Characteristics of Guitar Amplifier Output Transformers, and criteria for design

[Helmholtz]

Quote Originally Posted by Helmholtz View Post
Short and simple: For a given core, saturation determines the minimum number of primary turns for max. peak current to be expected.
Not exactly. Core saturation determines the number of turns and core area needed to withstand a certain amount of volt-time integral. That's not directly related to the peak output current to be expected. Primary magnetizing current is not the same as reflected secondary current. The two are only very indirectly linked. It's not in general possible to saturate a transformer core from the secondary. There is an exception to this for half wave rectification in small transformers with high resistance primary windings, and I have never seen this exception demonstrated.

As said, for a given core. Of course I meant max. peak value of magnetizing (not total) current. In a well coupled OT the (symmetrical) load current has no influence on saturation as primary and secondary load fluxes cancel. Magnetizing current is measured without load (open secondaries).

[nickb]

Most transformer practice was codified long before we had machine ways to handle systems of partial differential equations. The transformer designers of that time went off and made up charts and graphs of measure quantities of every transformer they built and use that as a way to iterate in on a design. The history book of transformer designs made and measured was a highly treasured intellectual asset of every transformer design shop. A great deal of transformer practice depends on knowing some kind of bounds on what you might see, then going off and designing with small variants and extrapolations of what you did see.

Don't underestimate the importance of this point. This is why I write the specs and hand them over to the transformer manufacturer to do the actual design. Not one failure so far equates to much happiness.

[R.G.]

As said, for a given core. Of course I meant max. peak value of magnetizing (not total) current. In a well coupled OT the (symmetrical) load current has no influence on saturation as primary and secondary load fluxes cancel. Magnetizing current is measured without load (open secondaries).

I think I misunderstood your comment, for which I apologize. I think I mentioned earlier that there are many ways to view a transformer's operation. I tend to view it from the standpoint of the volt-time integral for most core operations.

Yes, for a given core (which sets the magnetic path length and for a given stack of laminations sets the core area and also residual air gap) there is a maximum ampere-turn product you can allow. That ampere-turn MMF results in a flux density, which you want to keep out of the soft knee of saturation. So you keep adding turns until the designed criteria of maximum AC input voltage and minimum half-cycle time do not let a a magnetizing current flow that causes enough ampere turns to get the flux too deeply into saturation.

From the volt-time point of view, the core allows a current through that is the time integral of V(t)dt/L, and in a real world, non-simplified view includes the fact that L is actually L(t), or more accurately, L(i) where i is also time varying. I just view the magnetizing current in the primary as a result of volt-time. Raising the number of primary turns increases L by the square of the increase ratio whatever the "L" is as a time function, and that in turn slows down the magnetizing current that gets ramped up to.

[Malcolm Irving]

One experiment to find out more about the magnetising current in guitar amplifier output transformers would be to run tests with the OT secondary open-circuit! Horror of horrors - that is exactly what you shouldn't do to a tube guitar amp! But if we can guarantee to keep the input signal sinusoidal and only at low frequencies, I think it would be OK. The magnetising inductance would present a low impedance load (albeit inductive) to the tubes.

We could then measure the current going into the OT primary and see a pure magnetising current waveform (not obscured by load current reflected from the secondary).

Any volunteers?

[Helmholtz]

One experiment to find out more about the magnetising current in guitar amplifier output transformers would be to run tests with the OT secondary open-circuit! Horror of horrors - that is exactly what you shouldn't do to a tube guitar amp! But if we can guarantee to keep the input signal sinusoidal and only at low frequencies, I think it would be OK. The magnetising inductance would present a low impedance load (albeit inductive) to the tubes.

We could then measure the current going into the OT primary and see a pure magnetising current waveform (not obscured by load current reflected from the secondary).

Any volunteers?

No problem, if you disconnect the feedback loop and measure at low frequencies, where the peak current and the voltage-time-integral is maximal. Alternatively you could measure at line frequency with a step-up transformer and a variac. But the standard method is to measure the voltage-time-integral, which determines primary net flux and does not depend on the load.

