Cheapest power supply transformers?

[R.G.]

Why dont you wind your own transformers? It's not hard. A stack of iron, some wire and a bobbin... and the up side is you get just what you want not a big bundle of compromise.

... and also a puff of smoke and perhaps a puddle of molten copper, maybe a fire, if you miss some details.

Line isolation transformers, especially multi-hundred watt ones, are not a good beginner DIY project. It's something like mail ordering eggs to start your own poisonous-snake farm. There are some important details to know before you start.

[Enzo]

This is your snake. This is your snake on drugs. Any questions?

[jim1976]

Well if your dumb enough to plug it straight in... put a megger on it first and check the insulation, put a light bulb in the line even, its pretty basic? really cant see how you'd melt anything much if your finger is on the power switch. As a kid my father wound thousands of high to very high voltage transformers at home with a drill, no one died nothing burned down.

Lot of people wind pickups is not very different, c/t transformers are a little more tricky because you start at the c/t and wind two at once. The maths is easy too.

How do you design the perfect amp compromising on the most important parts?

[R.G.]

Well if your dumb enough to plug it straight in...

You don't have to be dumb to do that. Just uninformed that it's a bad idea.

You don't have to be dumb, just uninformed, to know that you need 1500V to 4000Vac isolation between primary and secondary and between primary and core, and either multilayer insulation or 8mm creepage and clearance to meet modern safety standards. A line isolation transformer probably works OK - the first time at least - if you don't do any primary/secondary insulation at all. But presumably one would want to use it a second, third... 243rd time, too.

put a megger on it first and check the insulation,

A beginner doesn't know what a megger is. Probably many people here don't.

put a light bulb in the line even, its pretty basic?

It is basic - once it's explained to you in detail. A beginner doesn't know this. Many of the people whose first post here is "My amp is [insert disaster here]!! OMG! What do I do? Can I fix it by replacing the fuse?" don't know that either.

really cant see how you'd melt anything much if your finger is on the power switch.

I've seen it happen. If the wall breaker happens to not trip or to be welded shut, the fault current in a primary short can dump several KW into one spot in the winding, Really Fast. And a beginning transformer hacker probably doesn't know that he needs to keep his finger on the switch, does he?

As a kid my father wound thousands of high to very high voltage transformers at home with a drill, no one died nothing burned down.

Yep. And one anecdote is worth the same as 1000 of them - your father may have been skilled, or just lucky. He certainly was experienced if he wound thousands of them. How many years did it take him to wind his first thousand? Did he have any training?

Lot of people wind pickups is not very different,

Except that you're using wire that's many times as thick, and putting deadly voltages on the winding after you're done.

c/t transformers are a little more tricky because you start at the c/t and wind two at once.

Hmmm. Tricky. Yep. Seems like that might need some learning and study before trying it the first time, maybe.

The maths is easy too.

I completely agree with this. Transformer math is simple arithmetic/algebra. But which equations do you use? Where did that beginner learn that?

How do you design the perfect amp compromising on the most important parts?

Good question. I'll take it in parts.
1. What is the perfect amp? By whose standards? If knowing how to do the perfect amp design is already known, I'm pretty sure that someone has already done it. If "perfect" is different per person, then by definition a DIY'er, particularly one without a lot of experience doesn't already know how to do it, does he? And is "perfect" one dimensional, multidimensionable, or even measurable? And if it's not measurable, how do you know when you're there? If it is measurable, what do you measure?

2. Compromising is the biggest fundamental step in design. Many design features have side effects that oppose one another. Design is fundamentally a process of compromise at many levels.

3. What *are* the most important parts, anyway? If you mean that the power transformer is (one of) the most important parts, I'm curious as to how you go there. Power transformers are one of those things where you mostly have to have enough but not too much. There are many ways to design alternate PTs that work just as well. But there are lots of ways to design one that's insufficient, too. I personally think that tube choice, circuit design and OT design are far more important than the PT.

4. How does the PT contribute to the perfect amp? This is really an extension of (1) above, but with more detail.

[jim1976]

I think there is just as much chance of any of this happening with a shorted transformer if your inexpedience and putting an amp together dont you think? He was talking of modifying a microwave transformer which are upwards of 2000v so you have to rewind the secondary why not just wind the whole thing?

