Two-Pole Low-Pass Filter Datasheet LPF2 V 4.10
001-13572 Rev. *J
Two-Pole Low-pass Filter
Copyright © 2002-2013 Cypress Semiconductor Corporation. All Rights Reserved.
PSoC® Blocks
Resources
Digital
Analog CT
API Memory (Bytes)
Analog SC
Flash
Pins (per
External I/O)
RAM
CY8C29/27/24xxx, CY8C23x33, CY8CLED04/08/16, CY8CLED0xD, CY8CLED0xG, CY8CTMG120,
CY8CTMA120, CY8C28x45, CY8CPLC20, CY8CLED16P01, CY8C28x43, CY8C28x52
0
0
2
120
0
1
Features and Overview
User-programmable corner frequency and damping ratio
Corner frequency ranging from 20 Hz to 180 kHz
Automated design for Bessel, Butterworth, and Chebychev filters
User-selected oversample ratio (OSR), ratio of sample frequency to corner frequency
Built-in polarity control
Built-in modulator for use in full-wave detection and frequency translation
The LPF2 User Module uses two switched-capacitor blocks to implement a general-purpose second order
low-pass filter. Corner frequency and damping ratio are functions of the ratios of programmable on-chip
capacitors and clock frequency; no external components are required. Selects your filter characteristics
and clock frequency; capacitor and clock divider values are automatically calculated in the design tool
(wizard). Multiple low-pass filters can be cascaded or combined with band-pass filters to achieve more
complex transfer functions.
The filter has programmable gain. The filter is biased at AGND, selected in the Global Resources, and the
gain is referred to this level.
Cypress Semiconductor Corporation
Document Number: 001-13572 Rev. *J
•
198 Champion Court
•
San Jose, CA 95134-1709
•
408-943-2600
Revised May 15, 2013
Two-Pole Low-pass Filter
DC and AC Electrical Characteristics
The following values are indicative of expected performance and based on initial characterization data.
Unless otherwise specified, all limits guaranteed for TJ =+25 °C, VDD = 5.0 V, Power HIGH, Opamp bias
LOW, output referenced to Analog Ground = 2*VBandGap. The default filter for tests is 2.0 kHz
Butterworth, sample rate = 200 kHz (column clock = 800 kHz), gain = 1, and modulation OFF.
Table 1.
Symbol
DC Electrical Performance Characteristics, VDD = 5.0 V
Description
Conditions
Min
Typ
Max
Units
Vos
Input offset voltage
Default filter
–
–
140
mV
TCVos
Temp coefficient, input
offset voltage
Default filter
–
–
50
µV/°C
G-Err1
Error, Gain = 1
Default filter, C1 = 1
–
–
1
%
G_Err2
Error, Gain = 2
Default filter, C1 = 2
–
–
1.5
G_Err4
Error, Gain = 4
Default filter, C1 = 4
–
–
2.0
Idd_LL
Operating current
Pwr = L, Bias = L
–
–
–
Idd_LH
Pwr = L, Bias = H
–
–
–
Idd_ML
Pwr = M, Bias = L
–
–
–
Idd_MH
Pwr = M, Bias = H
–
–
–
Idd_HL
Pwr = H, Bias = L
–
–
–
Idd_HH
Pwr = H, Bias = H
–
–
–
Table 2.
Symbol
DC Electrical Performance Characteristics, VDD = 3.3 V
Description
Conditions
Min
Typ
Max
Units
Vos
Input offset voltage
Default filter
–
–
140
mV
TCVos
Temp coefficient, input
offset voltage
Default filter
–
–
50
µV/°C
G-Err1
Error, Gain = 1
Default filter, C1 = 1
–
–
1
%
G_Err2
Error, Gain = 2
Default filter, C1 = 2
–
–
1.5
G_Err4
Error, Gain = 4
Default filter, C1 = 4
–
–
2.0
Idd_LL
Operating current
Pwr = L, Bias = L
–
–
–
Idd_LH
Pwr = L, Bias = H
–
–
–
Idd_ML
Pwr = M, Bias = L
–
–
–
Idd_MH
Pwr = M, Bias = H
–
–
–
Idd_HL
Pwr = H, Bias = L
–
–
–
Idd_HH
Pwr = H, Bias = H
–
–
–
Document Number: 001-13572 Rev. *J
Page 2 of 29
Two-Pole Low-pass Filter
Table 3.
AC Electrical Performance Characteristics, VDD = 5.0 V (Default filter for tests is 2.0 kHz Butterworth,
sample rate = 200 kHz (column clock = 800 kHz), gain = 1, modulation OFF)
Symbol
Description
Conditions
Min
Typ
Max
Units
fCORNER
Corner frequency
Min, OSR = 100 Max,
OSR = 6
1
–
160
kHz
fERR100
–3 dB cut-off freq
error
2.0 kHz Butterworth,
OSR = 100[1]
–
–
1.0
%[3]
fERR10
–3 dB Cut-off freq
error
2.0 kHz Butterworth,
OSR = 10[2]
–
–
5.0
%[3]
RespErr
Peak damping
2.0 kHz Chebychev,
OSR = 100[4]
–
–
0.5
dB
2.0 kHz Chebychev,
OSR = 10[5]
–
–
1.5
THD
Total harmonic
distortion
fIN = 400 Hz, VIN = 4 Vpp
–
–63
–
dB
Vn_spect
Inband noise
Pwr = L, Bias = L
–
–
–
nV/rtHz
Pwr = H, Bias = H
–
–
–
Pwr = L, Bias = L
–
–
–
Pwr = H, Bias = L
–
–
–
Vn_int
Integrated noise
nV/rtHz
Note
1. fERR100 Butterworth design C1=1, C2=1, C3=4, C4=23, CA=32,CB=32
2. fERR10 Butterworth design C1=8, C2=8, C3=17, C4=19, CA=32,CB=16
3. Scaled to 24.0 MHz clock, verify sysclk +/- 0.25%
4. fERR100 Chebychev design C1=1, C2=1, C3=4, C4=16, CA=32, CB=32
5. fERR10 Chebychev design C1=9, C2=9, C3=16, C4=16, CA=32, CB=16
Document Number: 001-13572 Rev. *J
Page 3 of 29
Two-Pole Low-pass Filter
Table 4.
