One for the transformer experts.

[retailer]

I understand that core losses in transformers are mostly a result of the core material ie composition of the steel and eddy currents within the core - as well as using thinner laminations the eddy currents can also be reduced by electrically insulating the laminations from each other - when re-purposing salvaged laminations they often need to be cleaned with a solvent of some type to remove built up varnish or pitch, this step will usually also remove any sort of varnish type surface treatment that was originally used to electrically insulate the laminations from each other - if these laminations are now used for a transformer how much will the efficiency of the transformer be reduced - I'm assuming a small mains transformers around 150W such as are found in domestic audio gear.

[Ed_Dinning]

Hi, others will have more theory for you, bit the voltages induced in the lams due to eddy currents are very small. On older lams the insulation was often an oxide coating, not easily displaced by solvents, in fact often "enhanced" by rust, to the detriment of the number of lam sets that could be re-fitted. This would increase the mag current a little.
Another effect that will increase mag current is the failure to butt the lams closely together. I had one rewind that had an overheated primary due to very poorly butted lams with gaps of 2mm or so

Ed

[kalee20]

The core losses are due to several things - eddy current losses (which is why the core is split into thin laminations rather than have a solid block of iron); hysteresis losses (magnetising and demagnetising the material absorbs some energy each cycle, and laminating the core makes no difference); magnetostriction losses (the core changes its dimensions minutely around each cycle, hence the audible 'hum').

Eddy current losses account for maybe half the total core losses in a typical transformer.

Insulating the laminations does help, but the insulation doesn't have to be very good. If you are worried about reusing laminations, and you've got them clean, a very light dust-coat from a spray can would do the trick. Use radiator paint or something that's good to 100°C would be my suggestion.

[dougietamson]

Here's muddying the waters...

A pair of transformers from a US made '80s 100W 4x6L6GC guitar amp, they attached the transformers to a mounting plate with a steel weld across all the laminations, then bolted to chassis, all now at the same potential.

I had hoped to add an 8ohm tap to the existing 2 and 4ohm taps on the OT but grinding the weld put me off.

EIA codes for the OT is 682 Electrical Windings, Inc., Chicago and date code 42 week 1981.
The PT is made by Basler Electric Company, Highland, Ill.

[Richardgr]

I suppose one option for those would be to reuse them with 6550As.

Assuming they are 2.5k impedance for parallel PP 6L6GC, then using it with a pair of 6550s and 8R on the secondary should mean the valves 'see' the expected 5k.

[retailer]

Thanks for the replies.
I've done a few 110V to 240V rewinds/conversions of US made power transformers, with no reported issues - all used in 60's and 70's made amps - all had a varnish coating on the laminations, once they'd been cleaned with solvent I had been doing as kalee20 suggested - a light dust coat of engine enamel - however the spray can is now empty and I'm doing a Fender Pro power transformer rewind/convertion and I wondering if I really need to insulate the laminations.
I've done the maths, finished the primary winding and with the laminations inserted, the primary draws 70mA, it's a good chunk of iron 21sq cm 3.5cm x 6.0cm, not sure if this is considered excessive.

[Ed_Dinning]

Hi Retailer, 70mA would be within the normal range for that type of transformer, but not knowing flux level it is run at and lam material it is difficult to be precise.
Try running the transformer on no load and see how hot it gets. Slightly warm is fine

You can check primary winding for current rating, but be aware that mag current (mainly) is in quadrature to load current so forms a vector sum

Ed