Simple op-amp based transformer mic pre

[Consul]

I've accumulated a crapload of parts over the years for projects that were never started. I've been doing an inventory recently, and I can't help but wonder if I have all the parts I need to make a pair of decent mic pres. I don't have any Jensens or 2520s, but I do have a pair of Shure 1:10 mic pre trafos (with shields), a tube full of 5532s, and some API output iron. What could possibly go wrong?

So, with that in mind, I knocked out the attached schematic (using DaveCAD). All it is is two inverting amp stages, with the first one (gain of 2) acting more or less like a buffer for the second main stage. So, a gain of 10 from the trafo, 2 from the first stage, and 1 to ~50 from the third. Here are my questions:

1) The cap shorting across the input after the trafo. I've seen this in a lot of schematics, and from what I can tell, it's there to kill any resonant peak caused by the fact that the secondary of the trafo and the input resistance and capacitance of the first gain stage basically make an LRC filter. Is that right? If so, I'm pretty sure I can figure out the value of the cap I need.

2) I'm wondering if I should make the first stage a unity gain buffer and just run the second stage with a gain from 1 to ~100. I did it this way thinking in terms of a) not wanting to use a ganged pot, and b) Gain Bandwidth Product, but that might not matter with the 5532. I haven't done that math yet.

3) The output trafos are those quadrafilar-wound API output thingies I got from a group buy (I think) quite some time ago. I'm pretty sure the 5532 can drive those, but want to make sure.

I would like to point out that I left a bunch of the usual stuff off the schematic, like the bypass caps for the power rails, phantom power, and all that. I'm also sure there are some tricks I can use to improve the circuit's stability. Any thoughts would be greatly appreciated.

This post was written under the influence of Jim Croce.

 

[usekgb]

You will actually get 20dB of gain from your transformer ( 20*log(Vout/Vin)).  I would say to run the two op amps with an equal amount of gain.  Put a log pot (25kohm?) in between the two op amps.  Run the op amps with around 20dB of gain each.  At full volume this would give you 60dB of gain, and the op amps will be in a nice linear range.  I believe the 5532 maxes out at +-18 volts for the supply, so that is what you will need for power.  If you want more headroom, use the opa2604, which will allow +-24v.  Just some thoughts.

 

[Consul]

I don't get the part about putting the gain pot in-between the amps. I'll have to try to draw that out to see for myself.

It also just occurred to me that I need to order log pots anyway (not to mention a case), so ordering ganged ones probably won't be that big of a deal. Everything else, though, I have on hand. I might have to play some games in the power supply to get a +48V rail, but that's not a big deal.

EDIT: Ganged pots won't work, since I need to run each stage with different input resistances. The only way out of this is to use one more op amp for the input buffer, then run two subsequent gain stages. Unfortunately, I don't have any 5534s, but I do have some TL074s...

I have to go to bed now, so I'll be visiting this thread again tomorrow evening.

 

[PRR]

Don't use inverting connection in low-noise stages. The hiss is worse.

Don't run the first stage at unity gain; then the second stage's hiss isn't overwhelmed, total hiss doubles.

Do NOT!! start a plan without goals. Do you need very low hiss? Do you need tons of gain? Do you need a wide range of gains (harpsichord to 5F6a)?

This will cover many situations. Start here then refine your real needs.

 

[gridcurrent]

The Steve Dove preamplifier design as shown on page 743 in "Handbook for Sound  Engineers" (1987) is a useful reference.  Attached is a "cheap and cheerful' version of his circuit, very adaptable to the parts you have on hand.
I would not load the secondary of the input transformer until you listen and/or test.

 

[usekgb]

Take a look at the JLM Dual 99v pre-amp http://www.jlmaudio.com/JLM99XMB.pdf.  Joe has done a lot of the work for you.  Strip out the servos, and change the power supplies to bipolar, and you're pretty much there.

 

[JohnRoberts]

I am too lazy to do the search for you but IIRC Sam Groner published several variant transformer input topologies in a thread on this forum. The designs were each inspected and discussed at the time.

From my background managing engineers, my advice is do not re-invent the wheel until after studying the prior art (this is a mature technology). If after you look at the prior art you are still compelled to roll your own, enjoy.

JR

 

[jdbakker]

I am too lazy to do the search for you but IIRC Sam Groner published several variant transformer input topologies in a thread on this forum. The designs were each inspected and discussed at the time.

http://groupdiy.com/index.php?topic=20038 (both transformer and semiconductor input)

JDB.
[and, of course, http://groupdiy.com/index.php?topic=16909]