Original UA console - 100D preamp, EQ, 101D program amp

[abbey road d enfer]

> the amount of NFB reduces at low frequency extremes which means there is a substantial peak in the amplifier response at some very low frequency.
The BA-3C has a dedicated NFB winding, a nice start. Assume that T-2 is "audio reasonable", not extended sub-sonic response.

Not to correct you, but in order not to confuse unsuspecting readers, a transformer does not have an intrinsic LF response. It's the combination of its inductance and the source impedance that drives it (including its own DCR) that defines it in first approximation.

 

[ruffrecords]

Not to correct you, but in order not to confuse unsuspecting readers, a transformer does not have an intrinsic LF response. It's the combination of its inductance and the source impedance that drives it (including its own DCR) that defines it in first approximation.

Strictly speaking it does, as does a capacitor. Both have a zero at zero Hz irrespective of whatever else they are connected to but you are correct, the practical LF response of a transformer is defined by the pole formed by the inductance and the circuit resistance.

Cheers

Ian

 

[emrr]

At best, hiss is constant at all NFB-gain settings. (In practice, total hiss might rise at very low gain, high NFB, because output and NFB loop hiss are constant and input hiss is gained less.)

And we circle around to the practical reality; you don't really hear hiss in a 40 dB design that approaches it's intended output level.  When I crank the BA-11 NFB up and reduce total gain to around 28 dB, there's not a hint of background noise to be found.  

Another practical observation, which I should bother to measure:   When doing spoken voice recordings with something like an Shure SM7, the modern Sytek preamp I own has more obvious hiss at full gain (65 dB quoted) than the majority of 1950ish units set for same gain.   At very least this demonstrates the value of the 'free' input transformer step up gain.   With the USA antiques, it's rare to find a case with less than 20 dB of input transformer gain for mic use.  Many of the later WE and one of the Collins use a 250:30K step up, and that's definitely the lowest quantity to be found from that era.  

BA-3C seems to be an all-purpose medium-level amp. The input pot is not optimal for very low-level work, unless there are two custom tapers.

Not custom, but available either dual carbon in one model or Daven stepped in another.  
I previously said I had removed the initial voltage divider in mine, so as to avoid reduction at that stage.  

The BA-3C has a dedicated NFB winding, a nice start.

I do like the variety of tertiary winding preamps I have, and feel there's an identifiable positive sonic contribution.  I realize from lit that this can cause feedback problems with multiple cascaded amps, but definitely not in today's single preamp world.  

 

[ruffrecords]

> the amount of NFB reduces at low frequency extremes which means there is a substantial peak in the amplifier response at some very low frequency.

Many ways to fix this. Often not pretty.

The BA-3C has a dedicated NFB winding, a nice start. Assume that T-2 is "audio reasonable", not extended sub-sonic response. Make C-6 5uFd-50uFd and C-7 1uFd-10uFd. Response is dominated by T-2 to well below 1Hz, 50:1 of excess gain can be blown-off without significant phase-shift. However leakage of such large caps ('specially in broadcast-reliability use) will be an issue. (In another module, RCA used 1uFd to block 200V then a second 1uFd to block the possible leakage of the first.)

Trouble is, all those fixes work only at one value of NFB  and an assumed value of open loop gain so you can forget it if you want to design a tube mic pre with variable gain by changing NFB which probably explains why most old tube mic pres have only a very limited gain range.

Cheers

Ian

 

[emrr]

Almost no one did variable NFB gain back then, and now it seems there's a small trend of units with about 10 dB NFB variable and a standard gain pot somewhere. 

In practice having a 10-15 dB NFB adjustment added to the typical 40 dB preamp with loop NFB is enough to adapt to modern source levels.  Ideal being a 20 dB spread that allows one to not use a 20 dB pad when levels are only slightly hot, or use a pad and get an extra 10 dB out of the amp when 20 dB pad is too much.

A unique combo is the gain control on the RCA OP-6. 

 

[emrr]

NFB control.  R15 strapping relates to need for 600 ohm output load capability, or not. 

 

[emrr]

Some interesting reading:

http://www.nutshellhifi.com/library/FindingCG.html

Lots of harmonic distortion comparisons and commentaries sprinkled around this site. 

 

[ruffrecords]

Thanks for the post of the Western Electric 141-A schematic. As this clearly demonstrates, the safe way to be able to vary closed loop gain by varying NFB is by having the loop closed at dc. Very low frequency stability problems are thus eliminated because there is no capacitor or transformer in the NFB network.

The topology is very reminiscent of the Pultec MB-1 which uses a 12AX7 instead of the 6J7 and 1/2 a 6SN7 that the 141-A uses. It looks like the transformer accounts for about 20dB of the overall gain and the amp provides 20 to 50dB.

Cheers

Ian

 

[lassoharp]

Some interesting reading:

http://www.nutshellhifi.com/library/FindingCG.html

Lots of harmonic distortion comparisons and commentaries sprinkled around this site. 

Good article.  I found the comments regarding non-bypassed cathode/local NFB very interesting.  In practice I've found that unless I need the extra gain I often prefer the sound of no bypass cap on V1 of a voltage amp.  Never played with that in power driver circuits. 

Surprised that dynamic Rp goes that high.  In the Collins 26, output stage driver is unbypassed, N7 - so dynamic Rp  ~34K?,  plate load R = 22K.  Same condition for V1 of same circuit.  Wiring a similar section on the bench - transformer driving the grids - bypass cap does not seem to give the usual 6-7db more gain.