AC Electrical Performance Characteristics, VDD = 3.3V (Default filter for tests is 2.0 kHz Butterworth,
sample rate = 200 kHz (column clock = 800 kHz), gain = 1, modulation OFF)
Symbol
Description
Conditions
Min
Typ
Max
Units
fCORNER
Corner frequency
Min, OSR = 100 Max,
OSR = 6
1
–
80
kHz
fERR100
–3 dB cut-off freq
error
2.0 kHz Butterworth,
OSR = 100[1]
–
–
1.0
%[3]
fERR10
–3 dB Cut-off freq
error
2.0 kHz Butterworth,
OSR = 10[2]
–
–
5.0
%[3]
RespErr
Peak damping
2.0 kHz Chebychev,
OSR = 100[4]
–
–
0.5
dB
2.0 kHz Chebychev,
OSR=10[5]
–
–
1.5
THD
Total harmonic
distortion
fIN = 400 Hz, VIN= 4 Vpp
–
–
–
dB
Vn_spect
Inband noise
Pwr = L, Bias = L
–
–
–
nV/rtHz
Pwr = H, Bias = H
–
–
–
Pwr = L, Bias = L
–
–
–
Pwr = H, Bias = L
–
–
–
Vn_int
Integrated noise
Document Number: 001-13572 Rev. *J
nV/rtHz
Page 4 of 29
Two-Pole Low-pass Filter
Typical Operating Characteristics
Figure 1.
Butterworth Pass-band Response,
OSR = 10
Figure 2.
1.0 dB Chebychev Pass-Band, Response
OSR = 10
Figure 3.
Butterworth Pass-band Response,
OSR = 100
Figure 4.
Chebychev Pass-band Response,
OSR = 100
Figure 5.
Butterworth Response, OSR = 10
Figure 6.
Chebychev Response, OSR = 10
Document Number: 001-13572 Rev. *J
Page 5 of 29
Two-Pole Low-pass Filter
Figure 7.
Butterworth Response, OSR = 100
Figure 8.
Figure 9.
Typical Noise Spectrum, 10 kHz Filter,
OSR = 100
Figure 10. Typical Noise Spectrum, 100 kHz Filter,
OSR = 20
Figure 11. Noise Spectrum at Power Setting
Document Number: 001-13572 Rev. *J
Chebychev Response, OSR = 100
Figure 12. Integrated Noise at Power Setting
Page 6 of 29
Two-Pole Low-pass Filter
Figure 13. Typical Vos vs Cap Value
fCORNER
kHz
Figure 14. Vos vs Temp
fSAMPLE
kHz
fCOLUMN
kHz
C1
C2
C3
C4
2.0
200
800
1
1
4
23
2.0
100
400
2
2
7
24
2.0
50
200
4
4
11
25
2.0
25
100
8
8
14
28
2.0
18
72
10
10
16
27
2.0
15
60
12
12
16
28
2.0
12
48
14
14
15
28
Filter Characteristic
In the frequency domain, a two-pole low-pass filter has the transfer function:
Equation 1
where:
Gain is the DC gain of the filter
d is the damping factor
ω0 is the natural frequency
ωn is the normalized corner frequency
s is the Laplace operator = 2*π*j*frequency
In-band performance is determined by the damping ratio and natural frequency. Bessel and Butterworth
filters are defined as having the –3.0-dB point at the natural frequency. A Chebychev filter has peaking or
ripple in the pass-band and gain equal to the DC value at the nominal corner frequency. Values for d and 0
are readily available in any filter design reference. Values for standard forms Bessel, Butterworth, 0.1-dB
Chebychev, and 1.0-dB Chebychev are built into the design tool. The filters have a roll-off of 12 dB for
each octave at frequencies up to half of the sample rate. The recommended OSR is between 6 and 100.
Document Number: 001-13572 Rev. *J
Page 7 of 29
Two-Pole Low-pass Filter
The filters can be cascaded to make more complex forms; see the LPF4 User Module datasheet and
AN67391 for examples.
The pass-band performance of standard filter types for a nominal corner frequency of 10 kHz is shown in
Figure 15. The 0.1-dB ripple Chebychev has a very flat pass-band, but less attenuation in the near out-ofband than other filters, as seen in Figure 16.
Figure 15. Low-pass Filter Pass-band Response
Figure 16. Low-pass Filter Response
Document Number: 001-13572 Rev. *J
Page 8 of 29
Two-Pole Low-pass Filter
The LPF2 switched capacitor filter does not follow the ideal form of the low-pass response. The switching
nature of the circuit results in a frequency dependent gain in the numerator of the equation. This reduces
the attenuation at higher frequencies, as shown by the example in Figure 17.
Figure 17. Switched Capacitor Filter Low-pass Response
Functional Description
An outline schematic of the LPF2 is shown in Figure 18. The filter uses two switched-capacitor blocks and
one or two column clocks (at the same frequency) depending on placement. Polarity and modulation
inputs are user-selectable.
