bq27501 golden image for single Varta Easypack L

Gyorgy,

On which step exactly did bqEasy fail?  Did you get any error messages?

If you have the log file for the discharge/relax period, could you post it so that I can take a look at it?

Also, if you haven't already, there are 2 application notes about bqEasy on the bq27501 product page which might be able to help you as well.

Cheers,

Charles

Hi,

My error message is the following:
Dataflash.GasGauging.State.UpdateStatus 0 does NOT = 2. Impedance track learning cycle was not completed correctly for both packs. Use Pack A and repeat.

My last log file:
8015.charge1_V3_10.zip

Thanks,

György

I made a nice cycle with long relaxation times, I got the VOK cleared, but still the error message with the request to do a cycle with the other pack.

My log:

2766.20100522_V3_10_part1.zip

Thanks,
György

Interesting...

I looked through your log file, and found that the RUP_DIS bit was set during a part of your discharge, which will cause this failure.  However, I wasn't able to figure out exactly why from these data.  I have a few suggestions and a request.

1. When the error message showed, was UPDATE STATUS 1 or 0?

2. Your charge cuts off at 100mA, is your taper current set value (in dataflash) set to a value larger than 100mA?

3. You might try discharging at a slightly slower rate.  We generally recommend C/5 for learning cycles.  You are a little higher than that, but I don't think this is really a problem.

To figure out exactly what is going on, I'd like to get a log of your dataflash during the learning cycle.  This should tell me exactly which part of your learning cycle fails and how.  If you set your dataflash to autoexport every 30 seconds or so, this should be fine.  If you don't already have a log, could you rerun the learning cycle with this logging enabled?

Thanks, and hopefully we can get this solved quickly.

Charles

One more thing:

Could you also include a dataflash image from the chip after the learning cycle fails?

Thanks,

Charles

Hi, here are my answers:

1. UPDATE STATUS = 0

2. Taper current in dataflash is 139mA

3. I tried to set discharge resistor value to achieve C/5 (264mA) at around 3.7V (nominal voltage), therefore it drops below this value at voltage falling.

But I have good news!
I tried to cheat a bit and made a disconnection/reconnection of the battery and I tried to run a new learning cycle, and luckily this solved the problem!
As I understood, the controller uses the other (unused) profile after reconnection.

The exact history:

- Learning cycle (20100522_V3_10_part1.log)
- Undervoltage lockout inside Varta battery (logged at the end of 20100522_V3_10.log)
- Shutting down of bqEvaluation software (host PC accidentally hibernated and EV2300 connection can be recovered only with restart - USB driver error?)
- "Defibrillating" (output turnon on) battery with short high voltage connection to it directly
- Start of bqEvaluation software
- Charger connection
- Battery disconnection for some tens of seconds
- Charge to full
- new Learning cycle
- Golden image creation :-)

Logfiles:
4846.LOGS.ZIP

Our HW runs Windows CE 6, please help me to transform the golden image to be able to program the chips at manufacturing by the onboard microcontroller on its I2C bus!

The golden image (please check if it is OK, or not):
0434.HDR3_0501_0123_GOLDEN.zip

Thanks,
György

That's fantastic, congratulations! 

It looks like the last failure occurred due to some freak glitch in the discharge cycle (if there are any discontinuities in the discharge cycle, this can cause a failure), and you're all set to go.

You're in luck with the WinCE 6.0, as I am currently developing the drivers necessary to communicate with and re-flash our gauge during production.  I'll let you know when I'm done with an alpha version to share with you.

If you're interested, I have just finished developing a re-flashing toolchain using an MSP430 for the purpose of going to production.  It does not involve WinCE, but it might be a good reference.

Cheers,

Charles

Hello again,

I've finished the driver development.  We used an AM3517 EVM for our base system and our code is based on that BSP, so it probably won't be 100% portable with your platform.  However, if you're interested, I can email you the project files for our driver and test application for your reference.

What platform are you using?  Also, what part of the driver development do you need the most help with?

Cheers,

Charles

Hello,

Richard, our SW developer had the following answer:

Hello!

We use the BSP of Atmel's AT91SAM9G45 board on Windows CE 6.0. We need the most help on how to communicate with the chip via i2c protocol.
It would be good to have your driver code and test application to see how it works and how difficult to port it to our platform.

Thank You in advance:
Richard

Thanks in Advance,
György