Low Voltage Photoelectric Smoke Detector ASIC with Interconnect

BL5980
Low Voltage Photoelectric Smoke Detector ASIC with
Interconnect and Timer Mode
1.0. INTRODUCTION
1.1 Features
Low Quiescent Current Consumption
Two AA Battery Operation
Internal Power On Reset
Local Alarm Memory
Internal Low Battery detector
Interconnect up to 40 Detectors
9 minute Timer for Sensitivity Control
Temporal or Continuous Horn Pattern
Low Battery and Chamber Test
Internal Infrared Emitter Diode (IRED) driver
Adjustable IRED Drive current
Adjustable Hush Sensitivity
Available in SOP16
1.2 Pin Diagrams
www.belling.com.cn
Page- 1 -
V1.1
BL5980
1.3 Description
The BL5980 is a low power, low voltage photoelectric type smoke detector IC. With minimal external
components, this circuit will provide all the required features for a photoelectric-type smoke detector.
A check for a low battery condition is performed every 86 seconds, and chamber integrity is tested once
every 43 seconds, when in Standby. The temporal horn pattern supports the NFPA 72 emergency
evacuation signal.
The BL5980 incorporates a gain-selectable photo amplifier for use with an infrared emitter/detector pair.
An internal oscillator strobes power to the smoke detection circuitry every 10 seconds, to keep the standby
current to a minimum. If smoke is sensed, the detection rate is increased to verify an Alarm condition. A
high gain mode is available for push button chamber testing.
An internal 9 minute timer can be used for a Reduced Sensitivity mode.
An interconnect pin allows multiple detectors to be connected such that, when one unit alarms, all units will
sound.
the BL5980 was designed for use in smoke detectors that comply with Underwriters Laboratory
Specification UL217 and UL268.
1.4 Ordering Information
Order Number
Package Type
Marking
Packing
BL5980
SOP-16
BL5980
XXXXX
Reel Tape
Tube
www.belling.com.cn
Page2
V1.1
BL5980
1.5 Typical Application Circuit
vss
D1
L1
10uH
3V AA Battery
C1
10uF
C4
4.7uF
LX(16)
VBST(15)
Current Sense
RS
LATCH CLK
VDD(3)
R1 C2
100 1uF
VSS
Comparator
PWM
VSS(1)
High
frequence
OSC
Power for low
voltage module
VSS
Boost Comparator
Precision
Reference
BIAS
Low Battery
Comparator
VSS
VDD
Power on
reset
IRCAP(11)
Trimmed
Oscilator
C3
100uF
Programmable
IRED Current
IRP(6)
D2
IRED(2)
R7
100
Photo
Integrator
D5
GLED(9)
Smoke
Comparator
from precision
reference
DAC
Programmable
Limits
Control Logic
and Timming
VBST
R6
330
VSS
IRN(7)
D3
IRED
DRIVER
D4
R5
330
To other Units
IO(12)
VSS
RLED(8)
Interconnect
C6
33uF
VBST
Horn Driver
To programme
every module
Level shift
Push-to-TEST/HUSH
TEST(4)
VDD
Test
Mode
TEST2(5) Identify
HB(13)
Horn
HS(14)
44bit
44bit
Allocation …… MTP
and
Cell
Lifetime
Calibration
Data
REG
LTD Data
C5
1nF
MTP
W/R ctrl
R3
R4 200K
1.5M
FEED(10)
Note
1: C2 should be located as close as possible to the device power pins, and C1 should be located as close as possible to
VSS.
2: R3, R4 and C5 are typical values and may be adjusted to maximize sound pressure.
3: Schottky diode D1 must have a maximum peak current rating of at least 1.5A. For best results it should have forward
voltage specification of less than 0.5V at 1A, and low reverse leakage.
4: Inductor L1 must have a maximum peak current rating of at least 1.5A.
5: DC-DC converter in High Boost mode (nominal VBST = 9.6V) can draw current pulses of greater than 1A, and is
therefore very sensitive to series resistance. Critical components of this resistance are the inductor DC resistance, the
internal resistance of the battery and the resistance in the connections from the inductor to the battery, from the inductor to
the LX pin and from the VSS pin to the battery. In order to function properly under full load at VDD= 2V, the total of the
inductor and interconnect resistances should not exceed 0.3Ω. The internal battery resistance should be no more than 0.5Ω,
and a low ESR capacitor of 10 µF or more should be connected in parallel with the battery, to average the current draw over
the boost converter cycle.
