Transformerless input tube preamp - is there such a thing

[Ilya]

Just out of curiosity, are there any decent circuits for mic preamps that feed directly into the tube circuit?
I've seen some examples of transformerless outputs (WE circuits shown by @emrr in one of his threads).

It would be super nice to have all tubes and no transformers. Just dreaming))

I haven't seen such input arrangements, but maybe there're some that I'm not aware of.

 

[emrr]

That Forsell / Millenia thing.

 

[Ilya]

You mean M2B? Indeed, I've looked it up and it doesn't have any transformers. Would be interesting to look at the input stage arrangement.

 

[JohnRoberts]

Tube mic preamps generally use step up input transformers.

JR

 

[analag]

I made one some years ago. It sounded good but it was noisy and complicated in comparison to the transformer input type. Fully differential with cross coupled feedback, stepped attenuators for precise matching. The input stage consisted of four paralled LSK170 cascoded into 6DJ8 per differential.

 

[k brown]

My pre based on the Pultec MB-1 was ironless in and out. But then I used it with custom-built unbalanced output mics. I configured the circuit for 30dB fixed gain; just right for most Classical recording situations. The pre was placed onstage at the mics, so unbalanced for only about 12' of HiQ double-shielded (mini-quad braid+foil) cable was never an issue (remember, the Sony C37A, and the Altec lipstic/Coke bottle mics were unbalanced between mic and pwr supply, because they were cathode follower-out; audio-out trafos were in the PSU). The MB-1 had cathode follower out, so it's impedence was low enough for my A/D converter which had 10 kOhm in Z. Unbalanced cables out of the pre at line level - also never a problem; routinely ran lines of 75' - 100'.

At the time I verified the practicality of this with Dave Royer, who also often did the same thing with his home-brew tube mics and 6AK5 mic pres.

 

[NOON]

Plenty of older preamps and mixers that were designed for high impedance mics like crystals. Not exactly hi fi though!

 

[rock soderstrom]

My pre based on the Pultec MB-1 was ironless in and out. But then I used it with custom-built unbalanced output mics. I configured the circuit for 30dB fixed gain; just right for most Classical recording situations. The pre was placed onstage at the mics, so unbalanced for only about 12' of HiQ double-shielded (mini-quad braid+foil) cable was never an issue (remember, the Sony C37A, and the Altec lipstic/Coke bottle mics were unbalanced between mic and pwr supply, because they were cathode follower-out; audio-out trafos were in the PSU). The MB-1 had cathode follower out, so it's impedence was low enough for my A/D converter which had 10 kOhm in Z. Unbalanced cables out of the pre at line level - also never a problem; routinely ran lines of 75' - 100'.

At the time I verified the practicality of this with Dave Royer, who also often did the same thing with his home-brew tube mics and 6AK5 mic pres.

With this approach you can achieve very good results. It is the only way I know to achieve high quality recordings with a pure tube solution without an input transformer.

 

[Dylan W]

There's Allen Wright's RTP3C phono stage, which I've had on my list to try for years but never got around to. Similar to analag's design, it's differential 2SK170 cascoded with 6922. The output stage is some mega-cascoded monstrosity. No global nfb and it's supposed to be built with a high performance shunt regulator. I recall reading that someone tried it for mic use and said it sounded incredible

 

[k brown]

With this approach you can achieve very good results. It is the only way I know to achieve high quality recordings with a pure tube solution without an input transformer.

Yeah, when I first tried it I expected some noise, but the system was absolutely silent. Only time I've ever gotten RF interference on a location recording was with FET mics.

My main system for many years was a pair of Oktava MK-012s with nothing in them but 6205 submini tubes, to the 2-channel trafo-less MB-1 pre.

 

[k brown]

There's Allen Wright's RTP3C phono stage, which I've had on my list to try for years but never got around to. Similar to analag's design, it's differential 2SK170 cascoded with 6922. The output stage is some mega-cascoded monstrosity. No global nfb and it's supposed to be built with a high performance shunt regulator. I recall reading that someone tried it for mic use and said it sounded incredibleView attachment 112301

I dunno - looks a bit like something designed by someone with too much time on their hands . . .

 

[Dylan W]

I dunno - looks a bit like something designed by someone with too much time on their hands . . .

