OPA567: Current Limit Iflag Problem

Hi Dominkovic,

Very good documentation of what you are observing with the OPA567 bridge circuit. Can you tell me the edge rate of the input pulse edge rate? It is hard to tell from the DSO diagrams. Additionally, would it be possible for you to try the same test with just a 2 Ohm resistive load on the output?

Regards, Thomas

Precision Amplifiers Applications

Hello Thomas,

thank you. The Edge rate (10%-90%) is About 100ns. I attached some DSO diagrams.

Blue: Input Voltage

Yellow: Iflag Signal

Green: Coil current

I tried the same test with just a 2Ohm resistive load on the Output --> no Current Limit triggering could be observed

How does the inductive load influence the current Limit protection and what could I do to Prevent this effect?

Could there be a high frequency current component that returns into the amp and triggers the current Limit.

Thanks,

Regards,

Branimir D.

Hello Brainmir,

This may be difficult to find. I suspect it has something to do with the higher frequency content of the input step and how it is traveling through the OPA567 BTL circuit with the R+L load, or back EMF due to addition of the inductive load. It may be that harmonics from the step are getting back into the OPA567 internal ILIM comparator circuit and causing the chatter - much like what can happen with a comparator that doesn't have any hysteresis. It would be helpful to know if slowing down the input step edge reduces, or eliminates the problem.

Here are some suggestions of things to try:

1. Add some capacitance from the IFLAG and ILIM pins to ground. You can try the IFLAG and ILIM pins separately. I suggest something on the order of 1 to 10 nF. See if that corrects the IFLAG behavior.
2. Add the back EMF diodes to each OPA567 output. See datasheet FIGURE 5. Output Protection Diode.

Do note that the current and voltage phase relationship is completely different for the purely resistive and the R+L loads. The current and voltage are in phase for the resistive load, but become shifted from each other as the frequency increases with the R+L load. I am not sure if that has something to do with the IFLAG behavior, but a point to keep in mind.

Regards, Thomas

Precision Amplifiers Applications Engineering

Hello Thomas,

thank you.

I already placed 4x 1N5819 as described in the datasheet, but no improvement.

I am on my holidays now. I will try to slow down the input step rise time and to put some caps on IFLAG and ILIM Pin when i come back in 10 days.

Regards,

Branimir

Hi Branimir,

Okay, adding the Schottky diodes to the OPA567 outputs didn't change anything so it looks like we can rule out back EMF. I didn't think there was a high probability that would resolve the issue, but it needed to be tested. It is always a good idea to have the diodes in place as a protection measure.

It is best to not leave an e2e inquiry open without responses happening for a long time such as 10 days. It can degrade our response metrics. If you can close this particular inquiry  for now and open a new e2e, referencing this one at that time, it helps keep our metrics from getting dinged.

Enjoy your holidays! I'll wait to hear from you until you get back.

Regards,, Thomas

Precision Amplifiers Applications Engineering