Active ribbon-mic

[Kingston]

I think this thread need a link to this one... both talking about the same product.

As far as I understood this 5 year old thread, this idea/circuit was developed to completion on this forum by PRR and tested and tweaked by others. It precedes Fethead by some years.

So not the same product. Let's make that very clear.

 

[zebra50]

Fair enough - probably my wording was bad, but the discussions are related. I'll amend my words to avoid misunderstanding.

 

[zebra50]

OK, Phantom power is easier than I thought:

Thread necrophilia!

I built something like this today in a little box, and it works well, giving about 6 to 9 dB extra gain, and for some mics a little more top end. Test were done with a Grampian GR2L (30 ohm) ribbon, a Melodium 42b (200 ohm) , and another unidentified ribbon, of which I posted here recently. Test preamp / mixer / headphone amp is a Yamaha 01V96 mixer.

I tried a few pairs of transistors, and found that the quietest in my parts box were a couple of MP2222. I matched these by hfe, and matched the pairs of resistors as close as I could. The noise seems to be better that cranking the preamp by the same amount, especially at the top of its reach.

With the bias resistors matched the base voltages were identical to the nearest millivolt, so I removed the input caps. Are there any issues with sub-millivolt DC across transformers? The circuit works just fine without these & I can't hear any problems. So long as there isn't a voltage difference across the tranny the ribbon should be safe.

 

[Rossi]

I have no experience with dropping the input caps. I've built other ribbon preamps for super-high ratio transformers that had no input caps,but those had FET inputs. With BJTs you usually use input caps.

That preamp should actually have about 10 dB gain, and super-low Rbb transistors are absolutely crucial for lowest noise.

 

[zebra50]

Thanks Rossi,

I have some 2n4401s, and a handful of THAT300 to try next, and I was also going to make the bias adjustable with a trimmer. I'll post back later with some results.

Stewart

 

[zebra50]

A quick update. 2n4401 works very well indeed, and gain is up around 10 to 12 dB. Best signal:noise so far.

THAT300, with parallel pairs, sounds good but is a little down on gain compared with the 2n4401 - I'll keep tweaking.

Rossi, what FETs did you try? I've set up a little matching circuit so was going to give this a go.

This is quite a useful little box, so thanks to everyone!

 

[Rossi]

My FET circuits were different than the one posted in this thread; they were for super-high ratio transformers. But I wasn't really happy with them. I'll give it another go sometime; I have some new ideas.

You may get slightly lower noise paralleling 2 of those 2N4401 per leg.
The best transistor for PRR's BJT circuit is the 2SC3329, but you'll have a hard time finding them. The 2SC2546 is almost as good and is easier to source. There's no need parallel those.

 

[zebra50]

Hi Rossi,

Thanks for the transistor recommendations. i'll look out for those.

My FET circuits were different than the one posted in this thread; they were for super-high ratio transformers.

I found this buffer circuit in one of the Royer patents (US 2006). It's designed for a 1:100 (or greater) transformer. Looks interesting. Of course patents sometimes have deliberate 'red herrings'. Are those MOSFETS?

 

[Rossi]

Hmm, interesting. Is there a circuit description in the patent?
Most active ribbon mic's I've seen had a dual FET input stage, using normal n-channel FETs.
The ones I built were unbalanced input, similar to a condenser mic circuit.

 

[clintrubber]

Hi,

They're junction FETs, according to the text ( http://www.google.com/patents?id=m2CbAAAAEBAJ&zoom=4&pg=PA1#v=onepage&q&f=false.
There's also a circuit description.


Have a good weekend,

  Peter

 

[PRR]

It is a simple Darlington emitter-follower, twice, except the first transistor is a JFET instead of a BJT.

It is not obviously novel; looks obvious. If you read the patent's I Claim you will probably find that they are very specific to a ribbon mike, not any broad claim on a push-pull JFET Darlington.

 

[zebra50]

Yes, the novelty claimed was for the way the transformer was arranged in this application.

 

[trubac]

Can we put Tantalum caps in there ?

 

[Kingston]

Hey thanks for the tip for that very useful layout software!

I tried the link on your image, but it's dead. Tried searching for it and finally found a working download link from http://diy-fever.com/software/ (or alternatively http://diy-fever.com/?project=software)


And about tantalums, why not, no worries about drying caps with them at least. Probably a great idea especially if you're planning on sinking the gadget into epoxy for extra protection.

Also see zebra50's comments above on running it without the 100uF input caps.

 

[trubac]

Yes it is a nice little software !
The idea wasn't epoxy but PlastiDip !

 

[zebra50]

Also see zebra50's comments above on running it without the 100uF input caps.

Try it and see what you think. My ears preferred it to using electros for the input. I didn't try tantalums.

If everything is well matched then there shouldn't really be any potential across the transformer.

(I have yet to convince myself that this is better than just having a clean quiet preamp with oodles of gain.)

 

[trubac]

Depends if you got a clean pre with 80 db gain ...
my Phoenix Audio DRS-8 is going to 80 db , but not without noises ...
And if I want to record medium to soft sources wiith my Melodium 42B, B&O BM4, Tannoy Barrel or Beyer M320, I know I will need at least 75db of clean gain.
Is the DRS + RibbonPre will be quieter than the DRS alone ? that is the question ...

I did try the AEA RPre and didn't like it ... maybe the Grace M101 or Neumann V476b or some Studer pres or AAE, Sonosax, ....

 

[zebra50]

You should definitely check this for yourself and let us know.

Like I said, I'm not yet convinced in either direction, but i have only used my ears, and haven't done a strict signal-to-noise test. I built three slightly different boosters with different transistors, and also tried the commercially available FETHEAD booster, which has a bit more gain.

What I found - or at least what I perceived - using my API A2D preamp was that for many ribbon mics there is very little difference between using a booster and just cranking the API. In both cases the noise from the ribbon mic is amplified, especially with an old oxidised ribbon and a non-optimised transformer.

But if you have a nice quiet mic then you'll be happy!

 

[trubac]

For those old ribbons, 1st thing to do is to rewire the ribbon in differential (stop using the body ofthe magnet as a conductor, but use wires), worked nicely with the 42B.

 

[zebra50]

That works for some mics. Sometimes the 'through-body' conduction works well, like in a healthy RCA 74. But I'll have another look at my melodium. I suppose it depends how well it was implemented, and what's gone rusty since!

This post by Marik is a really good explanation of the resistance noise problems:

http://www.groupdiy.com/index.php?topic=40920.msg507879#msg507879