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.

OPA547: OPA 547 : Problem with current limiting

Part Number: OPA547
Other Parts Discussed in Thread: OPA548

Hello,

I realized a 4Vrms 1.2kHz sinus by using an OPA547 (supplied in +/-15Vdc) with a gain of 3.

I have set the current limit to approximately 350mA by adding a 37k4 resistor between pin 3 and 4.

But when I test the current limitation by putting the output in  short circuit, I have the following behavior:

- When the sinus goes to negative voltage, current is limited to -350mA (functioning is OK)

- When the sinus goes to positive voltage, current ripples between 0 and 750mA at a 1MHz frequency

I have test the same function by injecting a positive or negative dc voltage and I have the same behavior

Is it normal?

  • Hello user4966318,

    The OPA547 current limit circuit should activate at nearly the same current setting for both the positive and negative limits over its output swing range.

    It is plausable that the positive output current limit circuit may have been damaged. Have you observed this behavior with more than one OPA547 device? Also, note that with the output shorted that the power dissipation can be very high. If the proper thermal requirements have not been observed the OPA547 output might be going in and out of thermal shutdown very rapidly which appears as oscillation. Does the OPA547 circuit work properly with the expected load?

    Regards, Thomas

    Precision Amplifiers Applications Engineering

  • Hello Thomas,

    Thanks for your answer,

    Following your message, I understand that there's  two current limit circuits in the component, one for positive current and one for negative, am I right?

    In this case, this could explain why the negative current limit works correctly and not the positive.

    Thomas Kuehl said:
    Does the OPA547 circuit work properly with the expected load?

    Yes the circuit work properly with the nominal load (200mA).

    As I had this problem on my two boards (with one of them which was never used), I thought that the problem came from my scheme.

    Following your answer I suspect now an issue with the soldering process.

    I'll replace the component on the board and make some tests


    Regards, Arnaud

  • Hello user4966318,

    You are absolutely correct about the OPA547 having two, independent current limit circuits; one for the positive swing and one for the negative swing. That is why the positive and negative current limits won't be precisely the same; however, they will be close. You may want to try checking the current limit behavior with a resistive load rather than a short, and see if the current limiting is more in line with what you expect. Using a resistive load will lessen the internal power dissipation compared to the output shorted condition.

    It is imperative that the thermal guidelines provided in the OPA547 datasheet be followed. If they aren't, it is very easy to get into situations where the thermal shutdown is activated unexpectedly.

    Regards, Thomas

    Precision Amplifiers Applications Engineering

  • Hello,

    First of all, thank's Thomas for your reactivity,

    I made some tests on the board with several load on the output in order to see the current's form,

    Here is the results (Voltage curve in Green, Current curve in Blue 1A=1V):

    At nominal current (20 Ohm) :                                                                                      

     With a 10 Ohm load:

    With a 5 Ohm load :                                                                                                      

      With a 3.3 Ohm load:

    Positive current limitation is functioning well when the load is above 5 Ohm (around), but when the load is smaller, the ripple come again.

    I have done this test one more time with a new component (TO-220 package with dissipator instead of DDPAK mounted on copper area for test above) wired on separate board and I found the same behavior.

    I don't find in the datasheet any note about a minimal load for the current's limitation to work.

    Could  you confirm if the current's limitation can work on short-circuit (it's mandatory for my function)?

    Thomas Kuehl said:

    It is imperative that the thermal guidelines provided in the OPA547 datasheet be followed. If they aren't, it is very easy to get into situations where the thermal shutdown is activated unexpectedly.

    During the test, I checked the E/S pin and the voltage was stable at -11.4Vdc (thermal protection not activated).

    Regards

    Arnaud

  • Hi Arnaud,

    So it appears that the noise/oscillation is observed on the output current waveform, both on the negative and positive portions? I do find it curious that the noise/oscillation doesn't appear on the voltage waveforms until the load resistance is down at 3.5 Ohms. It would seem that it would appear on both. What instrument are using to measure the output current, or are you using a current sense resistor?

    Would it be possible for you to share your schematic. I don't suspect that there is an issue with the way your are applying the OPA547, but I want to make sure there isn't something that hasn't been overlooked.

    Some memory I have of testing we did during the development of the OPA547/OPA548 EVM a couple of years ago is that we had observed a noise/oscillation under some output current limit conditions. I need to talk to the person that did the work and see if they have the same recollection.

    Regards, Thomas

    Precision Amplifiers Applications Engineering

  • Hello, Thomas,

    Thomas Kuehl said:
    So it appears that the noise/oscillation is observed on the output current waveform, both on the negative and positive portions? I do find it curious that the noise/oscillation doesn't appear on the voltage waveforms until the load resistance is down at 3.5 Ohms. It would seem that it would appear on both. What instrument are using to measure the output current, or are you using a current sense resistor?

    No, the noise you see on the blue curve, is mostly due to my current probe which is not dimensioned for that range of current (it's a 30A current probe) and my scope don't allow me to filter that noise.

    That's why we have this difference between the green and the blue curve.

    The green curve is most representative of the output's behavior.

    Here's an extract of my scheme:

    C708 and C709 are 100nF capacitors

    R708 was updated to 21k Ohm

    Regards,

    Arnaud

  • Hello Arnaud,

    Each of the two OPA547 output current limit circuits have local feedback loops. I suspect the positive current limit circuit is becoming unstable  the OPA547 output exhibits the oscillation on the positive peak with the 3.3-Ohm load. There are two things you might tray if the oscillation during certain load conditions is an issue:

    1. Try bypassing Pin 3, the current limit resistor connect pin with a 10 nF capacitor to ground.
    2. Add a series RC snubber from the OPA547 output to ground. Something on the order of 3.3-Ohms and a 10 nF in series. It should be connected right from the output pin to the ground plane close to the output pin.

    Please let me know if either of these make any difference.

    Regards, Thomas

    Precision