www.belling.com.cn
Page3
V1.1
BL5980
2.0. ELECTRICAL CHARACTERISTICS
2.1 Absolute Maximum Ratings
Supply Voltage ....................................................VDD=5.5V; VBST =13V
Input Voltage Range Except FEED, TEST......... VIN = -0.3V to VDD +0.3V
FEED Input Voltage Range................................ VINFD =-10 to +22V
TEST Input Voltage Range ............................... VINTEST =-.3V to VBST+.3V
Input Current except FEED.................................. IIN = 10 mA
Continuous Operating Current (HS, HB, VBST).. IO= 40 mA
Continuous Operating Current (IRED) ................IOIR= 300 mA
Operating Temperature ......................................TA = -10 to +60°C
Storage Temperature .........................................TSTG = -55 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 these or any other conditions above those indicated in the operation listings of
this specification is not implied. Exposure to maximum rating conditions for extended periods may affect device reliability
2.2 DC Electrical Characteristics
DC Electrical Characteristics: Unless otherwise indicated, all parameters apply at TA = -10 to +60°C, VDD
= 3V, VBST = 4.2V, Typical Application (unless otherwise noted)(Note 1, Note 2, Note 3)
Test
Parameter
Symbol
Min
Typ
Max Units Conditions
Pin
Supply Voltage
VDD
3
2
5
V
Supply Current
IDD1
3
1
2
µA
Standby Boost
Current
IBST1
15
100
nA
IRCAP Supply
Current
IIRCAP
11
500
µA
During smoke check
VBST1
15
3
3.6
4.2
V
VBST2
15
8.5
9.6
10.7
V
6
-200
200
pA
7
-200
200
pA
IRP = VDD or VSS
50
µA
FEED = 22V; VBST = 9V
FEED = -10V; VBST =
10.7V
IINOP
www.belling.com.cn
Standby, Inputs low, No
loads, Boost Off, No
smoke check
Standby, Inputs low, No
loads, Boost Off, No
smoke check
IRCAP charging for
Smoke Check, GLED
operation
IOUT = 40 mA
No local alarm, RLED
Operation, IOUT = 40
mA, IO as an input
IRP = VDD or VSS
Boost Voltage
Input Leakage
Operating
IIHF
10
IILF
10
20
-50
-15
Page4
µA
V1.1
BL5980
Input Voltage
Low
VIL1
10
2.7
V
VIL2
12
800
mV
IO Hysteresis
VHYST1
12
IPD1
4,5
3
IPDIO1
12
20
IPDIO2
Input Pull Down
Current
Output Voltage
Low
Output High
Voltage
Output Current
IRED Current
Temperature
Coefficient
Low Battery
Alarm Voltage
www.belling.com.cn
150
FEED, VBST = 9V
No local alarm,IO as an
input
mV
30
µA
VIN = VDD
80
µA
VIN = VDD
12
140
µA
VIN = 15V
VOL1
13,14
500
mV
VOL2
8
300
mV
VOL3
9
300
mV
VOH1
13,14
8.5
IIOH1
12
-4
-5
mA
IIODMP
12
5
15
mA
IIRED50
2
45
50
IIRED100
2
90
10
V
100
55
110
mA
mA
IIRED150
2
135
150
165
mA
IIRED200
2
180
200
220
mA
0.5
%/℃
TCIRED
VLB1
3
2.05
2.1
2.15
V
VLB2
3
2.15
2.2
2.25
V
VLB3
3
2.25
2.3
2.35
V
VLB4
3
2.35
2.4
2.45
V
VLB5
3
2.45
2.5
2.55
V
VLB6
3
2.55
2.6
2.65
V
Page5
IOL = 16 mA, VBST = 9V
Alarm, VIO = 3V or VIO =
0V, VBST = 9V
At Conclusion of Local
Alarm or Test, VIO=1V
IRED on, VIRED = 1V,
VBST = 5V, IRCAP = 5V,
(50 mA option selected;
TA = 27°C)
IRED on, VIRED = 1V,
VBST = 5V, IRCAP = 5V,
(100 mA option selected;
TA = 27°C)
IRED on, VIRED = 1V,
VBST = 5V, IRCAP = 5V,
(150 mA option selected;
TA = 27°C)
IRED on, VIRED = 1V,
VBST = 5V, IRCAP = 5V,
(200 mA option selected;
TA = 27°C
VBST = 5V, IRCAP =
5V;Note 4
Falling Edge; 2.1V
nominal selected
Falling Edge; 2.2V
nominal selected
Falling Edge; 2.3V
nominal selected
Falling Edge; 2.4V
nominal selected
Falling Edge; 2.5V
nominal selected
Falling Edge; 2.6V
nominal selected
V1.1
BL5980
Low Battery
Hysteresis
VLB7
3
2.65
2.7
2.75
V
VLB8
3
2.75
2.8
2.85
V
VLBHYST
3
100
Falling Edge; 2.7V
nominal selected
Falling Edge; 2.8V
nominal selected
mV
Note:
1: Wherever a specific VBST value is listed under test conditions, the VBST is forced externally with the inductor
disconnected and the DC-DC converter NOT running.
2: Typical values are for design information only.
3: Limits over the specified temperature range are not production tested and are based on characterization data. Unless
otherwise stated, production test is at room temperature with guard-banded limits.
4: Not production tested.
2.3
AC Electrical Characteristics
AC Electrical Characteristics: Unless otherwise indicated, all parameters apply at TA = -10° to +60°C, VDD
= 3V, VBST = 4.2V, Typical Application (unless otherwise noted) (Note 1 to Note 4).
Test
Parameter
Symbol
Min
Typ
Max
Units Conditon
Pin
Time Base
Internal Clock
Period
TPCLK
9.8
10.4
11
ms
PROGSET, IO = high
Operating
RLED Indicator
On Time
Standby Period
TON1
8
9.8
10.4
11
ms
TPLED1
8
320
344
368
s
TPLED2A
8
470
500
530
ms
Local alarm condition with
temporal horn pattern
TPLED2B
8
625
667
710
ms
Local alarm condition with
continuous horn pattern
TPLED4
8
10
10.7
11.4
s
Timer mode, no local
alarm
TPLED0
8
s
Remote alarm only
TPLED3
9
40
43
46
s
Latched Alarm Condition,
LED enabled
TOFLED
9
1.25
1.33
1.41
s
Latched Alarm Condition,
LED enabled
TLALED
9
22.4
23.9
25.3
Hours
Latched Alarm Condition,
LED enabled
Local Alarm
Period
Hush Timer
Period
External Alarm
Period
LED IS NOT ON
Standby, no alarm
GLED Indicator
Latched Alarm
Period
Latched Alarm
Pulse Train (3x)
Off Time
Latched Alarm
LED Enabled
Duration
Smoke Check
www.belling.com.cn
Page6
V1.1
BL5980
Smoke Test
Period
with Temporal
Horn Pattern
TPER0A
2
10
10.7
11.4
s
TPER1A
2
1.88
2
2.12
s
TPER2A
2
0.94
1
1.06
s
TPER3A
2
0.94
1
1.06
s
235
250
265
ms
TPER4A
2
313
333
353
ms
Standby, no alarm
Standby, after one valid
smoke sample
Standby, after two
consecutive valid smoke
samples
Local Alarm (three
consecutive valid smoke
samples)
Push button test, >1