Figure 18. Switched-Capacitor Filter Block diagram
Document Number: 001-13572 Rev. *J
Page 9 of 29
Two-Pole Low-pass Filter
A switched-capacitor filter is a time-sampled circuit. The transfer function is derived in the time (z) domain
and translated to the frequency (s=jw) domain using the bi-linear transform. The transfer function of the
LPF2 is:
Equation 2
where:
fS is the sample rate (= column clock/4)
C1 through CB values are calculated and entered by the user or calculated in the wizard and
automatically entered.
Mapping equation 2 to equation 1, yields a set of design equations:
Equation 3
Equation 4
Equation 5
Deriving the capacitor and clock values is a non-trivial task which also includes compensation for
frequency distortion due to the bi-linear transform used to translate the transfer function from the original
z-domain to the s-domain. An outline of the direct method is included in Appendix 2; the preferred method
is to use the automated design procedure in the wizard.
Document Number: 001-13572 Rev. *J
Page 10 of 29
Two-Pole Low-pass Filter
Operational Limits
The maximum corner frequency of the LPF2 is 180 kHz at high power and high bias, determined by
opamp slewing and settling. The corner frequency has been tested down to 10 Hz at OSR = 100. The
ultimate low corner frequency may be limited by noise at elevated temperature and aliasing of external
signals. The user is advised to characterize these parameters in his own system.
The standard form of a low-pass filter has a response which goes to zero at infinite frequency. The
switched-capacitor filter response in comparison is asymptotic to a finite value. This is a mathematical
result (not a circuit "problem") of the (s/2fS)^2 in the numerator of the transfer function. Solving equation 2
for a unity gain filter when frequency (s=2 f) is very large, the attenuation is:
Equation 6
For OSR = 50 (2-kHz filter, 100-kHz sample rate), this is -48 dB at the sample frequency, about 20 dB less
attenuation that a non-switched-capacitor filter measured at the same frequency. Near in-band
performance of the filter is affected by the higher level asymptote at low OSR, resulting in less roll-off. The
projected response of the wizard has been tested and demonstrated to agree closely with measured
results for a wide range of frequencies and filter types.
Power Setting
The operating current for the LPF2 is set in the user's code using the APIs provided. The operational
bandwidth of the switched-capacitor filter is determined by opamp slew rate and settling time. Slew rate
and settling time are a function of opamp gain-bandwidth (GBW), which is a function of the power setting.
High-power settings result in increased bandwidth. There are a total of six active power settings at VDD =
5.0 V.
Table 5.
Power Settings; Vdd = 5.0 V
Power
Bias
Gain*Bandwidth
PLBL
Low
Low
0.75 MHz
PLBH
Low
High
1.5 MHz
PMBL
Medium
Low
2.25 MHz
PMBH
Medium
High
3.1 MHz
PHBL
High
Low
4.5 MHz
PHBH
High
High
9.0 MHz
PHBH (power = high, bias = high) is not available at Vdd = 3.3 V. For reliable operation, the GBW should
be at least twice the filter corner frequency multiplied by the gain. Excess GBW increases noise and does
not improve filter performance.
Document Number: 001-13572 Rev. *J
Page 11 of 29
Two-Pole Low-pass Filter
The ability of the filter to accurately deliver a signal is determined by the slew rate, which is also a function
of power and bias setting.
Figure 19. Maximum Signal vs Frequency
The slew rate of the analog column output buffer is 0.65 V/usec. For operation at high power, this results in
a lower maximum signal limit than the filter itself.
Polarity and Modulator Functions
The input switched-capacitor block has a polarity control parameter. The polarity can also be controlled by
a clock external to the block to form a modulator, multiplying the input by +1 or –1 at the clock rate. This
generates signals at the sum of the carrier and input frequencies and at the difference of the carrier and
input frequencies. Modulation carriers should have 50% duty cycle to minimize even carrier harmonic
aliases. The modulation process introduces carrier/signal pairs at odd harmonics of the carrier at a level
proportional to 1/n where n is the number of the carrier harmonic. The filter is usually designed to
attenuate these out-of-band harmonics.
A common usage of the filter is to generate a sine wave from a digital square wave. This can be done by
feeding the filter through a direct Port 2[0..3] input or through a PGA. This results in a sine wave that is
dependent on supply voltage. In order to make the sine wave a fixed level independent of supply, the
digital signal can drive the modulator and the input to the filter can be routed to either RefHi or RefLo.
Document Number: 001-13572 Rev. *J
Page 12 of 29
Two-Pole Low-pass Filter
Another common usage of the modulator is a detector. The input fed to the modulator/filter is also fed to a
comparator; the digital comparator output drives the modulator input of the filter. The result is full-wave
detection. The example shows a 40 kHz input signal, 100% AM modulated at 4.0 kHz.
Figure 20. Full-Wave Detector Waveforms
The modulator can be used to mix frequencies up or down. The example in Figure 21 shows a 132-kHz
carrier mixed with a 120-kHz local oscillator driven to the modulator input resulting in a clean waveform at
the difference frequency of 12 kHz.
Figure 21. Modulator, Switched-Capacitor Filter Waveforms
Noise
Noise sources in the switched capacitor filter include the opamp, AGND buffer and capacitor array. The
sampling nature of the switched-capacitor filter results in copies of the noise being added at each multiple
of the sampling frequency up to the unity gain-bandwidth of the filter's opamps. The best method for
reducing aliased noise is to reduce the bandwidth of the opamp; this limits the number of aliased copies of
the noise. The example of Figure 16 show the noise spectrum level for a 1.75 kHz low-pass filter operating
at each of the six power levels Power = Low, Bias = Low (PLBL) through Power = High, Bias = High
(PHBH). The noise is integrated across the band to get a total RMS noise level, as shown in Figure 19.
Higher power levels have higher integrated noise.
