My OpAmp Ckt Oscillates.

[darwint]

hello guys.

i got an op amp circuit that oscillates at the output at 80MHz, ~70mVp-p with both inputs at ground. I am just supposed to be getting the Vos at the output. AMplifiers gain is 10. I was thinking this might be due to some capacitance loading at the output. It'd be great to hear some of your ideas why my op amp oscillates.

darwin

[audioguru]

Hi Darwin,
Welcome to our forum. ;D
You didn't attach your schematic for us to see which opamp, if it has supply bypassing and how it could have a gain of 10 with both inputs grounded.
Just about any opamp will oscillate without supply bypass capacitors and if there is capacitance directly connected to its output. With a TL07x audio opamp I use a 100 ohm isolation resistor from its output to feed high capacitance shielded cable to avoid oscillation.

[darwint]

Thanks pal for the welcome and reply!

I still don't know how to post pics here, but it's basically an inv voltage feedback amplifier with Rf=10k and Rin=1k. The output has 100ohms load, and doesn't drive a capacitor. I said both inputs are grounded because I am testing the circuit for the Vos and the supply current when input is zero. The circuit is on some kind of a test board. Supplies are properly bypassed by 0.1uF. I am thinking that stray capacitance at the output might be causing it to oscillate, because the part itself has been tested on the bench without oscillation, with the theory behind that the stray capacitance adds additional phase shift to the amplifier causing it to be unstable. Is it correct to say that some noise being picked up by the part on the ckt system triggeres the oscillation? Do you agree with my theory?

[audioguru]

I still don't know how to post pics here

During your reply, click on Additional Options. Click on Browse to select your schematic from your pc then double click on the file's name to attach it to your reply.

it's basically a non-inv voltage feedback amplifier with Rf=10k and Rin=1k.

Then maybe its input is capacitively picking up its output.

he output has 100ohms load

Most opamps can't drive a load less than 2k ohms, so 100 ohms is extremely low, nearly a short. The output might be constantly current limiting that might slow it down causing positive feedback. Try a 2k ohm load instead.

The circuit is on some kind of a test board.

Breadboards cause many problems for high frequency parts. They have too much stray capacitance.
Some opamps oscillate if their gain is low so the negative feedback is high and they are mounted in a socket which increases stray capacitance and coupling. Don't use a socket.

Supplies are properly bypassed by 0.1uF.

A 0.1uF ceramic disc cap might resonate at 80MHz. Try a 1000pF of 2000pF ceramic disc with very short leads in parallel with them or instead of them.

I am thinking that stray capacitance at the output might be causing it to oscillate, because the part itself has been tested on the bench without oscillation, with the theory behind that the stray capacitance adds additional phase shift to the amplifier causing it to be unstable.

The opamp's datasheet will show how much capacitance on its output causes it to be unstable. We need its part number to see its datasheet.

Is it correct to say that some noise being picked up by the part on the ckt system triggeres the oscillation? Do you agree with my theory?

Not noise but its input picking up its output, or its output delayed by stray capacitance which causes its feedback to be positive at high frequencies, or its supplies not solid at very high frequencies.

[darwint]

you've been very helpful buddy, thanks so much!

i greatly appreciate your point by point reply. as a new member i hope i can also be a good contributor to this forum.

[darwint]

During your reply, click on Additional Options. Click on Browse to select your schematic from your pc then double click on the file's name to attach it to your reply.


Then maybe its input is capacitively picking up its output.


Most opamps can't drive a load less than 2k ohms, so 100 ohms is extremely low, nearly a short. The output might be constantly current limiting that might slow it down causing positive feedback. Try a 2k ohm load instead.


Breadboards cause many problems for high frequency parts. They have too much stray capacitance.
Some opamps oscillate if their gain is low so the negative feedback is high and they are mounted in a socket which increases stray capacitance and coupling. Don't use a socket.


A 0.1uF ceramic disc cap might resonate at 80MHz. Try a 1000pF of 2000pF ceramic disc with very short leads in parallel with them or instead of them.


The opamp's datasheet will show how much capacitance on its output causes it to be unstable. We need its part number to see its datasheet.


Not noise but its input picking up its output, or its output delayed by stray capacitance which causes its feedback to be positive at high frequencies, or its supplies not solid at very high frequencies.

Thanks again. The part is actually ad8390, a low power high output current differential amplifier, and i don't see anything about how much capacitance it can drive in the datasheet.

darwin

[darwint]

"Is it correct to say that some noise being picked up by the part on the ckt system triggeres the oscillation? Do you agree with my theory?

Not noise but its input picking up its output, or its output delayed by stray capacitance which causes its feedback to be positive at high frequencies, or its supplies not solid at very high frequencies. "

Honestly I am still a bit confused about "at very high frequencies" since I got both inputs at 0v or ground, as opposed to having a high freq input. Would appreciate if you can clarify that up further for me.

[darwint]

hi pal, this is the basic circuit. only my circuit has both inputs at gnd. another interesting find of mine is the ciruit won't oscillate at output with 1kohm load. the datasheet though guarantees the specs at 100ohm load.

thanks.

[audioguru]

Hi WildSwan,
You have an unusual opamp. Its output current is 20 times more than an ordinary opamp. It has differential outputs. It is very wideband.
The datasheet shows a pin called Vocm that could pickup the output and cause oscillation if it isn't bypassed with a capacitor to ground:

[darwint]

thanks again bro. you've been a great help. i found that the reason for the oscillation was stray capacitance at the output as low as 20pF which causes my opamp to become unstable.