Transformerless ribbon mic circuit

[fireworks]

Does anyone have a reference for ribbon mics active circuits that don't use transformers ?
The main problem with the ribbon mics is their low output impedance, of about 0.02 ohm, but I think it should be solvable.
So, any exiting circuits ?

[fireworks]

Nobody knows ?

Another question is this: if an active circuit would replace the transformer that the ribbon mics use for impedance matching, would that be a significant improvement ?
I think it might, since the transformers are usually too non linear if one is looking for minimum distortion but I'm looking for other opinions ...

[kevinkr]

Because of the extremely low output levels and very low source impedance the transformer is almost an ideal step up device translating the large available current into a higher voltage without introducing significant noise. Distortion is easily enough managed. In this case I think the transformer is the best choice from a noise performance standpoint. You may need to shield it to protect against stray magnetic fields, but this may also be a problem with the ribbon itself.

[Allen Wright]

Originally Posted by fireworks

The main problem with the ribbon mics is their low output impedance, of about 0.02 ohm

Nope, the ribbon's OPZ is no problem at all - what is the problem is the TINY output voltage...

Without using a transformer - which is against my religion - the noise aspect can be handled by a large parallel arrangement of common base arranged low noise bipolars.

Then the current output of the ribbon is fully utilised. A look over the prolific thread MMP here on DIYaudio should give you some ideas to start with.

Be aware you can't pass the bipolar's running currents throughthe ribbon, or you will either; 1 damage it or 2/ at least move it out of it's idea position wrt the magnets.

Regards, Allen

[fireworks]

Originally Posted by Allen Wright

Nope, the ribbon's OPZ is no problem at all - what is the problem is the TINY output voltage...

Without using a transformer - which is against my religion - the noise aspect can be handled by a large parallel arrangement of common base arranged low noise bipolars.

Then the current output of the ribbon is fully utilised. A look over the prolific thread MMP here on DIYaudio should give you some ideas to start with.

Be aware you can't pass the bipolar's running currents throughthe ribbon, or you will either; 1 damage it or 2/ at least move it out of it's idea position wrt the magnets.

Regards, Allen

Thanks for the suggestion, I'll take a look.

[LineSource]

Hi fireworks,

I've gotten good sound by adapting a low noise moving coil preamp to a ribbon mike. Take a look at Erno Borbely's:
"All JFET MM/MC Phone Preamp" part 1 and 2 which includes description, schematic and component values. Google the title for the pdf or visit Borbely website.

Borbely's website contains lower noise implementation topologies that Erno sold kits for in the past. All of Erno's articles are a great education.

The diyAudio parts forum has seveal suitable JFETs for sale.

I found it important to purchase several Id graded JFETs and then also hand match for good performance. I found it necessary to mount the preamp very, very close to the ribbon and to avoid any extra aluminum-copper metal diodes. I built my own ribbons and could run the ribbon foil directly to the JFET input. The diyaudio post by John Curl Blowtorch preamp is a good refresher on low noise techniques.

[rayfutrell]

You will need to parallel a lot of BJTs because of the base spreading resistance. Toshiba has transistors with a specified base spreading resistance of 2 ohms, but that resistance is still 100 times the impedance of the ribbon mike. This base spreading resistance is in series with the ribbon mike and will have a very detrimental effect on the SNR.

Ray

[scott wurcer]

Originally Posted by rayfutrell

You will need to parallel a lot of BJTs because of the base spreading resistance. Toshiba has transistors with a specified base spreading resistance of 2 ohms, but that resistance is still 100 times the impedance of the ribbon mike. This base spreading resistance is in series with the ribbon mike and will have a very detrimental effect on the SNR.

Ray

This is more of a noise figure way of looking at it. The ribbon has .02nV of noise. We hooked the classic RCA ribbon to a complimentary bi-polar preamp that only had two pairs of npn and pnp's paralleled and noise was not an issue. This was about .4 to .5nV input noise which technically compared to .02nV is a horrible noise figure.

You would be better off working backwards and computing the output signal for an equivalent SPL and using that number as a target. Once you are down to 10dB SPL noise you have a very quiet mic.

[fireworks]

Originally Posted by LineSource

Hi fireworks,

I've gotten good sound by adapting a low noise moving coil preamp to a ribbon mike. Take a look at Erno Borbely's:
"All JFET MM/MC Phone Preamp" part 1 and 2 which includes description, schematic and component values. Google the title for the pdf or visit Borbely website.

Borbely's website contains lower noise implementation topologies that Erno sold kits for in the past. All of Erno's articles are a great education.

The diyAudio parts forum has seveal suitable JFETs for sale.

I found it important to purchase several Id graded JFETs and then also hand match for good performance. I found it necessary to mount the preamp very, very close to the ribbon and to avoid any extra aluminum-copper metal diodes. I built my own ribbons and could run the ribbon foil directly to the JFET input. The diyaudio post by John Curl Blowtorch preamp is a good refresher on low noise techniques.

Thanks for the schematics and the other infos, I'll take a look at them.
I'm quite interested in your ribbon mics: do you have pictures or construction techniques or audio samples ? How noisy are your mics compared to an identical one with a transformer ?

[fireworks]

Originally Posted by scott wurcer

This is more of a noise figure way of looking at it. The ribbon has .02nV of noise. We hooked the classic RCA ribbon to a complimentary bi-polar preamp that only had two pairs of npn and pnp's paralleled and noise was not an issue. This was about .4 to .5nV input noise which technically compared to .02nV is a horrible noise figure.

You would be better off working backwards and computing the output signal for an equivalent SPL and using that number as a target. Once you are down to 10dB SPL noise you have a very quiet mic.

Scott, thanks for the info. Do you have a reference for the 0.02nV noise ? Can it be computed from some other available data ?