It definitely looks over-engineered, doesn't it. Was always curious how it sounded though

 

[abbey road d enfer]

The only problem with tube circuits for amplification of microphones is the adaptation of tube's OSI (Optimum Source Impedance) for noise performance.
Most tubes have an equivalent input noise resistance of about 20 000 ohms, resulting in about 2.5 uV noise voltage in a 20kHz BW. That's why, for noise performance optimization, transformers with a ratio of about 1:10 provide perfect matching when used with a 150-200 ohms source.
If one would use the approach of paralleling input devices, as often done with transistors, it would take 100 input tubes, which is not terribly practical.
Indeed, using high-impedance microphones would be a possible solution, but not without hinderments.
Actually, condenser mics are high-impedance type as the capsule hits the grid of the tube. Putting a xfmr in the mic to run low-Z signals to mixers/preamps is the most common way, but alternative solutions are posible. For example, one could add a phase-splitter and two cathode followers to run a balanced signal with a global sensitivity of about 400mV @ 94dBspl. The noise generated by the electronics would be about 100dB below, the overall dominant noise being that of the acoustic radiation impedance of the capsule, about 10dB higher.
Of course it wouldn't work with dynamic and ribbon mics.
That's why transformers are commonly used, but also grafting SS circuitry.

 

[murrayatuptown]

The old (50's?) Mullard amplifier & preamplifier (DIY) build book reissued roughly 20 years ago (Audio Amateur a publications?) was mostly, possibly completely, free of mic input transformers, but the claimed s/n was not at all inspiring.

I just pulled the book out of a box, flipped thru it, and put it back. I assumed it was from an era when home recording enthusiasts only had hi-z (crystal?) microphones.

The only thing I thought was inspiring was the coverage on planning tagboard layouts.

Murray

 

[abbey road d enfer]

I just pulled the book out of a box, flipped thru it, and put it back. I assumed it was from an era when home recording enthusiasts only had hi-z (crystal?) microphones.

Actually, many dynamic microphones of the era had built-in transformers with an unbalanced high-Z connection.

 

[thor.zmt]

Just out of curiosity, are there any decent circuits for mic preamps that feed directly into the tube circuit?
I've seen some examples of transformerless outputs (WE circuits shown by @emrr in one of his threads).

It would be super nice to have all tubes and no transformers. Just dreaming))

There are a few reasons. You need to supply phantom power. Noise may be an issue but less so.

I did both SE and BAL input MC Cartridge input Phonostages (that means max. input is ~ 2.5mV with 0.5mV @ 0dBVU) using only tubes. It does take selected low noise tubes like 6922, which are a LOT quieter than most 12AX7 etc.

They can be as much as 20dB quieter than 12AX7, if operated correctly (low anode voltage, high anode current).

Given that common Mic's average 10mV/94dB signal and usually do not get better than 20dB(A) self noise, with selected tubes it's fairly easy to match the noise.

But personally I would use a solid state "electronic transformer" instead. Selecting low noise tubes and finding good ones get's time consuming and expensive quickly. More expensive than transformers.

In effect make a "flying" Sziklai Pair Pre-Pre with (say) 14/20/26dB Gain and use 2N4403 as input (could also use low noise P-Channel J-Fet's, but not sure there is an advantage here) and nice N-Channel Mosfets for the second stage.

Use current mirrors against a negative voltage derived from the heater voltage and a resistor to ground as output and another current mirror/splitter (to ~ 55V) to bias the circuit.

Voila you have a balanced in, single ended out, direct coupled input / output "electronic transformer" that handles phantom power injection just fine and takes the place of Mic-Transformer in any tube Preamp.

It will sound cleaner than a transformer at high levels and probably not THAT different from a transformer at lower levels.

Next? How to make an se in, bal out step-down "electronic transformer" for the output...

Thor

 

[lassoharp]

Just out of curiosity, are there any decent circuits for mic preamps that feed directly into the tube circuit?
I've seen some examples of transformerless outputs (WE circuits shown by @emrr in one of his threads).

It would be super nice to have all tubes and no transformers. Just dreaming))

I haven't seen such input arrangements, but maybe there're some that I'm not aware of.

Depends on what your definition of "decent" means. Many budget line intercom amps or consumer "recording amps" used non input xformer front ends designed for hi-Z crystal or other budget friendly mics. I had one once, and after I recapped it I decided to just try it as is before adding any I/O iron. I plugged a Shure SM-7 directly into the unbalanced 1/4 input with no Z transformer . . . . so not the best combination for good S/N bc an SM-7 is a relatively low output dynamic. The output seemed low but it was just the circuit not passing every little transient like a chips of broken glass shot out of a cannon (like some modern mic-pre combos can do). But to my surprise the sound was quiet nice and more than acceptable for a typical vintage tube sound. No hiss problems and was way quieter than that little Shure and other comparable SS mini mixers common the early 70s.

Maybe check the old Rider catalogs. They always had a lot of consumer/industrial gear like that. A lot of those circuits used a Grid-Leak style input.

So the combo of high quality mic + el cheapo no iron tube amps preamps can work very well, even without doing the "proper" interface (Hi Z converter) for doing recording where the tube sound is what is desired.