2
\$\begingroup\$

There is a related question similar to mine: Op Amp input protection, but I found this other configuration:

op-amp

where the series resistor is placed after the TVS diodes. What is its purpose? The input signal is expected to be near DC and low-impedance. So I don't think it has to do with bandwidth or input impedance/offset.

It could be a further protection of the op-amp inputs if an ESD event happens on J8?

\$\endgroup\$
2
  • \$\begingroup\$ Maybe its role is to compensate the influence of the input bias currents... although the resistors have too low resistance... \$\endgroup\$ Commented Apr 17, 2021 at 8:33
  • 1
    \$\begingroup\$ It just limits current if the input exceeds the rails. \$\endgroup\$
    – jonk
    Commented Apr 17, 2021 at 8:34

2 Answers 2

3
\$\begingroup\$

where the series resistor is placed after the TVS diodes. What is its purpose?

The series resistor is chosen by looking at the maximum permissible input current of the op-amp. This is usually stated on most op-amp data sheets. Here's what your data sheet says: -

enter image description here

So, if your input voltage is clamped at 18 volts as per the SD05C data sheet: -

enter image description here

Then it's a bit of simple maths to figure out the current through R40 (1 kΩ). The voltage difference is 18 volts - 5 volts = 13 volts. And, the current through the resistor will be 13 mA. And that means the resistor value is about ten times lower than it should be.

What is its purpose?

Hence, the input resistor is to protect your op-amp input from receiving a current greater than that stated in the data sheet. If that current were exceeded then you stand a chance of damaging the op-amp. Regarding the value being too low (as per my calculation): -

I can't get inside the mind of the person who designed this circuit and, there may be some other restriction on voltage and current to the left on the input connector J8. That may limit the peak voltage to 7 volts at which point the current into the op-amp input would be limited to 2 mA.

My best guess is that this has been incorrectly designed using the reverse voltage drop of the TVS at 1 mA (6 volts) instead of the maximum clamping voltage (18 volts).

\$\endgroup\$
5
  • \$\begingroup\$ For ESD, the max clamp voltage should persist only for less than a microsecond. During this short time, larger input currents should be permissible. I understand the 2 mA limit to be a DC limit. So the lowish resistor could be dictated by noise reasons. \$\endgroup\$
    – tobalt
    Commented Apr 17, 2021 at 11:42
  • \$\begingroup\$ @tobalt that may be true for ESD events but it's not true for indirect lightning surges that TVS diodes also protect against - it can be in the order of 50 to 500 us depending on which spec you adhere to. And, no, you cannot advise that for less than a microsecond the current can be higher - that excess current will cause the device internal protection diodes to turn into thyristors and clamp the power rails and possibly burn the device. \$\endgroup\$
    – Andy aka
    Commented Apr 17, 2021 at 14:36
  • \$\begingroup\$ @Andyaka: It would be helpful if data sheets could supply more information. From a practical perspective, a device that would be damaged by applying 2.1mA for a nanosecond would unlikely to tolerate 2.0mA continuously. It would be helpful to have some kind of spec that would indicate e.g. that certain combinations of factors must not be exceeded, but the device's internal parasitic leakage, capacitance, and clamping would be sufficient to guard against damage from individual factors. For example, specify that if a device were connected to a 1uA current source, it wouldn't be guaranteed... \$\endgroup\$
    – supercat
    Commented Apr 17, 2021 at 17:57
  • \$\begingroup\$ ...to clamp to any specific voltage, but would be guaranteed to clamp or leak enough current to keep the voltage low enough to prevent damage. Likewise, if DC current values are within limits, it one could dump in additional transient of 10pC, no matter how quickly it's applied. \$\endgroup\$
    – supercat
    Commented Apr 17, 2021 at 18:00
  • \$\begingroup\$ I'm not sure if I need to respond to this other than to say, what is the purpose of your comment? It sounds like a rant about flaky data sheets but I can't really tell. \$\endgroup\$
    – Andy aka
    Commented Apr 17, 2021 at 18:10
3
\$\begingroup\$

It's more of a precaution about over-voltage in input signals. Although there is a TVS on the input line, however, if there is a spike on the signal line, the voltage can rise above the TVS nominal voltage or if the over-voltage is continued more than TVS power dissipation, it can burn up the TVS and lead to high voltage on Op-amp's input pin.

Every op-amp has a limit on the input signal voltage. If the signal exceeds that, it can break the op-amp. To limit the input current in such cases, it's good practice to put a small resistor in series with the input. It can lead to some errors (because of voltage drop on resistor), but it will be safer.

\$\endgroup\$

Your Answer

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

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