• State
  • Locked Locked
  • Replies 6 replies
  • Subscribers 49 subscribers
  • Views 1042 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.

Biasing of THS4631 and use as comparator

Saad Pervaiz
Prodigy 120 points
Other Parts Discussed in Thread: THS4631, LM119

Hi,

I am looking for a high speed comparator with very low input offset voltage, low propagation delay and high voltage operational capability (Supply and input differential voltage capability of >12.5 V). I was looking into using THS4631 for this purpose.

Are there any other comparators/amplifiers that have similar characteristics, i.e. input offset voltage <1mV, very high slew rate and high voltage operational capability.

Secondly, can THS4631 be operated with uneven supply voltage, i.e. can I use this with +12 and -3 supply voltages (so that I  can have near zero output voltages) and expect similar performance?

Thanks,

Saad Pervaiz

  • 0
    TI__Genius 12631 points

    Hello Saad,

    It is not recommended to use an op-amp as a comparator. There are 2 primary reasons for this

    1. There may be back-to-back diodes between the amplifiers inputs that will act as a clamp and load the previous stage.

    2. An opamp's output stage is not meant to be clamped and can take a long time (several micro-sec) to recover from on overload condition.

    These are rules of thumb. At the end of the day the architectures of amplifiers are different and its best to try this out in the lab rather than depending on the datasheet specified conditions to see what can be achieved in terms of speed and prop delay.

    The THS4631 can be used with asymmetric supplies like you mentioned, however what is your proposed threshold voltage and input swing?

    What kind of propagation delays are you looking for? Also are you planning on operating on 15V supplies? There seems to be some trouble with the Comparators web portal on ti.com. I will keep checking it periodically and once it is fixed I can send you a list of possible options.

    Regards,

    Samir  

  • 0 in reply to Samir Cherian
    Prodigy 120 points
    Hi Samir,

    Thanks for your reply.
    I plan to use +12V and -3V supply. I am using a negative supply just to get a clean logic zero.
    The propagation delays that I am looking for are less then 15ns.
    I intend to sense the drain-source voltage drop of a power FET, and use it to determine the direction of the current. So I want to sense very small voltage differences, of the order of ~1 mV. But when the FET turns off, the voltage between the input terminals will go to 12V.
  • 0 in reply to Saad Pervaiz
    TI__Genius 12631 points

    Saiid,

      1mV differences are going to be a challenge because of the offset voltage. For example the THS4631 has a 0.5mV offset voltage which is half your budget. Also, when an amplifier comes out of overload I am not sure how long the device is going to take to regain its original "linear" offset. We don't characterize the part this way so its something you are going to have to try out. The next thing to consider with the THS4631 is the input common-made and output swing limitations. The input common mode for +/-15V supplies is -13V to 12V...so for your case using +12/-3V supplies this is going to be +9 to -1V. This means that your inputs should stay within this range. On the outputs, the limitations are +/-11V on 30V supplies....this indicates that your output is only going to swing between 8V and +1V. Im not sure if this is acceptable. You were not very clear on what your input signal is going to look like...is it 1mV difference on 0V common-mode or some other common-mode (dc bias) voltage?

    The LM119 and LM319 are two high speed comparators which have around 100ns of prop delay, however the offset voltage are around 4mV (max). The response times usually get faster with increased overdrive.

    This is a tough situation where you need the speed of a comparator and the offset voltage of an opamp. The high supply condition also makes this a difficult situation.

    -Samir

  • 0 in reply to Samir Cherian
    Prodigy 120 points
    Samir,
    Yes, the 1mV is on a common mode of 0V. Since this is connected from Source to Drain of the Power FET, when the FET is off, the inputs are pulled apart by 12V. When the FET turns on, the inputs are separated by the IR drop in the FET.
  • 0 in reply to Saad Pervaiz
    TI__Genius 12631 points

    Saad, Ok just to be clear

    - When FET is OFF inputs are pulled apart by 12 V. Can I assume drain is connected to 12V?

    -When the FET turns on, the inputs are separated by the IR drop in the FET. If my above statement about drain connected to 12V is correct, then I assume the source will be some voltage close to 12V but lower than 12V....correct?

    Where does the 1mV signal on 0V common mode come in?

    -Samir

  • 0 in reply to Samir Cherian
    Prodigy 120 points
    Yeah, so when the FET is off, its Drain is connected to 12V. So the inputs of the comparator will be pulled apart by that much.
    But when the FET turns on, the drain and source of the FET are pulled apart by the IR drop. Source is always connected to the ground. So the drain is also pulled close to the ground from 12V.
    I want to sense the change in the direction of the current. Since my IR drop is small, I want a comparator that will be able to flip its state for very small input voltage differences. I think that capability is limited by the offset voltage at the input for such precise voltage differences. Thats why I want the input offset to be as low as possible.