1
\$\begingroup\$

I am trying to switch the ranges of an I/V amplifier, such that I can measure from 100nA to ~1mA of current up to 100kHz (similar to this question). I make the switching using an TS5A3357 analog switch, however I seem to be limited in bandwidth somehow, no matter what op-amp I use. I'm guessing it has to do with the ON/OFF capacitances of the analog switch, but I don't know how to mitigate that. Is an analog switch even possible in such application or would I need to use another component (such as a relay)?

Why is this thing not working?

enter image description here enter image description here

Also note that the low-frequency gain, even though it should be 0dB, is -3.8dB.

\$\endgroup\$
  • \$\begingroup\$ would a variable gain amplifier not work out much nicer for you? As you noted, analog switches are bandwidth limited. \$\endgroup\$ – Marcus Müller Apr 24 '17 at 7:00
  • 1
    \$\begingroup\$ Does your TS5A model work fine by itself? \$\endgroup\$ – winny Apr 24 '17 at 7:00
  • \$\begingroup\$ @Calin: Where is the TSA? All I see is a standard inverting OpAmp configuration with voltage source as signal source (I'd expect a current source). \$\endgroup\$ – Curd Apr 24 '17 at 7:13
  • \$\begingroup\$ @Curd The voltage source + resistance inject current into the inverting input. This way I simulate a current source, since the voltage source is a sinusoidal one with 1.65V offset and 100mV amplitude. The inverting input had its potential fixed at 1.65V. Sorry, should have mentioned that. It's not really a current source, it's a current to voltage converter, so wrong term used on my part, sorry. \$\endgroup\$ – Calin Apr 24 '17 at 7:35
  • \$\begingroup\$ @Winny seems like it \$\endgroup\$ – Calin Apr 24 '17 at 7:36
0
\$\begingroup\$

The common mode input range for the LF411 does not include the negative power rail (The LF411A version does so take note). You have the negative power rail connected to ground and your input signal is referenced to ground. This is a basic problem and might cause all (or most) of the problems you witness. Fix this first. Read the data sheet - the LF411 has a typical common mode range that is as low as -11.5 volts on a -15 volt power rail. This means it can't be relied upon when DC levels are within 3.5 volts of it's negative supply rail. An offset of 1.65 volts is not enough.

Ditto the output voltage swing - it cannot get to within 1.5 volts of either rail and this applies to both LF411 and LF411A.

Then, if you still have problems substitute a fixed resistance of 50 k as feedback resistor and see if this has a 3dB point of only 10 kHz.

\$\endgroup\$
  • \$\begingroup\$ Funny thing is, I used an ideal op-amp first with the same result. Then I tried this op-amp just to see if the frequency response is the same or if the 3db point shifted (it shifted barely). However, I will try what you advised once I get home, maybe I did a mistake somewhere, thank you! \$\endgroup\$ – Calin Apr 24 '17 at 9:13
0
\$\begingroup\$

If really capacitances of the analog switch are the cause and if it is ok to switch to a non-inverting OpAmp configuration using discrete transistors as switches is an alternative:

schematic

simulate this circuit – Schematic created using CircuitLab

(limitations: signal must be positive; amplification > 1)

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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