chamber detections
Push button test, no
chamber detections
In remote alarm
TPER5A
2
7.5
8
8.5
s
TPER0B
2
10
10.7
11.4
s
TPER1B
2
2.5
2.7
2.9
s
TPER2B
2
1.25
1.33
1.41
s
TPER3B
2
1.25
1.33
1.41
s
TPER4B
2
313
333
353
ms
TPER5B
2
10
10.7
11.4
s
In remote alarm
Chamber Test
Period
TPCT1
2
40
43
46
s
Standby, no alarm
Long Term Drift
Sample Period
TLTD
2
630
680
730
s
Standby, no alarm
LTD enabled
TPLB1
3
320
344
368
s
RLED on
TPLB2
3
80
86
92
s
RLED on
THPER1
13
40
43
46
s
Low battery, no alarm
THPER2
13
40
43
46
s
Chamber failure
THON1
13
9.8
10.4
11
ms
Low battery, no alarm
THON2
13
9.8
10.4
11
ms
Chamber failure
THOF1
13
305
325
345
ms
Failed chamber, no
alarm, 3x chirp option
THON2A
13
470
500
530
ms
Local or remote alarm
(Note 1)
Smoke
Test
Period
with Continuous
Horn Pattern
Standby, no alarm
Standby, after one valid
smoke sample
Standby, after two
consecutive valid smoke
samples
Local Alarm (three
consecutive valid smoke
samples)
Push button test
Low Battery
Low Battery
Sample Period
Horn Operation
Low Battery
Horn Period
Chamber Fail
Horn Period
Low Battery
Horn On Time
Chamber Fail
Horn On Time
Chamber Fail
Off Time
Alarm On Time
with Temporal
Horn Pattern
www.belling.com.cn
Page7
V1.1
BL5980
Local or remote alarm
(Note 1)
Local or remote alarm
(Note 1)
THOF2A
13
470
500
530
ms
THOF3A
13
1.4
1.5
1.6
s
Alarm On Time
with Continuous
Horn Pattern
THON2B
13
235
250
265
ms
Local or remote alarm
(Note 1)
Alarm Off Time
with Continuous
Horn Pattern
THOF2B
13
78
83
88
ms
Local or remote alarm
(Note 1)
THON4
13
9.8
10.4
11
ms
Alarm memory active,
push-to-test
THPER4
13
235
250
265
ms
Alarm memory active,
push-to-test
Alarm Off Time
with Temporal
Horn Pattern
Push-to-Test
Alarm Memory
On Time
Push-to-Test
Alarm Memory
Horn Period
Interconnect Signal Operation (IO)
IO Active Delay
Remote Alarm
Delay with
Temporal Horn
Pattern
Remote Alarm
Delay with
Continuous
Horn Pattern
IO Charge
Dump Duration
IO Filter
s
From start of local alarm
to IO active
1.25
s
No local alarm, from IO
active to alarm
572
785
ms
No local alarm, from IO
active to alarm
1.31
1.39
s
313
ms
At conclusion of local
alarm or test
Standby, no alarm
Min
No alarm
TIODLY1
12
0
TIODLY2A
12
0.78
1
TIODLY2B
12
380
TIODMP
12
1.23
TIOFILT
12
Hush Timer Operation
Hush Timer
Period
Low Battery
Hush Timer
Period
TTPER
8
8.6
9.1
TTPERLB
7.73
8.22
8.71
Hours No alarm
TEOL
314
334
354
Hours EOL Enabled; Standby
EOL
End-of-Life Age
Sample
Detection
IRED On Time
www.belling.com.cn
TIRON
2
2
2
2
100
200
300
400
Page8
µs
µs
µs
µs
Prog Bits 3,4 = 1,1
Prog Bits 3,4 = 0,1
Prog Bits 3,4 = 1,0
Prog Bits 3,4 = 0,0
V1.1
BL5980
Note:
1: See timing diagram for Horn Pattern (Figure 5-2).
2: TPCLK and TIRON are 100% production tested. All other AC parameters are verified by functional testing.
3: Typical values are for design information only.
4: Limits over the specified temperature range are not production tested, and are based on characterization data.
2.4 Temperature Characteristics
Parameters
Operating Temperature Range
Storage Temperature Range
Sym
Min
TA
TSTG
Typ
Max
Units
-10
60
℃
-55
125
℃
3.0. PIN DESCRIPTIONS
3.1 Pin Function Table
Pin No.
Symbol
Function
1
VSS
Connect to the negative supply voltage.
2
IRED
Provides a regulated and programmable pulsed current for the infrared emitter
diode.
3
VDD
Connect to the positive supply or battery voltage.
4
TEST
This input is used to invoke Test modes and the Timer mode. This input has an
internal pull-down.
5
TEST2
Test input for test and programming modes. This input has an internal pull-down.
6
IRP
Connect to the anode of the photo diode.
7
IRN
Connect to the cathode of the photo diode.
8
RLED
Open drain NMOS output, used to drive a visible LED. This pin provides load
current for the low battery test, and is a visual indicator for Alarm and Hush
modes.
9
GLED
Open drain NMOS output used to drive a visible LED to provide visual indication
of an Alarm Memory condition.
10
FEED
Usually connected to the feedback electrode through a current limiting resistor. If
not used, this pin must be connected to VDD or VSS.
11
IRCAP
Used to charge and monitor the IRED capacitor.
12
IO
This bidirectional pin provides the capability to interconnect many detectors in a
single system. This pin has an internal pull-down device and a charge dump
device.
13
HB
This pin is connected to the metal electrode of a piezoelectric transducer.
14
HS
This pin is a complementary output to HB, connected to the ceramic electrode of
the piezoelectric transducer.
15
VBST
Boosted voltage produced by DC-DC converter.
www.belling.com.cn
Page9
V1.1
BL5980
16
4.0.
4.1
LX
Open drain NMOS output, used to drive the boost converter inductor. The
inductor should be connected from this pin to the positive supply through a low
resistance path.
DEVICE DESCRIPTION
Standard Internal Timing
The internal oscillator is trimmed to ±6% tolerance. Once every 10 seconds, the boost
converter is powered up, the IRCAP is charged from VBST and then the detection circuitry is
active for 10 ms. Prior to completion of the 10 ms period, the IRED pulse is active for a
user-programmable duration of 100-400 µs. During this IRED pulse, the photo diode current is
integrated and then digitized. The result is compared to a limit value stored in MTP during
calibration to determine the photo chamber status. If a smoke condition is present, the period
to the next detection decreases, and additional checks are made.
