High Quality Intelligent Manhole Cover Supply Primary 3.6V Li ...

Each time we take a SOH measurement the value drops by 2% of full scale, but the batteries have not actually drained by 2%.

You will get efficient and thoughtful service from SUNJ ENERGY.

We are using the BQ chip.

The chip is loaded with the default BQ image from Texas Instruments v1.02 bld10, but with the following changes...
1)  The CC offset, Board Offset, Current, and Voltage are calibrated for our board.
2)  The chem ID is loaded into the device.
3)  The gauge is configured for EOS mode.

Our system consists of two pieces. 

1)  A battery pack containing two Tadiran TL- cells in parallel for a total of mAh of capacity.  The nominal battery voltage is 3.6V, but may dip as low as 3.0V at cold temperature.  In series with the batteries there is a 200milli-Ohm resistor inside the battery pack that is part of a separate protection circuit.

2)  A circuit card that sits in another enclosure, apart from the battery pack.  Our card will function properly with input voltage from 1.8V to 5.5V.  The BQ REGIN pin is sourced from a constant 3.3V that comes from a Texas Instruments TPS.  The BAT pin is tied directly to the output of the battery pack.  We are using a 100 milli-ohm sense resistor.  The peak current is around 50mA, and our sleep current is under 1mA.  The system is deisigned to momentarily support loads of a few hundred mA, but that is not happening during this test.

The circuit card normally sleeps for around 8 seconds.  Wakes up for 200ms, and then goes back to sleep.  The process repeats indefinitely.

Here is the software process we are using to read the SOH values.
1)  Assert the GE pin.
2)  Wait at least 4 seconds.
3)  Send GAUGE_START
4)  Wait for at least 10 seconds.  The circuit card will have been asleep and awake for at least part of those 10 seconds.
5)  Send GAUGE_STOP
6)  Wait at least 4 seconds.
7)  Poll the G_DONE bit at a rate of not more than once every 4 seconds for up to 1 minute.
8)  Wait at least 4 seconds.
9)  If the G_DONE bit is set within one minute then read the SOH value from the gauge.

If you want to learn more, please visit our website mAh li socl2 battery.

If I put a new set of batteries into our pack and send the NEW_BATTERY command.  The SOH measurements read 98% after the first measurement cycle, but falls to 0% within a few hours.  With the above specifed power useage profile the batteries should last over a year.

Using the same set of batteries.  If I send the NEW_BATTERY command again.  The process will repeat.  The gauge reports 98% and then slowly decays to 0% in a few hours.

Here are my questions...

1)  Will the 200mOhm resistor affect the accuracy of the EOS algorithm?
  a)  If so is there any sort of settings change that could be made to either calibrate it out or minimize its effect on the measurements.  Many practical systems will include some resistance due to cabling or fuses that must occur prior to the gauge, there must be some way to calibrate them out.

2)  Given the above information, is there some setting in the image file that I should be changing?

3)  Assertion of the GE pin is not in any way synchronized to peak current draw.  We guarantee that both high and low current draw will occur during the measurement cycle, but we don't guarantee which one will occur first.  Is this a problem?

As a side note:  its too bad this chip doesn't have a register where I could program a time between measurements and a measurement duration and then just have the chip automatically wake up and take the EOS measurements.  If it did then we could just read the SOH values without all the extra work of managing when to turn the gauge on and off (I could live with the extra 15uA of standby current).

Contact us to discuss your requirements of 3V li mno2 battery. Our experienced sales team can help you identify the options that best suit your needs.