Wow,, that was fun!! Seems like the guy was trying his damnest to make a stereo amp out of that mono amp with the least amount of design changes and cost..

You should check out the September 1959 issue of Electronics World. The cover story is an amplifier developed at CBS/Columbia. It has two output tubes, and two output transformers. One transformer is push-pull and the other is single ended. The single ended xfmr primary is in series with whe B+ lead of the P-P transformer. The secondary of the SE transformer is wired in series with a center tap on the secondary of the P-P transformer to ground. The inputs the amplifier are L+R and L-R. The P-P mode of the output stage amplfies L-R (Mono) and the SE mode of the output stage amplifies L-R. The wiring of the two transformers produces Left and Right signals between ground and either side of the P-P xfmr secondary. The amplifier shown in the ariticle is a handmade prototype, but this became a product and was widely used in several Columbia consoles. In the article, they discuss producing L-R and L+R by the way the phono pickup is wired, but in the unit which I have, this is accomplished in the Preamp/Control unit.

Tony, as long as it involves electronics, I'll be watching, because I'm getting back into the electronics field more as I'm getting closer to retirement to have something to occupy my mind and keep it busy. I used to 'piddle' with electronics some several years ago, but life got in the way for awhile. I'm learning lots, and, remembering some of what I learned several years ago. This was fun to watch! It will be VERY interesting to see if he (with your help) can actually get it working. Great video! :)

That's a very clever analysis that you have done here, and if you haven't hit the nail on the head, it seems close enough to me. I particularly liked your comparison to the method used to separate out the left and right channels in a multiplex adaptor. This scheme isn't the same, but the concept is...

Wow, Tony. Just as I thought I'd seen every single amp design, you find this! Can't wait for any follow-up from you or Greg! I like the new format. There was a lot of clipping in the audio, though :(

Strange and, nice explaination . Hope you find out about this thing. More of this type would be great.

Fantastic reverse engineering on that fascinating circuit, Tony. The ultralinear design is similar to the 1957 Eico HF-52 integrated amp I recently covered. That amp also uses EL-34s and tests with a clean output over 50 watts. Similar to the BOC, the Eico HF-52 was introduced right at the early days of stereo. At that time, Eico sold an amp-only version of the mono amp called the Eico HF-50. As I was working on the HF-52 integrated amp, some of my viewers assumed I’d get another HF-52 to create a stereo system, but I didn’t like the idea of having a separate preamp for left and right so I set it up as a mono system as nature intended. How interesting would it have been though, if Eico had sold a stereo version of the HF-52 which included just one mono output amp, but with a stereo preamp? Then, consumers could purchase a matching HF-50 amp to create a stereo system. Had I been George Gott, that’s the approach I would have taken instead of that “multiplexing” route.

I thought this video was good fun for two reasons... One, the mystery and electronics are interesting. Two, your enthusiasm is through the roof and contagious. So, now I am super-pumped, too, and starving for more, even though I know little about the engineering involved. Cool!

I don't know enough to really add anything to the conversation, but I'm already looking forward to the follow up video. Cheers!

Tony, maybe you could get him to send it to you. I think all of us would love watching you scratch your head at what they did 🤪 plus maybe make it a working amp/prototype. It would make a great series and maybe get to the bottom of the endless questions. At least my novice mind has endless questions watching your quick overview. Just a thought.

Great explanation, valuable information. I learned a lot from this video. And I downloaded the booklet, Williamson Amplifier, for further study.

Ha Tony. Look closely at the Macintosh c28 schematics split them consolidate. Yes that is one of the experiments but it is a the basis of c8 and c28 without all the additional features you have the right idea in it. The c28 is a full dual version separate left and right. It gives a enter idea of what is missing as well. Keep up the good work I always enjoy watching your work.

The advantage is, most of the power in the stereo signals is common (mono), so they can get away with a cheaper push-pull transformer that doesn't need to deal with DC bias and thus doesn't need a very large or gapped core. Then the small amount of stereo difference signal is handled by the common-mode transformer which does get some DC flux, but can be much smaller since there are probably much fewer turns than a regular single ended transformer. Ultimately you can get away with far smaller magnetic cores than for two mono SE amps, and similar performance as long as your stereo signals are mostly mono. I don't think the push-pull transformer would strictly have to be bifilar wound. As long as both primaries have good coupling, inductance will be similar for similar number of turns and the DC resistance would just be a bit higher on one of the primaries which is no big deal.

great video....watched it after supper, still thinking about it when i got up this morning!