Document Number: 001-13572 Rev. *J
Page 13 of 29
Two-Pole Low-pass Filter
Anti-aliasing
Because of the switching nature of the filter, signals around harmonics of the sample frequency can alias
into the pass-band. At the input, a simple single pole RC low-pass filter with corner frequency above the
LPF2's corner frequency and below the sample frequency can prevent this aliasing and reduce noise. At
the output (that is, analog column buffer output for going off-chip), a low-pass filter attenuates clock noise.
Placement
To implement a low-pass filter the user places an instance of the LPF2 in the analog array. A window
opens showing the four variations in topology and their allowed placements. Inputs to the filter can
connect to the A-Cap or B-Cap inputs of the switched-capacitor block; this allows a variety of nonoverlapping input selections. Placements are either vertical (LPF2VA and LPF2VB) in a single column or
horizontal (LPF2A and LPF2B) using two columns. When using horizontal topology filters, the columns
must have the same clock. A Design Rule Check (DRC) at build verifies that this is correctly done. Filters
using the modulator are allowed only when the input is on the A-Cap (LPF2A and LPF2VA). The filter may
be dragged from one allowed location to the next. Changes in topology are accessed by right-click on the
user module, then left-click on "Selection Options."
Figure 22. LPF2 Selection Options in PSoC Designer
Document Number: 001-13572 Rev. *J
Page 14 of 29
Two-Pole Low-pass Filter
Wizard/Design
The design objective is usually to meet response requirements and achieve the highest possible sample
rate for the best waveform fidelity and minimum aliasing of out-of-band signals. Other system
requirements such as shared clocks may determine available sample rates.
After placing the filter in the design, the user right-clicks on the user module and chooses "Filter Design
Wizard" from the menu. The user enters decimal values for gain, corner frequency, and sample rate
parameters and selects the filter type from Bessel, Butterworth, 0.1 dB Chebychev, and 1.0 dB Chebychev
in a pull-down list.
Figure 23. LPF2 Wizard in PSoC Designer
The nominal d and ω0 values are displayed along with the compensated values based on OSR. The
wizard does not directly calculate the capacitor values. Instead, it calculates d and ω0 for all combinations
of all capacitors and selects the set that gives the closest fit to the required compensated values. The
OSR, required column clock, divider value, OSR and capacitor values are displayed. The plot shows lines
for the nominal response of the filter, the expected response based on the available integer capacitor
values and the locations of the sample rate and sample rate/2. When the user selects OK, the capacitor
values are transferred to the parameter fields for the LPF2. The clock is separately set by the user and
may be one of the available system clocks (VC1 or VC2) or from the output of one of the digital blocks.
Note that for horizontally located filters, both columns must have the same clock source.
Capacitor values are integers. The calculation routine in the wizard finds the best fit, but there is no
guarantee that it is a perfect fit, especially for high OSR (>60) which uses small capacitor values (≤ 4).
Document Number: 001-13572 Rev. *J
Page 15 of 29
Two-Pole Low-pass Filter
Most of the time, the d and ω0 values will be within a few percent of the design point and the filter
response will track the nominal shape of the filter in the pass-band within 0.5 dB.
Parameters
Input
Inputs to the filter come from outputs of adjacent PSoC blocks, references and P2[0:3].
AnalogBus
The output of the filter can be connected to the analog output bus of the column containing the output
block.
CompBus
The output block's comparator output can be connected to comparator bus. This enables connection
to digital blocks or to an interrupt. Because the threshold for the internal comparator is fixed at AGND,
the comparator functions as a zero-crossing detector.
Capacitor Values C1, C2, C3, C4, CA, and CB
The values of these capacitor values and the column clock determine the frequency response of the
filter. C1, C2, C3, and C4 are integer values from 1 to 31. CA and CB are selectable as 16 or 32. You
can design the capacitor values by the direct numerical approach in Appendix 2 or, preferably, using
the design wizard that searches for the optimum set of capacitor values to meet your stated response
requirements.
Polarity
This selects the polarity of the output with respect to the input. Select Inverting to set the output to
invert the signal relative to AGND. When Gain is set greater than 1.0, the gain is relative to AGND.
When Gain is set to 1.0 and Polarity is set to Non-Inverting, the signal is not referenced to AGND. The
parameter applies only to topologies using the A-input.
Sample Frequency
The sample frequency is an essential part of the filter design but is not set in the LPF2 User Module.
The sample frequency is determined by the user to maintain the OSR in the range of 6 to 100. The
wizard specifies the divider ratio from the system clock based on the sample frequency. The user is
expected to instantiate a PWM or a counter at four times the sample frequency or use either VC1 or
VC2 clocks.
Modulator Clock
The modulator uses the polarity control in the A input of the input block (FLIN). This function is available only for topologies LPF2A and LPF2VA. It is not available for LPF2B and LPF2V because these
topologies use the B input of the input block, which does not have polarity control. There are eight
sources (including OFF) for the modulator:
None
GlobalOutEven_0
GlobalOutEven_1
Row_0_Broadcast
ComparatorBus_0
ComparatorBus_1
ComparatorBus_2
Document Number: 001-13572 Rev. *J
Page 16 of 29
Two-Pole Low-pass Filter
ComparatorBus_3
Mixers are constructed using clocks derived from the digital blocks routed through GlobaOutEven_0,
GlobalOutEven_1, and Row_0_Broadcast. A full-wave detector is typically constructed by using the
output of a PGA or BPF to drive the signal input of the filter, and using the output of a comparator or
the comparator output of the BPF to drive the modulator input of the LPF. The full-wave detector works
best when the BPF source and the LPF have the same clock frequency.
