Output Transformer Testing Guide / Resource dump

[Mr_bibbles]

Hi All,


I wanted to put something up about Output transformer testing with an AC signal. I have a JCM900 on the bench with low / strange output issues, I've replaced the bias caps that were failing (no change) and verified the preamp ok, worked my way up to the OT. For me at least, this has been a confusing set of tests, so I thought it would be good to post some information up here with the intent of clearing it up for novices, including an example amplifier. I won't go into the math, but will list it more as a "garden shed mechanic's guide". Since I'm still learning myself, I hope some of the old heads can correct me where I'm wrong and help to expand on the information here.

To start, there are already some good resources:

http://www.aikenamps.com/index.php/o...mers-explained

http://www.geofex.com/ampdbug/outtrans.htm

http://www.antiqueradios.com/forums/...?f=13&t=168025


But as for this amp in question, it failed the first test listed in the excellent GEOFEX site: Primary impedance. Across the leads (disconnected from circuit) I measured:

Center tap (B+) ---> Side 1 - 1.6 Ohms
Center tap (B+) ---> Side 2 - 17.2 Ohms

Primary impedance should be roughly equal, although since one wind is wrapped AROUND the other primary, there is sometimes a difference due to the same number of winds not having the same length of wire. BUT because it's much more than a 10-15% discrepancy, we can assume that it is bad, right?

Because Transformers are expensive, I also did an AC test. This was performed with a light bulb current limiter. I tapped a wire off of the amp's filament supply, and sent the wire to the 8 ohm tap. This sends a 3.15VAC signal into the transformer, and we can test the signal as it comes out either side. The signal on the secondary side should be roughly equivalent to it's taps (4/8/16 etc) in terms of ratio, so the 16 ohm tap should be higher than 3.15VAC and the 4 ohm tap should be lower. On the primary side, each half should be identical.

On my example amplifier (no tubes installed) I got-

Secondaries: 3.5 VAC on all taps.
Primaries: All leads measure 3.26 VAC on both sides AND center tap.

So in this case, it seems as if there is a short somewhere internally, as there is no step up or down anywhere present on the transformer, except for a slight difference from Secondary --> primary. It's my understanding that the voltage when viewed on the primary should be "stepped up" significantly from the secondary, (or vice versa, "stepped down" if you were to feed it from the primary) It's worth noting that this transformer also had a brown spot much like a cigarette burn in the center of the plastic-wrapped core, which adds to the hypothesis that something had cooked internally.

Again, I'm writing this post for my own education as much as for an example for people who are similarly confused, so if I'm off base I'd appreciate some scolding from some of our more distinguished members. Thanks!

[g1]

There is a lot of confusion here in terms of your voltage reference. It almost sounds like you are using a common reference for all your readings.
Can you state where you have placed each of your meter probes for all your readings?

[eschertron]

It's my understanding that the voltage when viewed on the primary should be "stepped up" significantly from the secondary, (or vice versa, "stepped down" if you were to feed it from the primary)

If I were to set this AC test up, I might apply my 3.15VAC to the 16 Ohm tap (16 and common) because the ratio 4k/16 is lower than 4k/4 (just picking numbers out of the air, your OT's primary may not be 4k) and the voltage generated across the primary leads will be lower (safety first, right?). Reading the primaries, then, should give you a voltage side-to-side, and half that voltage each side-to-CT. I might use the equation [(E1 squared)/Z1] = [(E2 squared)/Z2] to verify my voltage ratios are "stepping up" properly. Using this formula, I'd expect in the neighborhood of 50VAC on the 4k primary.

edit: bottom line, if the resistance readings are that far off... the prognosis is not good.

edit: I see now what G1 was getting at. The ground reference probe on you meter needs to be across whichever winding is under test. So secondary ground to 16R (for example), and primary red to blue (or whatever colors you have). Reading to a chassis ground just won't do.

[Mr_bibbles]

OK, so this is where I'm getting confused. Geofex says:

Identify a secondary winding, 8 ohm tap if it's available, and hook up one and only one winding to either 1/2 of the 6.3VCT or to the variac. Make D^&%ED sure the variac is all the way down if you're using one of those. Make sure that no other leads are connected (or shorted together, or touching your screwdriver on your bench or... well, you get the idea). Check for safety load resistors on the output jacks or secondary windings. These can cause you to erroneously think it is bad. There must be no loads on any winding. Put your voltmeter on the winding, and the current meter to measure the AC current through it, hook up the 3.15 VAC source, and turn on the AC - not to the amp, but to the AC feeding the winding you're driving. The voltmeter should measure 3.15 (or close) volts AC, the light bulb (if used) should NOT be lit brightly, and nothing should be humming or smoking ;-). There should be only a little current going through the winding. If the voltage is lower than 3 volts, or you are pulling amps of current, then there is a load on the transformer, internally since you have disconnected all the leads, meaning that there is an internal short. (Which means that you ought to be sure that there is no other load on the transformer, because you're going to throw it away if it has an internal short - be sure you're right!)

If all is well, no smoke, flames, loud hum, poultergeists, or lightning, measure the voltage that now appears on the other windings. The voltages will be equal to the ratios of the voltages that will appear on these windings in normal operations. For the primaries of output tubes, this can be up to a couple of hundred volts, so don't think that you can relax your safety precautions - it can still kill you, even driven with 3VAC on the secondary of the output transformer. The half-primaries of the output transformer should have identical voltages on them. Secondaries should have multiples of


So with the voltage you're putting on the tap you chose. If you used an 8 ohm tap, a 4 ohm tap will have about 2.3VAC on it, a 2 ohm tap will have 1.6VAC on it, and a 16 ohm tap will have about 4.5VAC on it.