4.2
Smoke Detection Circuit
The digitized photo amplifier integrator output is compared to the stored limit value at the
conclusion of the IRED pulse period. The IRED drive is all internal, and both the period and
current are user programmable. Three consecutive smoke detections will cause the device to
go into Alarm and activate the horn and interconnect circuits. In Alarm, the horn is driven at the
high boost voltage level, which is regulated based on an internal voltage reference, and
therefore results in consistent audibility over battery life. RLED will turn on for 10 ms at a 2 Hz
rate. In Local Alarm, the integration limit is internally decreased to provide alarm hysteresis.
The integrator has three separate gain settings:
• Normal and Hysteresis
• Reduced Sensitivity (HUSH)
• High Gain for Chamber Test and Push-to-Test
There are four separate sets of integration limits (all user programmable):
• Normal Detection
• Hysteresis
• HUSH
• Chamber Test and Push-to-Test modes
In addition, there are user selectable integrator gain settings to optimize detection levels (see
Table 4-1).
4.3
Supervisory Test
Once every 86 seconds, the status of the battery voltage is checked by enabling the boost
converter for 10 ms and comparing a fraction of the VDD voltage to an internal reference. In
each period of 344 seconds, the battery voltage is checked four times. Three checks are
unloaded and one check is performed with the RLED enabled, which provides a battery load.
The High Boost mode is active only for the loaded low battery test. In addition, once every 43
seconds the chamber is activated and a High Gain mode and chamber test limits are internally
selected. A check of the chamber is made by amplifying background reflections. The Low
Boost mode is used for the chamber test.
If either the low battery test or the chamber test fails, the horn will pulse on for 10 ms every 43
seconds, and will continue to pulse until the failing condition passes. If two consecutive
chamber tests fail, the horn will pulse on three times for 10 ms, separated by 330 ms every 43
seconds. Each of the two supervisory test audible indicators is separated by approximately 20
seconds.
As an option, a Low Battery Silence mode can be invoked. If a low battery condition exists, and
www.belling.com.cn
Page10
V1.1
BL5980
the TEST input is driven high, the RLED will turn on. If the TEST input is held for more than 0.5
second, the unit will enter the Push-to-test operation described in Section 4.4 “Push-to-Test
Operation (PTT)”. After the TEST input is driven low, the unit enters in Low Battery Hush mode,
and the 10 ms horn pulse is silenced for 8 hours. The activation of the test button will also
initiate the 9 minute Reduced Sensitivity mode described in Section 4.6 “Reduced Sensitivity
Mode”. At the end of the 8 hours, the audible indication will resume if the low battery condition
still exists.
4.4
Push-to-Test Operation (PTT)
If the TEST input pin is activated (VIH), the smoke detection rate increases to once every 250
ms after one internal clock cycle. In Push-to-Test, the photo amplifier High Gain mode is
selected, and background reflections are used to simulate a smoke condition. After the
required three consecutive detections, the device will go into a Local Alarm condition. When
the TEST input is driven low (VIL), the photo amplifier Normal Gain is selected, after one clock
cycle. The detection rate continues at once every 250 ms until three consecutive No Smoke
conditions are detected. At this point, the device returns to standby timing. In addition, after the
TEST input goes low, the device enters the HUSH mode (see Section 4.6 “Reduced Sensitivity
Mode”).
4.5
Interconnect Operation (I/O)
The bidirectional IO pin allows the interconnection of multiple detectors. In a Local Alarm
condition, this pin is driven high (High Boost) immediately through a constant current source.
Shorting this output to ground will not cause excessive current. The IO is ignored as input
during a Local Alarm.
The IO pin also has an NMOS discharge device that is active for 1.3 seconds after the
conclusion of any type of Local Alarm. This device helps to quickly discharge any capacitance
associated with the interconnect line.
If a remote, active high signal is detected, the device goes into Remote Alarm and the horn will
be active. RLED will be off, indicating a Remote Alarm condition. Internal protection circuitry
allows the signaling unit to have a higher supply voltage than the signaled unit, without
excessive current draw.
The interconnect input has a 336 ms nominal digital filter. This allows the interconnection to
other types of alarms (carbon monoxide, for example) that may have a pulsed interconnect
signal.
4.6
Reduced Sensitivity Mode
A Reduced Sensitivity or Hush mode is initiated by activating the TEST input (VIH). If the TEST
input is activated during a Local Alarm, the unit is immediately reset out of the alarm condition,
and the horn is silenced. When the TEST input is deactivated (VIL), the device enters into a
9-minute nominal Hush mode. During this period, the HUSH integration limit is selected. The
hush gain is user programmable. In Reduced Sensitivity mode, the RLED flashes for 10 ms
every 10 seconds to indicate that the mode is active. As an option, the Hush mode will be
cancelled if any of the following conditions exist:
•Reduced sensitivity threshold is exceeded (high smoke level)
•An interconnect alarm occurs
•TEST input is activated again.
4.7
Local Alarm Memory
An Alarm Memory feature allows easy identification of any unit that had previously been in a
Local Alarm condition. If a detector has entered a Local Alarm, when it exits that Local Alarm,
the Alarm Memory latch is set. Initially the GLED can be used to visually identify any unit that
www.belling.com.cn
Page11
V1.1
BL5980
had previously been in a Local Alarm condition. The GLED flashes three times spaced 1.3
seconds apart. This pattern will repeat every 43 seconds. The duration of the flash is 10 ms. In
order to preserve battery power, this visual indication will stop after a period of 24 hours. The
user will still be able to identify a unit with an active alarm memory by pressing the
Push-to-Test button. When this button is active, the horn will chirp for 10 ms every 250 ms.
If the Alarm Memory condition is set, then any time the Push-to-Test button is pressed and
released, the Alarm Memory latch is reset.
The initial 24 hour visual indication is not displayed if a low battery condition exists.