I’d go with Tony’s conjecture #3; they (Bigg of California) probably realized they could save money by just buying lots of standardized output transformers (1 for each channel) from whomever made them for Fender (or other mfgr’s of guitar amps), rather than have one fiddly common-mode xfmr + a big, custom, bifilar-wound output xfmr + a DPDT switch and a pot. Then they just have two mono-bloc amps to do stereo, with one shared power supply. In fact, if you just get rid of the circuitry to the right of C13/14, R16/17/18, then you just get exactly that - two mono-bloc amps with the 1st stage (V1) being a voltage gain stage, and the 2nd stage being a cathode follower (voltage unity/current gain) stage, whose outputs (from each cathode of V2) go through a 3.3K resistor to the (non-grounded) terminals of the left and right speaker terminals. 🤗

I recall seeing a design using this concept in Australian magazine "Radio Television and Hobbies" published in the 60s. Stereo was sweeping the audiophile world and many enthusiasts had very good mono equipment with heavy and expensive transformers. Ultralinear transformer tappings had come in to get triode performance out of beam tetrodes without loss of output power. It was reasoned that much of the stereo signal had a lot of common mode particularly the lower frequencies so a single large transformer transformer could handle these. The stereo effect was more pronounced at higher frequencies which could be handled by a smaller transformer acting as single ended. BTW, did anyone notice the error in the circuit? There is a short across the primary of the push-pull output transformer at what should be the centre tap.

The way you described it made me think of multiplex also. I bet you would like to have your hands on it to do some experimenting. I would!

It would be a great project to build a duplicate if the darn thing works per your diagnosis. I guess the main problem would be acquiring one of those bifilar wound transformers in a single package. Maybe it could be modified using two separate ot's . Keep it coming Dr. Tony.

I have a Bella, little black schnoodle girl, who likes to bark at squirrels, horses, etc, etc. These are the kind of weird and wonderful pieces of kit I dream about finding on boot sales or in junk shops.

That got me puzzled! I wonder what would happen in the real world. Would love to get it on the bench and properly traced, that would be really interesting! Keep up the good work Tony!

Paul Carlson of "Mr. Carlson's" Lab" would likely be able to help with this. The guy's a genius. See what his take on this is. [Edit: Tony asked for information and help on this amplifier design. Paul Carlson works professionally as a reverse engineer, and is basically an engineer on steroids. He also has designed his own test equipment. If there is anyone that can figure out a circuit and its purpose, it would be him. Besides, he is a very friendly guy.]

The fact that it has a common mode recovery circuit makes me think this is the analogue equivalent of mp3 joint stereo, or a really primitive form of phase, amplitude and frequency detection and " channel steering" where you have a mono "core" plus a bit of right and a mono "core" plus a bit of left. I think it was a method of getting an older mono design to work in stereo and gain a marketing advantage. I'm also speculating that it was a bit too technical for the user compared to the new fangled stereo amps of the times and didn't sell well as it may have had phase issues. Just my UK 5 pence worth.

On the back of a Brian Eno lp. He gives a schematic for third channel to be used. It’s a simple idea, you take the positive outputs from both stereo channels put a +\- 10 resistor inline and hook it up to a 3rd speaker. The record was apparently mixed with this in mind. I’ve tried it and it’s a very interesting way to get a clear spacial quality to the recording in your room. Seems like it is related conceptually at least to this amp.

It appears to me that he was trying to do 2 things at once. Make an amp that could be used as a mono or Stereo amp with only the flip of a couple of switches. And as a noise reduction strategy for audiophiles all at an affordable price. I'm amazed that I've never heard of the company or even owned one of his amps, as my dad was a big audiophile and this shop was only about 5 miles from our home. Good luck with the project! I think that you're on the right track.

Nice analysis of a strange "stereo mono amp" -with only one output transformer! As I understand it the schematic is drawn after the amp has been subject to "heavy modification". I doubt very much it can reproduce a stereo signal when i t has only one output transformer, even with the little "phasing transformer" in the output signal path. I agree that the OPT probably has two separate primaries, the way it is drawn would represent an AC short on it. In my opinion the best would be to convert it back to the circuit of that other George Gott circuit

I have no idea whats going on but when a transformer has " Sample " written on it alarm bells go off, that screams prototype !....cheers.