[Malcolm Irving]

... Alternatively you could measure at line frequency with a step-up transformer and a variac. … .

Well yes, but what I was interested to find out more about is how much the output tubes can drive the OT into saturation. (I should have said that.)

[Helmholtz]

Well yes, but what I was interested to find out more about is how much the output tubes can drive the OT into saturation. (I should have said that.)

Just measure the max. V/f-value the amp can produce at the primary with a (speaker) load and adjust the variac accordingly. I mention speaker load because the speaker impedance rises considerably (up to a factor of maybe 10) at the bass resonance, allowing primary voltage to rise.
But if you have a scope with an integrating function, it is simpler to measure the max. primary voltage-time-integral over a half-wave.

[SoulFetish]

Don't underestimate the importance of this point. This is why I write the specs and hand them over to the transformer manufacturer to do the actual design. Not one failure so far equates to much happiness.

Could you elaborate on this point. I'm realizing that the best way to learn and understand the concepts of transformer design and operation intuitively is to actually build one(or more). But, I would like a professional, working transformer by the end of June. So I may have one wound for me while I read over the texts and materials and hopefully continue this conversation.

[nickb]

Could you elaborate on this point. I'm realizing that the best way to learn and understand the concepts of transformer design and operation intuitively is to actually build one(or more). But, I would like a professional, working transformer by the end of June. So I may have one wound for me while I read over the texts and materials and hopefully continue this conversation.

There's not much to say really (sorry to disappoint). I've always been able to find suitable OPTs off the shelf so it's really PTs that are of interest which I don't think is your focus. For these I specify the nominal primary and secondary voltages, the max VA and max primary voltage, the frequency (50/60 Hz), then each secondary voltage, regulation and current. Also mechanical stuff: general construction i.e drop thru', maximum dimensions etc of course. All one or two off replacements so cost is not is not the most important factor.

Well there is an exception which is a funny little story. Dude came to me in a panic with his dead (very expensive) boutique amp as he has a big recording session in a couple of days and he MUST have this amp. Naturally no schematics. Turns out the OPT has a short. The manufacturer and winder are both out of stock. The design used a couple of bottles running a single ended class A and each one drives a separate winding on the OPT. The only thing I can find that will physically fit is a push pull type. So I explain it all to the customer and he's happy for me to modify it to class AB push-pull with about twice the output power and then change it all back once crisis is over and I can get the original part. I did the work. He did the session and never came back to get it reinstated. He was happy with how it was. I'm saying no more. Draw your own conclusions.


PS: Here is a pretty good practical guide and a great insight into the process Wolpert_Audio_Xfmr_Design_Manual.pdf

[J M Fahey]

Could you elaborate on this point. I'm realizing that the best way to learn and understand the concepts of transformer design and operation intuitively is to actually build one(or more).

Well, not really

That will give you some practical skills, but nothing on Theory or new design knowkledge.

You would need to build *many* , varying winding turns per volt, interleaving, core material, core stacking, gap if introduced on purpose, and then measure and compare them all, to see what affects what and how much.

That is not happening for any standard Tube amp builder, not even for Commercial "wind by the book" or "clone standard stuff" winders, I bet only a few Pro winders, think MM, Hammond, Schumacher, Drake, Heyboer, Dagnall, etc. will do that.

Way too high for us, mere Mortals.

[J M Fahey]

There's not much to say really (sorry to disappoint). I've always been able to find suitable OPTs off the shelf so it's really PTs that are of interest which I don't think is your focus. For these I specify the nominal primary and secondary voltages, the max VA and max primary voltage, the frequency (50/60 Hz), then each secondary voltage, regulation and current. Also mechanical stuff: general construction i.e drop thru', maximum dimensions etc or course. All one or two off replacements so cost is not is not the most important factor.

