• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

Advise please on, er, output trafo advice!

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Greetings to the forum.

So, a young man called to see the grumpy old electronics geyser (that would be me....), for advice on building a valve audio amp. After a short discussion it becomes apparent that the young man is to all intent an educated novice, driven more by enthusiasm than knowledge. The discussion eventually gets onto the subject of the output stage where the young man states that he intends to use a class B push pull topology that utilizes TWO output transformers, one for each output valve, with the secondaries wired in series so as to combine the two phases back together!

I tried to explain the technicality of why this approach wouldn't be satisfactory, and in fact be a waste of two perfectly good SE output transformers, (assuming that's what he had in mind to use), only for my pearls of wisdom (!) to be received by indignation and disbelief on the part of the novice.

Has anyone a short explanation, in layman's terms, that would satisfy the newbie without his having to learn the minutiae of transformer working principals? So he realizes that I'm not just trying to urinate on his chips.

I'm just getting too old for all this crud!

Norman.
 
Moderator
Joined 2011
I tried to explain the technicality of why this approach wouldn't be satisfactory, and in fact be a waste of two perfectly good SE output transformers, (assuming that's what he had in mind to use), only for my pearls of wisdom (!) to be received by indignation and disbelief on the part of the novice.

I say experience is the best teacher, let him do that and find out for himself.
 

PRR

Member
Joined 2003
Paid Member
Radiotron Designers Manual, P-P output transformer section, emphasizes "tight coupling" between the two sides or you get gross rattles.

That's not the whole story but shows why the two sides MUST "work as one".

All the basic variations were discovered and investigated by "geysers" even older than you. It is unlikely he has invented a New Thing first time out. Have him show you an existing design which follows his plan.

Or just build it. On his money. (Problem may be that when it sucks, he'll blame the geyser.)
 

45

Member
Joined 2008
Radiotron Designers Manual, P-P output transformer section, emphasizes "tight coupling" between the two sides or you get gross rattles.

That's not the whole story but shows why the two sides MUST "work as one".

All the basic variations were discovered and investigated by "geysers" even older than you. It is unlikely he has invented a New Thing first time out. Have him show you an existing design which follows his plan.

Or just build it. On his money. (Problem may be that when it sucks, he'll blame the geyser.)

Or one could just point out that the tight coupling needed for class B operation was exactly the problem solved by McIntosh. So if such amps with their complicated transformers are still popular and his idea has never seen the light there is a valid reason....:D
No need to waste money.
 
Disabled Account
Joined 2013
What would you say this is? Certainly not Class B, both has same output but out of phase and output are added, bridging amp or similar, workable?
 

Attachments

  • 193957ogr19v44xvx0mejz.jpg
    193957ogr19v44xvx0mejz.jpg
    112.4 KB · Views: 204
Greetings to the forum.

So, a young man called to see the grumpy old electronics geyser (that would be me....), for advice on building a valve audio amp. After a short discussion it becomes apparent that the young man is to all intent an educated novice, driven more by enthusiasm than knowledge. The discussion eventually gets onto the subject of the output stage where the young man states that he intends to use a class B push pull topology that utilizes TWO output transformers, one for each output valve, with the secondaries wired in series so as to combine the two phases back together!

I tried to explain the technicality of why this approach wouldn't be satisfactory, and in fact be a waste of two perfectly good SE output transformers, (assuming that's what he had in mind to use), only for my pearls of wisdom (!) to be received by indignation and disbelief on the part of the novice.

Has anyone a short explanation, in layman's terms, that would satisfy the newbie without his having to learn the minutiae of transformer working principals? So he realizes that I'm not just trying to urinate on his chips.

I'm just getting too old for all this crud!

Norman.
Two issues to start with :
- the transformers will carry power, thus needs an airgap with following reduction in power
- where to take feedback ? One transformer ? both (how)?

in addition money. About twice the cost.
 
Thanks for the comments.

I've just been visited by the young man who was adamant about his idea with the SE transformer double up for class B push pull. It transpired that he has acquired a box of them NOS, and this afternoon brought them to show me that they were real pukka audio components.

And indeed they were.