4.8
End-of-Life Indicator
As an option, after every 14 days of continuous operation, the device will read a stored age
count from the MTP and increment this count. After 10 years of powered operation, an audible
warning will occur indicating that the unit should be replaced. This indicator will be similar to
the chamber test failure warning in that the horn will pulse on three times for 10 ms separated
by 330 ms every 43 seconds. This indicator will be separated from the low battery indicator by
approximately 20 seconds.
4.9
Photo Chamber Long Term Drift Adjustment
As an option, the design includes a Long Term Drift Adjustment for the photo chamber. If this
option is selected, during calibration a normal no-smoke baseline integration measurement is
made and stored in MTP. During normal operation, a new baseline is calculated by making 64
integration measurements over a period of 12 hours. These measurements are averaged and
compared to the original baseline stored during calibration to calculate the long term drift. All
four limits stored during calibration are adjusted by this drift factor. Drift sampling is suspended
during Hush, Local Smoke and Remote Smoke conditions.
www.belling.com.cn
Page12
V1.1
BL5980
5.0. USER PROGRAMMING MODES
TABLE 5-1: PARAMETRIC PROGRAMMING
Parametric Programming
Programming
Range
IRED Period
IRED Current Sink
Low Battery Detection Voltage
100-400 μs
100 μs
50-200 mA
50 mA
2.1–2.8V
100 mV
Typical Maximum Input Current (nA)
Photo Detection Limits
Normal/Hysteresis
Hush
Chamber Test
100μs
GF
GF
GF
GF
GF
GF
GF
GF
GF
GF
GF
GF
Resolution
=
=
=
=
=
=
=
=
=
=
=
=
1
2
3
4
1
2
3
4
1
2
3
4
200μs
58
29
14.5
7.2
116
58
29
14.5
29
14.5
7.2
3.6
29
14.5
7.2
3.6
58
29
14.5
7.2
14.5
7.2
3.6
1.8
300μs
400μs
19.4
9.6
4.8
2.4
38.8
19.4
9.6
4.8
9.6
4.8
2.4
1.2
14.5
7.2
3.6
1.8
29
14.5
7.2
3.6
7.2
3.6
1.8
0.9
Note:
Note:
1: GF is the user selectable Photo Integration Gain Factor. Once selected, it applies to all modes of
operation. For example, if GF = 1 and integration time is selected to be 100 μs, the ranges will be
as follows: Normal/Hysteresis = 58 nA, Hush = 116 nA, Chamber Test = 29 nA.
2: Nominal measurement resolution in each case will be 1/63 of the maximum input range.
3: The same current resolution and ranges applies to the limits.
TABLE 5-2: FEATURES PROGRAMMING
Features
Options
Tone Select
Continuous or NFPA Tone
10 Year End-of-life Indicator
Enable/Disable
Photo Chamber Long Term Drift Adjustment
Enable/Disable
Low Battery Hush
Enable/Disable
Hush Options
Option 1: Hush mode is not cancelled for any reason. If the
test button is pushed during Hush, the unit reverts to Normal
Sensitivity to test the unit, but when it comes out of test,
resumes in Hush where it left off.
Option 2: The Hush mode is cancelled if the Reduced Sensitivity threshold is exceeded (high smoke level), and if
an external (interconnect alarm) is signaled. If the test button is pushed during Hush, after the test is executed, the
Hush mode is terminated.
www.belling.com.cn
Page13
V1.1
BL5980
5.1 Calibration and Programming Procedures
Thirteen separate programming and test modes are available for user customization. To enter
these modes, after power-up, TEST2 must be driven to VDD and held at that level. The TEST
input is then clocked to step through the modes. FEED and IO are reconfigured to become test
mode inputs, while RLED, GLED and HB become test mode outputs. The test mode functions
for each pin are outlined inTable5-3.
TABLE 5-3: TEST MODE FUNCTIONS
Mode Description
TEST
TEST Data TEST2
Clock
FEED
IO
RLED
GLED
HB
VBST
VSS
0
VBST
VSS
FEED
VDD
VSS
IO
—
—
RLED
—
—
GLED
—
—
HB
T0
VIH
VIL
Horn Test
T1
Low Battery test
1
not used
VDD
BoostEn
LBstrb
RLEDen
GLEDen
LBout
Photo Gain Factor(2bits)
2
ProgData
VDD
ProgCLK
ProgEn 14 bits
RLED
GLED
HB
Integ Time (2bits)
IRED Current (2bits)
2
2
2
ProgData
ProgData
ProgData
VDD
VDD
VDD
ProgCLK
ProgCLK
ProgCLK
ProgEn 14 bits
ProgEn 14 bits
ProgEn 14 bits
RLED
RLED
RLED
GLED
GLED
GLED
HB
HB
HB
LB Hush Enable (1bit)
2
2
2
ProgData
ProgData
ProgData
VDD
VDD
VDD
ProgCLK
ProgCLK
ProgCLK
ProgEn 14 bits
ProgEn 14 bits
ProgEn 14 bits
RLED
RLED
RLED
GLED
GLED
GLED
HB
HB
HB
T3
EOL Enable(1bit)
Tone Select (1bit)
Norm Lim Set (6bits)
2
2
3
ProgData
ProgData
not used
VDD
VDD
VDD
ProgCLK
ProgCLK
CalCLK
ProgEn 14 bits
ProgEn 14 bits
IntLat(3)
RLED
RLED
Gamp
GLED
GLED
IntegOut
HB
HB
SmkComp(1)
T4
Hyst Lim Set (6bits)
4
not used
VDD
CalCLK
IntLat(3)
Gamp
IntegOut
SmkComp(1)
T5
Hush Lim Set (6bits)
5
not used
VDD
CalCLK
IntLat(3)
Gamp
IntegOut
SmkComp(1)
6
not used
VDD
CalCLK
Gamp
IntegOut
SmkComp(1)
7
8
9
10
11
12
ProgData
not used
not used
not used
not used
not used
VDD
VDD
VDD
VDD
VDD
VDD
ProgCLK
MeasEn
MeasEn
MeasEn
MeasEn
MeasEn
IntLat(3,4)/ProgEn 24
bits
ProgEn
ProgEn 30 bits
not used
not used
not used
not used
RLED
Gamp
Gamp
Gamp
Gamp
Gamp
GLED
IntegOut
IntegOut
IntegOut
IntegOut
IntegOut
Serial Out
HB
SCMP( 2)
SCMP( 2)
SCMP( 2)
SCMP( 2)
T2
Low Battery Trip (3bits)
LTD Enable (1bit)
Hush Option (1bit)
T6
T7
T8
T9
T10
T11
T12
Ch Test Lim Set(6bits)
Serial Read/Write
LTD Baseline (6 bits)
Norm Lim Check
Hyst Lim Check
Hush Lim Check
Ch Test Lim Check
VDD
VSS
HornEn
VDD
VSS
VDD
Note
1: SmkComp (HB) – digital comparator output (high if Gamp < IntegOut; low if Gamp > IntegOut)
2: SCMP (HB) – digital output representing comparison of measurement value and associated limit. Signal is valid only after MeasEn
has been asserted and measurement has been made. (SCMP high if measured value > limit; low if measured value < limit).
