Kustom 150 power supply 50% high

[olddawg]

@olddawg - earlier in the thread, I indicate that I have already rebuilt the output stage - twice - with rather immediate failure upon running it up to full voltage. I wasn't monitoring the B+ / B- at the time, but it did not draw excessive current up until the point when the line voltage was near 120, and the outputs failed.

I did not say run it up to 120vac. I said run the variac up monitoring the rails until you get the schematic specified rail voltages and see how the amp works, look at the ripple voltage, etc. I have seen power transformers short windings and do this before but usually they get hot. However, this is a BIG transformer. If it works with the variac set lower and the rails correct with no excessive hum you might be able to replace the choke with a big resistor or try to attenuate the B+ later. Test the inductor. You may have a bad transformer.

[Steve Conner]

No, if the power transformer had shorted turns, it would smoke, stink, blow fuses, and generally fry.

The point I was trying to make is that, if the choke input supply is working, the response to the variac will be non-linear. The rail voltage will start off alarmingly high, but as you keep increasing the line voltage, you'll hit a point where the choke magnetizes and the rail voltage stops rising. So you can't extrapolate from the behaviour at low line voltages.

In any case, as others have said, a choke input supply in a solid-state amp was generally a crazy idea, and you probably don't want to put too much effort into fixing it!

And that inductor is far too big to be some sort of snubber. It's a filter choke for sure. It should have relatively few turns of thick wire, making it easy to rewind. When you dismantled it, did it smell bad?

[jbreher]

@olddawg - I think I understand your suggested approach. However, I have already rebuilt the ouput stage twice. I think I'd rather get the power supply section working before going back to the output. Plus, I've already unwound the inductor - I'm kind of committed now

I had earlier speculated that the primary may have a shorted primary, leading to less turns in the primary, leading to a change in the turns ratio, leading to an increased output voltage.

I now think this is an impossibility. Here is my reasoning -- please chime in as to whether or not you agree.

If there is one or more shorted turns in the primary, the number of turns that have shorted are indeed effectively subtracted from the primary's turn count. However, those windings don't just go away. They are still there in the transformer. As they are no longer part of the primary, these 'orphan' turns essentially become an additional secondary. Further, the terminals of this secondary are shorted to each other -- the same as if they were a designed-in secondary, and you had soldered these leads together.

With any primary current applied, there is a voltage built up across this phantom secondary. The only load on this phantom secondary is the resistance of the winding itself. Accordingly, a huge current circulates in this phantom secondary.

The massive current is reflected to the primary, causing the primary to draw massive current as well.

The amp is not drawing such massive current. Accordingly, there is no shorted secondary. Accordingly, there is no shorted primary.

Make sense?

[jbreher]

@Steve - I understand your point about the non-linearity of Vin - Vout relationship. I *think* I have shown that the supply is not working properly, as I have modeled the quiescent load due to the output stage as 320 mA on each rail. Under this condition, the supply was still very high (+/- 60 Vdc rails for a 120 Vac line). Anyone disagree that the quiescent load on each rail, due to the output stage, is most probably below 320 mA?

I did not notice that the inductor smelled particularly bad. However, once I removed it, I saw varnish residue on the chassis. As I disassembled it, it looked pretty cooked. Several steps of the disassembly process were preceded by an acetone bath to loosen old bonds and such, so this may have reduced the smelliness. The outer coil all looked pretty decent as I unwound it. However, the inner coil was very blackened.

[jbreher]

Soliciting rewinding advice

Well, I think I've proven that the inductor is toast. I am now committed to rewinding it. Not having done this before, I am soliciting tips and advice.

After stripping the coils, I measured the diameter of the wire every 4' with a micrometer. I measured an average of 0.0350", in a range of 0.333" - 0.376". I am thinking this makes the wire 20 AWG. Anyone have an opinion on that?

MWS (MWS Wire Industries - Magnet Wire Specifications) suggests the most common insulation formulation to be their Single-insulated Polyurethane Nylon-155. This would have a nominal diameter of 0.0334" in 20 AWG.

How worried should I be about nicking the insulation upon doing the winding proper? Should I opt for the Heavy-insulated wire to help guard against this possibility? (0.0346" dia nominal)

Upon disassembly, each layer of windings was separated from the next by a layer of tape. What should I use for this? Is Kapton suitable?

I am concerned about my ability to get the windings to lay nice & neat on the former. What is the best way to do this? Should I wind it freehand, or should I make some sort of hand-cranked jig to hold and rotate the former?

[J M Fahey]

Build some hand-cranked machine, even out of wood.
It should turn a bolt, say, 3/8" x the necessary length , where you slip a piece of wood (or plywood pieces stacked), a "rectangular cube" if you wish, to fill the core snugly.
It should slip in-out easily but not too loose.
You wind on it, hand-guiding, every layer you insulate with thin Mylar.
Write down every 10 turns, to avoid losing count, it's very easy to do so.
If you rewind the main transformer, (my suggestion), you leave the primary undisturbed, but unwind the secondary and rewind it with the same wire, but use 70% of the original turns.
Example: original 120Tx0.7=84T new.
If you can use new wire, even better.
Good luck.

