• State
  • Locked Locked
  • Replies 8 replies
  • Subscribers 65 subscribers
  • Views 2989 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

BQ34Z100-G1 - current calibration problem

Slawomir Tyborczyk
Prodigy 125 points
Other Parts Discussed in Thread: BQSTUDIO, BQ34Z100EVM

Hi, I recently bought bq34z100evm in order to test its possibilities for use in a custom project. I'm using the evalboard in combination with TI's EV2300 USB interface, bqStudio v 1.3.52 and a connected 7Ah 12V PB Acid battery. The EVM is connected via I2C bus and seen in program as bq34z100G1 0100_0_16.

My problem concerns current calibration - I should first run the CC and board offset calibration, with battery connected, but without any load or charging. As far as i know, this step should set the current meter in the Dashboard tab to zero. Instead, the current meter shows about 200 mA, even with current shunt resistor shorted. Current calibration with connected load also doesn't seem to work properly. The only thing I managed was the succesful voltage calibration.

Any ideas?

  • 0
    Prodigy 125 points

    Help, anyone? I'm starting to think that no one has ever run this chip without any problems while reading this forum.

  • 0 in reply to Slawomir Tyborczyk
    Intellectual 720 points

    I had no problems calibrating the current on the bq34z100evm like described in SLUU904A (www.ti.com/.../sluu904).
    Without load bqStudio shows 0mA.
    I'm sorry i can't really help. I would check if the batteries are connected correctly as already described somewhere else in the forum or in the document linked above and double check if the Jumpers are correctly set on the demoboard.

  • 0 in reply to Sebastian M
    Prodigy 125 points
    Yeah, I double-checked all the connections, went carefully through the calibration procedure described in SLUU904A and still, without load, bqStudio shows around 205 mA.
    I'm starting to wonder if the evalboard is damaged.
  • 0 in reply to Slawomir Tyborczyk
    Expert 1132 points

    I've used the Eval module  with EV2300 as well as the chip on my own PCB without this particular problem. I have had issues with some external test instrumentation injecting a DC offset that caused a similar problem but I believe you indicated that you didn't have anything else connected but the battery and the EV2300 right?

    I would suspect that you have B- and Pack- swapped but you indicated you checked the connections thoroughly.

    If not I suggest checking to be sure you have a good USB cable. If you loose ground there, it could possibly cause a ground/current loop.

  • 0 in reply to BatMan
    TI__Guru 73830 points
    The device should report 0mA when there is no load present. Make use the the battery is connected between BAT+ and BAT- and the jumpers are configured properly. If you can zero the current with the CC Offset and Board Offset calibration features, then the gauge may be damaged. You could try removing the device from the EVM and soldering a new one.
  • 0 in reply to ThomasCosby
    Prodigy 125 points

    I replaced the IC earlier previous week, but with no effect.

    The photo I took on Friday explains everything.

    I was about to sent the EVM back to Mouser but decided to take a final look to the board, especially the current measurement circuit. The lower 100 ohm series resistor is barely soldered to the left pad. The resistance between two pads was about 2 Mohms. No wonder the gauge didn't work properly. After applying a fresh portion of solder current calibration run smoothly.

    In general, the quality of solder joints on the PCB is terrible. What's more, it didn't work out-of-the box. I don't know if TI has any decent standard of quality checking, but IMO this PCB should actually be resoldered. Damn PB-free soldering. I bought an EVM to speed up development stage of my project and lost a week of work trying to figure out what I'm doing wrong. Never thought you could apparently sell untested boards like this one.

  • 0 in reply to Slawomir Tyborczyk
    TI__Guru 73830 points

    Slawomir,

    I am sorry about the problems and that it turned out to be a faulty EVM. The EVMs are tested by our CM and this defect should have been caught. Can you provide a picture of the bar code on the back of the EVM and label on the box? We will investigate the failure and add corrective actions.

    Tom