3: IntLat (IO) – digital input used for two purposes. If FEED is at a logic high level, then a low to high transition on IntLat will initiate
an integration cycle. If FEED is at a logic low level, then a low to high transition on IntLat will latch the present state of the limits
(GAMP level) for later storage. T2-T5 limits are latched, but not stored until ProgEn is asserted in T6 mode
4: At the end of T6 mode, in order to store the limits, the IO input must be pulsed twice consecutively with FEED held low. The first
pulse will latch the data and the second will store it in MTP.
5.2 User Selections
Prior to smoke calibration, the user must program the functional options and parametric selections. This
requires that 14 bits, representing selected values, be clocked in serially using TEST as a data input and
FEED as a clock input, and then be stored in the internal MTP.
The detailed steps are as follows:
1. Power up with bias conditions as shown in Figure 5-1. At power-up TEST = TEST2 = FEED = IO =
VSS.
2. Drive TEST2 input from VSS to VDD and hold at VDD through Step 5 below.
3. Using TEST as data and FEED as clock, shift in values as selected from Register 5-1.The minimum
www.belling.com.cn
Page14
V1.1
BL5980
pulse width for FEED is 10 µs, while the minimum pulse width for TEST is 100 µs. For example, for the
following options, the sequence would be:
data - 0 0 0 1 1 0 0 0 1 0 0 0 0 1
bit - 30 31 32 33 34 35 36 37 38 39 40 41 42 43
Photo Amp Gain Factor = 1
Integration Time
= 200 µs
IRED Current
= 100 mA
Low Battery Trip
= 2.2V
Long Term Drift, Low Battery Hush and EOL are all disabled
Hush Option = Never Cancel
Tone Select = Temporal
4. After shifting in data, pull IO input to VDD, then VSS (minimum pulse width of 100ms) to
store shift register contents into the memory.
5. If any changes are required, power down the part and return to Step 1. All bit values must be
reentered.
Note: For test mode T2 only 14 bits (bits 25-38) will be loaded. For test mode T8 all 44 bits (bits 0-43), will be loaded.
Figure 5-1 Nominal Application Circuit for Programming
REGISTER 5-1 CONFIGURATION AND CALIBRATION SETTINGS REGISTER
W
LTD
bit 39
www.belling.com.cn
W
LB0
W
LB1
W
LB2
W
TS
bit 43
W
EOL
W
LBH
W
HUSH
bit 40
W
IRC1
W
IRC0
W
IT1
W
IT0
bit 32
Page15
V1.1
BL5980
W
PAGF1
bit 31
W
PAGF0
W
NL5
W
NL4
W
NL3
W
NL2
W
NL1
W
NL0
bit 24
W
HYL5
bit 23
W
HYL4
W
HYL3
W
HYL2
W
HYL1
W
HYL0
W
HUL5
W
HUL4
bit 16
W
HUL3
bit15
W
HUL2
W
HUL1
W
HUL0
W
CTL5
W
CTL4
W
CTL3
W
CTL2
bit 8
W
CTL1
bit 7
W
CTL0
W
LTD5
W
LTD4
W
LTD3
W
LTD2
W
LTD1
W
LTD0
bit 0
Note: W=writable bit
bit 43
Tone Select
TS
0=Continuous Horn Pattern
1=Temporal Horn Pattern
bit 42
End of Life Enable
EOL
0=Disable
1=Enable
bit 41
Low Battery Hush Enable
LBH
0=Disable
1=Enable
bit 40
Hush Option
HUSH
0=Never Cancel
1=Cancelled for high smoke level, interconnect alarm, or second push of TEST button (as
described above Low Battery Hush Enable
bit 39
Long Term Drift Enable
LTD
0=Disable
1=Enable
bit 38-36
Low Battery Trip Point
LB0, LB1, LB2
000=2.1V
001=2.2V
010=2.3V
011=2.4V
100=2.5V
www.belling.com.cn
Page16
V1.1
BL5980
101=2.6V
110=2.7V
111=2.8V
bit 35-34
IRED Current
IRC1, IRC0
00=50mA
01=100mA
10=150mA
11=200mA
bit 33-32
Integration Time
IT1, IT0
00=400us
01=300us
10=200us
11=100us
bit 31-30
Photo Amplifier Gain Factor
PAGF1, PAGF0
00=1
01=2
10=3
11=4
bit 29~24
Normal Limits (Section 3.2)
NL5, NL4, NL3, NL2, NL1, NL0
000000=0
000001=1
…
111110=62
111111=63
bit 23-18
Hysteresis Limits (Section 3.2)
HYL5, HYL4, HYL3, HYL2, HYL1, HYL0
000000=0
000001=1
…
111110=62
111111=63
bit 17-12
Hush Limits (Section 3.6)
HUL5, HUL4, HUL3, HUL2, HUL1, HUL0
000000=0
000001=1
…
111110=62
111111=63
bit 11-6
Chamber Test Limits (Section 3.3)
CTL5, CTL4, CTL3, CTL2, CTL1, CTL0
000000=0
000001=1
…
www.belling.com.cn
Page17
V1.1
BL5980
111110=62
111111=63
bit 5-0
Long Tern Drift Sample (Section 3.9)
LTD5, LTD4, LTD3, LTD2, LTD1, LTD0
000000=0
000001=1
…
111110=62
111111=63
Figure 5-2: Timing Diagram for Mode T2
As an alternative to Figure 5-1, Figure 5-3 can be used to program while in the application circuit. Note that in addition to the five
programming supplies, connections to VSS are needed at TP1 and TP2.
www.belling.com.cn
Page18
V1.1
BL5980
Figure 5-3: Circuit for Programming in the Typical Application
5.3
Smoke Calibration
A separate calibration mode is entered for each measurement mode (Normal, Hysteresis, Hush
and Chamber Test) so that independent limits can be set for each. In all calibration modes, the
integrator output can be accessed at the GLED output.
