• State
  • Locked Locked
  • Replies 2 replies
  • Subscribers 46 subscribers
  • Views 1021 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.

TLV3701 Issues

Richard Elmquist
Mastermind 45250 points
Other Parts Discussed in Thread: TLV3701

I have a customer who is seeing the following:

We are currently using the TLV3701IDBVR comparator. When we startup the comparator with 3.3V it outputs a logic high (for 300us to 2400us) even though the negative input is driven high and the positive input it attached to ground. Additionally the power on time seems to vary widely based on the power down duration.

There are two scope shots that he sent:

CH1 = In+

CH2 = In-

Ch3 = Vcc

Ch4 = Output

The time frames in the scope shots are different. He explained it below:

The only difference is the time I waited between powering up the IC from the last power down. If I wait only 5 sec, I will see a longer time (2.4ms). If I wait 30 sec, I will see a shorter time (200us).

This seemed to imply to me that there was a large capacitive load. He describes the load below:

There are 3 IC that are on the load with no additional capacitance. The load is high impedance (>10M). The issue has been seen on 2 of 2 boards that we checked.

I do not understand how this is happening. Is there something internal to the IC that takes a certain time span before it outputs the proper output voltages based on those present at the input. The output does become proper, but shouldn’t it do this when it powers up also.

Please let me know if you have further questions for the customer.

Thanks for your help with this!

Richard Elmquist

  • 0
    TI__Guru* 93105 points

    Hi Richard,

    The power-up behavior of comparators is seldom described, or shown, in the datasheet. I suspect that in the earlier days of soild-stare circuits there was some time allowances for everything in a system to come up to a steady-state level and the very short start-up times of the indivdual ICs wasn't needed. But by the number of inquires we get about op-amp and comparator start-up it appears that the need to know just exactly when and how the device will start is changing.

    Every op-amp and comparator will have its own unique start-up behaviors and there certainly will be variances over time, between different production runs of the device. Since many datasheets don't provide the information the user may be surprised by what they observe in the first few tens-of-microseconds to milliseconds depending on the device. For many comparators, including the TLV3701, it is not unusual to find the output in the opposite state from expected, or flipping states before the correct steady-state output state is had. After that it functions as expected following a comparison of the two input levels.

    The TLV3701 start-up behavior observed by the customer doing a power ON-OFF-ON cycle has been observed before. What we think is occuring is that the first ON cyle has a particular start-up characteristic as produced in the upper DSO image. Then the power is shut OFF. That doesn't mean that all of the internal transistor junctions are immediately discharged of any remaining charges. Some junctions may have significant charge stored while others have lost all their charge. The charge distribution from comparator input to output stage is now different than when the device has been OFF for a long time. In that case, all of the charges have dissipated. These pre-charged nodes result in a different start-up characteristic the next time a power ON occurs - providing that happens shortly after the last power OFF.

    The timing results the customer observed, "The only difference is the time I waited between powering up the IC from the last power down. If I wait only 5 sec, I will see a longer time (2.4ms). If I wait 30 sec, I will see a shorter time (200us)," supports our understanding.

    There isn't anything that can be done to change the comparator's behavior because it indicative of the comparator design, and the semiconductor technology used to fabricate the device.

    Regards, Thomas

    PA - Linear Applications Engineering

  • 0 in reply to Thomas Kuehl
    TI__Mastermind 45250 points

    Thomas,

    Thanks for your thorough response!

    I thought that this was the case, but needed to hear it from the experts.

    Thanks again.

    Richard Elmquist