36 dB
4
-36.090 dB
35
-39 dB
3
-38.588 dB
36
-42 dB
2
-42.110 dB
37
-48 dB
1
-48.131 dB
38
Mute
0
Mute
Legend:
Note 1:
Calculated Attenuation Value Color Code: Black → Above Target Value; Red → Below Target Value
Desired Attenuation Value Color Code: Magenta → Skipped Desired Attenuation Value(s).
Attenuation values do not include errors from digital potentiometer errors, such as Full-Scale Error or
Zero-Scale Error.
 2013-2015 Microchip Technology Inc.
DS20005207B-page 65
MCP41HVX1
8.11
Design Considerations
8.11.2
LAYOUT CONSIDERATIONS
In the design of a system with the MCP41HVX1
devices, the following considerations should be taken
into account:
In the design of a system with the MCP41HVX1
devices, the following layout considerations should be
taken into account:
• Power Supply Considerations
• Layout Considerations
• Noise
• PCB Area Requirements
• Power Dissipation
8.11.1
POWER SUPPLY
CONSIDERATIONS
8.11.2.1
The typical application will require a bypass capacitor
in order to filter high-frequency noise, which can be
induced onto the power supply’s traces. The bypass
capacitor helps to minimize the effect of these noise
sources on signal integrity. Figure 8-12 illustrates an
appropriate bypass strategy.
In this example, the recommended bypass capacitor
value is 0.1 µF. This capacitor should be placed as close
(within 4 mm) to the device power pin (VL) as possible.
The power source supplying these devices should be
as clean as possible. If the application circuit has
separate digital and analog power supplies, V+ and Vshould reside on the analog plane.
VDD
0.1 µF
VL
V+
Noise
Inductively-coupled AC transients and digital switching
noise can degrade the input and output signal integrity,
potentially masking the MCP41HVX1’s performance.
Careful board layout minimizes these effects and
increases the Signal-to-Noise Ratio (SNR). Multi-layer
boards utilizing a low-inductance ground plane,
isolated inputs, isolated outputs and proper decoupling
are critical to achieving the performance that the
silicon is capable of providing. Particularly harsh
environments may require shielding of critical signals.
If low noise is desired, breadboards and wire-wrapped
boards are not recommended.
8.11.2.2
PCB Area Requirements
In some applications, PCB area is a criteria for device
selection. Table 8-4 shows the package dimensions
and area for the different package options. The table
also shows the relative area factor compared to the
smallest area. The VQFN package is the suggested
package for space critical applications.
PACKAGE FOOTPRINT (1)
TABLE 8-4:
Package
B
DGND
FIGURE 8-12:
Connections.
DS20005207B-page 66
SCK
14
TSSOP
ST
5.10
6.40
32.64
1.31
20
VQFN
MQ
5.00
5.00
25.00
1
Dimensions
(mm)
Code
X
Y
Note 1: Does not include recommended land
pattern dimensions.
CS
V-
Relative Area
W
Type
Area (mm2)
SDI
SDO
MCP41HVXX
A
Package Footprint
Pins
V-
PIC® Microcontroller
0.1 µF
0.1 µF
VSS
Typical Microcontroller
 2013-2015 Microchip Technology Inc.
MCP41HVX1
8.11.3
RESISTOR TEMPERATURE
COEFFICIENT
Characterization curves of the resistor temperature
coefficient (Tempco) are shown in the device
characterization graphs.
These curves show that the resistor network is
designed to correct for the change in resistance as
temperature increases. This technique reduces the
end-to-end change in RAB resistance.
8.11.3.1
Power Dissipation
The power dissipation of the high-voltage digital potentiometer will most likely be determined by the power
dissipation through the resistor networks.
Table 8-5 shows the power dissipation through the resistor ladder (RAB) when Terminal A = +18V and Terminal
B = -18V. This is not the worst case power dissipation
based on the 25 mA terminal current specification.
Table 8-6 shows the worst-case current (per resistor
network), which is independent of the RAB value).
 2013-2015 Microchip Technology Inc.
TABLE 8-5:
RAB POWER DISSIPATION
RAB Resistance ()
Max.
| VA | + |VB |
= (V)
Power
(mW)(1)
Typical
Min.
5,000
4,000
6,000
36
324
10,000
8,000
12,000
36
162
50,000
40,000
60,000
36
32.4
100,000
80,000
120,000
36
16.2
Note 1: Power = V × I = V2/RAB(MIN).
TABLE 8-6:
RAB ()
(Typical)
RBW POWER DISSIPATION
| VW | + |VB | = (V)
IBW
(mA)(2)
Power
(mW)(1)
5,000
36
25
900
10,000
36
12.5
450
50,000
36
6.5
234
100,000
36
6.5
234
Note 1:
2:
Power = V × I.
See Electrical Specifications (max IW).
DS20005207B-page 67
MCP41HVX1
9.0
DEVICE OPTIONS
9.1
Standard Options
9.1.1
9.2
Custom options can be made available.
9.2.1
POR/BOR WIPER SETTING
The default wiper setting (mid-scale) is indicated to the
customer in three digit suffixes: -202, -502, -103 and
-503. Table 9-1 indicates the device’s default settings.
DEFAULT POR/BOR WIPER
SETTING SELECTION
Typical
RAB
Value
Package
Code
TABLE 9-1:
5.0 k
-502
Mid-Scale
10.0 k
-103
Mid-Scale
50.0 k
-503
Mid-Scale
100.0 k
-104
Mid-Scale
Default
Device
Wiper
POR Wiper
Resolution Code
Setting
DS20005207B-page 68
Custom Options
8-bit
7Fh
7-bit
3Fh
8-bit
7Fh
7-bit
3Fh
8-bit
7Fh
7-bit
3Fh
8-bit
7Fh
7-bit
3Fh
CUSTOM WIPER VALUE ON
POR/BOR EVENT
Customers can specify a custom wiper setting via the
NSCAR process.
Note 1: Non-Recurring
Engineering
(NRE)
charges and minimum ordering requirements apply for custom orders. Please
contact Microchip sales for additional
information.
2: A custom device will be assigned custom
device marking.
 2013-2015 Microchip Technology Inc.
