Output Transformer Question: Marshall JCM900 50W

[jpglyde]

I have a Marshall JCM900 50W (model 4500) with low output. It's only putting out a few watts (measured by turning it up to 11 and not experiencing any hearing loss, temporary or otherwise, with my head a couple feet from the speaker)

I've checked all I know to check. I can't find any bad resistors, diodes, etc in the amp section. The preamp section is sending a full (28Vpp) signal into the screens of the 6L6 power tubes and the power supply voltages are normal.

The only think I found odd is when I do a DC resistance test on the primary of the output transformer. I'm reading 34 ohms between center tap and one side and 40 ohms between center tap and the other side.

Does the fact that these readings aren't equal mean the transformer could be internally shorted? I don't know what else to check.

[eschertron]

I have am amp with a shorting jack on the speaker. If i don't get the plug in just right, the shorting jack doesn't "un-short" and I get no (or low) levels at the speaker. Mystified me until I figgered it out.

Start from the speaker end of the signal chain. Try a different speaker, verify impedance. Verify the speaker jack. Verify the impedance selector switch (if there is one). If everything on the secondary side is good (clean and tight), then move on to the OT. Do you know how to check the turns ratio of the tranny? If you don't, then look it up. I always have to look it up to remember the relationships of voltage/turns/impedance. Sucks getting old This check requires a few minutes with a soldering iron, and can be down without power on the amp if you can get a small 12v-ish project transformer.

If everything checks out to that point, I'd put in new tubes.

*random thought. I know you said you checked all... but if you didn't, verify the screen voltages at idle, along with the plate voltage.

[52 Bill]

The only think I found odd is when I do a DC resistance test on the primary of the output transformer. I'm reading 34 ohms between center tap and one side and 40 ohms between center tap and the other side.

Does the fact that these readings aren't equal mean the transformer could be internally shorted? I don't know what else to check.

To answer your question directly, no. The two halves of the primary winding are rarely exactly the same dc resistance. This is due to the fact that the length of the wire increases as the size of the winding gets larger.

This does not mean that your transformer doesn't have a shorted turn, but that is not the way to test for it. Go to R.G.'s site and build the transformer tester that has there. It only takes a few small parts and a battery, and you can test for shorted turns.

[J M Fahey]

I have am amp with a shorting jack on the speaker. If i don't get the plug in just right, the shorting jack doesn't "un-short" and I get no (or low) levels at the speaker. Mystified me until I figgered it out.

Well, that's eactly what happens when you have a shorted turn in a transformer: "you have an internal shorting jack ... which never gets unshorted" .

Since that short is not *zero* ohms but a low value, the amp manages to drive it (with great difficulty) and also the speaker ... wasting tons of power.

So amps with shorted OT do sound (which is confusing) and put out a few watts, but weak and anemic.

[Gregg]

The way those OTs are wound the primaries' DC resistances should be very close in value if not almost the same. The best way to check if you have shorted primary is to measure the inductance between two ends of the primary and of both halves separately. Although the DC resistance could be different the inductance of the two halves should be the same.

[J M Fahey]

Sorry but that does not show the problem.

Maybe you expect to find one winding with less than normal inductance and say: "bingo!! this is the bad one!!!" but it does not work that way.

All windings (secondary also) are magnetically in parallel (they are all wound on te same core) and a shorted turn in one kills inductance in all .

You can tell you have a shorted turn, you can not tell where.

Similar to reading many parallel power transistors or a +/- 15V rail and finding it shorted, you can tell there is a short but not where it is, and for the same reason.

Reread R G Keen's shorted turn indicator and see it says about the same.

[Enzo]

Maany people expect that a transformer with a "shorted turn" will conveniently measure different resistance. So instead of 40 ohms and 40 ohms, they think it will read 40 ohms and 2 ohms or 20 ohms. And that COULD happen. A turn can also sshort to the frame. That will cause trouble, but often not alter resistance readings. But a more common shorting failure is one turn shorts to the turn next to it. And your meter will NEVER detect that. The reduction in resistance would be whatever one turn of wire measures - maybe 8 or 9 inch piece of solid wire.


Some OTs are wound in a manner that does leave the two halves at roughly the same resistance, but others are simple winds, with a center tap tacked on halfway through the wind. Those will have ssubstantially different resistances on the halves. Someone mentioned that above - the inner turns are shorter, and the outer turns longer, so the outer half of the winding will have more resistance. fortunately transformers work on turns rations, not resistance.

[Gregg]

Sorry but that does not show the problem.

Maybe not always but sometimes it does. I've had couple of cases like this in the past. From what I remember one half measured around 2H the other one 1,3H or something. The OT had like 6-7H of overall inductance where at least 9H was expected. It turned out to be a nice hole in the insulation between one of the primaries and a secondary.
The JCM900 100W OTs are wound according to the classic 1/4-1/2-1/4 primaries configuration which is one of the ways to get the two halves with almost identical DCR. I suppose the 50W version is wound the same way but if it's not then 34 and 40 Ohms are still in the ballpark.