A question concerning op amps.

[MV]

So i play guitar and i've been into electronics for a while. I have learned about op amps and done many simulations with them. However I've never accually put this stuff in reality. So i bought a breadboard last week.
I had some tl071 from before, so i put one on breadboard and created a simple non-inverting buffer. Connected to 9+ power supply and no biasing. It worked but the volume pot had a terrible stratchy noise and when i turned it all up, there was no signal at the output of the op amp. However, an input cap an a shut resistor after the cap solved the problem. I still don't understand why the signal on the output disappeared when i turned volume all the way when there were no cap and resistor.
However as i proceeded to improve my buffer, i tried to bias it, but first i put a cap on the output, to block dc i was going to have with biasing. However there is no sound from my amp. As i remove capacitor there  is sound again. Then I put a 1k resistor in series instead of cap --  no sound, except when i set input pretty high, apmlifier does horrible noises. I put a 220k shunt resistor after that -- still no sound. I could not understand why there is no signal with a small resistance in series, so i put a trimmer in place of the resistor. When i increase resistance from zero a little more, there is a pop and there is no sound again. I cannot understand why does the resistance in series block the signal?
Here's the schematic of working and non-working breadboard.


Thank you,
Mile

[Kipper4]

Schematic would help here i think

[MV]

Schematic would help here i think

Done

[GGBB]

I can't explain the technical details of why you heard what you heard, but at least part of the reason why it does not work properly is because you have "no biasing."  You actually do have biasing - the 220K resistor to ground biases the opamp to 0V.  At the very least, this results in a half-wave rectified input signal.

From the AMZ Buffers page:

You have this (where Vr = 0V):


You need to do this (or provide an actual Vr=+4.5V above):


You also need to have an output cap to block DC.

[MV]

I've tried biasing. Since i had biased signal on the output i had to put cap on the output. And as I stated before, there is no sound when i have cap/resistor on the output. Just "farting" when i play louder.
Thanks anyway, I appriciate it.

[GGBB]

I've tried biasing. Since i had biased signal on the output i had to put cap on the output. And as I stated before, there is no sound when i have cap/resistor on the output. Just "farting" when i play louder.
Thanks anyway, I appriciate it.

So the schematic you posted - is that with "no biasing" or was that your attempt at biasing?  Either way it is wrong and It wouldn't surprise me if the sound you got out of it was like farting (half-wave rectified signal).  Have you tried it like either of the diagrams I posted?

[MV]

That's no biasing. I tried biasing exactly like the second schematic you posted. The same result, since there is an output cap.
Thanks anyway.

[GGBB]

That's no biasing. I tried biasing exactly like the second schematic you posted. The same result, since there is an output cap.
Thanks anyway.

The buffer in the diagram works.  The only thing that you would normally find that isn't on the diagram is a resistor to ground after the output cap - usually around 10K or a volume pot in its place.  If that doesn't work for you, it's not the circuit that's the problem - check your components, wiring/breadboard connections, post your voltage measurements and clear pictures.  Basically follow the debugging procedures described here: http://www.diystompboxes.com/smfforum/index.php?topic=29816.0.

[Mike Burgundy]

For now, start over - that's the beauty of a breadboard.
Build the schematic exactly like the "biased" one that GGBB posted.
Opamps don't need bias like transistor stages do - they "assume" ground is in the middle of their supply (this is important) and *follow* that on their output. Tie the + input to ground or V+ and the output will attempt to be exactly 0V or V+ respectively.

Here's the trick: most opamps (such  as the TL07x) need a dual supply. You can achieve this by using two batteries.


If the inputs are "DC-biased" at 0V, output sits at 0V with no signal, and with signal applied you get maximum voltage swing on the output, both positive and negative.

If you want to use one battery, you can "fool" the opamp into believing it has a dual supply by using a voltage divider to get 1/2 of your battery voltage, sowith a regular 9V block that's 4.5V. Call that Vref.
If you feed that to the + input through a large-ish resistor, the opamp will assume that voltage as it's baseline output and swing any applied signal around that. The largeish resistor needs to be small enough to keep DC on the input at Vref but large enough so the Vref voltage cannot soak up your signal (imagine a resistor here of 0Ohms) All this does mean there is a DC voltage on both in- and outputs which can screw with other bits of the circuit - for one, it will make pots scratch. This is where blocking caps come in, they let through AC(signal) but not DC.

If you use the schematic you posted, the opamp is clamped to its lower rail (in this case 0V, or as close as it can get to that) and won't work properly if at all.

Does that help?

EDIT: GGBB's schem uses a voltage devider directly on the + input. That'll work great, but it's often a better idea to make Vref once, treat that as a new voltage supply, and feed that to all the points needed through the large resistors mentioned. But this schem works just fine for a single opamp.

[Gurner]

As others have alluded to, the likelihood is that you've had something connected wrong (especially in the light this is your first exploratory breadboard scenario)....if you've really connected as you say (like the second diagram in GGBB's post above)....then taking DC voltage readings at each opamp pin will help us in getting to the bottom of where your problem lay.

[MV]

Thanks guys, a bipolar supply solved everything.
Have a nice day!

[PRR]

> there is an output cap.

That alone means nothing. (You could tuck a cap into a guitar cord.)

> a bipolar supply solved everything.

But uni-polar is sometimes more convenient.

So it may be good to see the basic problem.

Audio swings both ways.

If you have a one-side-grounded power supply, it can only swing one way.

And most opamps can't swing ALL the way to the rail. If input is ground-referenced, and swings from zero to +1V thru zero to -1V and back, they never come-alive. If pushed harder, say from zero to +2V thru zero to -2V and back, that bit from +1V to +2V gets through, the rest won't.

> Just "farting" when i play louder.

Just so.

The trick is to set the signal pins halfway between the rails. If you have bipolar supplies, that naturally falls on your "zero" volt point and things are simple. If you have only zero V and +9V, you bias-up the signal pins to +4.5V. Check with pencil/paper and on breadboard with meter.

The "halfway" or "4.5V" does not have to be exact. You are trying to slide a 3 foot wide car into a 9 foot wide garage. With NO bias, that's like aiming for the door-frame. Not good. You want to aim for the center, 4.5 feet from either side. But with a 3 foot wide car in a 9 foot opening you can come in quite a bit off center and not hit anything.

[tubegeek]

But with a 3 foot wide car in a 9 foot opening you can come in quite a bit off center and not hit anything.

Says the man with a big scratch down the side of his car....