MCP41HVX1
10.0
DEVELOPMENT SUPPORT
10.1
Development Tools
10.2
Technical Documentation
Several additional technical documents are available to
assist you in your design and development. These
technical documents include Application Notes,
Technical Briefs, and Design Guides. Table 10-2 shows
some of these documents.
Several development tools are available to assist in your
design and evaluation of the MCP41HVX1 devices. The
currently available tools are shown in Table 10-1.
Figure 10-1 shows how the TSSOP20EV bond-out
PCB can be populated to easily evaluate the
MCP41HVX1 devices. Evaluations can use the
PICkit™ Serial Analyzer to control the position of the
volatile wiper and state of the TCON register.
Figure 10-2 shows how the SOIC14EV bond-out PCB
can be populated to evaluate the MCP41HVX1
devices. The use of the PICkit Serial Analyzer would
require blue wire since the header H1 is not compatibly
connected.
These boards may be purchased directly from the
Microchip web site at www.microchip.com.
TABLE 10-1:
DEVELOPMENT TOOLS
Board Name
Part #
20-pin TSSOP and SSOP Evaluation Board
TSSOP20EV
14-pin SOIC/TSSOP/DIP Evaluation Board
SOIC14EV
TABLE 10-2:
Can easily interface to PICkit Serial Analyzer
(Order #: DV164122)
TECHNICAL DOCUMENTATION
Application
Note Number
TB3073
Comment
Title
Literature #
Implementing a 10-bit Digital Potentiometer with an 8-bit Digital Potentiometer
DS93073
AN1316
Using Digital Potentiometers for Programmable Amplifier Gain
DS01316
AN1080
Understanding Digital Potentiometers Resistor Variations
DS01080
AN737
Using Digital Potentiometers to Design Low-Pass Adjustable Filters
DS00737
AN692
Using a Digital Potentiometer to Optimize a Precision Single Supply Photo Detect
DS00692
AN691
Optimizing the Digital Potentiometer in Precision Circuits
DS00691
AN219
Comparing Digital Potentiometers to Mechanical Potentiometers
DS00219
—
Digital Potentiometer Design Guide
DS22017
—
Signal Chain Design Guide
DS21825
—
Analog Solutions for Automotive Applications Design Guide
DS01005
 2013-2015 Microchip Technology Inc.
DS20005207B-page 69
MCP41HVX1
MCP41HVX1-xxxE/ST
installed in U3 (bottom 14 pins of TSSOP-20 footprint)
Connected to
Digital Ground
(DGND) Plane
Connected to
Digital Power (VL) Plane
1.0 µF
VL
0
V+
SCK
41HVX1
P0A
CS
SDI
P0W
P0B
0
SDO
WLAT
SHDN
V0
0
P0A pin shorted
(jumpered) to
V+ pin
0
Through-hole Test
Point (Orange)
Wiper 0
P0B pin shorted
(jumpered) to
V- pin
DGND
NC
Four blue wire jumpers to connect
PICkit™ Serial interface (SPI) to device pins
FIGURE 10-1:
DS20005207B-page 70
1x6 male header, with 90° right angle
Digital Potentiometer Evaluation Board Circuit Using TSSOP20EV.
 2013-2015 Microchip Technology Inc.
1.0 µF
P0W
VDGND
NC
SHDN
0
0
FIGURE 10-2:
0
WLAT
SDO
SDI
P0B
MCP41HVX1
CS
P0A
SCK
VL
0
V+
MCP41HVX1
Digital Potentiometer Evaluation Board Circuit Using SOIC14EV.
 2013-2015 Microchip Technology Inc.
DS20005207B-page 71
MCP41HVX1
NOTES:
DS20005207B-page 72
 2013-2015 Microchip Technology Inc.
MCP41HVX1
11.0
PACKAGING INFORMATION
11.1
Package Marking Information
Example
14-Lead TSSOP (4.4 mm)
XXXXXXXX
YYWW
NNN
41H51502
E524
256
Part Number
Code
Part Number
Code
MCP41HV51-502E/ST
MCP41HV51-103E/ST
41H51502
MCP41HV31-502E/ST
41H31502
41H51103
MCP41HV31-103E/ST
41H31103
MCP41HV51-503E/ST
MCP41HV51-104E/ST
41H51503
MCP41HV31-503E/ST
41H31503
41H51104
MCP41HV31-104E/ST
41H31104
20-Lead VQFN (5x5x0.9 mm)
Example
PIN 1
PIN 1
41HV31
502E/MQ
e3
1524256
Part Number
Code
Part Number
Code
MCP41HV51-502E/MQ
502E/MQ
MCP41HV31-502E/MQ
502E/MQ
MCP41HV51-103E/MQ
103E/MQ
MCP41HV31-103E/MQ
103E/MQ
MCP41HV51-503E/MQ
503E/MQ
MCP41HV31-503E/MQ
503E/MQ
MCP41HV51-104E/MQ
104E/MQ
MCP41HV31-104E/MQ
104E/MQ
Legend: XX...X
Y
YY
WW
NNN
e3
*
Note:
Customer-specific information
Year code (last digit of calendar year)
Year code (last 2 digits of calendar year)
Week code (week of January 1 is week ‘01’)
Alphanumeric traceability code
RoHS Compliant JEDEC designator for Matte Tin (Sn)
This package is RoHS Compliant. The RoHS Compliant
JEDEC designator ( e3 ) can be found on the outer packaging
for this package.
In the event the full Microchip part number cannot be marked on one line, it will
be carried over to the next line, thus limiting the number of available
characters for customer-specific information.
 2013-2015 Microchip Technology Inc.
DS20005207B-page 73
MCP41HVX1
Note:
For the most current package drawings, please see the Microchip Packaging Specification located at
http://www.microchip.com/packaging
DS20005207B-page 74
 2013-2015 Microchip Technology Inc.
MCP41HVX1
Note:
For the most current package drawings, please see the Microchip Packaging Specification located at
http://www.microchip.com/packaging
 2013-2015 Microchip Technology Inc.
DS20005207B-page 75
MCP41HVX1
Note:
For the most current package drawings, please see the Microchip Packaging Specification located at
http://www.microchip.com/packaging
DS20005207B-page 76
 2013-2015 Microchip Technology Inc.
