The $5 preamp

[Derringer]

I breadboarded one of these and it worked great.
I built one with a pair of the preamps and it has low gain and oscillates like mad.


https://www.audiomasterclass.com/newsletter/the-famous-5-preamp-everything-you-need-to-know

here is the data sheet for the chip used
https://www.audiomasterclass.com/pdf/ina217.pdf

page 7 shows the basic preamp stage schematic.

On the breadboard and in the build, I used a LT1054 to make +/- 15v.
The breadboard only had one preamp.
I also realize now that I never grounded pin 5 on the breadboard ... and it worked great. Zero noise, plenty of gain.
I used 10uf bipolar input caps and a pair of 3.3 uf polar caps to make ~1.6uf bipolar output cap.
I didn't bother with the extra opamp to control the output DC, I just used the cap.
I tested an sm57 with vocals and a kick mic on a kick drum to test my lows. Again, it sounded fine. I seemed to have full range, or full enough range, frequency representation.
The output of the preamp was driving my interface which has 10K input resistance.


The one I built, tried to build, has two of these preamps, same power supply, but this time I did connect pin 5 on each opamp to ground. The gain is real low, it oscillates ... something's not right.

Aside, from finding poor solder joints, bad connections etc, might I need more filtering than those 0.1uf caps to ground off the rails since I have two of these connected to the same power supply?

Thanks

[mth5044]

Are you using 1 LT1054 per channel, or 1 to power both? Perhaps it can't supply enough current for both ina217.

If you pull one of the 217's, does the other channel work as before?

[rankot]

Did you use datasheet guidelines for layot? Do you have one PS filtering cap for each amplifier?

[Derringer]

haven't had a chance to fiddle with it yet, but yes, the first thing I will try is to pop one chip out and see how it does.

power-wise, the Lt1054 sheet says it'll produce 100ma. I am using one to power both preamp chips. The ina217 sheet says they use about 12 ma, if I'm reading it right. There was no power sag when I took readings. That was with no signal applied though.

I didn't follow layout guidelines on the breadboard and it behaved perfectly. I do have a bunch of long leads and no shielded cable in the current build. If all else is sound, I figured I'd replace the gain pots with trimmers to eliminate the flying leads, and drill holes in the enclosure to access the trimmers.

Each chip has it's own 0.1uf cap to ground on each of its rails

[antonis]

Each chip has it's own 0.1uf cap to ground on each of its rails

Sometimes it's more preferable to directly decouple supply rails by only one 100nF cap between them..
(to prevent "garbage" from GND insertion..) 

[Derringer]

With one chip removed, each preamp circuit seems to perform as it did during breadboard tests.
I tested each chip in its respective preamp position and got equal results.
So, the chips seem sound, and the wiring to each channel seems sound.

Think I should clip a pair of the decoupling caps?

How about my input wiring?
I designed it to use xlr to trs 1/4"cables, so I have trs 1/4" jacks on the preamp box for the inputs.
They are switching jacks. I wired them so that all inputs are grounded when no plug is inserted. When a plug is inserted, the shield remains grounded and the tip and ring break free from the ground and flow to the "+" and "-" inputs of the respective circuit. I connected across the shield lugs so they are always connected to ground AND jumpered the shield lugs of one jack to the other. The ground for the shield goes to a chassis ground which is tied directly to the power input ground.

So, when one jack is being used and the other isn't, the one that is not in use has its inputs grounded which connects to the ground of the jack being used. I don't see why there would be an issue with this ... but maybe I'm wrong?

[Sooner Boomer]

The chip is an instrumentation amplifier.   https://en.wikipedia.org/wiki/File:Op-Amp_Instrumentation_Amplifier.svg  Has anyone tried other chips specifically marketed as instrument amps?

[PRR]

> Has anyone tried other chips specifically marketed as instrument amps?

There are several various optimizations of "instrument amps".

DC accuracy (we don't care)
Gain accuracy (we hardly care)
Low hiss current (moot on lo-Z mikes).
Low hiss voltage (what we want)
Narrow bandwidth to control noise (bad)

[Rob Strand]

DC accuracy (we don't care)

It's interesting to note that putting AC coupling caps on an instrument preamp's inputs can screw-up the ability for the inputs to cancel common mode signals.   The cap tolerances cause an imbalance.

[Sooner Boomer]

There are several various optimizations of "instrument amps".

(snip...)
Narrow bandwidth to control noise (bad)

Are you saying that there is a configuration, an "optimization" of an instrument amp that gives low bandwidth?  Please explain.

[Derringer]

I'm probably just going to leave well enough alone and use it as a single channel preamp since my interface has 4 extra inputs the require an external preamp. This will take care of one, my mixer can handle two more (assuming I can just hard pan two inputs left and right and use the left/right outs of my mixer to amplify those interface inputs, and the 4th input will be dedicated to a DI guitar stomp I made a while ago because I guess my drummer needs to hear me playing for whatever reason when we record.