BQ34110EVM-796: Current calibration and measuring Issues

Hello,

Does this happen on all your boards? Can you outline the steps you take to calibrate the gauge? your offset seems very high.

Sincerely,

Wyatt Keller

Good morning Wyatt,

this is the only eval board I have.

I followed exactly the manual for the calibration, step by step.

Steps I followed:

1) Unseal and Unseal full access

2) CAL_Toggle to set CAL_EN high

3) In Calibration tab of BQstudio, performed only Temperature calibration for external sensor

4) Calibrated voltage only by measuring the battery terminals and wrote that value in Applied Voltage field (in mV)

5) Calibrated only CC Offset

6) Calibrated only Board Offset

7) Using electronic load of 2Amps, calibrated current only and wrote -2000mA in Applied Current field.

8) Sealed the gauge

9) Reset the gauge

When measuring discharge current the value is correct, but when charging for example 1Amp (1000mA), the measurement near 10000mA. 10 times higher.

If I calibrate current using positive current (charging), then while charging the current value is correct, but when discharging, the value is divided by 10. For example discharging at -2Amps, the value reported is -200mA.

I have attached the export file *.gg.csv of the data memory setting for you reference if you want to look at it. My battery is 3 cells 10.8V 20Ah Li-ion.

Thanks you

Regards

Max L.

BQ34110 Data memory for Current Cal Issue.gg.csv

Hello Max,

Can you try importing the defaults in the data memory tab and write all to see if the issue is still occurring? I would also try testing with another EVM to make sure the board is not damaged.

Sincerely,

Wyatt Keller

Hello Wyatt,

where can I get the defaults for all data memory? I mean the default *.gg.csv

Thank you

Regards

Max L.

Hello,

You can import the defaults by going to the bqStudio > data memory tab > import > import defaults.

Sincerely,

Wyatt Keller

Good morning Wyatt,

I loaded the defaults and the behavior is the same.

Were you able to test your board?

Thank you

Regards

Max L.

Hello Wyatt,

We have an old second board.

I connected with same setup and I am able to read all DATA memory using BQstudio. But when I try to write the complete settings or any single para meter gives me an error saying " A read of data written failed comparison". Nothing is written. When I read all para meter again, it shows the ones originally on the IC. 

Also when calibrating current and read current during charge or discharge it is always at 32767mA. It updates the voltage but the current is always 32767mA, even when not charging or discharging.

Is there anything I am missing? Or maybe that one is damaged too?

Thank you

Regards

Max

Hello Max,

Are you using the correct .bqz for the gauge firmware? is the gauge unsealed and full access when you try to edit the parameters? Those are some of the common issues that could cause this.

Sincerely,

Wyatt Keller

Hello Wyatt, yes the device is unsealed and full access.

When I open BQstudio, automatically enters to the IC GUI. I don't have to select, but I do know that in config folder of BQstudio there are two bqz's

0110_0_01-bq34110.bqz

0110_0_02-bq34110.bqz

Should I leave only one?

Thank you

Hello Max,

No that should be fine to have both if it auto detects, let me check with my colleague about this issue.

Was the gauge ever operational? Or did this start randomly?

Sincerely,

Wyatt Keller

Hello Wyatt,

we never tested full operation until now.

Now I have two eval boards that don't work.

I received already 2 replacement IC's. If you tell me I can replace them on the board, but need to be sure that this would fix the issues

Do you think we can replace the IC's?

Thank you

Max L.

Hello Wyatt,

The connection of battery/EVM board/Charger is correct according the EVM guide.

Attached is registers screenshot for Registers and Bit registers

Hello again Wyatt, please notice the current field is not corrected after I performed calibration. I toggled CAL_EN before perforning calibration.

Current field is always 32767mA no matter is is charging or discharging.

Thank

Hello Wyatt,

there is not a wrong connection. If I use the other board I have, I get measurements but the problem I told you at beginning,

In summary I have two Eval boards:

1) The first one gives always maximum value and does not allow to write data memory registers, and does not calibrate

2) the second one allows to read/write data memory registers and perfom calibration but for current for discharging reads fine, but for charging multiplies value by 10

Both using same connection as the Manual

So don't know what else to do here

Regards

Max L.

Hello Wyatt,

thank you for your answer.

I will do some more tests tomorrow and let you know, If I don't have a good progress then will replace the IC. I will get new ones today afternoon.

Thank you so much again

Regards

Max L.

Good morning Wyatt,

hey look, I replaced the IC on the board that was not writing updating the data memory. Now i can write on it.

But after calibration using -2000mA, when measuring the values in Register window looks fine, but when charging the current value is multiplied x10 + something else.

So now I have two boards performing identical, same issue both. Please look the screenshots. Any advice will be really helpful.

Thank you

Hello Max,

Can you re-flash the .srec firmware to the gauge with the most recent? Also make sure all the connections on the EVM are correct, can you share an image of how it is connected?

Sincerely,

Wyatt Keller

Good morning Wyatt,

thank you for your advices.

I programmed the 34110 with the lastest .srec found in the TI web site ( bq34110_v0_02_build_9.srec ).

When I clicked on the Execute FW button in the Programming window said at the left low corner on the screen ( Failed to execute firmware - No acknowledge from device)

So I did RESET and read current values before calibration just as reference:

BEFORE CALIBRATION:

I noticed that by default CC Gain is 9.999 mOhm and CC Delta 10.000 mOhm

1) When run charging the current almost match my charger, meaning 0.5A and the Current value on  the Registers window is 494mA, which is fair close.

2) When running discharge at constant current of 2A, the Current value on the Registers window is -223mA, that's bad since should be close to -2000mA. It is dividing by 10.

AFTER -2000mA CURRENT CALIBRATION:

I noticed when reading Data Memory that CC Gain changed to 1.125mOhm, CC Delta changed to 1.137mOhm, CC Offset changed to -45.99mA

1) When discharging at constant current of 2A, the Current value on the Registers window is now -2000mA, which is excellent.

2) When running charge at current of 520mA, the Current value on the Registers window is now 4750, that's bad since should match the real current of 520mA. It is multiplying by 10 and substracting the CC Offset multiplied by 10

4750 + (-45.99*10) = 5210 mA  this is 10 times the real value of 520mA

I do not know what else to do, if there is some other settings I am missing

REGARDING THE CONNECTION I HAVE

Please check picture below: Let me explain how is it made:

• Battery connected to Bat + and Bat -
• PC controlled Batt Load connected to Bat + and Pack -. This device is also a monitor of charging/discharging AmpHr, Current and Voltage
• Charger is connected to Bat + and Pack - , but in the middle a PC controlled relay to connect or disconnect the charger to the system. So when running discharge using the BATT LOAD, the charger is electrically isolated.
• The PC software used to handle the Batt Load and the relay controls both simultaneously.

Please let me know what else to do if you have any idea.

Thank you so much for your help

Regards

Max L.

Hello Wyatt,

Thank for your reply. I am going to connect a multimeter in both paths, charger and load in order to verify the current supplied or drained.

Will let you know soon.

Thank you

Regards

Max L.