Mic preamp input transformer alternative?

Status
Not open for further replies.
Jump to Latest
A

Andyroid

Member
Joined 2010
#1
I built a mic pre a while ago and I'm pretty happy with the results (18 months of tweaking and total rebuilding etc). The gain stages are currently center biased and sound clean as a whistle. The thing is, the Lundahl 1538 input transformer smears the top end far too much. I think it's fine for rhythm/backing tracks but solo/upfront/vocal tracks seem to need much more bite and clarity.

I would like to add a switch to choose between the input TX and some transparent alternative. I would like a simple drop in alternative to a transformer that can connect to the phantom power/-20dB pad/first gain stage just like the TX does. But I don't know where to begin.

Any suggestions and schematics would be greatly appreciated!

Thanks,
Andy
 
#2
Are you quite certain that a high quality transformer like Lundahl is smearing the top end? I don't use mic input transformers myself but mainly because I find good ones to be expensive. The alternative is to AC couple after the phantom supply to XLR pins 2 and 3 but if your amplifier is single ended, this may not be acceptable as you will lose the signal balance and thus all the noise rejection that your transformer gives you.
 
#3
The amp is single ended, a fully balanced pre is a project for another time. My pre has a DI jack socket that switches out the transformer and through that I can hear the pure sound of the valve gain stages. True, my test signal may be at fault, I have to use a DI box to get a signal into the two inputs (a Palmer active DI, best quality one I have).

To be honest, I may have exaggerated a little in order to make my point clear about the smearing.

But I would still like the option of a transformerless input that provides ~2k impedance. I'm not sure what you mean by AC coupling after the phantom power, just large capacitors convening at a suitably small grid leak?

Cheers
 
#9
Andyroid said:
Any suggestions and schematics would be greatly appreciated
Click to expand...
The frequency response of a transformer depends on the loading. You could experiment with a zobel network across the secondary to alter the bandwith.

Freshness of sound is sometimes achieved when 3rd and 5th harmonics are stressed. One could achieve this by lowering 2nd harmonic (from cancellation in a PP design) or by altering the load so the harmonic spectrum changes. Penthode amplification would present a different spectrum too.
 
#11
I suspect transformer loading is probably suboptimal. What have you tried as transformer loading so far ?

I have found that loading is very critical for good input transformer performance and the overall response can be tuned by varying it by even small impedance differences. Lundahls are noted for been clean on the top end so I don't think a change of transformer to another brand is likely to help.

Shoog
 
#12
Transformers resonate, hence the bandwith property. By loading the transfer characteristic flattens (if it was raised) or dips (if it was flat). Avoid interaction of the high pole with amplification stage resonance as they add up rapidly. Have a look at a good MC phono stage design with stepup transformer.
 
#13
Thanks guys, I've got a 38k load resistor which makes a ~1k5 input impedance. I've been thinking about introducing a zobel network, but square wave analysis shows no overshoot or anything like that so I never bothered. I don't think there is any 'problem' with the circuit, response is apparently flat up to and above 48k (where my shareware audio analyser stops measuring). I've just been on a quest for ultimate clarity and larger than life sound. I'll give the unbalanced transformerless input a go, and then put my efforts into further improving my rooms acoustic treatment.

Thanks for the suggestions, I can barely believe I couldn't figure out the unbalanced input thing. Suppose I was convinced the only way to get rid of the TX would be complicated. Couldn't see the wood for the trees!

Andy
 
Last edited:
#17
Andyroid said:
Thanks for the suggestions, I can barely believe I couldn't figure out the unbalanced input thing. Suppose I was convinced the only way to get rid of the TX would be complicated. Couldn't see the wood for the trees!

Andy
Click to expand...

If you do get rid of the transformer you will certainly worsen the noise figure of the mic preamp. If you have a relatively high input level then this may not be a problem but it is something to be aware of.

Cheers

Ian
 
#18
painted said:
Which one? It's not different than others 😉
Click to expand...

Painted, I don't know what you mean; no mention has been made of the output stage, or the valves used. If the 2nd stage were a common cathode 12ax7 coupled into the 10k input of a typical recorder with moderately high input capacitance then I don't think the high frequency response would be terribly good......
 
Last edited:
#19
The preamp goes LL1538 - EF86 diode biased(with local negative feedback) - AC coupled ECC81 cathode follower - 47k gain pot - ECC81 diode biased common gain stage - DC coupled 5814A (ECC82 equivalent) white cathode follower - 10uF coupling cap => audio interface is a 10k load. There is no noise floor to speak of, even using an sm58 my computer's quiet fan (in another room) drowns out any hiss and there is zero hum. The amazing lack of self noise genuinely surprises me, as this prototype layout has been twisted and warped so many times, it's a complete mess inside. I can't wait to rebuild with a proper layout and it already sounds better than any other pre I own.

I'll look into THAT corp, the InGenious IC looks interesting. Hopefully I'll achieve the fabled 'wire with gain'. Not had chance to try the simple pin2 coupling to V1 grid idea yet (been hanging more acoustic foam this evening.)

I'll let you guy know how it works out.
Cheers,
Andy
 
Last edited:
Status
Not open for further replies.
Top Bottom