MCP41HVX1
20-Lead Plastic Quad Flat, No Lead Package (MQ) – 5x5x1.0 mm Body [VQFN]
With 0.40 mm Contact Length
Note:
For the most current package drawings, please see the Microchip Packaging Specification located at
http://www.microchip.com/packaging
D
A
B
N
NOTE 1
1
2
E
(DATUM B)
(DATUM A)
2X
0.20 C
2X
TOP VIEW
0.20 C
0.10 C
C
SEATING
PLANE
A1
A
20X
(A3)
0.08 C
SIDE VIEW
0.10
C A B
D2
0.10
C A B
E2
2
1
NOTE 1
K
N
L
e
BOTTOM VIEW
20X b
0.10
0.05
C A B
C
Microchip Technology Drawing C04-139C (MQ) Sheet 1 of 2
 2013-2015 Microchip Technology Inc.
DS20005207B-page 77
MCP41HVX1
20-Lead Plastic Quad Flat, No Lead Package (MQ) – 5x5x1.0 mm Body [VQFN]
With 0.40 mm Contact Length
Note:
For the most current package drawings, please see the Microchip Packaging Specification located at
http://www.microchip.com/packaging
Units
Dimension Limits
Number of Terminals
N
e
Pitch
Overall Height
A
Standoff
A1
(A3)
Contact Thickness
Overall Length
D
Exposed Pad Length
D2
Overall Width
E
Exposed Pad Width
E2
Contact Width
b
Contact Length
L
Contact-to-Exposed Pad
K
MIN
0.80
0.00
3.15
3.15
0.25
0.35
0.20
MILLIMETERS
NOM
20
0.65 BSC
0.90
0.02
0.20 REF
5.00 BSC
3.25
5.00 BSC
3.25
0.30
0.40
-
MAX
1.00
0.05
3.35
3.35
0.35
0.45
-
Notes:
1. Pin 1 visual index feature may vary, but must be located within the hatched area.
2. Package is saw singulated
3. Dimensioning and tolerancing per ASME Y14.5M
BSC: Basic Dimension. Theoretically exact value shown without tolerances.
REF: Reference Dimension, usually without tolerance, for information purposes only.
Microchip Technology Drawing C04-139C (MQ) Sheet 2 of 2
DS20005207B-page 78
 2013-2015 Microchip Technology Inc.
MCP41HVX1
20-Lead Plastic Quad Flat, No Lead Package (MQ) – 5x5x1.0 mm Body [VQFN]
With 0.40 mm Contact Length
Note:
For the most current package drawings, please see the Microchip Packaging Specification located at
http://www.microchip.com/packaging
C1
X2
EV
20
1
C2
ØV
2
Y2
G
EV
Y1
E
X1
SILK SCREEN
RECOMMENDED LAND PATTERN
Units
Dimension Limits
Contact Pitch
E
W2
Optional Center Pad Width
Optional Center Pad Length
T2
Contact Pad Spacing
C1
C2
Contact Pad Spacing
Contact Pad Width (X20)
X1
Contact Pad Length (X20)
Y1
Distance Between Pads
G
Thermal Via Diameter
V
Thermal Via Pitch
EV
MIN
MILLIMETERS
NOM
0.65 BSC
MAX
3.35
3.35
4.50
4.50
0.40
0.55
0.20
0.30
1.00
Notes:
1. Dimensioning and tolerancing per ASME Y14.5M
BSC: Basic Dimension. Theoretically exact value shown without tolerances.
2. For best soldering results, thermal vias, if used, should be filled or tented to avoid solder loss during
reflow process
Microchip Technology Drawing C04-2139B (MQ)
 2013-2015 Microchip Technology Inc.
DS20005207B-page 79
MCP41HVX1
NOTES:
DS20005207B-page 80
 2013-2015 Microchip Technology Inc.
MCP41HVX1
APPENDIX A:
REVISION HISTORY
Revision B (June 2015)
• Test limits in Section 1.0 “Electrical Characteristics” were corrected. The following specifications were updated:
- Full-Scale Error
- Zero-Scale Error
- Potentiometer Differential Nonlinearity(10,
17
) (see Appendix B.13)
- Rheostat Integral Nonlinearity(12,13,14,17)
(see Appendix B.5)
- Rheostat Differential Nonlinearity
(12,13,14,17) (see Appendix B.5)
Note:
APPENDIX B:
This appendix discusses the terminology used in this
document and describes how a parameter is measured.
B.1
Potentiometer (Voltage Divider)
The potentiometer configuration is when all three
terminals of the device are tied to different nodes in the
circuit. This allows the potentiometer to output a
voltage proportional to the input voltage. This
configuration is sometimes called voltage divider
mode. The potentiometer is used to provide a variable
voltage by adjusting the wiper position between the two
endpoints as shown in Figure B-1. Reversing the
polarity of the A and B terminals will not affect
operation.
Devices tested after the product marking
Date Code of June 30, 2015 are tested to
these new limits.
• Corrected the packaging diagram for the VQFN
package. The 5 x 5 mm VQFN package is specified, but the 4 x 4 mm QFN package information
was shown.
• Updated Device Features table to include
MCP45HVX1 devices.
• Added Section 8.10 “Implementing Log Steps
with a Linear Digital Potentiometer”.
Revision A (May 2013)
• Original Release of this Document.
TERMINOLOGY
V1
A
V3
W
B
V2
FIGURE B-1:
POTENTIOMETER
CONFIGURATION.
The temperature coefficient of the RAB resistors is
minimal by design. In this configuration, the resistors all
change uniformly, so minimal variation should be seen.
B.2
Rheostat (Variable Resistor)
The rheostat configuration is when two of the three digital potentiometer’s terminals are used as a resistive
element in the circuit. With Terminal W (wiper) and
either Terminal A or Terminal B, a variable resistor is
created. The resistance will depend on the tap setting
of the wiper (and the wiper’s resistance). The
resistance is controlled by changing the wiper setting.
Figure B-2 shows the two possible resistors that can be
used. Reversing the polarity of the A and B terminals
will not affect operation.
A
W
RAW or
RBW
B
Resistor
FIGURE B-2:
RHEOSTAT
CONFIGURATION.
 2013-2015 Microchip Technology Inc.