[R.G.]

Well, I think I've proven that the inductor is toast. I am now committed to rewinding it. Not having done this before, I am soliciting tips and advice.

And may God have mercy on your soul...

After stripping the coils, I measured the diameter of the wire every 4' with a micrometer. I measured an average of 0.0350", in a range of 0.333" - 0.376". I am thinking this makes the wire 20 AWG. Anyone have an opinion on that?

Yep, #20, 0.033 single insulation to 0.035" heavy insulation.

How worried should I be about nicking the insulation upon doing the winding proper? Should I opt for the Heavy-insulated wire to help guard against this possibility? (0.0346" dia nominal)

It's usually not an issue unless you leave a sharp edge where it can be nicked as you wind.

Upon disassembly, each layer of windings was separated from the next by a layer of tape. What should I use for this? Is Kapton suitable?

Kapton is eminently suitable, and a good choice. If you can find it... Kapton film between layers is even better, as it supports the end turns better.

I am concerned about my ability to get the windings to lay nice & neat on the former. What is the best way to do this? Should I wind it freehand, or should I make some sort of hand-cranked jig to hold and rotate the former?

I have been winding transformers on and off since 1974. The handiest thing for good results is to set up a cranking jig that will rotate the cardboard center tube as you guide the wire onto it, and even better if it has some kind of plates/sheets at the ends of the tube to keep the wire from spilling off the ends. It's hard (was for me at least) to make each layer reverse neatly at the end of a layer without sliding off the stack of layers.

I hope you counted turns as you un-wound it.

The construction is one variant of swinging choke, so maybe they ... did... do that. Ugh!

One neat trick is to computed the mean length of a turn of wire from the dimensions of the core and windows, then measure off and wind two wires side by side at the same time. This bifilar winding ensures the same number of turns and the same resistance for both windings, and pays back the extra manual dexterity needed with only having to turn the crank half as much.

By disassembling the choke, you now know a world of useful information about how to re-wind one. It's all about tape.

If you rewind the main transformer, (my suggestion),

I respectfully disagree with that on the basis that safety isolation/insulation of the primary is endangered by messing with the PT at all. The choke, if that's the problem, is a much simpler and less demanding winding task, in most cases. It's only if the PT has separate and side-by-side primary and secondary sections that you have much chance of not compromising the safety considerations of a PT. I personally would never recommend someone rewind a PT unless it had separate, isolated primary and secondary sections. The theory is easy, but safety considerations will bite when you least expect them.

[jbreher]

Kapton is eminently suitable, and a good choice. If you can find it... Kapton film between layers is even better, as it supports the end turns better.

I don't quite get what you are saying here. Tape vs. film? What is the difference?

I have been winding transformers on and off since 1974. The handiest thing for good results is to set up a cranking jig that will rotate the cardboard center tube as you guide the wire onto it, and even better if it has some kind of plates/sheets at the ends of the tube to keep the wire from spilling off the ends.

I am already envisioning a mechanism to do this.

I hope you counted turns as you un-wound it.

Most certainly. Each layer.

Wire questions: What type do I want? Polyester with Nylon overcoat? Should I get a bondable overlayer?

And where do I go for small quantities of the right stuff? All the sources I've turned up so far seem to either have a 10-lb minimum, or a cruddy selection. I need less than a pound.

[R.G.]

I don't quite get what you are saying here. Tape vs. film? What is the difference?

You're right, tape is film, but usually (a) thinner (b) longer, way longer and (c ) has an adhesive on one side. What I meant by film was layer insulation, a single turn of thicker, non sticky stuff between either each layer or between sections.

Wire questions: What type do I want? Polyester with Nylon overcoat? Should I get a bondable overlayer?

Get "magnet wire". any will work, and you don't need the fancy coatings. So called "heavy film insulated" stuff is fine. You don't need "solderable", which implies the film melts and lets solder through, or bondable overcoat, etc.

And where do I go for small quantities of the right stuff? All the sources I've turned up so far seem to either have a 10-lb minimum, or a cruddy selection. I need less than a pound.

A really good place is an alternator rewinding shop. Every moderate-to-large city has a place that does nothing but rewind alternators, electric motors, starters, etc. Suitable approached, these guys will usually help.

Or you can type "magnet wire" into google. These guys don't look bad, although I've never used them:
Magnet Wire - $4.74 , they have 20 ga for $10 for a half-pound.

Ebay works too, if you're careful.

The insulating stuff is the hard part. This can be made easier once you recognize that clear film laser printer sheets are mylar/polyester film about 0.007" thick. A batch of that stuff makes good layer and outer-edge protective stuff.

The real hard part is tape. But I found this:
CS Hyde Co., IL, USA: High Temperature Polyester (PET) Masking Tape with Silicone Adhesive: Green, Teal, Blue Polyester Tape

[Enzo]

I don't know if it is suited to this application, but one of my all time tape favorites is Scotch 27 - "glass cloth electrical tape." Seems like it might work here.

It is woven of glass fiber - no, it is not loose itchy fiberglass type fiber. The adhesive likes heat, which sets it. it stands up to high temperatures, a good insulator. I keep a couple rolls next to me.