Well there is an ideal transformer design and it is what your project requires no more no less, with the correct voltages and wire size for the current your using. The less compromise the better the closer you are to your design, by perfect I mean achieving what you set out to do in the first place. If you set out to build a 200watt class b monster and end up with a Marshall 100watt clone is it just as you say how the design fundamentals work?

Dad never had any experience winding transformers, he's an engineer. The only reason he did it was to save money... to quote the topic "cheapest power supply transformer"

You seem pretty knowledgeable in design and building, it wouldn't be a big step to try knocking a couple up both input and output, you certainly understand the safety issues.

[Enzo]

Hi Jim, welcome to the forum.

RG is in fact an engineer himself. We have a good number of them here. Engineers, that is.

[R.G.]

I think there is just as much chance of any of this happening with a shorted transformer if your inexpedience and putting an amp together dont you think?

Yes. There are many ways to shoot yourself in the foot.

He was talking of modifying a microwave transformer which are upwards of 2000v so you have to rewind the secondary why not just wind the whole thing?

Well, to start with, I think that a beginner messing with a 1-2KW transformer that puts out 2000V is a Bad Idea, too. Again, the operant word here is "beginner". Someone who's trained, who understands the requirements and safety issues is a big boy and on their own. Presumably someone who is educated can make their own decisions.

On the topic of MOTs, (1) they're big; way too big for an optimal amplifier power transformer, and quite heavy; (2) many of them use internal magnetic shunts for secondary current limiting; these have to be removed (nondestructively to the rest of the transformer); (3) MOT laminations are often sealed together with a bead of weld along the stack. This makes destacking and restacking impossible, and I consider hand-threading the wires into a welded stack to be a silly exercise. Better to work the hours and buy one than spend the time, get it wrong a couple of times, burn up materials, and so on.

Well there is an ideal transformer design and it is what your project requires no more no less, with the correct voltages and wire size for the current your using.

That is, by definition, a compromise. More iron versus more copper? Heavier, but lower iron and copper losses, but also higher cost in materials? Lighter, more economical of iron and copper, but running hot and more on the edge of reliability? Higher temperature insulation class to let it run up at 150C to 180C internal, but so hot that it blisters you when you touch the core? I designed transformers for a living for a while. There is no single optimum design for a transformer in most cases. It depends on what you set out to do and how much you can spend. The O/P was optimizing for lowest cost. In that case, Enzo had it right: get an older amp and scavenge the PT out of it. That's by far the cheapest solution. But then the voltage, currents, size and other ratings may not be perfect for the circuits. Compromise. It's what's for dinner.

The less compromise the better the closer you are to your design,

Design is the art of visualizing a desired result and arranging available parts and processes to get there. It's not a hypothetical "perfect circuit" that you have to achieve. Less compromise is not better. Making a better/cheaper/faster set of compromises is what it's about.

by perfect I mean achieving what you set out to do in the first place. If you set out to build a 200watt class b monster and end up with a Marshall 100watt clone is it just as you say how the design fundamentals work?

The problem people get into is thinking they're visualizing a result but actually visualizing a path. Building a 200W class B monster is trivial, if you decide that solid state amps are OK and that 200W was what you wanted. Oh? You have to have TUBES in there? Ah, so it's really a 200W TUBE amp. That's different. Many people have a difficult time saying exactly what it is they want. There is a lot of technical discipline in writing down the desired results. If you write down the results you want, then achieve them, you've done a "meets requirements" job. Once you know what meets requirements, you can start talking about getting there more cheaply, with higher performance/reliability/yada/yada, and with a shorter time between start and finish.

Artists worry about perfection. Engineers worry about "works right, on schedule, and to budget".

Dad never had any experience winding transformers, he's an engineer.

Cool. I'd probably enjoy talking to him about it. I'm an engineer and spent some years designing power supplies and the necessary transformers for a living.

The only reason he did it was to save money...