The Gamp output voltage, which represents the smoke detection level, can be accessed at the
RLED output. The SmkComp output voltage is the result of the comparison of Gamp with the
integrator output, and can be accessed at HB. The FEED input can be clocked to step up the
smoke detection level at RLED. Once the desired smoke threshold is reached, the TEST input is
pulsed low to high to store the result.
The procedure is described in the following steps:
1.Power up with the bias conditions shown in Figure5-1.
2.Drive TEST2 input from VSS to VDD to enter the Programming mode. TEST2 should remain at
VDD through Step8described below.
3.Apply three clock pulses to TEST input to enterT3 mode. This initiates the calibration mode for
Normal Limits setting. The Integrator output should appear at GLED and the smoke detection level
at RLED.
4.At this point clock FEED to increase the smoke detection level as needed. Pulling IO high with
FEED to increase the smoke detection level as needed. Pulling IO high with FEED at a logic high
level will initiate an integration. The integrator output signal should appear at GLED. The
sequence of incrementing the limit, performing an integration and monitoring the HB output for the
resulting comparison can be repeated until the desired threshold is reached. Once the desired
smoke threshold is reached, with FEED held low, the IO input should be pulsed low to high to latch
the smoke detection level.
5.Apply a fourth clock pulses to TEST input to enter T4mode.Thisinitiatesthecalibrationmode for
Hysteresis Limits. The sequence in Step 4 should be repeated to set the Hysteresis Limit.
6.Apply clock pulse to TEST input again to enter T5mode and initiate calibration for Hush Limits.
Repeat Step 4to set the Hush Limit.
7.Apply clock pulse to TEST input a sixth time to enterT6 mode and initiate calibration for
Chamber Test Limits. Repeat Step 4 to set the Chamber Test Limit.
8.After pulsing the IO input to latch the Chamber Test Limit, the IO must be pulsed low to high a
second time to store the limits in memory.
www.belling.com.cn
Page19
V1.1
BL5980
Figure 5-4: Timing Diagram for Modes T3 to T6
5.4 Serial Read/Write
As an alternative to the steps in Section 5.3 “Smoke Calibration”, if the system has been well
characterized, the limits and baseline can be entered directly from a serial read/write calibration
mode.
To enter this mode, follow these steps:
1. Set up the application as shown in Figure 5-1.
2. Drive TEST2 input from VSS to VDD to enter in Programming mode. TEST2 should remain at
VDD until all data has been entered.
3. Clock the TEST input to mode T8 (High = VBST, Low = VSS, 8 clocks). This enables the Serial
Read/Write mode.
4. TEST now acts as a data input (High = VDD, Low = VSS). FEED acts as the clock input (High =
VBST, Low = VSS). Clock in the limits, LTD baseline, functional and parametric options. The data
sequence should be as follows:
6 bit LTD sample (LSB first)
6 bit Chamber Test Limits (LSB first)
6 bit Hush Limits (LSB first)
6 bit Hysteresis Limits (LSB first),
6 bit Normal Limits (LSB first)
Then, the data sequence follows the pattern described in Register 5-1:
2 bit Photo Amp Gain Factor
2 bit Integration Time
2 bit IRED current
3 bit Low Battery Trip Point
1 bit Long Term Drift Enable
1 bit Hush Option
1 bit Low Battery Hush Enable
1 bit EOL enable
1 bit Tone Select
A serial data output is available at HB.
5. After all 44 bits have been entered, pulse IO to store into the MTP memory.
www.belling.com.cn
Page20
V1.1
BL5980
Figure 5-5: Timing Diagram for Mode T7
5.5 LTD Baseline Measurement
If the Long Term Drift Adjustment is enabled, a Long Term Drift Baseline must be set. If an
accurate value is known based on previous chamber characterization, it can be loaded above in
T7 with the serial data. If not, zeros can be entered as placeholders in T7 and a long term drift
(LTD) baseline measurement must be made. To do this, the unit should be connected to its smoke
chamber and placed in a no-smoke condition. To enable the baseline measurement, pull TEST
from VSS to VBST again (or a total of 8 times) and return to VSS. Once the chamber is clear,
pulse FEED low to high to make the baseline measurement. The duration of this pulse should be
at least 2ms.
After baseline LTD measurement has been made, pulse IO with FEED held low to store the result
in memory.
www.belling.com.cn
Page21
V1.1
BL5980
Figure 5-6: Timing Diagram for Mode T8
5.6 Limits Verification
After all limits and LTD baseline have been entered and stored into the memory, additional test
modes are available to verify if the limits are functioning as expected. Table 5-4 describes several
verification tests.
TABLE 5-4: LIMITS VERIFICATION DESCRIPTION
Limit
Test Description
Normal Limits
Clock TEST to Mode T9 (9 clocks). With appropriate smoke level in chamber,
pull FEED to VDD and hold for at least 1 ms. The HB output will indicate the
detection status (High = smoke detected).
Hysteresis Limits
Clock TEST to Mode T10 (10 clocks). Pulse FEED and monitor HB as in
Normal Limits case.
Hush Limits
Clock TEST to Mode T11 (11 clocks). Pulse FEED and monitor HB.