Applications Programming Interface (API)
The API routines are given as part of the user module to enable you to deal with the module at a higher
level. Entry points are provided to initialize the LPF2 User Module, change power settings, and disable the
user module. This section specifies the interface to each function together with related constants provided
by the “include” files.
Note
In this, as in all user module APIs, the values of the A and X register may be altered by calling an API
function. It is the responsibility of the calling function to preserve the values of A and X before the call if
those values are required after the call. This “registers are volatile” policy was selected for efficiency
reasons and has been in force since version 1.0 of PSoC Designer. The C compiler automatically takes
care of this requirement. Assembly language programmers must ensure their code observes the policy,
too. Though some user module API function may leave A and X unchanged, there is no guarantee they
will do so in the future.
For Large Memory Model devices, it is also the caller's responsibility to preserve any value in the
CUR_PP, IDX_PP, MVR_PP, and MVW_PP registers. Even though some of these registers may not be
modified now, there is no guarantee that will remain the case in future releases.
LPF2_Start
Description:
Performs all required initialization for this user module and sets the power level for the switched
capacitor PSoC blocks.
C Prototype:
void
LPF2_Start(BYTE bPowerSetting)
Assembly:
mov
A, bPowerSetting
lcall LPF2_Start
Parameters:
bPowerSetting: One byte that specifies the power level to both analog PSoC blocks. Following reset
and configuration, the PSoC blocks assigned to the instrumentation amplifier are powered down.
Symbolic names provided in C and assembly, and their associated values, are listed in the following
table:
Document Number: 001-13572 Rev. *J
Page 17 of 29
Two-Pole Low-pass Filter
Symbolic Name
Value
LPF2_OFF
0
LPF2_LOWPOWER
1
LPF2_MEDPOWER
2
LPF2_HIGHPOWER
3
Note
For proper performance, filters with corner frequencies above 40 kHz should (1) use
LPF2_HIGHPOWER and (2) set the global parameter “Opamp Bias” to High in the Global Parameters window.
Return Value:
None
Side Effects:
The A and X registers may be altered by this function.
LPF2_SetPower
Description:
Sets the power level for the switched capacitor PSoC blocks. May be used to turn the blocks in the
user module off and on.
C Prototype:
void
LPF2_SetPower(BYTE bPowerSetting)
Assembly:
mov
A, bPowerSetting
lcall LPF2_SetPower
Parameters:
bPowerSetting: Same as the bPowerSetting used for the Start entry point.
Return Value:
None
Side Effects:
The A and X registers may be altered by this function.
LPF2_SetCA, SetCB
Description:
Sets the value of the feedback capacitors in the user module FLIN block (CA) and FLOUT block (CB).
This allows on-the-fly modification of the low-pass filter transfer function.
C Prototype:
void
void
LPF2_SetCA(BYTE FEEDBACK_CONSTANT)
LPF2_SetCB(BYTE FEEDBACK_CONSTANT)
Document Number: 001-13572 Rev. *J
Page 18 of 29
Two-Pole Low-pass Filter
Assembly:
mov
A, FEEDBACK_CONSTANT
lcall LPF2_SetCA
; or, call
LPF2_SetCB
Parameters:
FEEDBACK_CONSTANT: One byte that specifies the size of the feedback capacitors CA or CB (see
the LPF2 Block Diagram). Symbolic names are given in the C and assembly include files; their associated values are listed in the following table:
Symbolic Name
Value
LPF2_FEEDBACK_16
0x00
LPF2_FEEDBACK_32
0x01
Return Values:
None
Side Effects:
The A and X registers may be altered by this function.
LPF2_SetC1, SetC2, SetC3, and SetC4
Description:
Sets the value of specific capacitors in the user module. This allows adjustment of gain by modifying
C1 and altering filter transfer characteristics by adjusting the other values.
C Prototype:
void
void
void
void
LPF2_SetC1(BYTE
LPF2_SetC2(BYTE
LPF2_SetC3(BYTE
LPF2_SetC4(BYTE
bCapValue)
bCapValue)
bCapValue)
bCapValue)
Assembly:
mov
A, bCapValue
lcall LPF2_SetC1
; or, call
LPF2_SetC2 (or SetC3 or SetC4)
Parameters:
CapValue: Integer value from 1 to 31 for C1, C2, C3, and C4 (see the LPF2 Block Diagram). Values
outside this range are truncated modulo 32.
Return Values:
None
Side Effects:
The A and X registers may be altered by this function.
Document Number: 001-13572 Rev. *J
Page 19 of 29
Two-Pole Low-pass Filter
LPF2_SetPolarity (A-Input Topology Filters Only)
Description:
Sets the polarity of the output signal by selecting whether to invert or not to invert the input signal on
FLIN. This allows on-the-fly modification of the low-pass filter output polarity. This function applies
only to A-input topology filters.
C Prototype:
void
LPF2_SetPolarity(BYTE FEEDBACK_CONSTANT)
Assembly:
mov
A, FEEDBACK_CONSTANT
lcall LPF2_SetPolarity
Parameters:
POLARITY_CONSTANT: One byte that specifies whether to invert or not to invert. Symbolic names
are given in the C and assembly include files; their associated values are listed in the following table:
Symbolic Name
Value
LPF2_POLARITY_INVERTING
0x00
LPF2_POLARITY_NON_INVERTING
0x01
Return Values:
None
Side Effects:
The A and X registers may be altered by this function.
LPF2_Stop
Description:
Powers off the user module.
C Prototype:
void
LPF2_Stop(void)
Assembly:
lcall
LPF2_Stop
Parameters:
None
Return Value:
None
Side Effects:
None.