Hi Tony back in the late 70s a circuit close to what you are describing was used with bipolar transistors used in automotive audio systems to create higher output voltage to drive a 3 speaker system that produced coupled mono for one speaker and stereo for the other two speakers with the phase relations to higher volume signals altering each channel, sort of using IM distortion effects to produce a type of moving center of balance . transformers produce the higher output impedance needed for 8 ohm speakers with thin wiring just in reverse to tube circuits requiring lower output impedances. it is kind of was a pre 1980s quad system sound and distortion was too high for anything other than cheap automotive aftermarket stuff that hung below dash. There were two output Transformers were inter-connected and tapped windings from both transformers fed the 3rd speaker. I service a lot of these for Sears, Western Auto and White's Automotive stores. Any grounding of the 3rd speaker wire was destructive to both channel's output transistors. the schematics never show transformer windings and I never really tried to find out the exact internal winding connections because PC board traces failed before the transformers were damaged. This concept might have come from some of the work in designs of amps from the 50s and 60s you were describing . I was in my early years of College (EE) and the NRI School of Communications Electronics Certificate for FCC license. Did this work of the side for extra Funding. These Systems Left the Market in the early 80s when IC technology replaced the discreet components in automotive Audio and Automotive Audio moved to HIFI and much more power. Like Your Channel you are Keeping it Real.

It seems that the designer’s goal was to have a mono push- pull/ stereo single ended amp. Clearly can only use one output transformer, so I believe that the mono/ stereo switch should be wired-up to manage the signal phases with the aid of common mode transformer. Convergence pot seems to be meant as channels separation adjustment and perhaps please inversion feedback for push-pull (???). The mono stereo switch must be wired much differently than it is in the schematic. The common mode transformer should be meant to recover part of the signal lost in stereo mode due the self cancellation in output transformer (L and R signals are not equal). The output transformer may have both primary and secondary windings separate and connected by the switch. Signal phase understanding here is essential (and complicate) but very interesting as unusual amp. Thanks for sharing this unknown little beast.

Like, Like, Like, will watch again and think about what is happening. Thanks for the look.

Yes you can have stereo output from a single pp OT and smaller modulating transformer exactly as in your schematic. I have old german stereo tube radio with almost same schematic design and i asure you it reproduces very noticeable stereo. In my opinion there is significant crosstalk between the channels, because they couldn't achieve perfect canceling of signals in the modulating transormer. You can see the schematic of the radio and see the similarities and differences because there are such. It is Wega 513. It's speakers layout is 2.1. 1 subwoofer in pushpull and 2 midhighs in se mode - it is very interesting and the radio sound great 😉

Hey Tony, first time commenting I guess. It appears to me they are using what they had to "mimic" a stereo output. I would guess these units would be circa 1959-1961. There was a move by all the major US companies to convert to stereo. The transistor units were around the corner and they were using their mono tube stock to convert to stereo, if only for a couple of years. These units are very similar to the Zenith 5B23 (mono push pull) and 5B27 (single ended stereo). I have both of these units and it is obvious they used the same exact chassis for both units. It appears BOC was doing the same thing. If it were me, I would get new OT's and use a different schematic to give those EL34's a good life. Thanks for the great content!

Looks like you have run into dome prototype work! Too bad it may not be complete!. I think the output xfrm pases throuh the difference of the two channels while the " common mode", passes the summ. On the sec , you got L-R and R-L. Then you add (subtract?) the L+R from the other xfmr, and voilá! You got L and R on each channel, provided the the turns ratio, and phases are correct. Kind of too much trouble to achieve stereo!. I use a similar summ and diff to achieve separation of ultrasound audio in a doppler medical device, to distinguish the blood sounds toward and away, respect of the transducer direction, many years ago.

6:27 - At first I thought that was a 1/4 jack where the STATIC BALANCE" control was. At least it wasn't converted into a guitar amp! WHEW!

The only thing bothering me about the "prototype" hypothesis is the production-style labeled faceplate.

Excellent video Tony! Even more excellent Mystery. Your explanations make sense to me, but like you, I have a lot of questions...most are the same you have. I wish you could post a URL of the schematic diagram of the unit as received. Would love to be able to print one out and study it. I would love to see the unit working...and a view of the waveforms on an oscilloscope.

