Trouble with power transformer: uneven voltages

[Tom Phillips]

Hi Carlo,

This situation just keeps getting worse.
I'll respond to the fit problem in the other thread.The construction of the smaller transformer looks unconventional like it was made by someone just getting in the business.

As to the voltage imbalance issue, the results don't make sense to me. i.e "exactly the same readings but inverted in the other direction"
Are you willing to investigate that a little more?

Regards,
Tom

[Satamax]

Wel, if they are indeed two different taps, the same thing that happens when you short out the secondary of any other power transformer.

Hi Enzo, another daft idea! Virtual centertap? Like we do for the heaters? Two resistors, or even easier two 1N4007 diodes?

[Enzo]

Tom I am pretty sure what he means is that the one end was 50v high before, and switching "center taps" makes the other end the one that is 50v high. Exact same problem, we just moved it over. The trouble is that neither "center tap" is in the center.

Virtual centertap?

I think if we sealed all the holes on my car, we could flip it over and have a boat. OR we could just get a proper boat to start with. I don't see the problem as how to hook this thing up, I see the problem as this is not the right transformer to use.

[slidincharlie (Carlo P)]

Tom I am pretty sure what he means is that the one end was 50v high before, and switching "center taps" makes the other end the one that is 50v high. Exact same problem, we just moved it over.

That's correct. Sorry if I wasn't clear in my previous post.

I don't see the problem as how to hook this thing up, I see the problem as this is not the right transformer to use.

I have just redone a few readings.
Besides that 50V or so unbalance between the two halves of the HV secondary, the B+ is good. I read 420 VDC on the plates with 20 mA bias current.
If the described unbalance is going to "harm" the amp (see one of Tom Phillips' posts above) and stress the filter caps, or if the amp will not sound as expected anyway, I'm going to replace this transformer and buy a new one. I want to sleep happy and worryless, and have a functional and safe amp...
Seeing the problem from a different perspective, should I trash this tranny or is it good for other uses, considering the unbalance?

[Steve Conner]

To me, it sounds like the transformer maker misunderstood what the customer was asking for, and made a transformer with no true center tap and two bias taps halfway either side of where the center tap should have been. Weber either never noticed, or decided to try and cut his losses by selling them anyway.

It could be made to work by grounding each "center tap" via a diode (one cathode to each center tap, anodes to ground) but that's a kludge, and then you don't have a bias voltage: the only voltage you can rectify from the bias tap will be positive.

[matndona]

If you have access to a voltmeter with an Inductance setting you could get to the heart of the matter pretty fast. By voltage test, it looks like the center tapped winding has an uneven number of turns on the two halves. Sounds like somebody set the winding machine counter wrong.

On this type of coil normally the first half <inner> winding to the CT will have a lower DCR than the 2nd half <outer> winding as its got fewer feet of wire for the same # turns. So DCR might not tell the story. Inductance though is pretty much related to turns on the core. I'd expect the two halves to be within 1% inductance easy if wound correctly.

[Tom Phillips]

I’d like to summarize the situation and offer a little more food for thought.
We are dealing with a Weber supplied P/N 025130EU power transformer that is supposed to be a drop in replacement for a Fender Deluxe Reverb amp. This is the second one that Carlo has installed in the amp. The first one had overheating issues that were discussed in another old thread.

There is no doubt that this second transformer is faulty. It was manufactured incorrectly. When it is hooked up per the supplied spec sheet, the two halves of the high voltage AC secondary are 46 V out of balance. Ted Webor told Carlo “There is nothing wrong with the transformer. It will work fine, even though the voltages are not exactly the same.” Several of us have disagreed with that and Alex (Thread entry #9) gave a good explanation of why you wouldn’t want to use the transformer as is.

Carlo posted photos of the two transformers he has received in thread entry #13. The core size, end ball and construction technique are different for each. There was either a significant change at the manufacturer or Ted Weber changed suppliers. This is just an interesting point and does not help solve the problem.

