6
\$\begingroup\$

I was taking a look on an op-amp datasheet (OPA2810) and I saw the following parameters: "Allowable input differential voltage".

The op-amp woud be used as a difference op-amp, so as there will be a feedback there is no reason to have a voltage difference between the inverting input and the non-inverting input. Nevertheless, at power on the difference could be higher than what is allowed before the op-amp sets its output to set no difference voltage between the non-inverting input and the inverting input.

Will it damage the op-amp if the allowable input differential voltage is exceeded during power on?

enter image description here

enter image description here

\$\endgroup\$
7
  • 1
    \$\begingroup\$ What is the part number? Are those abs-max or recommended parameters? Generally exceeding the rate specifications of a part is not wise - if you operate a part beyond what it was designed/tested for there is no guarantee that it will continue working as intended. \$\endgroup\$
    – Tom Carpenter
    Dec 17, 2022 at 14:22
  • 1
    \$\begingroup\$ There is not enough information to formulate an answer, \$\endgroup\$
    – RussellH
    Dec 17, 2022 at 14:31
  • \$\begingroup\$ Read note (1) under section 7.1 Absolute Maximum Ratings. I believe the answer to your question is clearly stated there. \$\endgroup\$
    – user324996
    Dec 17, 2022 at 14:51
  • 1
    \$\begingroup\$ I edited in the link provided by @TomCarpenter, without which the post is quite incomplete (always state the source of the material you post, BTW - that's also to address copyright issues and fair use). \$\endgroup\$
    – LorenzoDonati4Ukraine-OnStrike
    Dec 19, 2022 at 13:00
  • 1
    \$\begingroup\$ "at power on the difference could be higher" are you sure of this, or is this just a theoretical worry? Did you actually analyze your circuit and determine that during power-up the input differential voltage can actually grow too large? \$\endgroup\$
    – LorenzoDonati4Ukraine-OnStrike
    Dec 19, 2022 at 14:21

3 Answers 3

Reset to default
6
\$\begingroup\$

According to the absolute maximum specifications, the differential input voltage can be at most, the lesser of \$\pm7[\mathrm{V}]\$ and \$\pm V_s[\mathrm{V}]\$. So if you have a supply of \$<7[\mathrm{V}]\$ total (where total means \$V_s = V_{s+} - V_{s-}\$) then you can't exceed that supply voltage.

These are absolute maximum ratings. As per the datasheet, stresses beyond these limits may damage the device. This could be immediate blue smoke, it could be a degredation in future performance, it could be reduction in part lifetime.

It doesn't matter if it's just "at power on". The manufacturer have stated this is not a condition you should operate the part in at any time. It is not possible to say whether the part will survive.

TL;DR; You should either select a different part that meets your requirements, or design your circuit to prevent the part being used outside its recommended operating conditions.

\$\endgroup\$
0
3
\$\begingroup\$

Tom Carpenter provided the right answer to your question. I'll add that if your circuit topology cannot guarantee that at power-up the max differential voltage limit can be satisfied, instead of changing the circuit or the part, as suggested by Tom Carpenter, you could add some input protection circuitry.

The simplest that come to mind is a pair of antiparallel rectifier diodes placed across the inverting and non-inverting inputs:

schematic

simulate this circuit – Schematic created using CircuitLab

This will clamp the differential voltage to about 0.7V until the circuit feedback loop kicks in.

Of course you should analyze the overall impact of those diodes. For example, their leakage current could affect the circuit behavior, especially on precision design. They will also add some capacitance, which may affect bandwidth.

When leakage is a problem often a diode-connected BJT is used, since its BC junction behaves as a low leakage rectifier (this technique is used in many measurement instruments, like DMMs).

\$\endgroup\$
4
  • \$\begingroup\$ A resistor in series? \$\endgroup\$
    – Circuit fantasist
    Dec 19, 2022 at 14:10
  • 1
    \$\begingroup\$ @Circuitfantasist I didn't show an entire circuit protection scheme, so there are lots of details missing, of course (series resistors to limit current is just one detail). Without seeing the OP schematic any further detail is just a shot in the dark. Mine is just a pointer toward the complex issue of input protection. \$\endgroup\$
    – LorenzoDonati4Ukraine-OnStrike
    Dec 19, 2022 at 14:17
  • \$\begingroup\$ This TI (Burr-Brown) application note suggests a diode-connected JFET for some applications. \$\endgroup\$
    – Theodore
    Dec 19, 2022 at 15:11
  • 1
    \$\begingroup\$ @Theodore Yep, that's another alternative. The problem is discrete JFETs are an endangered species nowadays, with a very little assortment to choose from. OTOH, discrete BJTs are alive and kicking, and even jellybean parts like the 2N3904 exhibit incredibly low leakage in this application (~1pA@10V for the BC junction). \$\endgroup\$
    – LorenzoDonati4Ukraine-OnStrike
    Dec 19, 2022 at 16:26
2
\$\begingroup\$

There is a risk of damage only if "ideal" voltage sources (circuits with very low resistance) are connected to the two op-amp inputs. Current is what damages devices. In practice, there is almost always some resistance in series protecting the input.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.