OPA541: Damage at startup with an inductive load

Hi Gilbert,

Please provide the following information:

1. How do +Vs and -Vs rise in relation to each other, and individually during power up?
2. Is a signal present at the "Signal" input before the power supplies are applied? 
3. What are the "Signal" characteristics?
4. Is R67 (RCL) 0 Ohms, or some other value?
5. Have you observed the voltage across the output load during both start-up and normal operation with a DSO?
6. Are there power supply decoupling capacitors connected at the OPA541 +Vs (+24 V) and - Vs (-24 V) device pins to ground? If so, what are the capacitor values?
7. How were you able to determine that the damage is being incurred by the OPA541 Sense Current function?

The fact that damage is occurring with this highly inductive load suggests that a high-voltage, back EMF is being generated by the load which is exceeding the breakdown capacity of the OPA541 internal current sense circuit. Your responses to my questions should help us determine if that is happening.

Regards, Thomas

Precision Amplifiers Applications Engineering

Hello Thomas,

Here are our measurments :

1. How do +Vs and -Vs rise in relation to each other, and individually during power up?

+Vs rise 60ms after -Vs after power up

2. Is a signal present at the "Signal" input before the power supplies are applied?

Yes during the setup of the power supplies the input is set to -12volts during 200ms

3. What are the "Signal" characteristics?

the signal input is -12volts during 200ms and after 3.3volts

4. Is R67 (RCL) 0 Ohms, or some other value?

R67=150mili ohm

5. Have you observed the voltage across the output load during both start-up and normal operation with a DSO?

no we have to do the job

6. Are there power supply decoupling capacitors connected at the OPA541 +Vs (+24 V) and - Vs (-24 V) device pins to ground? If so, what are the capacitor values?

10µF with in parallel 100nF

7. How were you able to determine that the damage is being incurred by the OPA541 Sense Current function?

It was just a supposition. I have an OP541 that is KO I can make mesure  of impedance of pin if you want.

Gilbert DURAND

HI Gilbert,

Here are some thoughts based on your responses:

2. Is a signal present at the "Signal" input before the power supplies are applied?

Yes during the setup of the power supplies the input is set to -12volts during 200ms.

As Kai mentioned this could be the problem. An high input current can sometimes flow in unexpected places internal to an op amp if an input voltage is applied before the power supplies are applied, or come up to their minimum levels. You might try increasing R69, R70, R85 and R86 by a factor of 10x and see if that fixes the problem. Increasing the resistances would limit the input current that might flow during power up. Better yet is if the input voltage weren't brought up to level until the OPA541 is fully powered.

5. Have you observed the voltage across the output load during both start-up and normal operation with a DSO?

no we have to do the job

That information should help establish if the damage is due to an output condition, or other.

I do see from your schematic that you included the back EMF diodes from the output pins to each supply rail as shown in Figure 13. Clamping Output for EMF-Generating Loads. That is a good practice.

One thing that we have learned about driving a complex inductive load with a power op amp such as the OPA541 is that better over-voltage circuit protection is had by adding transient voltage suppressor (TVS) diodes from each supply pin to ground. Their function is similar to that of a zener diode, but they are more rugged and designed for repeated electrical surges in the circuits where they are applied.

Should a back EMF cause one or the other Schottky diodes on the output to conduct the current flowing through the diode would seek a path to flow through. If the voltage at the op amp supply pin rises high enough to cause the TVS diode to momentarily turn on, clamping the voltage to the diode's breakdown voltage and providing a path for the EMF surge current.

A TVS diode such as the Littelfuse 1KSMB27C looks like it would be a good choice for the +/-24 V supplies. The TVS diodes are connected to the op amp power supply lines as shown here:

Regards, Thomas

Precision Amplifiers Applications Engineering