How To Determine Power Transformer Specs

[Riviera2dice]

I have a friend who's trying to "reverse engineer" the specs of a power transformer.

Specifically he'd like to find the ampacity of the heater tap. He has a chassis that he's thinking of adding some pre-amp stages to and just wants to be sure he doesn't exceed his PT's specs.

I'm familiar with determining impedance on OT taps, but I'm a little stumped on the PT question.

Anyone have any help on this?

[Steve Conner]

On a PT the rating has nothing to do with the impedance, it's the winding resistance and permissible temperature rise that limits it.

The only way I know of to do that is to measure the winding resistance cold, then load the transformer to your proposed load for several hours until it reaches a steady temperature, then measure the resistance again hot, and use the formula for the temperature coefficient of copper to work out how hot the winding is running. Then compare that temperature to the rated temperature for the class of insulating materials that the transformer is made with. Allow a little margin for hot spots.

You will need a specialized 4 terminal resistance bridge or other method: standard DMMs can't accurately read the resistance of a heater winding.

If you don't know the insulation class, the answer is probably "A few more preamp tubes than the original application will probably be OK- unless you spat on it and it sizzled", but this answer comes with no warranty.

[loudthud]

If you can measure the diameter of the windings, that might be an important clue. Can the transformer be taken apart enough to get a caliper on the wires?

[Riviera2dice]

Thanks to both of you for responding!!

Steve, your answer is sort of what I was thinking. Hook it up to the load and see if the tranny heats up. Also, that it ought to be able to handle 1 or 2 extra pre-amp tubes. He says his theoretical (calculated) change is from 2.2 amps to 2.8 amps for the heater circuit. I looked at a 350-350 tranny I have on the shelf and it's rated for 3 amps on the heater circuit.

Loudthud, I'll pass that along and see if he can measure it. Now, you mean the diameter of the primary winding, right? The size of the "clump" of wound wire.

[loudthud]

The wire size of each winding is indicative of that winding's current rating so we would like to know the diameter of the heater winding wire.

A shorthand method of estimating the current capacity of a wire goes something like this: In the US, you take the diameter of the wire in 1/1000ths of an inch, square that number and you get what are called "circular mills". I don't know what the "rule of thumb" is for transformers, but for a wire in air it's something like 1 amp per 1000 circular mills and for a wire in a bundle it's 700 ma per 1000 circular mills. Perhaps someone on this board more familiar with transformer winding will give is a better number. I assume there is a similar rule of thumb for metric wire sizes.

[Delta362]

Riv was asking for me while I was away from the computer. The amp is an old Admiral 5T4A. The amp came with two EL84, one 12AX7, one 6AV6, and one 5Y3. I am going to replace the 6AV6 with a 12AX7. Then I'd like to add another 2 12AX7 tubes (totally 4) and make the amp into a two channel version.

The PT heater wire is solid core, not braided, either 20 or 22awg. It's pretty much the same size as my 1/2W metal film resistor leads. My concern was having enough juice in the PT to supply the heater elements. Also, if I replace the 5Y3 with 1N4007 diodes instead, that would take away the loading on the 5V output of the PT, which should also reduce the operating temperature of the PT. Do you see any problems with that?

[andrew4566]

I have done what you are trying without problems. I was taking a chance but for .6 amps I felt it was reasonable risk. My thought was that if I was overloading the winding the heater voltage should drop noticably, so when testing the amp (after it was built) I measured the heater voltage under varying load conditions and it didn't drop below 6.1v. haven't had any issues since.

I did also use diode rectification (as you mentioned) and therefore didn't use the 2amps of 5v current...may have helped?

-Andy

[Delta362]

Thanks for the confirmation, Andrew! Now I just need to finish designing the circuitry so I can get started on the layout.