I wonder if this amp was used in some kind of specialized application. First of all, the unusual labeling "Dynamic Balance", "Static Balance", and "Dampening Factor" are not the typical things you see on a typical amp. Usually you see "Volume", "Tone", "Balance" (Single word), and sometimes equalization pots. I wonder if this is actually NOT supposed to be a stereo output but instead control whether the two input channels are separate or mixed somehow. It appears that the controls are designed to control two different inputs somehow, and perhaps it was used in some kind of PA, DJ, or even a special effects application that involved control from two different audio sources and maybe the "Stereo" output is from both sources together. I also wonder if this is some kind of multiplex encoder or decoder and the output actually was fed into another unit or something or if there was a speaker with a demodulation or filter circuit in it that "decodes" the mixed signal into two channels somehow.

A single ended stereo amp that can be bridged to a push-pull mono amp?

He must of used some program to actually draw out that schematic. It's so precious. Every resistor is identical. I would like to know which program was used. It looks vintage style. I'm not an engineer either, but have hand drawn many schematics of amps that I built.

This is a version of something call Stereo Simplex or Stereoplex. This was described in a 1960 Popular Electronics. Heathkit also made a phonograph kit using this type of amplifier. The phono cartridge is wired out of phase for this to work.

When Bella barked, it got the attention of our Cavalier, Gibson. Started wagging his tail and looking for her!

BRILLIANT ANALYSYS !

Needs more cowbell. :D

Very nice content. I really enjoy your circuit description, I look forward to more content like this. Interesting analysis of the output transformer ; bifilar windings. I hope you can get this amplifier operating, it would be nice to see it perform. I’m sure the design would work, but with how much separation and distortion vs power output? Keep up the great videos.

Very impressive reverse engineering from a “noobie”. Sorry I’m of no help regarding this one but I’ll be watching

So, if I understand your explanation, in single input channel mode, you get two outputs - each one a bit out of phase with each other... I.e. simulated stereo. So you hook up a mono source and get synthesized stereo output. Possibly this was built during the market changeover from mono to stereo HiFi. That sounds like an excellent idea for getting extra mileage from those mono records everyone surely had and didn't want to shell out to buy again in stereo...

Definitely an interesting design, I've never seen anything like this. The way SW1 and SW2 are wired makes no sense to me, I'm pretty sure it's an aftermarket mod. The differential L-R signal output looks a bit like the rear channels on 1970s era pseudo-quadrophonic amps/receivers. Common mode transformer is an interesting thing, and it really needs to be beefy as most audio signal power incl. low frequencies (bass/kickdrum) is gonna be found in the common mode (unless that bass/kick is panned out in the recording), plus it needs to have a slot, since DC is going through it. Moreover, it's gonna introduce hum to the speakers from the power supply's ripple, and it's gonna be there even in a perfectly biased amp. Replacing C11 with a larger one or even making a choke-input DC filter (choke wired between 5V4 cathode and C11+C12) should help.

I can see how it would work, but I can't see it working very well, certainly not well enough to waste a nice output transformer on the attempt, however interesting it might be. Also, no real HiFi circuit would run the input tube in grid-leak bias using a grounded cathode. You usually only find that in guitar amps where they're trying to save parts and squeeze the maximum gain from one gain stage.

Great. Now I'm going to have dreams all night trying to figure out that output transformer arrangement! ;)

That looks like a "center channel" amplifier. With a stereo input, it would amplify the DIFFERENCE signal between the L and R channels.

thats very nice of you tony

I hope it's not wired like Greg's schematic or it's going to blow up when it's plugged in! He shows a ground tap on the 117 V primary of the power trans. but he is missing the ground center tap on the high voltage winding by the 5V4.

Brilliant!

I can't imagine this amp being anywhere near "Fifty" as the nameplate indicates. Seems rather convoluted to try to obtain a stereo output by using the output transformer and the output tubes connected this way to avoid using two output transformers. I also can't imagine the output of either channel (?) being clean and undistorted. Wow, so interesting!

I also wonder if this amp was used in some kind of signal processing application, RF or iF transmission, TV audio or signal (inside the station), or something possibly.

There are some crazy arrangements out there in the world of amplifiers. Danelectro made a guitar amp badged as a Silvertone, that had a PP output running through two OTs in parallel, each having one speaker connected to the secondary. I don't know if this was to save costs because they had a lot of lower powered OTs for smaller amps or what the deal was, but pretty interesting to say the least. At first glance at the amplifier, you would think it was a stereo guitar amp (which they do make as well, but also a strange animal).

