Calibration
80
SLUUBD3D – September 2015 – Revised September 2018
Copyright © 2015–2018, Texas Instruments Incorporated
Manufacture Production
1. Apply a known voltage in mV to the battery terminals on which the resistor divider is connected.
2. Divide the known battery voltage by the number of cells configured in
AFE Cell Map
. This would be
Actual Avg Cell Voltage.
3. Read
ExtAveCellVoltage()
.
4. Calculate the Ext Cell Divider Gain.
•
New
Ext Cell Divider Gain
= (Old
Ext Cell Divider Gain
× Actual Avg Cell Voltage)/Measured
Avg Cell Voltage
5. Update
Ext Cell Divider Gain
in data flash.
12.2.3 VAUX Voltage Calibration
The bq78350-R1 can be configured with an auxiliary voltage measurement input. This measurement has
its own calibration procedure, as follows:
1. Apply a known voltage in mV to the VAUX input.
2. Read
VAUXVoltage()
.
3. Calculate the VAUX Gain.
•
New
VAUX Gain
= (Old
VAUX Gain
/VAUX Voltage)
4. Update
VAUX Gain
in data flash.
12.2.4 Voltage Calibration Data Flash
(1)
Setting this value to 0 causes the gauge to use the internal factory calibration default.
Class
Subclass
Name
Type
Length in Bytes
Min
Max
Default
Unit
Calibration
Voltage
Cell1 Offset to
Cell15 Offset
Integer
1
–128
127
0
(1)
—
Calibration
VAUX Voltage
VAUX Gain
Integer
4
0
10000
5000
—
Calibration
Ext Cell Voltage
Ext Ave Divider
Gain
Integer
2
0
32767
5000
—
12.2.5 Current Calibration
12.2.5.1 CC Offset Calibration
1. Apply a known current of 0 mA, and ensure no current is flowing through the sense resistor connected
between the SRP and SRN pins.
2. If
ManufacturerStatus()[CAL]
= 0, send 0x002D to
ManufacturerAccess()
to enable the
[CAL]
flag.
3. Send 0xF082 to
ManufacturerAccess()
to enable factory calibrated CC output on
ManufacturerData()
.
4. Poll
ManufacturerData()
until ZZ increments by 2 before reading data.
5. Obtain the factory calibrated conversion readings of current from
ManufacturerData()
:
•
FCAL
CC
= GGgg of
ManufacturerData()
Is FCAL
CC
< 0x8000? If yes, use FCAL
CC
; otherwise, FCAL
CC
= –(0xFFFF – AAaa + 0x0001).
6. Average several readings for higher accuracy. Poll
ManufacturerData()
until ZZ increments to indicate
that updated values are available:
•
FCAL
CC
= [FCAL
CC
(reading n) + … + FCAL
CC
(reading 1)]/n
7. Read
Coulomb Counter Offset Samples
from data flash.
8. Calculate offset value:
•
CC offset = FCAL
CC
× (Coulomb Counter Offset Samples)
9. Write the new
CC Offset
value to data flash.
10. Re-check the current reading and if it is not accurate, repeat the steps.
11. Send 0x002D to
ManufacturerAccess()
to clear the
[CAL]
flag if all calibration is complete.