• State
  • Locked Locked
  • Replies 4 replies
  • Subscribers 47 subscribers
  • Views 437 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

TLV9062: Bad batch consistency

wei zhang
Prodigy 20 points
Part Number: TLV9062
Other Parts Discussed in Thread: TMS320F28035, AMP-PDK-EVM

Tool/software:

TLV9062IDR OPA circuit.docx

TLV9062IDR is used in our micro-inverter project as current sense and AC voltage sense.  The circuit is as attached .

About 4000pcs TLV9062 are used in one batch.  over 30pcs units failed in the board level test without inputting the power load.

We measured the impendence from each pin of TLV9062 unit to GND on the board and take the chip from board, mearsure the naked chip,  The impendence of each pin of suspicious chips is same with the normal chip.

Mostly the failure happened in the two case:

1, the current sense of zero load current.

2, AC line voltage detection. (230Vac)

The detail parameters are referenced at attached.

Pls help to check if the design has some potential risk in batch production due to the dispertion of OPA parameters. In the datasheet , normally the typical parameters are filled without min/max value.

How to optimize the TLV9062 design for current and AC line voltage detection?

  • 0
    TI__Mastermind 28570 points

    Hi Wei,

    What is after the 510ohm resistors of these circuits? It says ADC but is there some capacitor or other circuitry? 

    When there's no load, I'm assuming that I1- is 0V? 

    From what you described, it's almost as if the op-amp isn't able to have a common mode voltage close to the ground supply rail. Usually that only happens when the op amp is driving a lot of current. I don't think that's what's happening here but I'd be interested in seeing if there's a voltage drop across the 510 ohm resistors. 

    We have an AMP-PDK-EVM that you can use to isolate the op amps and see if there's any other differences. We are hoping to have the AMP-PDK-SOIC-8 socket card hopefully next week. This would be very helpful for debugging the parts. 

    Best Regards,

    Robert Clifton 

  • 0 in reply to Robert Clifton56
    Prodigy 20 points

    Hello Clifton,

    Very appreciated for your reply. I am looking forward to get more professional teaching .

    TMS320F28035 is after the 510 ohm resistors of these circuits. It's ADC pin (pin 17) of 28035 ,and there is one capacitor - 10nF (C148).

    Yes, I1- is 0V ,when there is no load.

    I described that op-amp isn't able to have a common mode voltage close to the ground supply rail. Ideally when input of op-amp is 0V, Vout of opa-amp is 0v. Usually we can get 0.01V for batch production by the previous 2000 sets mass-production.
    But sometimes we can find 0.5Vdc for some abnormal case.

    Is it possibly caused by the discrete properties of Ib, Vos, Ios (all these parameters will after the Vos of Opa-amp)? I am looking at the problem in the batch production. My product will use 100ku+/month TLV9062 in the later. My concern is the dispersing .

    Will measure the voltage drop across the 510 ohm resistor and feedback to you next week.

    Best Regards,

    Dave

  • 0 in reply to wei zhang
    Prodigy 20 points

    PIN 12 17 18 of TMS320F28035 are used as ADC input. Pls notes it.

  • 0 in reply to wei zhang
    TI__Mastermind 28570 points

    Hi Wei,

    Would you be interested in using the AMP-PDK-EVM to help debug this? It would help isolate what might be causing this. 

    The datasheet says under very low output current operation, the op amp can get very close to the supply rails. See screenshot from the datasheet below:

    One idea is since you see 500mV from the rail, that would suggest that the op amp is sinking or sourcing current. However there's a few other explanations such as the input to the op amp is 0.5V. I'll await for you to read out the voltage drop across the 510ohm resistor before eliminating any possibilities. 

    Best Regards, 

    Robert Clifton