Chamber Test Limits
Clock TEST to Mode T12 (12 clocks). Pulse FEED and monitor HB.
www.belling.com.cn
Page22
V1.1
BL5980
Figure 5-7: Timing Diagram for Modes T9-12
5.7 Horn Test
This mode allows the horn to be enabled indefinitely for audibility testing. To enable, the TEST2
input should be driven to the Vdd level and TEST should be held at Vss. The IO pin is configured
as horn enable.
Figure 5-8: Timing Diagram for Mode T0
5.8 Low Battery Test
This mode allows the user to enable the internal low battery circuitry to perform a low battery test.
To enter this mode, follow these steps:
1. Power up with the bias conditions shown in Figure 5-1.
2. Drive TEST2 input from VSS to VDD to enter the Programming mode. TEST2 should remain at
Vdd through the following steps
3. Apply one clock pulse to the TEST input to enter the T1 mode
4. Drive the IO input from VSS to Vdd. This will enable the boost converter and turn on the RLED
www.belling.com.cn
Page23
V1.1
BL5980
driver.
5. Monitor the HB output for the low battery comparator status.
Figure 5-9: Timing Diagram for Mode T1
6.0. APPLICATION NOTES
6.1 Standby Current Calculation and Battery Life
The supply current shown in the DC Electrical Characteristics table is only one component of the
average standby current and, in most cases, can be a small fraction of the total, because power
consumption generally occurs in relatively infrequent bursts and depends on many external factors.
These include the values selected for IRED current and integration time, the VBST and IR
capacitor sizes and leakages, the VBAT level, and the magnitude of any external resistances that
will adversely affect the boost converter efficiency.
A calculation of the standby current for the battery life is shown in Table 5-1, based on the
following parameters:
VBAT = 3
VBST1 = 3.6
VBST2 = 9
Boost capacitor size = 4.70E-06
Boost Efficiency = 8.50E-01
IRED on time = 2.000E-04
IRED Current = 1.000E-01
TABLE 6-1: STANDBY CURRENT CALCULATION
IDD Component Voltage
(V)
Current
(A)
Fixed IDD
3
1.00E-06
Photo Detection Current
Chamber
test
www.belling.com.cn
3.6
1.00E-03
Duration
(s)
Energy
(J)
Period (s)
1
1.0E-02
3.60E-05
Page24
43
Average
IDD
Power (W) Contribut
ion (A)
IBAT
(uA)
3.00E-06
1.00E-06
1.0
9.85E-07
3.28E-07
0.3
V1.1
BL5980
(excluding IR
drive)
IR drive during
3.6
0.10
Chamber Test
3.6
1.00E-03
Smoke
Detection
(excluding IR
drive)
IR drive during
3.6
0.10
Smoke
Detection
Low Battery Check Current
Loaded Test
Load
9
2.00E-02
Boost
Vbst1 to
Vbst2
Unloaded Test
Load
3.6
1.00E-04
2.00E-04
7.20E-05
43
1.97E-06
6.57E-07
0.7
1.00E-02
3.60E-05
10.75
3.94E-06
1.31E-06
1.3
2.00E-04
7.20E-05
10.75
7.88E-06
2.63E-06
2.6
1.00E-02
1.80E-03
6.85E-05
344.00
344.00
6.16E-06
2.34E-07
2.05E-06
7.81E-08
2.1
0.1
1.00E-02
3.60E-06
43.00
9.85E-08
Total
3.28E-08
8.09E-06
0.0
8.1
The following paragraphs explain the components in Table 6-1.
6.1.1 Fixed IDD
The IDD is the Supply Current shown in the DC Electrical Characteristics table.
6.1.2 PHOTO DETECTION CURRENT
Photo Detection Current is the current draw due to the smoke testing every 10.75 seconds, and
the chamber test every 43 seconds. The current for both the IR diode and the internal
measurement circuitry comes primarily from VBST, so the average current must be scaled for both
on-time and boost voltage.
The contribution to IBAT is determined by first calculating the energy consumed by each
component, given its duration. An average power is then calculated based on the period of the
event and the boost converter efficiency (assumed to be 85% in this case). An IBAT contribution is
then calculated based on this average power and the given VBAT. For example, the IR drive
contribution during chamber test is detailed in Equation 6-1:
EQUATION 6-1:
6.1.3 LOW BATTERY CHECK CURRENT
The Low Battery Check Current is the current required for the low battery test. It includes both the
loaded (RLED on) and unloaded (RLED Off) tests. The boost component of the loaded test
represents the cost of charging the boost capacitor to the higher voltage level. This has a fixed
cost for every loaded check, because the capacitor is gradually discharged during subsequent
operations, and the energy is generally not recovered. The other calculations are similar to those
shown in Equation 6-1. The unloaded test has a minimal contribution because it involves only
some internal reference and comparator circuitry.
www.belling.com.cn
Page25
V1.1
BL5980
6.1.4 BATTERY LIFE
When estimating the battery life, several additional factors must be considered. These include
battery resistance, battery self discharge rate, capacitor leakages and the effect of the operating
temperature on all of these characteristics. Some number of false alarms and user tests should
also be included in any calculation.
For ten year applications, a 3V spiral wound lithium manganese dioxide battery with a laser seal is
recommended. These can be found with capacities of 1400 to 1600 mAh.
www.belling.com.cn
Page26
V1.1
BL5980
6.1.5
Functional Timing Diagrams
Figure 6-1 BL5980 Timing Diagram – Standby, No Alarm, Low Supply Test Failure and
Chamber Test Failure.
www.belling.com.cn
Page27
V1.1
BL5980
Figure 6-2: BL5980 Timing Diagram – Local
Alarm with Temporal Horn Pattern, Local Alarm with International Horn Pattern, Interconnect as
Input with Temporal Horn Pattern and Interconnect as Input with International Horn Pattern.
www.belling.com.cn
Page28
V1.1
BL5980
Figure 6-3: BL5980 Timing Diagram – Alarm Memory and Hush Timer.
www.belling.com.cn
Page29
V1.1
BL5980
7.0. PACKAGING INFORMATION
www.belling.com.cn
Page30
V1.1