OpAmp as reference buffer driving 16 thermistors and the ADC REF input - how?

[AQUAMAN]

I have 16 thermistors in parallel that I want to drive with 2.5V
I also want this 2.5V to be the reference input to an ADC for ratiometric measurement

The thermistors however have a 330n Cap beside them. I have used an opamp with an NPN to get a high current. They are 10k thermistors and at the highest temperature they will be down to about 300 Ohm
My problem is that when I put a cap on the REF pin of the ADC, the op amp starts oscillating. How do I stop this?

[ledtester]

My guess is that the 16 330n caps presents a large capacitive load for the op amp.

This article is pretty good on the theory and possible remedies:

https://www.analog.com/en/analog-dialogue/articles/ask-the-applications-engineer-25.html

I can't say though, which one you should try in your case. Which op-amp are you using?

The end of the article mentions some Analog Devices op-amps which have "unlimited" capacitive drive capability... I wonder to what extent that's true in practice. For a slow slew rate applications (which yours is) the article suggests they should work out ok.

Update: I see your problem happens when you add the ADC cap:

My problem is that when I put a cap on the REF pin of the ADC

How big is this cap?

[AQUAMAN]

100nF
Im just using LTSpice to simulate at the moment

[Doctorandus_P]

Remove the inductor between the emitter of your transistor and the non-inverting input of your opamp.

More serious:
Your NPN transistor also adds a lot of amplification to your opamp, which already has 100dB or so of gain and that reduces stability margins and can easily lead to oscillations.

Your circuit is very much like a current source, and these are well known to need some extra tinkering to get stable. To give you plenty idea's of how to proceed, have a look at:

https://duckduckgo.com/?q=opamp+current+source+stability&t=hd&va=u&iax=images&ia=images

I would also swap the thermistors and their load resistors.
GND is always dirty and you've just made a clean reference, so make use of that by placing your thermistors directly to your reference voltage. I don't know what ADC you use, but a lot of them do not like to measure signals close to GND

Also, if your whole system is radiometric, you do not have worry about your voltage reference. Why not simply take a LM317L, or a voltage regulator with 3V3 output? (and use 3V3 as reference).

Have you also considered self heating of your thermistors by their exitation currents?

If you only need to measure temperature once a second or so, then you can make use of that by disabling the power supply for them inbetween measurements. This can reduce the dutycycle (and therefore the self-heating) of the thermistors by a few orders of magnitude.

[Kleinstein]

The OPs with unlimited cap drive are still not a good solution to drive a highly capacitive load. They don't oscillate with a capacitive load, but they are still rather close to oscillation and thus show quite some ringing. Just a bit on the other side of the edge, but still not good.

The usualy cicuit for driving capacitive loads works well for fixed capacitance, but than starts to fail when the capacitance goes higher than the design value.


If the capacitor at the thermistors is as shown it would not be a problem for the OP - there is still quite some resistance in series. Capacitive load is a problem with low (e.g. < 100 Ohms) series resistance.

[richard.cs]

Remove the inductor between the emitter of your transistor and the non-inverting input of your opamp.

I think that's a badly-drawn resistor.

More serious:
Your NPN transistor also adds a lot of amplification to your opamp, which already has 100dB or so of gain and that reduces stability margins and can easily lead to oscillations.

It's an emitter follower, a voltage source with unity voltage gain. It has current gain of course, but the opamp has voltage feedback so this shouldn't affect stability. In fact, the current gain helps stabilise the capacitive load because the opamp "sees" a capacitive load as being h_FE times smaller than it really is (to change the voltage on the reference capacitor by 1 Volt requires only 1/h_FE as much current from the opamp, hence by C=q/V C is smaller). It does add some delay, but unless it's a very fast opamp it won't be dominant.

The last circuit in the A.D. app note ledtester posted is a good one which I have used several times, (in conjunction with a NPN follower in my case), it works well.

PS 100 dB is a factor of 10^10 in power and 10^5 in current/voltage. There's no way a single bipolar has that much gain.

[Terry Bites]

Use a composite amplifier put a high current opamp in the fb loop of a precision opamp: see https://www.analog.com/en/analog-dialogue/articles/composite-amplifiers-high-output-drive-capability-with-precision.html.  or https://www.ti.com/lit/pdf/sboa002?keyMatch=COMPOSITE%20AMPLIFIER&tisearch=search-everything
This keeps the heat out side your precision opamp chip too.
The LT6658 can output multiple 150mA @ 2.5V referrences

[mikerj]

Is the op-amp oscillating, or is your emitter follower oscillating? An emitter follower driving a capacitive load can be prone to instability, a base resistor would be a wise addition to this circuit.