Presumably then he did have at least a working knowledge of the safety and performance issues. Good. Otherwise, those transformers remain a safety issue every time they're turned on. To me, it's the idea that my work can endanger everyone who ever uses it that keeps me awake at night. It's a big responsibility.

to quote the topic "cheapest power supply transformer"

Yep. We've also seen that "cheapest power supply transformer" that works well and is safe is probably not involved with rewinding.

You seem pretty knowledgeable in design and building, it wouldn't be a big step to try knocking a couple up both input and output, you certainly understand the safety issues.

It's not a big step, and I have done that. In fact I've done it enough to know that doing it well is a huge amount of work even after you know how. "Cheap" can imply both time and money.

[jim1976]

I concede R.G its totally to expensive and beyond mere mortals. Out of interest have you ever seen power supplies that dont use transformers? just rectified straight off the mains! thats cheap but also a ticket to the next life! Just to pick your brain a bit and of topic do you know of any research/information on original 5e3 transformers output and power, what sounds good what sounds bad, I believe there is a fare amount of variation in the production. I'm from new zealand and I really have to build my own transformers because they are so expensive because of the freight and then there is the 60hz 120v thing and where 50hz 240ish

Thanks Enzo for the welcome got to say this is a great site there is so many ideas on here and great advice too.

[J M Fahey]

Dad never had any experience winding transformers, he's an engineer.

Well, that sure helps
But even more important is the "old days" bit.
Not to start talking like most old guys (not that I'm not one of them ), but "in the old days" Electronics (and a lot of other fun activities) were much more hands on, and unless you had a Wall Street Dad or something, *building* something was expected from you, so Magazines (and some Books) had detailed instructions on how to make things.
And yes, handwinding with a hand drill mounted on a table was a classic.
Plus precise instructions on how to wind evenly, how to insulate layer by layer, lamination stacking, external wires, mounting, even varnishing .
And probably you could chat (face to face, not via Facebook or MSN ) with some experienced old guy who would explain or even help you.
At least I could profit , at 15 or 16 y.o. , from the experience of a 50 or 60 y.o. neighbour who had been repairing radios (no TVs way back then) since *he* was 20. And so the torch was passed on.
But today, who can you ask?
Modern schematics seldom identify a transformer, at most they suggest some shop and code to order it.

[jim1976]

So true about the old days they just did it... we have it so easy today. Popular Mechanics and magazines like that had so many "how to" articles like winding transformers to building a gyrocopter. I think army surplus stores had a lot to do with it too people would look at all this cheap technology and think hmmm what could I make with that.. and the old hams and radio repairmen that would just give you there time so freely... Did you know Leo Fender was an accountant?

[R.G.]

Just to pick your brain a bit and of topic do you know of any research/information on original 5e3 transformers output and power, what sounds good what sounds bad, I believe there is a fare amount of variation in the production.

Let me start with the idea that a power transformer feeding a rectifier/filter cap network pretty much chops the incoming AC into short, sharp pulses to feed the filter cap, and then whatever is in the filter cap feeds the amp. Other than the filter cap not doing its job well and RF pulses from the rectifiers turning off, there's not a good basis for saying one PT 'sounds' better than another. There are extraneous issues, possibly magnetic leakage from the core, or unbalanced "centertapped" windings. But in general, I don't know how two transformers that produce the same output voltages with similar internal resistances and flux densities would sound different. I can't come up with any good theoretical way that could happen.

The music world is shot through with people trying to sell microscopic differences in one thing or another as better. Some of the claims are merely questionable, others sound like outright fraud.

With that out of my system, no, I don't have a recipe for 5e3 transformers.

Frankly, if I was after the original 5e3 sound and believed transformers would give it to me, I would find a 5e3 that you like the sound of and measure the transformers. You can do a really good job of reproducing a PT by measuring the winding resistances, turns (= voltage) ratios, primary inductance and leakage inductances. Likewise, measure the lamination size and stack. The transformer calculations you spoke of should tell you some reasonable estimates of what windings are where. It will take some paper-and-pencil time, but it's not difficult. In truth, 50/60Hz power transformer design is a bit boring once you've done a few, and especially if you already know the core size that works.

