Hello,
I have some questions regarding op amps, and how to reduce noise.

My circuit is built on a breadboard. (Yes, the nylon plastic things, and it is a larger, 3 ic wide board.)

I'm using TL081 op amps in a circuit. The first is configured as a wien oscillator. It puts out a nice sine wave at around 5 kHz, about 10 volts peak to peak. I then pass this signal through a resistor network; a 100 k to an amp configured as a non inverting amplifier, and a 560 ohm resistor to ground.

The second stage has a 6.1 - 10k feedback resistor to the inverting input. A pair of probes measures a resistance which is connected between the inverting input and ground (in the feedback loop.) It all works very well, except there is a lot of noise, coming from the non inverting amp.

I had the circuit working, and left it run while I went to work. When I returned home that evening, and was all set to show a friend, the output was bouncing wildly. It stopped when I shut down the power supply for a few minutes. I then added a few .01uF caps at all my V+ and V- supplies for the chips. Also, I found I could reduce the noise a bit further by adding a .01uF cap across the probes.

I disconnected the oscillator, and the output of the non inverting amp was bouncing with only the power to the chip, and the feedback resistor.
What is most troubling, is that if I just power up a TL081, no resistors at all, or just a 560 ohm on the non inverting input, the chip is still putting out some erratic pulses on pin 6 (output.)

Are these pulses normal in these op amps? How does one deal with this noise? Is it because my circuit is on the breadboard, and it should work better when I try the same circuit on a prototype board? I could handle a bit of noise, but the output is bouncing 50 - 100 milli-volts every time this noise occurs, which will make reading it with an ADC very difficult.

Can anyone shed some light for me?

Thanks for listening,
Robert

 

the problems from breadboard may be due to the loose connection of the leads with the BB. i 'll prefer to rewire that part to another part of the BB .
BTW a 100K + 560 to gnd (as i understood from the description) will atteunate the signal very much.

__________________
Gods own Country
Incredible !ndia

www.flickr.com/photos/_akg/

"Give a man a fish, and he will eat for a day. Teach that man to fish, and he will eat for a lifetime."

 

Thanks for the reply akg,

Yes, that is my intention, so it will not overload the non inverting amp, and so that it will not be so much to remove the offset later. What I'm wondering, though, is if that is too much attenuation? I have reduced the 100k to 60k, and increased the 560ohm to as much as a few k, with no noticible improvements. I think it did quiet down a bit too, when I added a small cap in paralell with the 560 ohm. I'm going to try increasing the value of that cap, and see what happens.

A big part of the problem is my limited analog knowlege, but I'm changing that now. :cry:

Regards,
Robert

 

good luck.. try changing the position.

__________________
Gods own Country
Incredible !ndia

www.flickr.com/photos/_akg/

"Give a man a fish, and he will eat for a day. Teach that man to fish, and he will eat for a lifetime."

 

If you have an overall gain of 10,000 the noise is the internal noise of the op amp. Otherwise, the problem could be a bad solder joint or other connection. I am trying to make a noise generator and find that a gain of 2000 is not enuf!

__________________
see my website: www.geocities.com/russlk

 

Robert give ur ckt to russlk

__________________
Gods own Country
Incredible !ndia

www.flickr.com/photos/_akg/

"Give a man a fish, and he will eat for a day. Teach that man to fish, and he will eat for a lifetime."

 

Hi,
Ha, ha, perhaps I should try to make a noise generator, and you should try to make something noiseless! (Amplifiers which oscillate, and oscillators that amplify!)

Then it is likely noise from this huge chunk of nylon the circuit is made on. I'm putting on a proto board now, just to see...

Also, a lot of it was from the probe, and using a shield got rid of some noise, but the output still swings too much.

Yes, Russell can have it! :lol:

Regards,
Robert

 

for 5kc you are using a hi freq amp for some reason. use an lm324 for 5 kc next time. the noise is probably ground loops look at your ground and make sure they go to power at one point only. it would help next time is to provide a schematic to look at.

 

With its gain set to 1000, an LM324 's frequency response drops above only 500Hz. and makes plenty of noise.

__________________
Uncle $crooge