DS20005207A-page 81
MCP41HVX1
B.3
Resolution
EQUATION B-2:
The resolution is the number of wiper output states that
divide the full-scale range. For the 8-bit digital
potentiometer, the resolution is 28, meaning the digital
potentiometer wiper code ranges from 0 to 255.
B.4
RS CALCULATION
R AB
RS  Ideal  = ---------------N
2 –1
or
 VA – VB 
------------------------I AB
-------------------------N
2 –1
Measured
 VW  @FS  – VW  @ZS  
--------------------------------------------------------------I AB
RS  Measured  = --------------------------------------------------------------N
2 –1
Where:
2N - 1
Where:
VW = Voltage on Terminal W pin
The resistance step size (RS) equates to one LSb of the
resistor ladder. Equation B-1 shows the calculation for
the step resistance (RS).
Ideal
 VW – VA 
R W  Measured  = --------------------------IWB
VA = Voltage on Terminal A pin
Step Resistance (RS)
EQUATION B-1:
= 255 (MCP41HV51/61)
IWB = Measured current through W and B pins
The wiper resistance in potentiometer-generated
voltage divider applications is not a significant source
of error (it does not effect the output voltage seen on
the W pin).
The wiper resistance in rheostat applications can
create significant nonlinearity as the wiper is moved
toward zero scale (00h). The lower the nominal
resistance, the greater the possible error.
B.6
RZS Resistance
The analog switch between the resistor ladder and the
Terminal B pin introduces a resistance, which we call
the Zero-Scale resistance (RZS). Equation B-3 shows
how to calculate this resistance.
EQUATION B-3:
VB = Voltage on Terminal B pin
IAB = Measured Current through
A and B pins
VW(@FS) = Measured Voltage on W pin at
Full-Scale code (FFh or 7Fh)
Where:
VW(@ZS) = Voltage on Terminal W pin at
Zero-Scale wiper code
VB = Voltage on Terminal B pin
IAB = Measured Current through
A and B pins
VW(@ZS) = Measured Voltage on W pin at
Zero-Scale code (00h)
Wiper Resistance
Wiper resistance is the series resistance of the analog
switch that connects the selected resistor ladder node
to the wiper terminal common signal (see Figure 5-1).
A value in the volatile wiper register selects which
analog switch to close, connecting the W terminal to
the selected node of the resistor ladder.
The resistance is dependent on the voltages on the
analog switch source, gate, and drain nodes, as well as
the device’s wiper code, temperature, and the current
through the switch. As the device voltage decreases,
the wiper resistance increases.
The wiper resistance is measured by forcing a current
through the W and B terminals (IWB) and measuring the
voltage on the W and A terminals (VW and VA).
Equation B-2 shows how to calculate this resistance.
DS20005207A-page 82
RZS CALCULATION
 VW  @ZS  – VB 
R ZS  Measured  = -------------------------------------------IAB
= 127 (MCP41HV31/41)
VA = Voltage on Terminal A pin
B.5
RW CALCULATION
B.7
RFS Resistance
The analog switch between the resistor ladder and the
Terminal A pin introduces a resistance, which we call
the Full-Scale resistance (RFS). Equation B-4 shows
how to calculate this resistance.
EQUATION B-4:
RFS CALCULATION
 VA – V W  @FS  
RFS  Measured  = -------------------------------------------IAB
Where:
VA = Voltage on Terminal A pin
VW(@FS) = Voltage on Terminal W pin
at Full-Scale wiper code
IAB = Measured Current through
A and B pins
 2013-2015 Microchip Technology Inc.
MCP41HVX1
B.8
Least Significant Bit (LSb)
This is the difference between two successive codes
(either in resistance or voltage). For a given output
range it is divided by the resolution of the device
(Equation B-5).
EQUATION B-5:
B.9
Monotonic Operation
Monotonic operation means that the device’s output
(resistance (RBW) or voltage (VW)) increases with every
one code step (LSb) increment of the wiper register.
VS64
0x40
LSb CALCULATION
VS63
In Resistance
RAB
-------------LSb  Ideal  =
N
2 –1
In Voltage
VA – VB
-----------------N
2 –1
Measured
Wiper Code
0x3F
Ideal
 VW  @FS  – V W  @ZS  
--------------------------------------------------------------I
V W  @FS  – V W  @ZS 
AB
LSb  Measured  = --------------------------------------------------------------- ---------------------------------------------------------N
N
2 –1
2 –1
VAB = Measured Voltage between A and B
pins
FIGURE B-3:
THEORETICAL VW
OUTPUT VS. CODE (MONOTONIC
OPERATION).
IAB = Measured Current through A and B
pins
RS63
0x3F
VW(@FS) = Measured Voltage on W pin at
Full-Scale code (FFh or 7Fh)
RS62
0x3E
Digital Input Code
VW(@ZS) = Measured Voltage on W pin at
Zero-Scale code (00h)
VS0
VW (@ tap)
n=?
VW = VSn + VZS(@ Tap 0)
n=0
Voltage (VW ~= VOUT)
= 127 (MCP41HV31/41)
VB = Voltage on Terminal B pin
VS1
0x02
0x00
= 255 (MCP41HV51/61)
VA = Voltage on Terminal A pin
VS3
0x03
0x01
Where:
2N - 1
0x3E
0x3D
RS3
0x03
RS1
0x02
0x01
0x00
RS0
RW
(@ tap)
n=?
RBW =
RSn + RW(@ Tap n)
n=0
Resistance (RBW)
FIGURE B-4:
THEORETICAL RBW
OUTPUT VS. CODE (MONOTONIC
OPERATION).
 2013-2015 Microchip Technology Inc.
DS20005207A-page 83
MCP41HVX1
B.10
Full-Scale Error (EFS)
B.11
The Full-Scale Error (see Figure B-5) is the error of
the VW pin relative to the expected VW voltage
(theoretical) for the maximum device wiper register
code (code FFh for 8-bit and code 7Fh for 7-bit), see
Equation B-6. The error is dependent on the resistive
load on the VOUT pin (and where that load is tied to,
such as VSS or VDD). For loads (to VSS) greater than
specified, the full-scale error will be greater.
The error is determined by the theoretical voltage step
size to give an error in LSb.