Well there is an exception which is a funny little story. Dude came to me in a panic with his dead (very expensive) boutique amp as he has a big recording session in a couple of days and he MUST have this amp. Naturally no schematics. Turns out the OPT has a short. The manufacturer and winder are both out of stock. The design used a couple of bottles running a single ended class A and each one drives a separate winding on the OPT. The only thing I can find that will physically fit is a push pull type. So I explain it all to the customer and he's happy for me to modify it to class AB push-pull with about twice the output power and then change it all back once crisis is over ans I can get the original part. I did the work. He did the session and never came back to get it reinstated. He was happy with how it was. I'm saying no more. Draw your own conclusions.

Happens all the time, meaning the Booteek babble is usually that, babble, no real improvement (in at all) over "things properly done the established way".

He must have found that regular Push Pull is better (certainly louder) than his clunky "side by side single ended" arrangement.
Definitely louder , and strains transformer core way less.
FWIW I know two guys who carried the single ended concept (which I admit has an interesting flavour of its own) a little beyond its comfortable area (which is, say, a single 6V6/EL84 for very good 5W); one is a Brazilian friend who makes a single KT88 "20W" amp. Must actually be 14/15W at most.
Everybody "loves" the idea .... but their "wallet votes" say the contrary, he hardly sold any, beyond the original order.
290 | AcedoAudio Amplificadores Valvulados
It sounds good, but maybe the fact that it uses a humongous OT (old style Fender OT), has only 14/15W , not enough to play along a (Brazilian or Argentine) Drummer, and is quite expensive for what it is, has conspired against sales.
More important, I fail to hear a sonic advantage over a similar, conventional PP 15W amp, which sounds as good and is way less expensive.

And now an Argentine colleague is offering a similar , "15/18W" one with *three* 6V6 in parallel.
Oh well, I wish him luck ... heŽll need it.

[Helmholtz]

...and don't underestimate the safety aspects like creepage and clearance distances, hi-pot testing and the likes.

[J M Fahey]

...and don't underestimate the safety aspects like creepage and clearance distances, hi-pot testing and the likes.

I very much doubt booteek makers worry about that

[R.G.]

Juan's right, all counts.

Winding transformers is a different skill from designing transformers. The design process involves picking core material type, core lamination size and stack, copper wire sizes, division of windings into sections and layers, computing the build height and width interatively, picking layer materials, and choosing insulation classes for wire, former, and layer insulation.

Once you get the design done, you can start procuring materials and finally winding.

The materials are not prohibitively expensive, but they are hard to find in small quantities. So while you're messing with your design, it would be good to be finding suppliers. It's hard to buy just enough of the materials. iron core laminations, magnet wires, interlayer insulation sheet, core former tubes, insulating tape, and so on. Minimum quantities will be a constant misery on all of these.

You're probably going to need to make or buy some kind of winding machine. Both hands will be busy guiding the wire where it goes on each layer, taping things in place, and so on. It is entirely possible to build something like this, but it's a PITA. I had access to a prototyping transformer winder at my work way back when, along with a selection of core sizes, magnet wire, insulating sheet and tapes. And it was still a lot of work winding a transformer.

I've done some transformer winding for myself, in my garage. Juan's done more of this than I have, I think, so he can provide a good bit of advice on the practicalities. My take on it is that is that you have to really LIKE the process of winding the things and doing all that work, buying and stockpiling all the supplies and so on, to do the several iterations of each transformer design that have historically been necessary. That is, you would need to be into transformer design and building as a hobby nor business.

I must admit - I'm not. I have some understanding of and experience with design and winding, but I don't love it enough to make a hobby of it.

[SoulFetish]

Well, not really

That will give you some practical skills, but nothing on Theory or new design knowkledge.

You would need to build *many* , varying winding turns per volt, interleaving, core material, core stacking, gap if introduced on purpose, and then measure and compare them all, to see what affects what and how much.

Yeah, but that just sounds like a lot of work