The kind of things that would have been used to drive a four inch speaker in a sixties TV set, E,I, laminations gapped with paper shim, crimped frame and two part plastic bobbin, secondary over primary.

I've suggested that if he wanted to use them for PP, better results would be likely if they were incorporated into a series output pair style circuit, (transformers de-gapped?) If his first plan failed ;)

Any how, I've sold some surplus valves and bases, and he's gone home vowing to prove the old geysers wrong.
 
Could pull the I lams off the two xfmrs (grind off the excess outer lamination crimp band) and mate the two E sections together with a long SS hose clamp(s) around the outer edge of the crimp band, pulled up tight. No air gap needed then for P-P. Should work better (higher primary inductance) than the two SE xfmrs used separately.
 
Could pull the I lams off the two xfmrs (grind off the excess outer lamination crimp band) and mate the two E sections together with a long SS hose clamp(s) around the outer edge of the crimp band, pulled up tight. No air gap needed then for P-P. Should work better (higher primary inductance) than the two SE xfmrs used separately.

Greetings smoking-amp.

That is an intriguing idea, has it been done successfully before? I'm personally interested, not to pass it on to my young friend though, as it seems he is the Messiah of audio, come to lead all of us other poor ignorant creatures from our blind fumbling attempts at hollow state perfection :eek:

PRR said:
"Radiotron Designers Manual"

Thanks for that, I have a PDF of the fourth ed. that I emailed the relevant pages from to him. he replied. "Thanks. But now that I've read it, its WRONG".

I replied that I've been working with this technology since the late nineteen forties and that the book is one of the better on the subject, his return was that he had recently passed his higher national diploma in electrical and electronic engineering and therefore his knowledge was at a higher level of refinement.

Having never once constructed a circuit with a valve I'm guessing. I wonder if there are any books on how to drive a car?

Well that's him cut off!

Trolls are everywhere these days.

I hope he reads this, AND, I wish I'd charged more for those parts.

Thanks Guys,
 
Free Grid
We have all had moments like that.
I thought 'I've cut him off' very funny.
Glad you liked it Allen4411 I do like a good biasing reference. He can return when he is a little less negative.
Alan4411 said:
''No man was ever so completely skilled in the conduct of life, as not to receive new information from age and experience.'' J. Swift
I will remember that in case I talk to youthful arrogance again.

It does seem to be silly season. A chap from Lincoln drove here this morning, round trip of approx. 140 miles, to pick up a Mazda UU5 rectifier that I had (NIB), for his radio restoration project. He refused to pay the £10 I was asking as being "too much" and drove home empty handed! You just can't help some people.:confused:
 
Greetings smoking-amp.

That is an intriguing idea, has it been done successfully before?

So called "cut cores", or C cores, are made similarly. A wound strip of continuous lamination is wound on a rectangular form with some surface binder or spot welded to hold.

Then it gets cut into two U sections, which are then inserted into a bobbin. Sometimes two such cores get used per bobbin for more area. (around opposite sides of bobbin) The "cut" however is ground and polished to a mirror finish before it is ready to use. And a steel tension band is also used around the core to hold it together.

So the mated E sections will not perform as well without the grind and polish operations. But one should expect performance at least half way in-between SE and normal P-P xfmrs as far as primary inductance goes or coupling from P to P. Clearly an improvement over just the two SE xfmrs, since there will be no paper gap layer.

The well made P-P case might have 10X the primary inductance of the SE xmfr, maybe can expect 4X for the bonded E cores. The tension of the SS hose clamp bands around the E's will make a difference.

Could belt sand and polish the E Core faces, but a lot of messy work for a smallish gain. Would be difficult to avoid rounding the surfaces or edges too, without some automated equipment. Might make it worse.

Also, the original SE bobbins will likely have been wound with extra turns to make up for the low effective permeability of the gapped core. This will end up as extra winding resistance for the P-P core case. And the extra turns will produce more leakage L and distributed C compared to an optimally wound P-P xfmr.
 
Last edited:
,.....the original SE bobbins will likely have been wound with extra turns to make up for the low effective permeability of the gapped core. This will end up as extra winding resistance for the P-P core case. And the extra turns will produce more leakage L and distributed C compared to an optimally wound P-P xfmr.