Zero-Scale Error (EZS)
The Zero-Scale Error (see Figure B-6) is the difference
between the ideal and measured VOUT voltage with the
Wiper register code equal to 00h (Equation B-7). The
error is dependent on the resistive load on the VOUT pin
(and where that load is tied to, such as VSS or VDD). For
loads (to VDD) greater than specified, the zero-scale
error will be greater.
The error is determined by the theoretical voltage step
size to give an error in LSb.
Note:
Note:
Analog switch leakage increases with
temperature. This leakage increases
substantially at higher temperatures
(> ~100°C).
As analog switch leakage increases, the
full-scale output value decreases, which
increases the full-scale error.
EQUATION B-6:
FULL-SCALE ERROR
Analog switch leakage increases with
temperature. This leakage increases
substantially at higher temperatures
(> ~100°C).
As analog switch leakage increases the
zero-scale output value decreases, which
decreases the zero-scale error.
EQUATION B-7:
ZERO SCALE ERROR
VW  @ZS 
E ZS = ------------------------------------V LSb  IDEAL 
VW  @FS  – VA
E FS = --------------------------------------VLSb  IDEAL 
Where:
Where:
EFS
=
Expressed in LSb
VW@FS)
=
The VW voltage when the wiper
register code is at full-scale
VIDEAL(@FS)
=
The ideal output voltage when the
wiper register code is at full-scale
VLSb(IDEAL)
=
The theoretical voltage step size
EFS
=
Expressed in LSb
VW@ZS)
=
the VW voltage when the wiper
register code is at zero-scale
VLSb(IDEAL)
=
the theoretical voltage step size
VW
VA
VFS
VA
VFS
VW
Actual
Transfer
Function
VZS
VB
Full-Scale
Error (EFS)
Ideal Transfer
Function
0
Full-Scale
Wiper Code
FIGURE B-5:
EXAMPLE.
DS20005207A-page 84
VZS
VB
0
Zero-Scale
Error (EZS)
FIGURE B-6:
EXAMPLE.
Actual
Transfer
Function
Ideal Transfer
Function
Full-Scale
Wiper Code
ZERO-SCALE ERROR
FULL-SCALE ERROR
 2013-2015 Microchip Technology Inc.
MCP41HVX1
B.12
Integral Nonlinearity (P-INL)
Potentiometer Configuration
The Potentiometer Integral nonlinearity (P-INL) error is
the maximum deviation of an actual VW transfer
function from an ideal transfer function (straight line).
In the MCP41HVX1, P-INL is calculated using the zeroscale and full-scale wiper code end points. P-INL is
expressed in LSb. P-INL is also called relative
accuracy. Equation B-8 shows how to calculate the PINL error in LSb, and Figure B-7 shows an example of
P-INL accuracy.
Positive P-INL means higher VW voltage than ideal.
Negative P-INL means lower VW voltage than ideal.
Note:
Analog switch leakage increases with
temperature. This leakage increases
substantially at higher temperatures
(> ~100°C).
As analog switch leakage increases, the
wiper output voltage (VW) decreases,
which affects the INL Error.
EQUATION B-8:
P-INL ERROR
 VW  @Code  –  VLSb  Measured   Code  
E INL = ----------------------------------------------------------------------------------------------------------------VLSb  Measured 
Where:
INL = Expressed in LSb
Code = Wiper Register Value
VW(@Code) = The measured VW output voltage
with a given Wiper register code
VLSb = For Ideal:
VAB /Resolution
For Measured:
(VW(@FS) - VW(@ZS))/255
B.13
Differential Nonlinearity (P-DNL)
Potentiometer Configuration
The Potentiometer Differential nonlinearity (P-DNL)
error (see Figure B-8) is the measure of VW step size
between codes. The ideal step size between codes is
1 LSb. A P-DNL error of zero would imply that every
code is exactly 1 LSb wide. If the P-DNL error is less
than 1 LSb, the Digital Potentiometer guarantees
monotonic output and no missing codes. The P-DNL
error between any two adjacent codes is calculated in
Equation B-9.
P-DNL error is the measure of variations in code widths
from the ideal code width.
Note:
Analog switch leakage increases with
temperature. This leakage increases
substantially at higher temperatures
(> ~100°C).
As analog switch leakage increases, the
wiper output voltage (VW) decreases,
which affects the DNL Error.
EQUATION B-9:
E
DNL
P-DNL ERROR
 V W  code = n + 1  – V W  code = n  – V LSb  Measured  
= ----------------------------------------------------------------------------------------------------------------------------------------------------V LSb  Measured 
Where:
DNL = Expressed in LSb
VW(Code = n) = The measured VW output voltage
with a given Wiper register code
VLSb = For Ideal:
VAB /Resolution
For Measured:
(VW(@FS) - VW(@ZS))/# of RS
INL < 0
111
110
111
Actual
transfer
function
110
101
101
Wiper
Code
Actual
transfer
function
Wiper 100
Code
011
100
011
Ideal transfer
function
010
010
Wide code, > 1 LSb
001
001
000
000
INL < 0
VW Output Voltage
FIGURE B-7:
Ideal transfer
function
P-INL ACCURACY.
 2013-2015 Microchip Technology Inc.
Narrow code < 1 LSb
VW Output Voltage
FIGURE B-8:
P-DNL ACCURACY.
DS20005207A-page 85
MCP41HVX1
B.14
Integral Nonlinearity (R-INL)
Rheostat Configuration
The Rheostat Integral nonlinearity (R-INL) error is the
maximum deviation of an actual RBW transfer function
from an ideal transfer function (straight line).
In the MCP41HVX1, INL is calculated using the ZeroScale and Full-Scale wiper code end points. R-INL is
expressed in LSb. R-INL is also called relative
accuracy. Equation B-10 shows how to calculate the RINL error in LSb and Figure B-9 shows an example of
R-INL accuracy.
Positive R-INL means higher VOUT voltage than ideal.
Negative R-INL means lower VOUT voltage than ideal.
EQUATION B-10:
R-INL ERROR
 R BW  @code  – R BW  Ideal  
EINL = ----------------------------------------------------------------------------R LSb  Ideal 
Where:
B.15
Differential Nonlinearity (R-DNL)
Rheostat Configuration
The Rheostat Differential nonlinearity (R-DNL) error
(see Figure B-10) is the measure of RBW step size
between codes in actual transfer function. The ideal
step size between codes is 1 LSb. A R-DNL error of
zero would imply that every code is exactly 1 LSb wide.