If you can only measure the 60Hz version, you need to put in 6/5 as much stack for the lower frequency and the same number of turns on the secondary. The primary is (obviously) about twice the turns of smaller wire.

OTs are different. You would do the same measurements, but add a high frequency resonance test. The primary inductance and leakage inductances tell you how many interleaves you must have, but don't give you the winding order. The high frequency resonance tells you the distributed capacitance, but it's difficult to design for that directly.

A set of good mechanical measurements, plus the winding resistances, primary and leakage inductances (usually from open circuit and short circuit inductance measurements) and a high frequency resonance measurement will let you come up with a workable solution in short order. But you may have to wind a couple or a few to be happy with the results. The PT will probably be OK on the first trial, but the OT will possibly be complicated.

Perhaps best would be to find a technician that has repaired a 5e3 and replaced the PT or OT or both, and buy him enough beer to give you the dead transformers. Then you can UN wind them, learning where and how big the wires are and how many turns they have, and have the knowledge. I know New Zealand is not a huge place, but this might well exist there.

[diagrammatiks]

I'd say that trying to build a run of 5 amps with a budget of 200 dollars for parts is a part dream unless you start stripping out fairly important things out of that budget...

You can't do power transformer and output transformer for anything over like a 5 watt single ended amp for less then 60...

then chassis casts, component costs, small parts costs. Things like grommets, proper isolating jacks, tube sockets, switches, and pots really add up.

Unless you plan on buying things a thousand at a time I'd really suggest reevaluating your budget.

[jim1976]

I really agree with you about that R.G I dont know how many hairs can be split using the same hardware. The interesting things I here about the 50s amps is it might be the 1 in 100 that really does have the magic, the same is said about strats its only that 1 in 100 that really has magic. Have you looked at the pickups in that 1 in 100 strat they were mostly from Abigail Ybarra I really dont think it was her magical winding, it was her magical winder! the counter was out all her pickups were under wound, sometimes as low as 4.9k like Stevie Rays No1. If you look at the custom shop Abby pickups nothing is over 6k, but that was production in that era.

So thinking about it why is there only 1 in 100 5e3's with the magic, I suspect its because of poor QC in transformer production. A few turns up or a few turns down changes everything, I happened to be looking at tubedata for a 6v6 if you increase or decrease the load resistance from the optimum on the tube the 2nd harmonics rise alot or the the 3rd harmonics rise increasing the overall total harmonic distortion. But if you look at it in respect to the plate current a higher load resistance it drops the plate current reducing the load on the power transformer, less p/t current less sag better amp. This is one of the main reasons I think people "need" to build there own transformers otherwise that illusive magical point cant be found!

I have never seen many early fenders here if you do see one it has been imported recently and older repairmen are even rarer haha. In that era things were protected so if could be made here you couldn't import it, strangely that made for local innovation and "kiwi amps" are very different from anything in your part of the world mostly there class B with a tube buffer to lower the impedance and increase the drive voltage for the power tubes! I have one here it has around 650v on the plate of a 6l6! and its getting around 80watts from a pair! If your interested I have a schematic.

I think we need to start a output transformer investigation thread, even a simple weight verse voltage ratio from simply pluging it into the mains, maybe even some dc resistance readings. Along with some tube distortion analysis data...

[Enzo]

One thing about 5E3 era amps that is untrue today is that they were built wil loose tolerance parts. The resistors were 20% types. The caps had loose tolerances as well. And of course the filter caps with tolerances like +80/-20%. SO two amps side by side could have an 80k measured 100k plate resistor and the next one have a 120k measured 100k plate resistor and both amps are in tolerance with good parts. All the little differences added up to make an amp what it was. The circuit then could wind up all over the map.


I would expect more consistency from the transformers of that era than from all the other parts. I don't think it was some guy sitting there winding by hand. Set the machine up for X turns and hit the START switch.

[jim1976]

I guess its all going to add up...

But look at all the replacement power transformers for the 5e3 voltages are all different... what about that 8000k to 8ohm output transformer? was it really that... I really think taking some measurements might make you think again.