Op Amp Input Protection

[juz_ad]

I was given some advice last year on another forum about some basic, general purpose input protection methods for op amps. The schematic above shows what I ended up with and is based on a TL07x that is part of a circuit powered by a +/-12V supply.

Inputs are designed to *expect* an AC or DC signal of up to +/-5 Volts from control voltage through to audio frequencies.

It should serve two main functions: 1) If somebody plugs something into the Inverting input that goes above/below the supply rails e.g. a -15V signal into the input while IC1 is powered by +/-12V and 2) If somebody connects some type of input to the Inverting Input.

I know that's a wide brief. It's not designed to be bullet-proof and I know that trying to protect against non-specific things is difficult, I also acknowledge that there are possibly multiple techniques to choose from and *lots* of opinions.

I wanted to run it by the people here. Am I protecting against anything with this? Causing more problems than I'm solving? Missing something? I'd appreciate your input.

Thanks.

 

[jiripolivka]

The diodes on the left do protect the opamp from voltage peaks exceeding the +/-12 V rails. The diodes between opamp inputs limit the difference voltage to +/- their forward voltages. The opamp itself clamps output voltage to +/- 12 V rails. What else you expect ? The TL series opamps are also protected for load shorts; you can improve opamp behavior with capacitive loads by adding a series resistor ~100 Ohms from output to load.

 

[juz_ad]

That's my understanding of it as well, however a couple of people have suggested that with an input at audio frequencies there could be some current leakage problems using this config.

How much of a problem? I don't know - that's one of the things I'm trying to get some second/third opinions on.

Thanks.

 

[jiripolivka]

That's my understanding of it as well, however a couple of people have suggested that with an input at audio frequencies there could be some current leakage problems using this config.

How much of a problem? I don't know - that's one of the things I'm trying to get some second/third opinions on.

Thanks.

If your application is audio only, then you should put an input capacitor in series with the 100k resistor, and there will be no current leakage. The same capacitor at the output. Depending on the lowest frequency limit, 0.1 to 1.0 uF can be good.
To amplify the DC component by opamps, under high-gain conditions, input current and offsets are sometimes a problem.

 

[juz_ad]

If your application is audio only, then you should put an input capacitor in series with the 100k resistor

Unfortunately the input needs to be DC coupled, It needs to accept everything from DC control voltages through to AC voltages at audio frequencies within a +/-5V range.

 

[jiripolivka]

Then the diodes in your schematic can protect the opamp from any damage; only adjust the gain to prevent clamping.

 

[erikl]

... with an input at audio frequencies there could be some current leakage problems using this config.
How much of a problem?

No problem; your opAmp is very well protected.

 

[0by1]

Output is limited to -0.7v to 0.7v (with unity gain), is that ok?

 

[erikl]

Output is limited to -0.7v to 0.7v (with unity gain), is that ok?

No, it is not limited to +/-0.7V at the output, only at the input! In this inverting opAmp circuit, the input difference is max. -Vout/Aol ; Aol being the open loop gain of the opAmp. I.e. with ±5V output voltage you have only -/+5mV input difference (if Aol is only 1000; usually it's much larger).