If the R-DNL error is less than 1 LSb, the RBW Resistance guarantees monotonic output and no missing
codes. The R-DNL error between any two adjacent
codes is calculated in Equation B-11.
R-DNL error is the measure of variations in code widths
from the ideal code width. A R-DNL error of zero would
imply that every code is exactly 1 LSb wide.
EQUATION B-11:
R-DNL ERROR
  VOUT  code = n + 1  – V OUT  code = n   – VLSb  Measured  
EDNL = -------------------------------------------------------------------------------------------------------------------------------------------------------------------------V
LSB  Measured 
Where:
INL = Expressed in LSb
RBW(Code = n) = The measured RBW resistance
with a given wiper register code
RLSb = For Ideal:
RAB /Resolution
For Measured:
RBW(@FS) /# of RS
DNL = Expressed in LSb
RBW(Code = n) = The measured RBW resistance
with a given wiper register code
RLSb = For Ideal:
RAB /Resolution
For Measured:
RBW(@FS) /# of RS
INL < 0
111
110
Actual
transfer
function
111
110
101
Wiper
Code
101
100
011
010
Ideal transfer
function
Actual
transfer
function
Wiper 100
Code
011
010
001
001
000
000
INL < 0
RBW Resistance
FIGURE B-9:
DS20005207A-page 86
R-INL ACCURACY.
Ideal transfer
function
Wide code, > 1 LSb
Narrow code < 1 LSb
RBW Resistance
FIGURE B-10:
R-DNL ACCURACY.
 2013-2015 Microchip Technology Inc.
MCP41HVX1
B.16
Total Unadjusted Error (ET)
The Total Unadjusted Error (ET) is the difference
between the ideal and measured VW voltage.
Typically, calibration of the output voltage is
implemented to improve system performance.
The error in bits is determined by the theoretical voltage
step size to give an error in LSb.
Equation B-12 shows the Total Unadjusted Error
calculation.
Note:
Analog switch leakage increases with
temperature. This leakage increases
substantially at higher temperatures
(> ~100°C).
As analog switch leakage increases, the
wiper output voltage (VW) decreases,
which affects the total Unadjusted Error.
EQUATION B-12:
TOTAL UNADJUSTED
ERROR CALCULATION
 V W_Actual  @code  – VW_Ideal  @Code  
E T = -----------------------------------------------------------------------------------------------------------V LSb  Ideal 
Where:
ET = Expressed in LSb
VW_Actual(@code) = The measured W pin output
voltage at the specified code
VW_Ideal(@code) = The calculated W pin output
voltage at the specified code
(code × VLSb(Ideal))
VLSb(Ideal) = VAB/# RS
8-bit = VAB/255
7-bit = VAB/127
B.17
Settling Time
The settling time is the time delay required for the VW
voltage to settle into its new output value. This time is
measured from the start of code transition to when the
VW voltage is within the specified accuracy. It is related
to the RC characteristics of the resistor ladder and
wiper switches.
In the MCP41HVX1, the settling time is a measure of
the time delay until the VW voltage reaches within 0.5
LSb of its final value, when the volatile wiper register
changes from zero-scale to full-scale (or full-scale to
zero-scale).
 2013-2015 Microchip Technology Inc.
B.18
Major-Code Transition Glitch
Major-code transition glitch is the impulse energy
injected into the Wiper pin when the code in the Wiper
register changes state. It is normally specified as the
area of the glitch in nV-Sec, and is measured when the
digital code is changed by 1 LSb at the major carry transition (Example: 01111111 to 10000000, or
10000000 to 01111111).
B.19
Digital Feedthrough
The Digital feedthrough is the glitch that appears at the
analog output caused by coupling from the digital input
pins of the device. The area of the glitch is expressed
in nV-Sec, and is measured with a full-scale change
(Example: all 0s to all 1s and vice versa) on the digital
input pins. The digital feedthrough is measured when
the digital potentiometer is not being written to the output register.
B.20
Power-Supply Sensitivity (PSS)
PSS indicates how the output (VW or RBW) of the digital
potentiometer is affected by changes in the supply voltage. PSS is the ratio of the change in VW to a change
in VDD for mid-scale output of the digital potentiometer.
The VW is measured while the VDD is varied from 5.5V
to 2.7V as a step, and expressed in %/%, which is the
% change of the VW output voltage with respect to the
% change of the VDD voltage.
EQUATION B-13:
PSS CALCULATION
 V W  @5.5V  – V W  @27V  
---------------------------------------------------------------------V W  @5.5V 
PSS = ---------------------------------------------------------------------- 5.5V – 2.7V 
--------------------------------5.5V
Where:
PSS = Expressed in %/%
VW(@5.5V) = The measured VW output voltage
with VDD = 5.5V
VW(@2.7V) = The measured VW output voltage
with VDD = 2.7V
B.21
Power-Supply Rejection Ratio
(PSRR)
PSRR indicates how the output of the digital potentiometer is affected by changes in the supply voltage. PSRR
is the ratio of the change in VW to a change in VDD for
full-scale output of the digital potentiometer. The VW is
measured while the VDD is varied ±10% (VA and VB
voltages held constant), and expressed in dB or µV/V.
DS20005207A-page 87
MCP41HVX1
B.22
Ratiometric Temperature
Coefficient
The ratiometric temperature coefficient quantifies the
error in the ratio RAW/RWB due to temperature drift.
This is typically the critical error when using a digital
potentiometer in a voltage divider configuration.
B.23
Absolute Temperature Coefficient
The absolute temperature coefficient quantifies the
error in the end-to-end resistance (Nominal resistance
RAB) due to temperature drift. This is typically the
critical error when using the device in an adjustable
resistor configuration.
Characterization curves of the resistor temperature
coefficient (Tempco) are shown in Section 2.0 “Typical Performance Curves”.
B.24
-3 dB Bandwidth
This is the frequency of the signal at the A terminal that
causes the voltage at the W pin to be -3 dB from its
expected value, based on its wiper code. The expected
value is determined by the static voltage value on the A
Terminal and the wiper-code value. The output decreases
due to the RC characteristics of the resistor network.
