Marshall JCM900 Output Issue

[Mr_bibbles]

Got a JCM900 4100 on the bench: http://www.thetubestore.com/lib/thet...-Schematic.pdf

Owner is in a very very loud psychedelic rock band that uses a lot of pedals including loopers. Symptoms are very low, grainy output.

Swapped Power tubes a pair at a time (until I get a matched quad in the mail) no change. Pre-amp out sounds fine, Plate / screen voltages check out, no leakage from coupling caps. Replaced V3, no change. Traced the (healthy sounding) signal with an audio probe all the way up until the impedance selector, where it dies out to a whisper. Signal sounds good on the grid of the output tubes even. Now trying to use the Output Transformer diagnostics tool from R.G. Keen's lovely site but I can't wrap my head around it....

Checked resistance between primary B+ (brown) to Plates (pin 3) - 2.2k, same as wirewound resistors.

ACcording to the diagnostic page,"Plate windings are almost always in the tens to a few hundred ohms of resistance. A reading much over 1K is pretty sure sign that the winding is burned open."

Does this mean it's toast?

Primary
Brown ---> Red 16 ohms
Brown ---> White .9 ohms
Red ---> White 17 ohms

Secondaries to primaries: all OL


RG also states: "Do a similar test on the secondaries - look for open windings, or shorts to the core." Do these following readings count as open windings? The readings sometimes dip to .0, but I recorded most as .1 if I saw it at all.

Secondary resistance (NO LEADS CONNECTED)

Green (16) ---> Yellow (8) .1
Green ---> Purple (4) .1
Green ----> Common .2
Yellow ----> Purple .1
yellow ---> Common .1
Purple ---> common .2


Please help me understand simple OT testing. Thanks!

[The Dude]

The center tap of the OT primary should read approximately the same to either outside. It seems one side of yours is shorted.

Edit: If those readings are with the primary connected, disconnect it and re-test resistance readings to double check/verify it's not something else in the circuit.

[Mr_bibbles]

Thanks as always dude. I'll check that tomorrow when I get into the shop. Can you help me clarify the B+ to plate testing as well?

"Plate windings are almost always in the tens to a few hundred ohms of resistance. A reading much over 1K is pretty sure sign that the winding is burned open."

On the Schematic, it seems like B+ is connected to plates through only the wirewound resistors AND through the primary windings. I assume the primary windings are going to be much lower than 2.2k, so the reading here should indicate the Transformer itself's resistance right? By disconnecting the leads from it (gunna do this first thing tomorrow) we'll know the primary resistance without a shadow of a doubt, but ideally this should work with the leads connected, right?

[The Dude]

The plates connect directly to the OT windings via the red and white wires (1 wire for each tube pair- no resistors). The brown wire is the center tap. B+ connects there. You have tested the windings with this:

"Primary
Brown ---> Red 16 ohms
Brown ---> White .9 ohms
Red ---> White 17 ohms"

If any part of the transformer measured open or very high in resistance in circuit, we would know it's bad without disconnecting it. Because we are seeing a near short across half of it, we need to verify it is the transformer itself that is shorted and not a component across the winding. It could be socket, solder bridge, etc., etc. If you suspect a bad transformer, you want do be positive. It's an expensive part.

[g1]

Sounds like you are measuring to pin4 and reading the 2K2 screen resistors.
Can you disconnect the white wire and measure resistance from brown to white?

[Ted]

16 ohms is about right for one side of a JCM900 O/P transformer. The most likely short path is across primary windings inside transformer. To confirm, unsolder red, brown and white wires from transformer and measure again.

If you have a signal generator try putting 1V at 50Hz into the 8 ohm speaker output (With the amp powered down and disconnected from mains, obvs). You should see the same voltage on each primary plate connection. Can't remember exact voltage offhand, but between 10 and 30 V. If the two sides are wildly unbalanced (like > 30%), you need a new transformer.

[Mr_bibbles]

Great! So total failure would indicate the transformer as bad without having to disconnect it, (though you always should because it's an expensive part) while partial failure would need additional testing.

Disconnected measurements:

Brown ---> Red 16 ohms
Brown ---> White 1 ohm

The transformer primary is shorted and needs replacement.

[Mr_bibbles]

I do not have a signal generator It's on my christmas list if any of you guys are feeling generous

[Mr_bibbles]

Sounds like you are measuring to pin4 and reading the 2K2 screen resistors.
Can you disconnect the white wire and measure resistance from brown to white?

Measurements posted above, but I was measuring pin 3. Disconnected white ---> Brown is OL.

[g1]

Disconnected measurements:

Brown ---> White 1 ohm

The transformer primary is shorted and needs replacement.

Disconnected white ---> Brown is OL.

I hope one of those statements is a typo.
Either case would be bad I suppose.

[Mr_bibbles]

I think I took your directions a bit too literally:

"Sounds like you are measuring to pin4 and reading the 2K2 screen resistors.
Can you disconnect the white wire and measure resistance from brown to white?"

Disconnected white wire -----> attached Brown wire = OL
Disconnected White LUG -----> Disconnected Brown LUG = 1ohm

So I will get some new iron on order Thanks guys.

[Mick Bailey]

In the meantime revisit RGs transformer testing article. Just a battery and neon could save you loads of trouble. FWIW, even if the transformer reads 'good' on your meter, it can still have a shorted turn. You cannot measure this with a DMM.