Welding on an output transformer

[cyclecamper]

I bought a used and abused broken Peavey Tour VB-2, which uses 6 EL34's to make 225 watts. Or would, if it worked.

It looks like the output transformer broke loose, and flew around and broke tubes etc.

So somebody welded it back on its mounting plate (while uninstalled). Which sounded crazy to me until I found out that Peavey also welds them to the same plate.

The funky output transformer it the only thing that 'looks' wrong, except the middle keyed phenolic pin is broken out of most of the EL34's (which really shouldn't matter).

I would have thought they used a laminate core not just for assembly but to prevent eddy currents in the core. Wouldn't welding across the core in multiple places be a bad thing? I'd imagine there's also a big risk of cooking off the insulation, or even melting some copper windings! Or inducing enough current to blow thin wires?

I can't imagine why someone would decide to weld a transformer core instead of just
putting a plate across the top and 4 long bolts.

 

[Enzo]

But it doesn't shunt the core. Peavey has welded on mounting plates for many years, no problems have come from it.

 

[cyclecamper]

Peavey has extensive experience. Yet their weld didn't hold, their weld broke, and the transformer went flying, wreaking havoc. And now I have very very low confidence in the unauthorized repair welds. Unfortunately, really thorough testing of an output transformer is very difficult. I still can't see the sense of welding it, when better mounting methods seem simple. Then again, I'm prejudiced by looking right at a catastrophic failure.

 

[sreten]

Hi, So do some drilling and add nuts, bolts and washers, rgds, sreten.

 

[Enzo]

THAT weld broke. Tens of thousands of their amps did not break the weld. There is a limit to the abuse something can take before it breaks, even a well made something. There is nothing you can make that someone else cannot destroy.

That is a big heavy piece of iron. Bolt it down, and someone can drop the amp, and the momentum of that iron can pull the bolt nuts right through the chassis metal. But whatever horor story we can come up with, 99.99% of them will never face it.

The welded plate is reliable and sturdy - even if one broke. Bolts and clamps are reliable and sturdy too. They will work fine instead of welding. There is no need to try to reweld it unless you are a professional welder.

Not all failures are systematic or due to a design flaw.


In my 27 years running a Peavey service center, I have had one broken weld plate. I contacted my inside guy at the factory and told him I didn;t feel it should have come apart, as ther was no sign of abuse on the cab. He agreed and sent me a replacement transformer. The amp was long out of warranty, but they covered it anyway.

 

[Andrew Eckhardt]

Welding a transformer core without TIG and some application specific experience is bound to come out messy. Besides, welding it a second time would usually occur after a varnish dip which would make a clean, strong weld impossible. I'd mount it mechanically if I could.. Or replace as suggested is probably better if the original mounting is still possible.

 

[JMFahey]

Agree.
The original Peavey welding looks TIG made by some automated machine, looks *very* clean and smooth, doubt a human could easily do it that way.
Well, maybe he could if he had to weld 1000 transformers a day and had some device to guide the torch.
I've tried to rewind such transformers and gave up, could not pull them apart in a clean way, so I chose to wind a new one on a fresh conventional core.
And mounted conventionally with L brackets and bolts.
By the way, I also contact cement the core to the chassis if possible, helps a lot against vibration.

 

[cyclecamper]

I'm really reluctant to use the output transformer that's been re-welded by some shop that was too incompetent to finish the repair, yet the real Peavey part is expensive.

What's the load impedance an EL34 likes to push, as opposed to what a KT88 likes to work out into? I have a brand new output transformer meant for a 400 watt Trace Elliot V6 with 6 KT88s, instead of this 225-watt Peavey's 6 EL34's. Mechanically there's mounting issues again, but electronically would it be likely to work?

At the least, welding across the core laminations makes it nearly impossible to rewind.

The 400-watt transformer from the Trace Elliot V6 isn't much bigger, but the core is thicker with more laminations. I figure that's a good thing for bass.

I finally made some really hefty mounts for putting the Trace Elliot V6 transformer in the Peavey. I'll post up some pics this weekend before I finish botling it down, because I'd like some opinions before I do. I did have to change its orientation to accomodate the larger transformer. I know it's generally bad to mount the power and output transformer in the same orientation, but they're so far away from each other I hope it doesn't matter. I also had to mount it vertically, which challenges the mounts even more. And I mounted it raised a little off the chassis, which further challenges the mounts. But it is strong, and sinks some heat from the core, and has a bit more cooling surface area. I hope it's not too close to the output tubes or preamp tubes now.

I finally wired in the 400-watt trace Elliot output transformer.

