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.

TLV3402: Glitch At Start-Up

Part Number: TLV3402
Other Parts Discussed in Thread: TLV3401, TLV3501, LMV7235

I have a customer that is seeing the following issue with the TLV3402:

There is a glitch at 25 ms where the OUT signal is pulled down. The output is pulled up with a 10 kOhm resistor. We observed this behavior consistently at -40 C but also sometime at room temp.

This is how it is configured:

Vcc is tied to +2.7V rail.
IN+ is tied to a +2.048V voltage reference (REF3320AI) with a 100 kohm resistor.
IN- is tied to the output of an opamp (TLV2401I) that is monitoring a 60 Hz signal with a 47 kohm resistor.
OUT is pulled up with a 10 kohm resistor to +2.7V. OUT is also tied to an inverter input (74LVX14) through a 499 Ohm resistor.

The waveform I sent has the +2.7V rail (Vcc). As for the IN- glitch, this is actually caused by the output of the comparator which drops before the glitch on IN-. As for the unused comparator, we tried tying one input high and one low but this did not correct the problem.

They have removed the part from the application, but I am wondering if this is “glitch” is something that we have seen before. They say the new comparator is not showing this issue. It would seem to me that something else is happening, but there does not appear to be any issue with the pull-up voltage. The (+) and (-) inputs do not seem to ever reach the point where the output would change and yet you can see from the scope shot that it is indeed happening.

Thanks for your help with this!

Richard Elmquist

  • Hi Richard,

    What is the "new" comparator? Is it faster? Higher supply current?

    My guess is that they are seeing the start-up time of the comparator (not to be confused with the prop delay).

    The TLV3401 is a <1uA comparator, which means it is inherently slow (30us) and has low internal biasing currents (in the nA's), thus there is not a lot of current to kick-start the bias circuit (which usually depends on a leakage to start). During this start-up time the output is not yet valid. My guess is that the internal biasing is finally coming up at 26ms, has an incorrect output at first, then corrects itself about 1ms later when it is finally is fully operational and awake.

    This is not unusual behavior. Higher supply current (faster) devices should start up proportionally faster due to higher internal currents.

    If the application does not require <1uA supply currents (and a 10K pullup suggests it does not), then a faster device should be used instead of a nanopower device.

    They can try a LMV7235 or TLV3501 (P-P output - no output pullup needed), which are faster and should start up much faster.

    Regards,

  • Paul,

    This is what I thought, but thanks for the thorough explanation.

    I will make the product selections to the customer.

    Thanks for your help!

    Richard Elmquist