OT Weirdness Bogen CHB20A?

[Phil_S]

I bought the OT from a CHB20A on eBay for cheap ($16 shipped), so I'm not concerned about getting burned. I am trying to test it to determine that it is OK and for some idea of how to best use it. The CHB20A uses a pair of 6GW8/ECL86 in PP. The OT has two secondaries, 70V, and 4Ω-8Ω-16Ω-25V. I'm ignoring the 70V winding for now as I can't think of a use for it.

Ordinarily, I'd test by applying ~7V to the secondaries, meter the output on the primary, calc the turns ratio and derive the primary impedance. I use an old filament transformer that I think must be 6.3V and 25.2V; unloaded 7V and 35V.

When I supply 7V to the secondaries, supply voltage drops to 2V and primary output is so low I don't trust what the meter says. Concerned that a 5V drop indicates the OT is pulling current (low secondary DCR, under 1.5Ω), I tried using a resistor in series, but no better luck was had; voltage was in mV on both secondary and primary. I didn't like what I saw, which was highest primary voltage with the source on the 16Ω tap, and lowest primary voltage with the source on the 4Ω tap -- with the calculation suggesting the primary impedance is changing as I change taps. This didn't make sense to me. I repeated the secondary input tests with two different supplies, including a 6.3V-9.5A filament supply from an organ transformer. In round numbers the results I got were:
16Ω:6.4kΩ
8Ω:2.8kΩ
4Ω:1.2KΩ

I changed to supplying 31V (same supply transformer, at least input was constant throughout the test at 31.00-31.04V) to the primary. Secondary output and calc of primary was as follows:
25V tap 0.761v
16Ω 0.401v 96K
8Ω 0.296v 88K
4Ω 0.141 194K (I think maybe too low to be reliable; 16 and 8 tests are reasonably in sync)
Ignoring the 4Ω results, this tells me the primary is 10x what is expected. It should be in the 8K-9K range from reading the tube data sheet.

Testing with input to the primary is so different from testing with input to the secondary, I feel lost. The only reading I "liked" at all is 16Ω:6.4KΩ.

At this point, I am wishing I had a signal generator to send 400Hz through it, but I don't. I have a scope that supplies a 1KHz test signal. I have several meters and at least two will measure frequency.

My gut is telling me the testing is all wrong, yet I don't like what I see. Any suggestions here would be appreciated.

BTW, ALT+234 will generate the omega character!

Thanks.

Phil

[rdjones]

First, using a 400Hz signal vs line freq is only going to provide a slightly more accurate reading, and only in a case where the transformer has poor low frequency response.

Your readings seem to indicate something shorted somewhere.
Are all the continuity measurements reasonable ?

reddog

[Phil_S]

Hi RD. Thank you for the reply.

There is continuity for sure because I get ohms and voltage and no smoke.

The DCR at 353Ω blue to brown on the primary seems to be OK. It is not symmetrical, and I'll assume it is not interleaved. Blue to red is 145.3Ω and brown to red is 208.5Ω.

Secondary DCR is more problematic. With the Fluke 8025B:
0-25V tap - 1.7Ω DCR
0-16Ω tap = .6Ω DCR
0-8Ω tap = .6Ω or .5Ω DCR
0-4Ω tap = .4Ω DCR

With the Fluke 8600A:
0-25V tap - 1.6Ω DCR
0-16Ω tap = .40Ω DCR
8-16Ω tap = .46Ω DCR
0-8Ω tap = .44Ω
4-8Ω tap = .27Ω DCR
0-4Ω tap = .30Ω DCR

The readings are so close and all under 1Ω so that I don't think they can be relied upon very much. I suspect the wire extending the winding outside the OT is messing with the true readings.

Well, one last attempt. To compensate for the low range, I used a 5Ω resistor in series. This lets the meter move to a better spot hopefully. The resistor is 5.34Ω, which I subtracted from all the readings below with the 8600A:
0-25V tap - 1.46Ω DCR
0-16Ω tap = .23Ω DCR
8-16Ω tap = .30Ω DCR
0-8Ω tap = .24Ω
4-8Ω tap = .1Ω DCR
0-4Ω tap = .14Ω DCR

Something not right about the relationship of the 8Ω to the 16Ω taps? Not sure such low numbers can carry that significance. Where does the voltage drop from 7V to 2V play into this (original post)?