In entry #4 Carlo posted the open circuit secondary voltage readings. The readings show:
1) One side of the HV secondary is 46V higher than the other. Therefore, the “center tap” is NOT in the center.
2) The bias tap is at 46V with respect to the wire color coded as the center tap

Given the fact that the imbalance voltage is exactly the same as the bias tap voltage, we speculated that the bias tap and the center tape wires may have been misconnected inside the core. Carlo took the voltage readings again (See thread entry #6) and reported the following:
“Pretended that the RED/BLU bias tap was the center tap.
Result (with the wires in air): I have exactly the same readings but inverted: the wire that gave me the higher VAC reading now gives me the lower, and viceversa.”

This still doesn’t seem possible to me and the reason I am continuing my discussion is because of the possibility that it’s still a mislabeling of the RED/YEL center tap lead wire and the BLU/RED bias tap wire. Normally, I’d just say to send it back but it’s currently on the other side of the ocean. The discussion may be just academic at this point but it’s an interesting challenge to figure out what’s going on here so please bear with me and read on.

Remember that the voltage on one side of the winding is high by exactly the same voltage as the bias tap reading. If you re-measure temporarily using the bias lead as the center tap, I don’t see how the exact same 46V imbalance would move to the other side. I submit that either the two sides will now be in balance OR the imbalance would be twice as much. (i.e. 92 VAC). If “Neither of those is the center tap” as Enzo speculated in response #10 then the complete 46V imbalance would not have moved from one side to the other. The voltage imbalance for each measurement would have been a different split. Right?

So Carlo – I still wonder if you got the two red wires mixed up when you took the second set of readings. Easy to do since they are both red. If you are willing, I’d like to know what the voltages on each secondary lead is if you measure them all with respect to one red lead. (Put your common test lead on one red wire and leave it there as you read the voltage on each of the other leads of the HV secondary winding) Do you have the time to indulge me?

Cheers,
Tom

[Satamax]

I have attached the schematic of the tranny, in case it is of any help.

I have read again all voltages with all wires in the air, to find where the bug is (use the schematic as a reference):

RED to RED: 698 VAC
1st RED to RED/YEL (ct): 326 VAC
2nd RED to RED/YEL (ct): 372 VAC

RED/WHT to RED/WHT: 555 VAC
1st RED/WHT to RED/YEL (ct): 300 VAC
2 nd RED/WHT to RED/YEL (ct): 255 VAC
(I don't use the red/wht secondary)

RED/BLU to RED/WHT (ct): 46 VAC.

The filament heaters (6.3V and 5V) read ok.

With he exception of the RED/BLU bias tap, the other HV secondaries are higher than specs and, most important, with one side unbalanced, correct?

May I use this transformer, or should I return it to Weber (for the second time!)?
Is there a way to correct this?

Carlo, daft me, but i come with another crazy idea! On your diagram, if you measure from the top red/white to the red/yellow and from the bottom red to red yellow, aren't the voltages a bit more equal?

Hth.

Max.

[Ptron]

To me, it sounds like the transformer maker misunderstood what the customer was asking for, and made a transformer with no true center tap and two bias taps halfway either side of where the center tap should have been. .

I'm with you, Steve. This is exactly what I was thinking. It even looks that way in the diagram.

[Tom Phillips]

I'm with you, Steve. This is exactly what I was thinking. It even looks that way in the diagram.

Maybe so but I would hope that there were more detailed and clear specifications supplied to the transformer manufacturer when Ted Weber placed the order. And then a check of the first article before going ahead with the order.

Anyway, the nature of the beast that Carlo currently has will be more clear to me if he takes the latest set of voltage reading I requested with all the readings referenced to the same point.
Cheers,
Tom

[Marvino]