Document Number: 001-13572 Rev. *J
Page 20 of 29
Two-Pole Low-pass Filter
Sample Firmware Source Code
The capacitor values for the filter are set by the user in the Wizard or by assigning values in the user
module's parameter window. In C, using the low-pass filter is as simple as using the Start API to begin
operation and calling the Stop API to complete operation.
//
// This sample shows how to create a Low Pass Filter with corner frequency 1kHz.
//
// OVERVIEW:
//
// The LPF2 input/output can be routed to any analog pin or adjacent analog block
// depending on placement.
// In this example the LPF input is routed to P0[5] and the output is routed to P0[3].
//
//The following changes need to be made to the default settings in the Device Editor:
//
//
1. Choose the LPF2VA MUM configuration of the LPF2 user module.
//
2. Place it onto ASC10 and ASD20 blocks.
//
3. Rename the User Module's instance name to LPF2.
//
4. Run the LPF2 Wizard from the context menu.
//
- Set the F corner (Hz) parameter to 1000
//
- Set the F sample (Hz) parameter to 100000
//
- Leave the Gain and Filter Type parameters by default
//
("0.0" and "Butterworth" respectively)
//
- Click the "OK" button
//
5. Set the LPF2's Input parameter to ACB00.
//
6. Set the LPF2's AnalogBus parameter to AnalogOutBus_0.
//
7. Set the LPF2's Polarity parameter to Non-Inverting.
//
8. Leave the rest of UM parameters by default.
//
9. Set the AnalogColumn_Clock_0 to VC2 (on the interconnect view)
//
10. Set the AnalogOutBuf_0 to Port_0_3 (on the interconnect view).
//
11. Place the PGA UM onto ACB00 block.
//
12. Rename User Module's instance name to PGA.
//
13. Set the PGA's Gain parameter to 1.000.
//
14. Set the PGA's Input parameter to AnalogColumn_InputMUX_0.
//
15. Set the PGA's Reference parameter to AGND.
//
16. Set the PGA's AnalogBus parameter to Disable.
//
17. Set the AnalogColumn_InputMux_0 to Port_0_5 (on the interconnect view)
//
// CONFIGURATION DETAILS:
//
//
1. The UM's instance names have to be shortened to LPF2 and PGA.
//
2. The Analog Column clock should be 400 kHz to get the 1kHz low Pass Filter
//
with Over Sampling Ration = 100.
//
//
//
1. Set the VC1=SysClk/N to 15
//
2. Set the VC2=VC1/N to 4
//
3. Set the A_Buff_Power to High
//
// USER MODULE PARAMETER SETTINGS:
//
// ------------------------------------------------------------------------------// UM
Parameter
Value
Comments
// -------------------------------------------------------------------------------
Document Number: 001-13572 Rev. *J
Page 21 of 29
Two-Pole Low-pass Filter
// LPF2
Name
LPF2
UM's instance name
//
C1
1
Set by Wizard
//
C2
1
Set by Wizard
//
C3
4
Set by Wizard
//
C4
22
Set by Wizard
//
CA
32
Set by Wizard
//
CB
32
Set by Wizard
//
Input
ACB00
//
AnalogBus
AnalogOutBus_0
//
ÑompBus
Disable
Default value
//
Polarity
Non-Inverting
//
Modulator Clock
None
Default value
//
// PGA
Name
PGA
UM's instance name
//
Gain
1.000
//
Input
AnalogColumn_InputMUX_0
//
Reference
AGND
//
AnalogBus
Disable
// ------------------------------------------------------------------------------/* Code begins here */
#include <m8c.h>
#include "PSoCAPI.h"
// part specific constants and macros
// PSoC API definitions for all User Modules
void main(void)
{
// M8C_EnableGInt ;
PGA_Start(PGA_HIGHPOWER);
LPF2_Start(LPF2_HIGHPOWER);
}
// Uncomment this line to enable Global Interrupts
// Turn on the PGA
// Turn on the LPF
The equivalent assembly language is:
;
; This sample shows how to create a Low Pass Filter with corner frequency 1kHz.
;
; OVERVIEW:
;
; The LPF2 input/output can be routed to any analog pin or adjacent analog block
; depending on placement.
; In this example the LPF input is routed to P0[5] and the output is routed to P0[3].
;
;The following changes need to be made to the default settings in the Device Editor:
;
;
1. Choose the LPF2VA MUM configuration of the LPF2 user module.
;
2. Place it onto ASC10 and ASD20 blocks.
;
3. Rename the User Module's instance name to LPF2.
;
4. Run the LPF2 Wizard from the context menu.
;
- Set the F corner (Hz) parameter to 1000
;
- Set the F sample (Hz) parameter to 100000
;
- Leave the Gain and Filter Type parameters by default
;
("0.0" and "Butterworth" respectively)
;
- Click the "OK" button
;
5. Set the LPF2's Input parameter to ACB00.
;
6. Set the LPF2's AnalogBus parameter to AnalogOutBus_0.
Document Number: 001-13572 Rev. *J
Page 22 of 29
Two-Pole Low-pass Filter
;
;
;
;
;
;
;
;
;
;
;
;
;
;
;
;
;
;
;
;
;
;
;
;
;
;
;
;
;
;
;
;
;
;
;
;
;
;
;
;
;
;
;
;
;
;
;
;
7. Set the LPF2's Polarity parameter to Non-Inverting.
8. Leave the rest of UM parameters by default.
9. Set the AnalogColumn_Clock_0 to VC2 (on the interconnect view)
10. Set the AnalogOutBuf_0 to Port_0_3 (on the interconnect view).
11. Place the PGA UM onto ACB00 block.
12. Rename User Module's instance name to PGA.
13. Set the PGA's Gain parameter to 1.000.
14. Set the PGA's Input parameter to AnalogColumn_InputMUX_0.
15. Set the PGA's Reference parameter to AGND.
16. Set the PGA's AnalogBus parameter to Disable.
17. Set the AnalogColumn_InputMux_0 to Port_0_5 (on the interconnect view)
CONFIGURATION DETAILS:
1. The UM's instance names have to be shortened to LPF2 and PGA.
2. The Analog Column clock should be 400 kHz to get the 1kHz low Pass Filter with
Over Sampling Ration = 100.