B.25
Resistor Noise Density (eN_WB)
This is the random noise generated by the device’s
internal resistances. It is specified as a spectral density
(voltage per square root Hertz).
DS20005207A-page 88
 2013-2015 Microchip Technology Inc.
MCP41HVX1
PRODUCT IDENTIFICATION SYSTEM
To order or obtain information, e.g., on pricing or delivery, refer to the factory or the listed sales office.
PART NO.
Device
XXX
X
/XX
Resistance Temperature Package
Version
Range
Device:
MCP41HV31:
Single Potentiometer (7-bit) with
SPI Interface
MCP41HV31T: Single Potentiometer (7-bit) with
SPI Interface (Tape and Reel)
MCP41HV51: Single Potentiometer (8-bit) with
SPI Interface
MCP41HV51T: Single Potentiometer (8-bit) with
SPI Interface (Tape and Reel)
Examples:
a)
b)
c)
d)
a)
b)
Resistance
Version:
502 = 5 k
103 = 10 k
503 = 50 k
104 = 100 k
Temperature
Range:
E
Package:
ST = 14-Lead Plastic Thin Shrink Small Outline,
4.4 mm Body
MQ = 20-Lead Plastic Quad Flat, No Lead
Package, 5 x 5 x 0.9 mm Body
c)
d)
MCP41HV51T-502E/ST
5 k, 8-bit, 14-LD TSSOP.
MCP41HV51T-103E/ST
10 k, 8-bit, 14-LD TSSOP.
MCP41HV31T-503E/ST
50 k, 7-bit, 14-LD TSSOP.
MCP41HV31T-104E/MQ
100 k, 7-bit, 20-LD VQFN (5x5).
MCP41HV51T-502E/MQ
5 k, 8-bit, 20-LD VQFN (5x5).
MCP41HV51T-103E/MQ
10 k, 8-bit, 20-LD VQFN (5x5).
MCP41HV31T-503E/MQ
50 k, 7-bit, 20-LD VQFN (5x5).
MCP41HV31T-104E/MQ
100 k, 7-bit, 20-LD VQFN (5x5).
= -40°C to +125°C
 2013-2015 Microchip Technology Inc.
DS20005207B-page 89
MCP41HVX1
NOTES:
DS20005207B-page 90
 2013-2015 Microchip Technology Inc.
Note the following details of the code protection feature on Microchip devices:
•
Microchip products meet the specification contained in their particular Microchip Data Sheet.
•
Microchip believes that its family of products is one of the most secure families of its kind on the market today, when used in the
intended manner and under normal conditions.
•
There are dishonest and possibly illegal methods used to breach the code protection feature. All of these methods, to our
knowledge, require using the Microchip products in a manner outside the operating specifications contained in Microchip’s Data
Sheets. Most likely, the person doing so is engaged in theft of intellectual property.
•
Microchip is willing to work with the customer who is concerned about the integrity of their code.
•
Neither Microchip nor any other semiconductor manufacturer can guarantee the security of their code. Code protection does not
mean that we are guaranteeing the product as “unbreakable.”
Code protection is constantly evolving. We at Microchip are committed to continuously improving the code protection features of our
products. Attempts to break Microchip’s code protection feature may be a violation of the Digital Millennium Copyright Act. If such acts
allow unauthorized access to your software or other copyrighted work, you may have a right to sue for relief under that Act.
Information contained in this publication regarding device
applications and the like is provided only for your convenience
and may be superseded by updates. It is your responsibility to
ensure that your application meets with your specifications.
MICROCHIP MAKES NO REPRESENTATIONS OR
WARRANTIES OF ANY KIND WHETHER EXPRESS OR
IMPLIED, WRITTEN OR ORAL, STATUTORY OR
OTHERWISE, RELATED TO THE INFORMATION,
INCLUDING BUT NOT LIMITED TO ITS CONDITION,
QUALITY, PERFORMANCE, MERCHANTABILITY OR
FITNESS FOR PURPOSE. Microchip disclaims all liability
arising from this information and its use. Use of Microchip
devices in life support and/or safety applications is entirely at
the buyer’s risk, and the buyer agrees to defend, indemnify and
hold harmless Microchip from any and all damages, claims,
suits, or expenses resulting from such use. No licenses are
conveyed, implicitly or otherwise, under any Microchip
intellectual property rights.
Trademarks
The Microchip name and logo, the Microchip logo, dsPIC,
FlashFlex, flexPWR, JukeBlox, KEELOQ, KEELOQ logo, Kleer,
LANCheck, MediaLB, MOST, MOST logo, MPLAB,
OptoLyzer, PIC, PICSTART, PIC32 logo, RightTouch, SpyNIC,
SST, SST Logo, SuperFlash and UNI/O are registered
trademarks of Microchip Technology Incorporated in the
U.S.A. and other countries.
The Embedded Control Solutions Company and mTouch are
registered trademarks of Microchip Technology Incorporated
in the U.S.A.
Analog-for-the-Digital Age, BodyCom, chipKIT, chipKIT logo,
CodeGuard, dsPICDEM, dsPICDEM.net, ECAN, In-Circuit
Serial Programming, ICSP, Inter-Chip Connectivity, KleerNet,
KleerNet logo, MiWi, MPASM, MPF, MPLAB Certified logo,
MPLIB, MPLINK, MultiTRAK, NetDetach, Omniscient Code
Generation, PICDEM, PICDEM.net, PICkit, PICtail,
RightTouch logo, REAL ICE, SQI, Serial Quad I/O, Total
Endurance, TSHARC, USBCheck, VariSense, ViewSpan,
WiperLock, Wireless DNA, and ZENA are trademarks of
Microchip Technology Incorporated in the U.S.A. and other
countries.
SQTP is a service mark of Microchip Technology Incorporated
in the U.S.A.
Silicon Storage Technology is a registered trademark of
Microchip Technology Inc. in other countries.
GestIC is a registered trademarks of Microchip Technology
Germany II GmbH & Co. KG, a subsidiary of Microchip
Technology Inc., in other countries.
All other trademarks mentioned herein are property of their
respective companies.
© 2013-2015, Microchip Technology Incorporated, Printed in
the U.S.A., All Rights Reserved.
