Linear Optocoupler Noise

[dreaquil]

I'm working on a linear optocoupler circuit for a project as isolation. For this im using a vishay IL300. I set up the following circuit with a 7.2V supply voltage and started probing around to see my signal. (R1=R2=20k and R3=100R) The way i understood it, the opamp is in negative feedback so the inputs to the opamp should be the same. (I'm using an lm324) When i probe the inputs, the inverting input has a huge amount of noise on it which is obviously coming out at the output. I've tried adding a 200nF cap to the supply cause i thought it needed decoupling but it didnt help. Any ideas what im doing wrong?

EDIT: The blue is the inverting input (which should be tracking it) for a 500mV 1kHz sinewave with a 200mV dc offset

[Alex Eisenhut]

We're missing some info here, what is the waveform you're feeding in at Vin? And what have you done with the two other opamps in the LM324 package? You can't leave them floating I think.
You also need to check that the input common mode is respected. ie no signals within 1.5V of VCC.
Never mind the scope, what is Vin and what are the DC levels at?
If Vin is a a generator that outputs a bipolar signal, you're toast right there.

You can't just copy high-level "idea" schematics from a datasheet like that. They're more like a block-diagram.

[David Hess]

I think you are coupling 60 Hz line noise into the inverting input with the oscilloscope probe; the impedance at the inverting input is not very low.  External loop gain through the optocoupler may be compromising the stability of the amplifier.

You might try adding a feedback capacitor between the operational amplifier output and inverting input.

[Alex Eisenhut]

I'm working on a linear optocoupler circuit for a project as isolation. For this im using a vishay IL300. I set up the following circuit with a 7.2V supply voltage and started probing around to see my signal. (R1=R2=20k and R3=100R) The way i understood it, the opamp is in negative feedback so the inputs to the opamp should be the same. (I'm using an lm324) When i probe the inputs, the inverting input has a huge amount of noise on it which is obviously coming out at the output. I've tried adding a 200nF cap to the supply cause i thought it needed decoupling but it didnt help. Any ideas what im doing wrong?

EDIT: The blue is the inverting input (which should be tracking it) for a 500mV 1kHz sinewave with a 200mV dc offset

Is that peak or p-p? Anyways you've got negative signal at the input, that's no good. Maybe the 324 takes time to recover from a negative input, although it's not very big? Is the CTR OK for the resistors?

Hmm, why don't you just drive the LED directly from the generator and look just at the output of the right amp?
Please also make sure you don't leave the other 2 opamps floating in the package, who knows what those two guys are doing?

Um, also check your grounding. 7.2V is an awful weird voltage, is that a battery?

[T3sl4co1l]

Try maybe 100pF from op-amp output to -in.

Tim

[dreaquil]

Thanks for all the great replies.. Hadn't thought of any of this stuff
 I will be testing all this stuff this afternoon and will get back to you. We
The 500mV is pp. The voltage is a battery as this will be going into an RC car project for current sensing.

[Whales]

he voltage is a battery as this will be going into an RC car project for current sensing.

Be wary of your batt's lowest voltage -- make sure your circuit still works

[Alex Eisenhut]

Thanks for all the great replies.. Hadn't thought of any of this stuff
 I will be testing all this stuff this afternoon and will get back to you. We
The 500mV is pp. The voltage is a battery as this will be going into an RC car project for current sensing.

What I was getting at is to make sure you actually have your grounds sorted out. A battery is floating , and you have an optical isolator too, maybe your scope was floating relative to what you were trying to measure? But then I didn't see the magical 16.7ms or 20ms sine wave of hum.

The further thought occurs, if it's going into a battery system, why do you need optocoupling?

[dannyf]

The opamp on the right is configured as a follower - not all opamps are unity gain stable.