Those distribution charts are nice.  He doesn't go into the specifics of subjective sonic impressions of each.  I'm inclined to ask if all the others except maybe the transformer coupled would give the sonic impression that something is missing in comparison to the cascade.

   

 

[emrr]

Thanks for the post of the Western Electric 141-A schematic. As this clearly demonstrates, the safe way to be able to vary closed loop gain by varying NFB is by having the loop closed at dc. Very low frequency stability problems are thus eliminated because there is no capacitor or transformer in the NFB network.

I refrained from posting the following earlier, thinking that you possibly meant the entire loop, meaning no coupling caps either.  This is part of an RCA Photophone compressor output stage.

 

[ruffrecords]

I refrained from posting the following earlier, thinking that you possibly meant the entire loop, meaning no coupling caps either. 

That one is particularly good because, not only does the NFB network have a dc path but there is only one coupling capacitor in the open loop which means it is unconditionally stable at low frequencies.

Cheers

Ian

 

[emrr]

Digging around in some old mags, found a March '97 issue of Pro Audio Review with Bob Katz reviewing Fred Forssell's M-2A preamp.  This was a transformerless tube preamp, and there's a pretty glowing bench test with it.  Worth digging out, if you have it.   Review says it's fully differential, with PSU of +350/-330/+12/+48, uses 6072/12AT7/12AU7, little NFB, no overall NFB.

Maybe we can get Fred to talk a bit about this one, 14 years later?


----edit-----


I see it is the 1st version of the Millennia product:

http://www.mil-media.com/m-2b.html

 

[emrr]

When doing spoken voice recordings with something like an Shure SM7, the modern Sytek preamp I own has more obvious hiss at full gain (65 dB quoted) than the majority of 1950ish units set for same gain.   At very least this demonstrates the value of the 'free' input transformer step up gain.

Double checked this today, with a 200 ohm resistor across the input of each.  Noise above about 250 Hz is nearly identical in the Sytek's quietest channel, and the 1947 Gates SA-20 with a 1620 having average noise in the first stage.   Noise rises below 250 Hz in the tube unit, but is not apparent to the ear.    The Gates uses a 200:50K input transformer with no secondary load R; typical of the era.  This feeds a triode strapped 1620 which feeds a stepped 250K volume control.  There is only cathode bypass as local FB at the 1620. 

 

[PRR]

> the modern Sytek preamp I own has more obvious hiss at full gain (65 dB quoted) than the majority of 1950ish units set for same gain.  At very least this demonstrates the value of the 'free' input transformer step up gain.
> Double checked this today, with a 200 ohm resistor across the input of each.  Noise above about 250 Hz is nearly identical in the Sytek's quietest channel, and the 1947 Gates SA-20 with a 1620 having average noise in the first stage.  Noise rises below 250 Hz in the tube unit, but is not apparent to the ear.

Just to be sure what I am reading:

"Modern" and "vintage" have nearly the _same_ noise (above 250Hz)?

Your "modern" has a less-quiet channel?? Is it sick?

No, preamp noise below 250Hz has to be really bad to be heard directly. Typically blower-rumble and truck-leakage will mask it. A tube-swap marathon might find a bottle with lower 1/f rumble, but I say best left alone.

 

[emrr]

Yes, the modern transformerless preamp has about the same noise level as the 1947 piece.  It has increasing levels of hum as you try channels closer to the power supply.   I returned the first one, and the second was the same. I understand later model revisions improved this condition.  Noise in the Sytek is slightly better to the ear than the average modern transformerless preamp, and falls drastically with gain control set at 95% or less.  So long as you don't record quiet sources with low output mics, it is inaudible.   A ribbon always proves happier into a preamp with a step-up input transformer, than into the Sytek.  

http://www.sytek-audio-systems.com/products/preamps/mpx4/

 

[DOX-EMF]

Hello Gents !
Been awhile since last looked over the forum here and glad I did !
This has been an excellent thread !
Being an audio engineer for the last 30 years or so and also a Amateur Radio OP can really relate
to the Lynn Olson article especially ! Thanks Doug for pointing it out..

Recording was so much simpler when it was done with just one Mic direct to laquer. ;D

DOX

 

[szczepi]

did any of you actually built this pre?
I'm looking into it, but I am not sure what kind of transformers should I use. What should be the windings ratio?
Will any of Lundahl audio transrofmers do well?
http://www.lundahl.se/tubes.html

 

[AmpSim]

Hi
who has this pdf file? could u send it to me? I found nothing in the gmail....
thanks a lot!

 

[lassoharp]

Look in Technical Documents - UA

 

[Marì]

The 610 seems to me to be a 'cost reduced' version of the 100 series. The 610 has both tubes set up as two stage CC with NFB which allows the channel fader to be placed between the two stages. The limited 610 EQ looks like a very cut down version of the 100 series implemented in the NFB of the second tube and because there is only one tube driving the output transformer it has a 30K primary which throws away 17dB of gain and may go some way to explaining anecdotal reports of distortion in the 610 under some conditions.

Cheers

Ian

Great Knowledge of vintage consoles! I'm interested in this "two stage CC with NFB" cause I'm looking for the explanation of the circuit of
Langevinn 5116 B. It seems Class A but they say that is a push pull. Do you know if the 5116 B is a circuit the same topology of 610? I mean a two stage cross couple could utilizes a push pull for balancing , can you describe the Ua 610 a class A push pull? Thanks you very much!