It seems to be conflicting as it says disconnect ALL the transformer leads, hook up ONE AND ONLY ONE lead, etc, but then also check the output jacks for resistors? Shouldn't disconnecting all the leads do this by default? I'm guessing I'm missing something here...

So with what Eschertron said, the test should be done across a wind- meaning Secondary Common ----> Secondary 4/8/16 Taps or Primary End to end. Does this mean we feed "One and only one" lead with 3.15AC into HALF a winding, and then measure across it tosee what comes out the other side? COMMON must be attached to ground?


All leads disconnected except 3.25v on 8 ohm tap, COMMON tied to ground:

Common --> 16 Ohm 2.3VAC
Common --> 8 Ohm 1.85VAC
Common --> 4 Ohm 1.3VAC

When measuring the Primaries, I got 0VAC across all leads. Note that there is NOTHING (including primary/secondary common --> ground) attached during these tests... Should I have hooked something to ground?

Thanks again-

A Dingus

[g1]

Common does not need to be attached to ground, it is just the reference for that side of the OT. So black probe is put there for all secondary side measurements.
It seems your source voltage is getting loaded down when connected. Likely due to the shorted primary side. With a good PT it should hold up at the 3.15V or whatever your source voltage is.
As a learning experience this would have probably been better if the PT was good.

[Mr_bibbles]

Thanks G1! I'm sure this is going to be helpful to someone though, as for some reason this is one of the more confusing tests I've come across, even though it seems simple enough.

So, we can determine that this transformer, with both the mismatched impedances across the primary and the lack of AC voltage on the primaries when fed through the secondaries, is bunko. The Secondaries were giving me 0VAC when fed through the 8ohm tap and measured across common. Then, when I attached that common to ground, the voltages showed up as you'd expect them, with ratios reflecting their given values.

So,

1. G1, can you confirm that NOTHING should be hooked up to the transformer except the AC feed to a secondary tap?
2. How come when the common is not tied to ground, I say 0.0VAC across the primaries to common? Is there a problem with the transformer that would cause this to happen?

[Dave H]

It seems to be conflicting as it says disconnect ALL the transformer leads, hook up ONE AND ONLY ONE lead, etc,

Read it again. It doesn't say "hook up one and only one LEAD" It says "hook up one and only one WINDING" (a winding being two leads)

It looks like you had it connected with only one secondary lead connected to one side of the 3.15V and you were measuring with respect to chassis ground. This put 3.15V between the chassis and the OT secondary but it doesn't put 3.15V across a secondary winding meaning it will measure 3.15V from ground to any secondary tap (and perhaps the primary taps too via capacitive coupling)

I think the least confusing way of doing it is to disconnect all the OT wires then apply 3.15V between the OT secondary common tap and 8 ohm tap. Now measure between the OT primary CT and each plate tap. (Don't measure with respect to ground). The plate taps should both read the same (high) voltage.

[g1]

Then, when I attached that common to ground, the voltages showed up as you'd expect them, with ratios reflecting their given values.

Now I'm really confused. Secondary common is for reading voltages only on the secondary.
When you connect your voltage to the secondary, you measure the primary voltages from end to end of the primary leads. (and from each end of primary to primary C.T.)
Did you somehow end up getting voltage readings on the primary at any point in testing?

[Enzo]

Re-read post #7.

For testing the transformer, ALL wires should be disconnected. Nothing connected to the circuit, and nothing "grounded" to chassis. ALL voltage readings are taken from one transformer wire to another, NOT to chassis. Your meter is not connected to the chassis.

[nosaj]

Now I'm really confused. Secondary common is for reading voltages only on the secondary.
When you connect your voltage to the secondary, you measure the primary voltages from end to end of the primary leads. (and from each end of primary to primary C.T.)
Did you somehow end up getting voltage readings on the primary at any point in testing?

Honestly I think the OP should stop and do resistance tests between pri/sec leads and to the frame before using AC, isn't that usually first?

nosaj

[Enzo]

You are right. Normally we would expect that to have been done up front before we got down to actually "testing" the transformer.

[g1]

Ok, in the first post he mentioned it failed the first resistance test. I'm not sure if he did the rest of them, including resistance from windings to frame, so I'll try to do a rough rundown.
See attached picture of OT windings.
All wires of transformer must be disconnected from the circuit for testing.

It is important that none of the windings have any connection to the metal frame of the transformer. So meter on resistance, black probe to the frame. Touch each wire with the red probe. All of them should show open circuit, how ever that is displayed on your particular meter. Often it is shown as 'OL', but check the manual for your meter if you are not sure.
If any wire gives a resistance reading from wire to frame, the OT has a problem and is probably not safe to test with voltages.

For the secondary winding, put the black probe on common. Touch the red probe to each of the secondary taps, in the example shown it would be the 16, 8, and 4, wires. Each of them should show a low resistance.
For the primary side, measure the resistance between the 2 outside wires, A & B. It will probably be something between 50 and several hundred ohms. Then check from CT to A and from CT to B. Each of those readings should be roughly half of what you got between A and B.

For the voltage testing, you should have connected 3.15V (from half the heater winding) between the 8 ohm tap and common. That requires 2 connections (one of the heater wires from PT and the heater center-tap from the PT). In your case I guess the heater CT was connected to ground but that is not really relevant for the test.
Now you measure AC voltage between A and B, (careful, high voltage!). It will usually be 20 times (or more) what you applied to the secondary. From CT to A or from CT to B should be roughly half of what you measured between A and B.

I would suggest that you apply the voltage across the 2 outside wires of the secondary, so 16ohm tap to common (instead of 8 ohm to common). This will reduce the high voltage hazard a little, and will test all the secondary taps at the same time.
You don't need to worry about measuring the voltage on the other secondary taps, your main concern is what you get at the primary.