I suppose I should put some kind of cover sheath over the transformer wires to protect them from the heat long-term.

Interesting that the V6 output transformer uses the same size cores as the original, but each is a little thicker and the stack is twice as fat.

The new trans should be capable of some pretty low bass, lower than a bass guitar requires.

 

[cmjohnson]

Thread resurrection six years later. Because I'm on a rant about this right now.

Peavey had a BAD habit of welding plates to transformers, across the laminations.

YES, this shorts the cores and cranks hysteresis losses way up. The transformer still works but with huge losses in efficiency and it messes with the usable bandwidth as well. The transformer will run hotter than it would if they had not STUPIDLY welded the laminations.

Peavey's packaging engineer(s) clearly never talked to their electrical engineers.

Any Peavey amplifier equipped with a transformer that's been welded across the laminations can be improved by replacing the transformer(s) with similar types that have NOT been welded or otherwise had their laminations electrically shorted.

 

[trobbins]

cmjohnson, I think you should talk to an electrical engineer yourself.

Welding across a lamination stack typically has a minimal influence on transformer performance for large transformers or chokes. For a guitar amp with transformer laminations of that size I suggest it has no consequence at all.

Have you had a related problem ?

 

[cmjohnson]

I regularly deal with a number of EEs and every one of them laughed and shook their heads when I pointed out a few cases similar to this one, all various Peavey amps, over the years. Not one of them thinks that it's anything resembling good engineering practice to weld a transformer across its laminations. Such a questionable practice flies in the face of all the electronics education I've acquired since 1975.

 

[abraxalito]

Such a questionable practice flies in the face of all the electronics education I've acquired since 1975.

Its been a bit narrow, hasn't it? Your electronics education I mean. Since 1975 it has only concerned eddy currents in transformers and how laminations avoid them?

 

[trobbins]

It is most likely the initial shock and knee jerk reaction to seeing that style of construction for the first time. It does require some magnetics thought to appreciate what can change and how influential it is. Sort of like when people worry themselves silly about the outer surface of their power transformer having signs of rust.

It's closest equivalence is probably a mounting bolt that passes through a hole in the core and doesn't use insulated washers to avoid inducing a circulating current.

 

[Kay Pirinha]

See my answer here: The dumbest thing I've ever seen in amp construction....
Best regards!

 

[Osvaldo de Banfield]

I agree with cmjohnson. Welding the lamination becomes it loosier and too difficult to repair/disassemble.

 

[cmjohnson]

Questioning my electronics education is not a good idea. While I'm not a designing engineer, I do technician work in all phases of electronics from basic linear to multi-frequency, multi-phase switching power supplies to audio, video, RF, and digital troubleshooting and repair. I did seven years as the chief technician in a two way radio shop and have worked in avionics in addition to multiple various other positions. My education is very broad based. If it consumes electricity I can understand it and fix it when it's broken, at the component level. I'm not one of those "technicians" who fixes things by swapping out a circuit board.

Among other things I did the system design and oversaw the deployment of multiple radio systems in multiple countries. Digital trunked radio system design is a specialty art and I do it.

Abraxalito, you, on the other hand, have difficulty comprehending even the requirement and reasoning for transformer laminations. Basic first month of electronics education stuff. You should not be talking.

 

[Osvaldo de Banfield]

I work in repairing industrial equipment, I saw lots of this kind of inductors an transformers made in such a way, and sincerely I'm in doubt how the quality of the iron is altered after the welding process. Also, several inductors for choke input at the three phases systems use welded inductor WITHOUT GAP that is a not good design, IMHO.

 

[Mark Tillotson]

Welding on the outside of the core doesnt do anything. Welding on the inside and on the outside would create a conducting loop, but just doing the outside doesn't create any current loops. Draw it out if unsure...

 

[trobbins]

Engineering is often about managing compromises, and appreciating all aspects, not just technical.

The assumption is that a company made a poor engineering decision - but the reasoning presented is imho lacking in any engineering perspective. I see no balanced assessment of the likely considerations that went in to the decision to use welding for that application, or any evidence of performance comparisons of an example transformer that has been made the 'Peavey' way versus what has been considered the 'right' way.

Imho, if you lambast Peavey with multiple posts, then back it up with evidence and assessment, or accept the critique that heading a thread with 'the dumbest thing' comes back at you.

I think the topic is very valid to discuss, and tease out what degradation of performance it may incur, and why particular applications have come around to using it. I order in and manufacture with choke cores that are supplied with welded laminations - so for my particular application, I have been down a similar path, and do recall the knee-jerk concern when first seeing such a core.