PROJECT SETTINGS:
1. Set the VC1=SysClk/N to 15
2. Set the VC2=VC1/N to 4
3. Set the A_Buff_Power to High
USER MODULE PARAMETER SETTINGS:
------------------------------------------------------------------------------UM
Parameter
Value
Comments
------------------------------------------------------------------------------LPF2
Name
LPF2
UM's instance name
C1
1
Set by Wizard
C2
1
Set by Wizard
C3
4
Set by Wizard
C4
22
Set by Wizard
CA
32
Set by Wizard
CB
32
Set by Wizard
Input
ACB00
AnalogBus
AnalogOutBus_0
ÑompBus
Disable
Default value
Polarity
Non-Inverting
Modulator Clock
None
Default value
PGA
Name
PGA
UM's instance name
Gain
1.000
Input
AnalogColumn_InputMUX_0
Reference
AGND
AnalogBus
Disable
-------------------------------------------------------------------------------
; Code begins here
include "m8c.inc"
include "memory.inc"
include "PSoCAPI.inc"
; part specific constants and macros
; Constants & macros for SMM/LMM and Compiler
; PSoC API definitions for all User Modules
Document Number: 001-13572 Rev. *J
Page 23 of 29
Two-Pole Low-pass Filter
export _main
_main:
; M8C_EnableGInt ; Uncomment this line to enable Global Interrupts
mov
A, PGA_HIGHPOWER
lcall PGA_Start
mov
A, LPF2_HIGHPOWER
lcall LPF2_Start
; Insert your main assembly code here.
.terminate:
jmp .terminate
Note
The design equations show that gain is proportional to the value of C1, but the corner frequency
and damping do not depend on it. After the transfer function is chosen, the LPF2_SetC1 API function may be used to implement a programmable-gain control.
Appendix 1: Registers
The topology and placement of the LPF2 User Module determine half of the bits in the configuration
registers for the blocks used. Bit fields that are independent of placement are indicated by fixed values in
the register tables. Of the variable bitfields, most are determined by selection of input and transfer function
design. Definitions of variable bitfields used in the register definitions follow at the end of this appendix.
Further details of registers may be found in the Technical Reference Manual.
Horizontal A-Input Topology
Table 6.
Block FLIN, ASCxxCRx
Bit
7
6
0
5
Polarity
4
3
CR0
CA
CR1
Input
CR2
0
0
0
0
0
CR3
0
0
1
0
Feedback
2
1
0
C1
C2
0
0
0
Power
AMD_CR0
Table 7.
AMD_CR0 for modulator in Column 0 or 2
Bit
7
0
6
5
4
AMOD select column 2
3
0
2
1
0
AMOD select column 0
AMD_CR1
Table 8.
AMD_CR1 for Modulator in Column 1 or 3
Bit
7
0
6
5
AMOD select column 3
Document Number: 001-13572 Rev. *J
4
3
0
2
1
0
AMOD select column 1
Page 24 of 29
Two-Pole Low-pass Filter
Horizontal B-Input Topology
Table 9.
Block FLIN, ASCxxCRx
Bit
7
6
1
4
3
CR0
CA
CR1
Feedback
CR2
0
0
0
0
0
CR3
0
0
1
0
Input
Table 10.
0
5
2
1
0
C2
C1
0
0
0
Power
Block FOUT, ASDxxCRx
Bit
7
6
0
5
0
4
CR0
CB
CR1
FBIN
CR2
AnalogBus CompBus
0
C4
CR3
0
1
0
2
1
0
C3
0
0
3
0
0
0
1
0
1
Power
Vertical A-Input Topology
Table 11.
Block FLIN, ASCxxCRx
Bit
7
6
Polarity
4
CR0
CA
CR1
Input
CR2
0
0
0
C4
CR3
0
0
1
0
Table 12.
0
5
3
2
1
0
C1
C2
Feedback
Power
Block FLOUT, ASDxxCRx
Bit
7
6
CR0
CB
0
CR1
FBIN
CR2
AnalogBus CompBus
CR3
0
5
0
0
4
3
2
1
0
C3
0
0
0
0
1
0
0
0
0
0
0
1
0
0
1
Power
AMD_CR0
Table 13.
AMD_CR0 for Modulator in Column 0 or 2
Bit
7
0
6
5
AMOD select column 2
4
3
0
2
1
0
AMOD select column 0
AMD_CR1
Document Number: 001-13572 Rev. *J
Page 25 of 29
Two-Pole Low-pass Filter
Table 14.
AMD_CR1 for Modulator in Column 1 or 3
Bit
7
0
6
5
4
AMOD select column 3
3
0
2
1
0
AMOD select column 1
Vertical B-Input Topology
Table 15.
Block FLIN, ASCxxCRx
Bit
7
6
1
4
3
CR0
CA
CR1
Feedback
CR2
0
0
0
0
0
CR3
0
0
1
0
Input
Table 16.
0
5
2
1
0
C2
C1
0
0
0
Power
Block FLOUT, ASDxxCRx
Bit
7
6
0
5
Polarity
4
CR0
CB
CR1
FBIN
CR2
AnalogBus CompBus
0
C4
CR3
0
1
0
2
1
0
C3
0
0
3
0
0
0
0
1
Power
1
Variable BitField Definitions
The following definitions apply to all preceding register definitions:
CA and CB set the FLIN and FLOUT feedback capacitors, respectively, to either 16 or 32 units (see the
LPF2 Block Diagram). CA and CB are configured in the Design Wizard or directly in the Device Editor.