Hi Carlo,
With regard to your resistance measurements of the B+ winding from the CT. The difference in resistance is a normal result of the way the transformer is wound. Number of turns is directly related to voltage. Imagine winding the begining of the B+ winding, with X number of turns to the center tap. Now we have to wind the same number of turns on top of the bobbin we have just wound, however with an ever increasing winding diameter. Thus the second side of the CT has a longer "piece" of wire (more resistance) than the first half. So the ohms difference from CT is normal on any conventionally wound transformer. (Other more sophisticated transformers have side by side bobbins for equal wind length.) The small difference in voltage is negligable. Just don't exceed the quiescent VoltAmpere rating of the winding. I design my power supplies at about 65% of mfr's rating. This allows me to run comfortably in class A with very little or no heat. I recommend the Hammond 300 series transformers for the following reasons, parallel primary windings with international voltage taps; you get exactly what you want. Also the higher circular mils (wire size) per ampere allows for greater energy transfer and minimum heat. These transformer also have a 50v bias tap. AN ADDITIONAL CONCERN! Many replacement transformers have not been design for modern mains/line voltage. One big problem I've discovered is with filament winding loading and having exactly 6.3 VAC rms on all the filaments. I discovered the filament winding in one rebuild, running just over 7 volts! This leads to cathode overheating, stripping thus premature failure of the tubes' performance. Just some thoughts for you to consider. Best of luck, Marvino

[Enzo]

Tom, look at the thing this way. The two "center taps" are 46v apart. By measuring the voltage at either end or the whole winding, when we ground one center tap, that 46v is in series with one half, and when we ground the second center tap instead, then that 46v is on the other side. We are choosing which side to include the 46v with by which center tap we choose. I am assuming the voltages to each end from either side of the 46 are the same.

I can't see how the offset could be anything BUT the 46v of the center winding section.

[slidincharlie (Carlo P)]

Guys, sorry for responding late.
Yesterday I've been away from the pc. Further the notification system apparently did not work.
Later today, back from work, I'll post the new readings.

Thanks,
Carlo

P.S.: I publicly apologize for posting a piece of private correspondence between me and Ted Weber. I realize that it is not fair, but I thought that it added to the discussion. Sorry again.

[Rick Erickson]

{what would happen if Carlo grounded the two wires labelled as 45v? }

That would short the 45V (actually 90V [45V+45V]) winding which would burn out the transformer rather quickly.
I was dissapointed with the Tweed Deluxe chassis I ordered from Webber VST. I ordered a Princeton Power transformer at the same time, which should be the correct size for an original Tweed Deluxe - only to find when the chassis arrived, that it was cut out for a Deluxe Reverb PT.
I've emailed Ted about several issues, the 30% taper pots that should be 10%, (there are several including the 3MRA Speed pot) and the 1M switched Champ pots that should be 30% Log (A Taper) and are actually 30% Reverse Log (C Taper). On top of that it bugs me to no end that his pots have a 7/16" metric nut when the industry standard is 1/2"x32tpi.
Some of my complaints may be nit-picking, but I think it would be really great if someone could supply the correct parts for these vintage-style amps as well as for servicing the true vintage ones. It's not just Ted either - I ordered a Tweed Bassman Chassis from Mojo and was (A) charged twice when they didn't include it in my first order (this was never resolved - they took the stance that they couldn't possibly have made an error like that) and (B) The Chassis was so poorly made that I hate to use it for anything. It looks like a problem waiting to happen.
Generally my dealings with Webber VST has been much more positive than negative. I deal as little as possible with companies like Mojo and ARS Electronics (another story for another day) that have been customer service nightmares for me.

RE

[Marvino]

Marvino here, Gentlemen I'am really embarassed after looking at Ted Weber's spec and trans schem which Carlo posted and checked at the Weber site. The schem as drawn, makes the HV secondary appear as one winding symetrically tapped with respect to start and finish(red/red). If my observation is right, and the "red/yel CT" is grounded, use the "start" end with the dot RED/WHT lead to the rect plate, the RED/YEL as grounded CT and use the bottom RED lead to the other rect plate. You will get equal voltage at the rectifier plates and you can use the BLU/RED as a 45 volt bias tap. If the trans was tapped symetrically with respect to start and finish, I would not plan to load the winding near it's claimed 150 ma. I feel there will be some difference in winding inductances from the RED/YEL - RED/WHT vs the RED/YEL - RED leads, even though you only measured a 7 ohm difference. After saying all this, If I'm right, and the trans has good HV insulation, my feeling is that it was meant for bridge rectification. You can current limit the 45 volt leads and extract 45 volts to two reversed diodes connected to a cap resistor filter for your fixed bias pot. Please excuse my jumping in so quickly without taking a careful look at the info provided. Marvino