Odd problem with amp (OT gives funny readings)

[ElectronAvalanche]

Dear All,

I have this DIY amp here that blew the mains fuse after extended use. No signs of any burned parts. Amp simply blew the mains fuse when the PT broke down. Amp is a mix of Fender preamp and Marshall preamp with a SLO type Outupt section.

The PT was toast, installing a new PT was the initial cure. Firing up the amp with a light bulb limiter shows now glow of an 100W bulb with the Standbye on (no B+ to the amp).

Heater voltages with the light bulb limiter in series drops (from 6.3V to 5V)once all tubes are installed (which can of course be expected). Putting the B+ on, the bulb lights up about 30-40% and the B+ drops from 460V to below 220V on on the rectifier.

Now to the odd problem:
Measuring the resistance of the OT with all output tubes pulled and no power to the amp shows 130 Ohms from CT to the two outer primary windings (on both sides), but only 30 Ohms between the outer two primary windings (plate to plate winding should be 1k7 Ohms as per datasheet, I am measuring 130 Ohms). I would have expected to have this the other way around: largest resistance between two outer windings and half of that from outer windings to CT: Is this a sign of a toasted OT? I still need to run the amp without the bulb limiter. OT is a Dagnall Marshall type. Wiring of primary was not changed during the repair as is.

Any ideas welcome. Odd is that the two outer primary windings show symmetrical resistance readings. If there would be a short, I would assume resistance would be different.

Thanks,

Electron

[pompeiisneaks]

That does not sound good for the OT, no. It should be something like almost identical, but not between CT and each of the other OT leads, and then add those two up and that's what you get between the OT leads. Example. Say you got 130 each, it should add up to 260 between the two OT main leads. 30 means the OT may have a short, and that is why the PT blew too. Not sure, though, but it's a good possibility. Remember, copper wire isn't a perfect conductor, so the more distance it travels, the greater the resistance, and that's all an OT is, a very long wire with a connection in the middle for the CT. If you get less resistance between the two OT's leads then it means there's a shorter length of wire between them than the path to the CT itself.

~Phil

[Phil_S]

Some of how you express things doesn't seem quite right. I think you mean to say 130 ohms from outer end to CT on both sides, so you'd expect 260 ohms from end to end? "Winding" is a single loop of wire or a coil that has two ends and sometimes a center tap. Are you saying you expect 260 ohms end to end on the primary winding but only measure 30 ohms? Either you aren't measuring correctly, there's a short, or something else. The dim bulb suggests there's no short.

I want to pick "something else" as the problem. Are you measuring ohms from end to end with the CT connected? Are tubes pulled from sockets? I'd break the CT connection, pull tubes, and measure again. I expect you'll get 260 ohms end to end. This is a reasonable way to confirm if there is a short or not.

Don't be surprised if the two halves of the winding aren't exactly the same ohms. It is typical for there to be a small difference.

After you confirm about 260 ohms end to end, if there is nothing else stopping you, it's time to remove the dim bulb limiter and fire it up. Do this without tubes first, as the tube load will cause current to flow and that is when damage will occur if there's a problem. If you get reasonable voltage readings without tubes -- readings will all be too high, then it is time to put the tubes in the amp and try again.

Good luck. Let us know what happens.

[pompeiisneaks]

Some of how you express things doesn't seem quite right. I think you mean to say 130 ohms from outer end to CT on both sides, so you'd expect 260 ohms from end to end? "Winding" is a single loop of wire or a coil that has two ends and sometimes a center tap. Are you saying you expect 260 ohms end to end on the primary winding but only measure 30 ohms? Either you aren't measuring correctly, there's a short, or something else. The dim bulb suggests there's no short.

I want to pick "something else" as the problem. Are you measuring ohms from end to end with the CT connected? Are tubes pulled from sockets? I'd break the CT connection, pull tubes, and measure again. I expect you'll get 260 ohms end to end. This is a reasonable way to confirm if there is a short or not.

Don't be surprised if the two halves of the winding aren't exactly the same ohms. It is typical for there to be a small difference.

After you confirm about 260 ohms end to end, if there is nothing else stopping you, it's time to remove the dim bulb limiter and fire it up. Do this without tubes first, as the tube load will cause current to flow and that is when damage will occur if there's a problem. If you get reasonable voltage readings without tubes -- readings will all be too high, then it is time to put the tubes in the amp and try again.

Good luck. Let us know what happens.

Very probably not very clear. Let me try with a crappy pictogram.
Imagine the @ symbol are loops of wire between point A, the Center Tap (CT) and B (A and B are the leads at either end of the OT).

A@@@@CT@@@@B

I'm referring to the OT leads as A and B there above, CT is the center tap. If you read 130 ohms between A and CT and say 132 ohms between CT and B, then you should read 262 ohms between A and B. He's saying he's reading 30 ohms between A and B, but about 130 between either A and CT or B and CT.

Your idea that resistance between A and B being 30 due to the CT being connected does have some merit, if the CT is still connected it could create a parallel resistance between any other points on the chassis etc, causing the significant difference and it wasn't something I had thought of. you definitely need to remove the transformer taps from any connections to the chassis or sockets to be 100% sure you're not reading some other path. BUT if the A and B aren't connected, I don't see how that could happen directly as the circuit of the other half we're not 'seeing' can't loop back either A or B right?

OTOH if A and B are connected to the sockets and something's wrong there providing path to ground (which may have caused the transformer issues in the first place) then you could be getting this odd reading.
BTW I'm making up that 260 ohms, I don't know if it really is reading 130 per leg or not, that was his initial reading, and could be also wrong for the same reason you suggest.

~Phil

[ElectronAvalanche]

Thanks for the fast replies!