C1, C2, C3, and C4 set the capacitors to integer values between 1 and 32. Values are calculated in the
filter Design Wizard or are set directly in the Device Editor using values calculated using the direct design
procedure in Appendix 2.
Input controls the multiplexer that selects the input signal.The user module "Input" parameter determines
the value of this bitfield. The value of the Input parameter can be set using the routing features of the
Device Editor or entered in the LPF2 User Module parameters window.
AnalogBus enables connection of the filter output to the analog bus. The value of the AnalogBus
parameter is manually configured using the routing pull-downs in the Device Editor or entered in the
parameters window.
CompBus enables connection of the filter output to the comparator bus. The value of the CompBus
parameter is manually configured using the routing pull-downs in the Device Editor or entered in the
parameters window.
Feedback is the C2 feedback connection, automatically determined by placement of the LPF2 User
Module in the Device Editor.
FBIN is the connection from the FLIN output to the FLOUT input, automatically determined by placement
of the LPF2 User Module in the Device Editor.
Document Number: 001-13572 Rev. *J
Page 26 of 29
Two-Pole Low-pass Filter
Polarity controls whether the output of the filter is inverted or non-inverted. This bit can be configured
directly using Device Editor. This option is available only to A-input topology filters.
Power controls the On/Off state of the PSoC block and the operating current setting. It is set initially by
calling the user module API function LPF2_Start and can be modified by calling the functions
LPF2_SetPower and LPF2_Stop.
AMOD select column x enable source for analog modulator in appropriate column when modulator
function is set.
Appendix 2: Direct Numerical Design
Direct calculation, derived from equations 3, 4, and 5 yields a set of capacitor values that work well for
over-sample ratio (OSR) greater than 30. For lower OSR, the user is advised to use the calculation in the
wizard, which uses a more complete optimization method.
Given fCORNER, d, and OSR (fSAMPLE/fCORNER).
Calculate pre-warped corner frequency and damping
Equation 7
Equation 8
Calculate C2:
Equation 9
Calculate C4:
Equation 10
If C4 > 31, decrement C2, recalculate C4:
Equation 11
Document Number: 001-13572 Rev. *J
Page 27 of 29
Two-Pole Low-pass Filter
Calculate C1 from Gain:
Equation 12
Set clock frequency (PWM or clock resource) at fCORNER * OSR * 4. If using a horizontal placement, verify
that both columns have the same clock source. Set capacitor values and clock source in parameter
windows, select input, connect output, enter start commands in code (C or assembler), build, and test.
Document Number: 001-13572 Rev. *J
Page 28 of 29
Two-Pole Low-pass Filter
Version History
Version
Originator
Description
2.5
DHA
Added Version History.
3.00
DHA
Redesigned the user module wizard and updated the user module datasheet.
4.00
SEG
Improved capacitor value calculation algorithm in user module wizard.
Reworked user module datasheet.
4.10
MYKZ
1. Users can now store printer settings in User Module Wizard.
2. Corrected algorithm for calculating the wizard's chart data in the nominal series.
3. Improved algorithm of capacitor calculation in User Module Wizard.
Note
PSoC Designer 5.1 introduces a Version History in all user module datasheets. This section documents high level descriptions of the differences between the current and previous user module versions.
Document Number: 001-13572 Rev. *J
Revised May 15, 2013
Page 29 of 29
Copyright © 2002-2013 Cypress Semiconductor Corporation. The information contained herein is subject to change without notice. Cypress Semiconductor Corporation assumes no responsibility
for the use of any circuitry other than circuitry embodied in a Cypress product. Nor does it convey or imply any license under patent or other rights. Cypress products are not warranted nor intended
to be used for medical, life support, life saving, critical control or safety applications, unless pursuant to an express written agreement with Cypress. Furthermore, Cypress does not authorize its
products for use as critical components in life-support systems where a malfunction or failure may reasonably be expected to result in significant injury to the user. The inclusion of Cypress products
in life-support systems application implies that the manufacturer assumes all risk of such use and in doing so indemnifies Cypress against all charges.
PSoC Designer™ and Programmable System-on-Chip™ are trademarks and PSoC® is a registered trademark of Cypress Semiconductor Corp. All other trademarks or registered trademarks
referenced herein are property of the respective corporations.
Any Source Code (software and/or firmware) is owned by Cypress Semiconductor Corporation (Cypress) and is protected by and subject to worldwide patent protection (United States and foreign),
United States copyright laws and international treaty provisions. Cypress hereby grants to licensee a personal, non-exclusive, non-transferable license to copy, use, modify, create derivative works
of, and compile the Cypress Source Code and derivative works for the sole purpose of creating custom software and or firmware in support of licensee product to be used only in conjunction with
a Cypress integrated circuit as specified in the applicable agreement. Any reproduction, modification, translation, compilation, or representation of this Source Code except as specified above is
prohibited without the express written permission of Cypress.
Disclaimer: CYPRESS MAKES NO WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, WITH REGARD TO THIS MATERIAL, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE. Cypress reserves the right to make changes without further notice to the materials described herein. Cypress does not
assume any liability arising out of the application or use of any product or circuit described herein. Cypress does not authorize its products for use as critical components in life-support systems
where a malfunction or failure may reasonably be expected to result in significant injury to the user. The inclusion of Cypress' product in a life-support systems application implies that the manufacturer
assumes all risk of such use and in doing so indemnifies Cypress against all charges.
Use may be limited by and subject to the applicable Cypress software license agreement.