ISBN: 978-1-63277-544-3
QUALITY MANAGEMENT SYSTEM
CERTIFIED BY DNV
== ISO/TS 16949 ==
 2013-2015 Microchip Technology Inc.
Microchip received ISO/TS-16949:2009 certification for its worldwide
headquarters, design and wafer fabrication facilities in Chandler and
Tempe, Arizona; Gresham, Oregon and design centers in California
and India. The Company’s quality system processes and procedures
are for its PIC® MCUs and dsPIC® DSCs, KEELOQ® code hopping
devices, Serial EEPROMs, microperipherals, nonvolatile memory and
analog products. In addition, Microchip’s quality system for the design
and manufacture of development systems is ISO 9001:2000 certified.
DS20005207B-page 91
Worldwide Sales and Service
AMERICAS
ASIA/PACIFIC
ASIA/PACIFIC
EUROPE
Corporate Office
2355 West Chandler Blvd.
Chandler, AZ 85224-6199
Tel: 480-792-7200
Fax: 480-792-7277
Technical Support:
http://www.microchip.com/
support
Web Address:
www.microchip.com
Asia Pacific Office
Suites 3707-14, 37th Floor
Tower 6, The Gateway
Harbour City, Kowloon
Hong Kong
Tel: 852-2943-5100
Fax: 852-2401-3431
China - Xiamen
Tel: 86-592-2388138
Fax: 86-592-2388130
China - Zhuhai
Tel: 86-756-3210040
Fax: 86-756-3210049
Austria - Wels
Tel: 43-7242-2244-39
Fax: 43-7242-2244-393
Denmark - Copenhagen
Tel: 45-4450-2828
Fax: 45-4485-2829
India - Bangalore
Tel: 91-80-3090-4444
Fax: 91-80-3090-4123
France - Paris
Tel: 33-1-69-53-63-20
Fax: 33-1-69-30-90-79
India - New Delhi
Tel: 91-11-4160-8631
Fax: 91-11-4160-8632
Germany - Dusseldorf
Tel: 49-2129-3766400
Atlanta
Duluth, GA
Tel: 678-957-9614
Fax: 678-957-1455
Austin, TX
Tel: 512-257-3370
Boston
Westborough, MA
Tel: 774-760-0087
Fax: 774-760-0088
Chicago
Itasca, IL
Tel: 630-285-0071
Fax: 630-285-0075
Cleveland
Independence, OH
Tel: 216-447-0464
Fax: 216-447-0643
Australia - Sydney
Tel: 61-2-9868-6733
Fax: 61-2-9868-6755
China - Beijing
Tel: 86-10-8569-7000
Fax: 86-10-8528-2104
China - Chengdu
Tel: 86-28-8665-5511
Fax: 86-28-8665-7889
China - Chongqing
Tel: 86-23-8980-9588
Fax: 86-23-8980-9500
China - Dongguan
Tel: 86-769-8702-9880
China - Hangzhou
Tel: 86-571-8792-8115
Fax: 86-571-8792-8116
Germany - Munich
Tel: 49-89-627-144-0
Fax: 49-89-627-144-44
India - Pune
Tel: 91-20-3019-1500
Germany - Pforzheim
Tel: 49-7231-424750
Japan - Osaka
Tel: 81-6-6152-7160
Fax: 81-6-6152-9310
Italy - Milan
Tel: 39-0331-742611
Fax: 39-0331-466781
Japan - Tokyo
Tel: 81-3-6880- 3770
Fax: 81-3-6880-3771
Italy - Venice
Tel: 39-049-7625286
Korea - Daegu
Tel: 82-53-744-4301
Fax: 82-53-744-4302
Netherlands - Drunen
Tel: 31-416-690399
Fax: 31-416-690340
China - Hong Kong SAR
Tel: 852-2943-5100
Fax: 852-2401-3431
Korea - Seoul
Tel: 82-2-554-7200
Fax: 82-2-558-5932 or
82-2-558-5934
China - Nanjing
Tel: 86-25-8473-2460
Fax: 86-25-8473-2470
Malaysia - Kuala Lumpur
Tel: 60-3-6201-9857
Fax: 60-3-6201-9859
Detroit
Novi, MI
Tel: 248-848-4000
China - Qingdao
Tel: 86-532-8502-7355
Fax: 86-532-8502-7205
Malaysia - Penang
Tel: 60-4-227-8870
Fax: 60-4-227-4068
Houston, TX
Tel: 281-894-5983
China - Shanghai
Tel: 86-21-5407-5533
Fax: 86-21-5407-5066
Philippines - Manila
Tel: 63-2-634-9065
Fax: 63-2-634-9069
China - Shenyang
Tel: 86-24-2334-2829
Fax: 86-24-2334-2393
Singapore
Tel: 65-6334-8870
Fax: 65-6334-8850
China - Shenzhen
Tel: 86-755-8864-2200
Fax: 86-755-8203-1760
Taiwan - Hsin Chu
Tel: 886-3-5778-366
Fax: 886-3-5770-955
China - Wuhan
Tel: 86-27-5980-5300
Fax: 86-27-5980-5118
Taiwan - Kaohsiung
Tel: 886-7-213-7828
Dallas
Addison, TX
Tel: 972-818-7423
Fax: 972-818-2924
Indianapolis
Noblesville, IN
Tel: 317-773-8323
Fax: 317-773-5453
Los Angeles
Mission Viejo, CA
Tel: 949-462-9523
Fax: 949-462-9608
New York, NY
Tel: 631-435-6000
San Jose, CA
Tel: 408-735-9110
Canada - Toronto
Tel: 905-673-0699
Fax: 905-673-6509
China - Xian
Tel: 86-29-8833-7252
Fax: 86-29-8833-7256
Poland - Warsaw
Tel: 48-22-3325737
Spain - Madrid
Tel: 34-91-708-08-90
Fax: 34-91-708-08-91
Sweden - Stockholm
Tel: 46-8-5090-4654
UK - Wokingham
Tel: 44-118-921-5800
Fax: 44-118-921-5820
Taiwan - Taipei
Tel: 886-2-2508-8600
Fax: 886-2-2508-0102
Thailand - Bangkok
Tel: 66-2-694-1351
Fax: 66-2-694-1350
01/27/15
DS20005207B-page 92
 2013-2015 Microchip Technology Inc.