So I unsoldered the primaries and now get 15 Ohms and 17 Ohms from the two outer wires (plate wires red and brown, i.e. A and B according to Phils pictogram) to the CT (white ) and 30 Ohms between the two outer wires (measured between A and B). Seems there is a short in the primary of this OT, since 30 Ohms for the whole coil seems really a bit low. For the measurement in my original posting, the tubes were pulled, so there was only a connection between pins 3 of the octals of each pair and the corresponding half of the primary winding.

So I assume the primary of the OT had a short (either a power tube malfunction or the owner of the amp ran it with an open load) and the PT had to deliver too much current? The B+ fuse was intact btw, only the mains fuse blew.

I probably would need to apply a low AC voltage to the secondary side of the OT to be really sure the OT is defect, but the reading of 30 Ohms across the primary seems mighty suspect.

Odd thing is that the B+ fuse did not blow, at least the owner delivered the amp with an intact B+, the mains was missing. The PT smelled a bit burned and had some signs of overheating, the OT looks pretty normal from the outside. No part in the amp shows any sign of excessive heat damage.

So I guess a new OT needs to be ordered.

Could an open load result in a failure of the OT?

Tubes were Sovtek 5881 and they also do not look like they went thermal.

Thanks again for your insights and help,

Electron

[pompeiisneaks]

Actually your new reading seem to imply it's pretty much fine. The A side at 15 B side at 17, gives a total of 32 ohms pretty close to 30 that you're reading. Sounds to me like there may be some other problem like a path to ground or other kind of thing that shouldn't be there in the sockets. BUT since the PT fried, that makes me thing the poor thing did get some abuse and could have killed the OT as well.

~Phil

[labb]

Might want to give this a try: https://www.premierguitar.com/articles/ ... r-tester-1

[danman]

You mentioned in your OP that the data sheet listed the primary impedance at 1k7 which is what you expected to see. This number is most likely the reflected primary impedance that will be seen at the power tube plates. You will not actually see this figure when testing with your meter though.

[Reeltarded]

Mains, heaters.. B+ fuse?

[pdf64]

A shorted turn will suck all the energy out of a transformer, inductance will collapse, power output will be a tiny fraction of what it should be
Try RG's shorted turn tester http://www.geofex.com/FX_images/xform_test.gif

[tubeswell]

...
Measuring the resistance of the OT ... shows 130 Ohms from CT to the two outer primary windings (on both sides), ...(plate to plate winding should be 1k7 Ohms as per datasheet, I am measuring 130 Ohms). I would have expected to have this the other way around: largest resistance between two outer windings and half of that from outer windings to CT: Is this a sign of a toasted OT? ...

On a Superlead100 OT, 1k7 is the intended plate-to-plate impedance on the primary that you get when you have the appropriate speaker load connected to the secondary - it is not the DC-resistance of a winding (Impedance is resistance-to-AC, and transformers are 'AC-pass' devices that work by inductance, and inductance only occurs where there is a change in current over time - i.e. when there is AC/signal present).

It's pretty normal for the DC-resistance on one half of a push-pull primary to be higher than the other side, because one half of the primary is usually wound over the top of the other side of the primary, and so the wire is physically longer on one half (even when the primary and secondary are interleaved).

If you want to check the integrity of an OT, you can use RG Keen's shorted turn tester (as pdf64 suggested), or you hook a dummy load up to the secondary and pass a 100VAC source across the primary, and see if you get the expected amount of VAC on the secondary. (the Pr:Sec VAC ratio is the square root of the Pr:Sec Impedance ratio, so you will need to square the VAC ratio to get the impedance ratio (and then you will need to multiply the impedance ratio by the secondary load to get the primary impedance).

[ElectronAvalanche]

I stand corrected, the 1k7 primary is an impedance not DC resistance.

The Hammond Mfg. site states a DC resistance for a Marshall 100W replacement transformer of about 16Ohms between outside (plate) and CT. So my DC resitance measurement is actually not too far off.

https://www.hammfg.com/files/parts/pdf/1750U.pdf

But nonetheIess I threw together the RG Keen Transformer Test circuit. And at the primary side I see a dim but quite noticable flash when pressing the momentary switch. On a small PT I have here in my stack, on the mains side I get a much more brighter flash, on the secondary side of the small PT with 10V winding, the flash is as bright as the flash with the OT.

Shorting the primaries of the OT and then pushing the momentary switch results in no flash (as expected).

So at least the primary side seems to have no full short. I think I need to bite the bullet and check the amp without the light bulb limiter and see if I get signal on the speaker output.

Thanks again for all the input.

Electron

[dorrisant]

Put approximately 10vac across the brown and blue of the primary... Measure the ac at the secondary. Then you can verify whether the impedance has changed. If it looks much different at all, then the OT is probably toast.

[Phil_S]

Put approximately 10vac across the brown and blue of the primary... Measure the ac at the secondary. Then you can verify whether the impedance has changed. If it looks much different at all, then the OT is probably toast.

I'm with Tony on this one. As long as you've got it disconnected, there is no harm in seeing if the turns ratio is reasonable. If not reasonable, then you know the winding has a short or a break.

[Stevem]

Here's a simple test out for a OT.
Apply 10 vac to the primary and you voltmeter to the secondary.
With EL34 tubes this is what you should see at the output taps if the OT is good.
4 ohm = .35 vac.
8 ohm = .49 vac.
16 ohm = .69 vac
With 6L6 type tubes this changes to .
2 ohm = .22 vac.
4 ohm = .31 vac.
8 ohm = .44 vac.
16 ohm = 62 vac.
6V6 tubes or EL84.
4 ohm = .22 vac
8 ohm = .31 vac
16 ohm = .44 vac.
All of these voltages are for pairs of tubes!