Is there something they didn't teach us in basic electronics?

[Habeohtjt5555]

https://app.bitly.com/Bi756VogwGs/bitlinks/2ziFW78

 

[unclejed613]

in "basic" electronics, you are generally working with components that have been made in a factory. rewinding a transformer is a bit of an advanced skill. if the transformer is an "ultra-linear" type (which McIntosh uses in most of their amplifiers), you need to keep track of not just how many turns are on the bobbin, but the direction and spacing of the turns, and the thickness of the insulating tape between windings. it also takes a bit of skill to get the core apart for removal of the bobbin. this process is more difficult if the layers of E-I pieces alternate their orientation. it will take a lot of skill to reassemble the core. after the job is done, but before reattaching the end bells, the transformer should be dipped (soaked) in varnish to insure there's no core noise or magnetic slap in the windings.

 

[Diver300]

Of course there is some things that aren't included in basic electronics, and those include production engineering and economics.

The material costs are insignificant on a job like that. What you are paying for is the labour and skills to do the work. Rewinding a transformer is far more work than making a new one.

The laminations of a transformer are held together with the varnish. I've seen some transformers where they are welded together, so taking the thing apart will be very tedious.

4 or 5 windings isn't "pretty basic", it is on the more complicated end of the scale. On a valve amplifier, it's likely to mean 4 or 5 different sizes of wire, and the issues of getting all the windings in phase. It's also got to work at a wide range of frequencies, so stray inductance may be important, so the order of the windings needs to be kept. There may be screens between the windings.

You will expect the guy who asked $600 to do a neat job give you back a transformer that looks like new, and to guarantee the work, so he will factor that in as well.

As far as what the transformer does when working, it'll be doing exactly what you were taught in basic electronics.

 

[crutschow]

I think if you attempt to do that rewinding job you will soon realize why he wanted $600 for the job.

What they didn't teach you in basic electronics is how difficult it is to rewind a multi-winding transformer.

 

[dknguyen]

There's a lot of gotchas in winding transformers. The old guy at my work specializes in this. Along with the gotchas is that it takes a long time and you have to be careful and meticulous the entire time.

 

[MrAl]

Hi All,

A output transformer recently failed (McIntosh MA-230) and so far the only shop I found who said they could rewind it wants far too much for the task.

After all it's only a few hundred feet of wire right?

I'm thinking worse case, I could carefully dissassemble the transformer, mark the parts to insure an orderly reammembly,
remove the core and unwind it until I find the failed winding, or just replace everything and start new.

I'll need to build a jig with a turn counter and secure a source for the enameled wire. Pretty simple if you ask me. So why is this guy asking $600 for the job? It's a pretty basic transformer with 4 or 5 separate windings, but nothing really special.

Is there something they didn't teach us in basic electronics?

thanks
habeo

Hi,

As others have said, this is not an easy task. It sounds easy when you think about winding and unwinding, but you probably cant even get the windings off without cutting them off with a hack saw.

I did a few transformers in my time and a couple motors, or should i say attempted. The problem is that the old stuff is always vacuum varnished and that mans that every turn of wire is virtually 'glued' to every other turn, encased if you will. Even the laminations are done the same way so you may not even be able to get the thing apart without bending all the laminations.
It's a very very difficult task.

The most common way to fix this kind of thing is to look up the part number and order a new transformer, when that is possible of course. When it is not, start with a fresh set of laminations and go from there, if you dare.
Keep in mind that a transformers is not just a core and windings, there is also a coil form and layers of tape strategically placed so as to reduce the chance of arc over between windings that have significant voltage difference. There is also a safety issue where sometimes the secondary has to be physically isolated from the primary so it can not be wound on top of the primary coil.
So you see there is a little bit more too this than you might think at first.

One of the 'attempts' i mentioned above was for an oscilloscope. The high voltage secondary was at fault and so i had to figure out a way to fix this. After taking apart the transformer with various tools including hacksaw, i realized the only way i was going to get the multi turn secondary was to use SEPARATE transformers, and the only way to get the high voltage secondary was with multiple transformers with secondaries wired in series to get to the right voltage.
I think it took three or four separate transformers to get to the right voltages, but it worked and i used it for years after that. I also added a second channel through the use of a chopper because it was only single channel and it's the only scope i had at the time.

So you may want to rethink this. If you cant buy a replacement and you know what voltages you need you can always use multiple transformers to get all the secondary voltages.

 

[gophert]

If the old one failed, it likely failed for a reason.

The laminations have an oxide layer, poly film or paper separating them to prevent eddie currents. The aging, especially hot environments, thermal cycling and/or internal abrasion (magnetostriction) from high current use can cause the insulating layers to fail and hot spots start to form. If this happened, you'll be better off getting a new core. Rewinding a bad core could be a $600 waste.

 

[Diver300]

I was given a clock radio that had been blown when the mains voltage got too large. The transformer had centre-tapped windings at 4.5 and 9 V, to power the LEDs and the radio, plus a single 15 V winding to power the clock and to provide the 50 Hz.

The transformer was a standard frame size, so I used a centre-tapped 4.5 V replacement, added a voltage doubler for the radio, and then found space for a tiny, low power 15 V transformer for the clock.

I would never have tried rewinding a transformer.

 

[AnalogKid]

Separate from all of that, MacIntosh output transformers are *not* normal transformers. AudioGuru probably can speak to this in more detail, but I remember something about bi-filar windings, feedback windings, etc. Very high-end stuff at the time.

ak

 

[unclejed613]

If the old one failed, it likely failed for a reason.

in amplifiers, an output tube going gassy, cathode bypass cap shorting, or an internal problem causing the tube to go into full conduction can cause a transformer to burn out. also, running the amp without a speaker connected can cause internal breakdown of the transformer.

 

[schmitt trigger]

What Analog Kid said:
The McIntosh transformers were extremely complex. There were not trivial to assemble even when new.
Rebuilding one takes the difficulty to a whole new level.

 

[rjenkinsgb]

For info, the manuals and schematics for the MA-230 are here:
http://www.berners.ch/McIntosh/en/MA230.htm

The output transformers are an unusual design with two extra feedback windings. The two centre tapped ones would likely be bifilar wound for accurate symmetry & balance, as AnalogKid says.

See the articles here relating to possible MA-230 transformer failure - the first gives a contact phone number to buy new replacements (as of 2009).

http://audiokarma.org/forums/index.php?threads/output-transformer-repair.224974/
http://www.audiokarma.org/forums/index.php?threads/ma-230-output-xformer-rebuild-or-replace.224994/