[Phil_S]

Just picked this up at AX84.com. Maybe I should give it a try?
http://www.geofex.com/FX_images/xform_test.gif

[Phil_S]

I guess I talking to myself here, but I'll post anyway...it helps to believe someone is listening.
One more time around checking secondary DCR. This time, I put a 10Ω resistor in series, using the 8600A, which will show two digits after the decimal. After subtracting the 10Ω to get the assumed true DCR:
0-4Ω = .09Ω DCR
0-8Ω = .20Ω DCR
0-16Ω = .21Ω DCR

If there are 10 turns on the 4Ω tap, the 8Ω tap should have 14.1 turns, and the 16Ω tap should have 20 turns. Assuming the turns ratios, 0-16Ω should be double 0-4Ω, and it is, within reason, very close to that. 0-8Ω should be 1.4x 0-4Ω, and it is not. This suggests there is short somewhere on the secondary.

I think I am going to take it apart and attempt to rewind it. I don't need the 25V tap, so that can be eliminated. I may even be able to unspool the wire and reuse it for a rewind as the DCR between the 16Ω and 25V is 1.23Ω, suggesting there is 5x the amount of wire from 0-16Ω. Also, I don't need the 70V winding, so I can unspool that, as well and not put it back if that turns out to be the outer winding, which I am guessing it is.

Now I need to find a small fan motor or similar to help me spool the wire.

[rdjones]

I guess I talking to myself here, but I'll post anyway...it helps to believe someone is listening.

When you start to answer just make sure you don't argue with yourself

One more time around checking secondary DCR. This time, I put a 10Ω resistor in series, using the 8600A, which will show two digits after the decimal. After subtracting the 10Ω to get the assumed true DCR:
0-4Ω = .09Ω DCR
0-8Ω = .20Ω DCR
0-16Ω = .21Ω DCR

If there are 10 turns on the 4Ω tap, the 8Ω tap should have 14.1 turns, and the 16Ω tap should have 20 turns. Assuming the turns ratios, 0-16Ω should be double 0-4Ω, and it is, within reason, very close to that. 0-8Ω should be 1.4x 0-4Ω, and it is not. This suggests there is short somewhere on the secondary.

Agreed.
I also think your previous measurements on the primary show a bit more unbalance than is typical, but not unreasonable.
Can/did you measure the voltages on each side of the primary when exciting the secondary and compare side to side ?
Also just for the heck of it see what's on the 70V winding to make sure it's not shorted.

I think I am going to take it apart and attempt to rewind it. I don't need the 25V tap, so that can be eliminated. I may even be able to unspool the wire and reuse it for a rewind as the DCR between the 16Ω and 25V is 1.23Ω, suggesting there is 5x the amount of wire from 0-16Ω. Also, I don't need the 70V winding, so I can unspool that, as well and not put it back if that turns out to be the outer winding, which I am guessing it is.

Now I need to find a small fan motor or similar to help me spool the wire.

Good luck with that. BTDT

I have offered the opinion that on some more infamous transformers that include a high voltage line output tap (250 or 500 Ohm, 70V) it might contribute to the sound.

rd

[Phil_S]

When you start to answer just make sure you don't argue with yourself

You mean people do that?
Input to the 16Ω secondary gives symmetrical voltage on the primary. It was something like 20.49v and 20.47v. The 70V seems to be fine. Devil voltage in (6.66V ) puts out 9.47V on the primary, and better news that input voltage is appropriate in respect of the filament transformer being used as the supply. Interesting thought about what the 70V winding might contribute. I can always rewind it if not too many turns -- maybe I'll get lucky and it's on the inside of the other secondary. I am guessing by the size of the bobbin and that the 0-16Ω secondary will be in the neighborhood of 200 turns unless they used silly thin wire. Since I'll be hand winding, more than that is probably too many turns to keep track. This is supposed to be 20W rated, meaning there is, at least in theory, some lower limit on how fine the secondary wire gauge can be.

[Phil_S]

Contacted the seller. He gave a full refund within an hour! Now that it's free, I have no guilt at all about unwinding it. It will be a while 'til I get to it. Taking apart a tranny takes some time.

Contacting the seller in a non confrontational manner was the right thing to do. I told him that I wasn't upset, that I couldn't leave feedback of any kind because the OT was defective, and that I wasn't asking for anything. I suppose this plays into my basic outlook on eBay, which is that you shouldn't spend what you will feel bad about losing. This transaction was for less than $16. That makes it much easier to keep it in perspective.