Low frequency oscillation problem in opamp

[smuel]

Hi,

I've modified a guitar pedal design to include a gain stage at the output. There was already an opamp buffer at the output, so i've just converted it to a single supply (9V) non-inverting amp.

I've learnt the theory of opamp circuits at uni, but we've never really played around with them that much in real life. With an input signal of 200mVp-p, and resistors set to give a gain of 10, the output is no where near 10 times the input. Have i totally forgotten a basic piece of knowledge from 1st year?....

In the end, i had to set the gain at around 50 to get any kind of decent amplification, but the output is still only about 500mVp-p...

Also, there was some high frequency oscillation happening at a certain tone control, which i fixed by bypassing the feedback resistor with a 0.1uF ceramic, but now there is low frequency oscillation at about 430Hz when i role the tone control right back.

Any ideas on the oscillation?

Feel free to bag me out for not knowing how a simple opamp circuit works, i should definetely know this stuff by now. actually it's probably sitting in the back of my brain somewhere never to be found again...

Thanks a lot

Sam

 

[Kral]

Re: oscillating opamp

Sam,
It's difficult to dagnose your problem without a schematic, but here goes:
The gain of an inverting amplifier configuration is Rf/Ri. Rf is the feedback resistor, Ri is the input resistor. The gain of a non-inverting configuration is 1 + (Rf/Ri).
~
Having said that, all bets are off if your circuit is oscillating. Here are some precuations to take against oscillation:
1. Make sure thet the connections to the op-amp inputs are short. In case of the non-inverting configuration, you may have to provide a filter to prevent inadvertent positive feedback from the output to the non-inverting input, especially if the source impedance of the input is high.
.
2. Make sure that the op-amp power supply connection(s) are bypassed by capacitors to ground.
Regards,
Kral

 

[btbass]

oscillating opamp

Have you biased the non inverting input to half the supply rail? Have you got a bias path for the non inverting input? 0.1uF sounds much too large for the feedback resistor, if it was about 47K, a typical value would be 33pF.

 

[smuel]

Re: oscillating opamp

Yep, I've got the input biased at 4.5V. It's a 9V battery supply, with 2x680k resistors dividing it for the input. I have used a seperate 4.5V bias for the feedback virtual ground.

I spoke to one of my lecturers and he suggested sticking a small cap from the virtual ground to real ground.

I totally agree that the 0.1uF cap is too big, wasn't thinking straight.. I'll change it when i get home and see if that helps.

The low 430Hz oscillation is only present when there is a high impedance input to the opamp..

any other ideas?

Thanks a lot

Sam

 

[Kral]

Re: oscillating opamp

Sam,
430 Hz is too low a frequency to be caused by spurious capacitive feedback to the non-inverting terminal. If bypassing the "phantom ground" network with a capacitor, as your instructor suggests, does not help, I would suggest monitoring the battery supply voltage with a 'scope. If an AC waveform shows up on the supply, then the internal resistance of the supply is too high. The solution could be as simple as replacing the battery with a fresh one. Also, bypassing the battery supply with a large capacitor may help.
Regards,
Kral

 

[smuel]

Re: oscillating opamp

Thanks everyone for the advice. It turned out to be coming from the virtual ground. I stuck a 0.1uF cap from earth to virtual ground, and a 56pF cap across the 22k feedback resistor, and now it is working perfectly.

Opamps are much simpler on paper..

Thanks again,

Sam