It kind of looks like a "L - R" simulated stereo circuit. Dynaco did something like this back in the 70's with an "all passive" circuit to generate simulated "rear channel" output. I actually built stereo "blend" circuit using a quad op-amp. I used it to add "leakage" on older (stereo) recording to make them sound more natural and also to simulate "ambiance" with older mono recordings. Many early solid-state pre-amps (circa 1960's) used to have a " stereo blend" control so the listener could alter the imaging. If you listen to some early Stones recordings there's quite a bit of "unnatural" channel separation. RIP Charlie Watts.

This seems backwards. I'd think the larger output transformer for would be L+R, which would be handling the bass. The L-R would be the smaller one. However, being a prototype, it might be a "will this work?" proof of concept before doing it right.

Maybe the two channels are 'multiplexed' using a single-channel amp (and an 88.2 KHz clock) :)

So is this running class A? It makes sense then. Just looks like a class B. So you amplify the R and L. Summed on the same output transformer, The current draw on B+ will be picked up on the recovery transformer and the difference will be on the main transformer. Only issue is more power is in the smaller transformer. Yes, its a multiplex prototype.

SOUNDS RIGHT TO ME !

i have a German radio that has an output transformer that is wired kind of like this, one side of the pri is open and i cant find one. its weird for sure, no one know how to make one or ever want to rewind it.

Could the Primary and Secondary of the OT be two separate windings and SW2 switch them together for Mono and apart for the Stereo? This was my thinking from the beginning. Then it wouldn't matter if the signals were in phase or not. It would essentially be two SE mono blocks or a single PP mono block.

A bit of George C. Gott trivia via newspapers.youknowwhat: In June of 1949 he sued Earl "Madman" Muntz for defrauding him out of a television business they 'partnered' in. Muntz got him for $70,000 and also Gott's patent on a "'one knob control' invention" No word on how it turned out.

People today would not know how to use this amp, to many knobs to play with. Love to see this up and running and how much power will it put out.

Hi Tony, i have no experience with tube amplifiers, but is the selenium rectifier not drawn in the wrong direction, on both schematics ?

"Pseudo stereo" amp with undoubtedly poor channel to channel isolation and lots of distortion by using common output transformer. Tell me if I'm wrong.

Did they make any custom amps per a customer's request...?

Tony, take note of notes 3, 4 & 5 on the schematic; especially #3 & #5!

INCREDIBLY ASTUTE ! FASINATING !

Could you fix some guitar amps too please?

I think "convergence" is more likely to deal with different gain between the channels to properly mono out the signal. As such, it's probably closer to the dynamic balance.

I assume the test/operate switch is a standby switch?

Not sure that "long story short" was the right term, but very interesting.

I would put that circuit in micro-cap and run simulation tests to see what is correct and what needs to be changed, would give a full picture a bigger insight in what's going on like with the loose transformer wires what are they doing? With micro cap what is free now you can try any scenario you want Did he not write a journal a lot of these engineers wrote journals, might be some information in that or when he was working at Macintosh. did Mackintosh make anything similar to that?

DERIVING STEREO FROM THIS CIRCUIT WOULD BE A MIRICLE !

Hey @Electronics Old and New by M Caldeira mystery for you ;)

this could have a live chassis the transformers could be output transformers

KZfaq will not let me put in a link, If I had your email could send you the link and solve the mystery.

There are too many variables here. The switches are not properly labeled and the tubes, for this to work, would have to be matched at an insanely perfect level, or else you could HEAR the obvious errors... I think. My Brook 10C-3 would have sounded bizarre with unmatched output triodes, for example. Try explaining what "Class G bias" was! Not easy to do, whatever your background! A 2A3 was SUCH a "simple" tube!

That way of using sum and difference seems innovative enough to patent. Perhaps someone can try to find if a patent exists for it? Personally, if I were tasked to convert a push pull mono tube amp design to stereo, I would have the existing output transformer provide L+R with the speakers in series and add a smaller transformer for L-R between the center tap of the output and the midpoint of the speakers. L+R is generally a much bigger signal than L-R.

That's 1959 Popular Electronics April pg 45

Popular Electronics 1958 april pg 45

Test: simplex stereo

Give it to Paul Carlson. He might have some other insight..

Dun dun duuuunnnnnn.. . ...

Test

Theory disconnected from practice is boring to me....well...maybe others appreciate this