I was interested as it seemed on the face of it a good way to make an experimental transformer, and also the two primary and secondary windings are effectively side by side, they could be made electrically matched, if not ideally inductively coupled.

The transformers that I was shown also had two part bobbins that would make them a snap to modify if the extra L/C was a problem.

Probably not worth the effort in this case though, they just weren't big enough to be any real use.
 
Then it gets cut into two U sections, which are then inserted into a bobbin. Sometimes two such cores get used per bobbin for more area. (around opposite sides of bobbin) The "cut" however is ground and polished to a mirror finish before it is ready to use. And a steel tension band is also used around the core to hold it together.

They are also cut and wound so that they are 100% grain oriented. EI laminations only have 2/3 of grains oriented. If you join the two E's only about 50% because the non-oriented part in EI laminations is on the E.
 
I have a PDF of the fourth ed. that I emailed the relevant pages from to him. he replied. "Thanks. But now that I've read it, its WRONG".

I replied that I've been working with this technology since the late nineteen forties and that the book is one of the better on the subject, his return was that he had recently passed his higher national diploma in electrical and electronic engineering and therefore his knowledge was at a higher level of refinement.

So what?

It´s actually quite the contrary.

FWIW I started studying Engineering in 1969 and even way back then, tubes were not taught any more.

Only a few textbook lines mentioned them, everything was Solid State and Digital anything was the new Big Thing, go figure.

Only ones who studied them were those specializing on RF and Industrial Electronics, because Transmitters and, say, high power oscillators used in plastic soldering or metal heat treating still used Tubes, *large* ones, usually ceramic, largest fan or water cooled.

But Tube * amplifiers* ??? Preamps? ... old guy obsolete stuff.
Even TVs had gone solid state.

What´s my point? .... I VERY MUCH doubt your friend has received *any* training at all on Tube amplifiers and related stuff.

For that, and much else, you must always rely on the old accumulated knowledge from the 40's to 60's ... such as , for example, the venerable Radiotron Designer´s "Bible"

He does have a Degree, no doubt, and I respect it, but he got it on a *different* subject.
 
He does have a Degree, no doubt, and I respect it, but he got it on a *different* subject.

Yes, exactly.

When these "types" start waving sheepskins in my face it just tells me that they have just lost the *discussion*. I won't then continue, as it would be the definition of futility.

The first clue in this instance for me was when the young man used the contraction "HI-FI", in the same breath as "simple class B".

My parting shot with him was:- "My knowledge may be limited, but I've forgotten more about the technology from the forties to the eighties, than YOU could ever learn. I've also met a lot of PhDs who work in fast food outlets. And, thats Mister grumpy old git to you sonny boy." :D
 
Just for interest, as the method has been mentioned in the thread, I have a friend who built a stereo homebrew single ended amp design that he used the double C core method of transformer construction. The actual material he used for the cores was annealed steel banding as used for securing items for shipping!.. No doubt sub optimal, but, the amplifier did work well, and sounded clean and unobjectionable. As he said, if you don't try, you won't know.
 
Greetings smoking-amp.

That is an intriguing idea, has it been done successfully before? I'm personally interested, not to pass it on to my young friend though, as it seems he is the Messiah of audio, come to lead all of us other poor ignorant creatures from our blind fumbling attempts at hollow state perfection :eek:



Thanks for that, I have a PDF of the fourth ed. that I emailed the relevant pages from to him. he replied. "Thanks. But now that I've read it, its WRONG".

I replied that I've been working with this technology since the late nineteen forties and that the book is one of the better on the subject, his return was that he had recently passed his higher national diploma in electrical and electronic engineering and therefore his knowledge was at a higher level of refinement.

Having never once constructed a circuit with a valve I'm guessing. I wonder if there are any books on how to drive a car?

Well that's him cut off!

Trolls are everywhere these days.

I hope he reads this, AND, I wish I'd charged more for those parts.

Thanks Guys,

I'd be worried. Please tell him that the voltages involved in tube circuits can be dangerous.

Then get out of his way, let him re-invent the wheel, and hope he comes back some day when he sobered up.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.