Hot Power Transformer Question

[newperson]

Hello,
How hot should/can a power transformer get without a load?  I have an older silverface Champ that I am looking at for a friend that has a newer replacement input power transformer.  It is getting rather hot without any tubes rather quickly.  It has a slightly higher high line output 730ACV (650ACV according to data sheet) but other than that it passes all of RG's tests (http://www.geofex.com/ampdbug/pwrtrans.htm).  No fuses are being blown.

So before I hook up some tubes, I was wondering if it could pass all the tests above and still have an internal short causing it to heat up?

[R.G.]

It's getting hot with no load?

[PRR]

Champ PT with NO tubes installed should be somewhat warm after an hour.

Hot fast means there is some hidden load, or the PT was made bad. An internal short between turns can do this.

Did this PT ever work or has it been hot from the start?

> hot with no load?

A standard "Champ ...without any tubes" is just a pilot lamp and maybe 200 ohms across the 6V winding. Tops 1 Watt, compared to nearly 40 Watts when working. Very-very nearly no load.

[newperson]

That is what I thought. 

There is not any load to it (other than the lamp).  No tubes are hooked up and it gets hot within two mins.  It is not blowing the fuse though.  The pilot light comes on and is steady bright.

The transformer is a newer brand replacement.  The amp is new to me so I don't know it's history very well. 

So it is likely to have an internal short of some kind then?
Thank you for the replies.

[R.G.]

It is possible that it's hooked up wrong and is self heating. One possibility is that it has multiple primary taps the wrong ones are connected to the power line. Or that a wire or two is shorting to chassis, or any of a number of other things.

But no, it should not be heating up noticeably with only the pilot lamp in two minutes.

It is possible it would blow fuses if the tubes were in it.

I hesistate to say this, and DO NOT do it unless you are sure you already know how to do it safely, but you could measure the incoming AC current. If it's over about 2-5% of the full load power AND you are sure the secondaries are unloaded, the transformer has a problem. But check for external wiring issues before tossing it out.

Are you sure it's the right transformer to be in the amp?

[newperson]

Hi,
The specs from the manufacturer say this is the correct transformer.  I went ahead and pulled out the old wiring job and rewired it.  After the rewire it is not getting warm at all with or without the tubes.  So I think that solved the issue.  I did not see anything strange in his wiring job though. 

The amp is quite and not putting out full volume.  I see the 6v6 cathode resistor looks like it is getting very hot though.  It tests fine when powered down.  I am going to get a new 470 Ohm and do a recap to see if that helps the issue.  Could the power cause the resistor to change value and cause a lower output?  The volume pot changes smoothly for the full range of the pot. 

I have been reading over RG's tube amp flowchart for low output.  The voltage look correct with power and the parts all test fine within the amp without power.

Again,
Thanks for the replies

[iccaros]

Those champs have a issue with the cap on the cathode of the 6v6 when you change out the power supply or even use newer rectifier tubes, they put  a 25volt cap and the voltage sat at around 20 volts.. A slight increase can cause the cap to burst, the first sign is low volume, then you may see the 6v6 red plate, or get oscillation and distortion at low volumes .. I have fixed three of these this month.. all the same problem, new Chinese rectifier tube, drops less voltage.. I have been replacing them with 120v caps

[newperson]

I replaced it already with a 50v one.  I swapped out all three of them.  I just had to order the 20/20/40 filter caps with the 470ohm cathode resistor.  All three caps did test fine out of circuit. 

I am testing this amp with my older tubes.  I have always heard there can be issues with newer rectifier tubes, so I have never bought any.  The older ones usually are less expensive anyway.

Thank you for the tip,

[PRR]

> 6v6 cathode resistor looks like it is getting very hot

Very basic test on any cathode-biased tube amp: Cathode Current. Easily measured with cathode voltage and cathode resistor value. Champs tend to run 30mA-50mA. Your real goal is to get current times plate-cathode voltage above 10 Watts but never over 14 Watts (tho some amps beat the 12W 6V6 with 17W and work many hours).

What is the resistance and voltage on G1? With _NO_ power, G1 MUST go to ground through less than 1Meg (preferably <500K, tho Fender may cheat that). With power, G1 must be within 1 volt of ground.

A missing/failed grid resistor lets the grid "float" to "any" voltage. The cathode will follow. Often the grid floats positive, cathode voltage follows, you have BIG voltage and BIG current. This tends to go on until B+ sag gets severe, or until something smokes.

Alternatively the coupling cap from the driver stage can leak B+ onto G1 and pull it high. Snipping the cap is an instant fix (but to get sound you need a new cap).

[newperson]

For the low output volume it seems that the volume stops rising at 2 on the dial.  The rest starts to introduce a bit of distortion.  I have replaced all the parts above with no change.  Also all the parts still tested fine when removed. 

I am now convinced that the Output transformer has gone partly bad.  The grid of the 6V6 has plenty of signal with the audio probe while the output is much lower than the input signal.  The 6V6 grid is much louder than the input source.  RG's notes say that it is possible for the transformer to go partly bad and still have weak bad sounding output.  I have never encountered this before.

I unscrewed the output transformer and see the chassis is burned from the transformer.

My question is... Am I missing something to test before ordering a new OT?

[PRR]

> convinced that the Output transformer has gone partly bad.

This is VERY un-common.

> The 6V6 grid is much louder than the input source.

Sure better be. About 1,000:1 times stronger. It is hard to tell 1,000:1 from 100:1 without careful measurements.

> RG's notes say that it is possible for the transformer to go partly bad and still have weak bad sounding output.

True.

> Am I missing something to test before ordering a new OT?

Buying (and shipping!) iron should be a last resort.

Also a puzzling coincidence that first you suspect power transformer, but then:

"pulled out the old wiring job and rewired it.  After the rewire it is not getting warm ....  I did not see anything strange in his wiring job"

You fixed one "major" problem without actually finding it. I don't know if you are gifted or just lucky. Now another "major" problem. Which sure could be a wiring fault.

> the chassis is burned from the transformer.

YEE-OUCH!! I've seen tar leak out, and there's always encrusted dust and rust, but "burned"??

I just do not know. Since it is only a Champ, a new OT won't break the bank. However it may not fix the problem, money wasted. (And I'd want to keep Silverface Fender iron as long as possible.)

If it wuz me, I'd putz around a while, looking for trouble and changing things that don't cost much. Had an Ampeg VT40 (BIG beast!) hanging around for 9 years before I nailed a unique fix. (Nobody needs the barn-blowing power and fragility of a VT40 these days; cut-back to half power means cheaper PT and much lower stress.)

[R.G.]

I'm with Paul - replacing a transformer is the last resort.

You can effectively test transformers. This amounts to using a resistance test to see that everything that's supposed to be connected is and everything that's not supposed to be isn't. Then you do the magic, secret transformer test with a battery and a neon bulb to find if any windings are internally shorted. If it passes both of those, you put a voltage on a winding and see if the voltage ratios are correct on the other windings. If it passes all that, only high voltage internal arcs under high temperatures or vibration remain as failures.

[newperson]

I agree with all of the above.  I will mess around with it a bit more.  I tested the 'simple test' method from RG's site, but I do not have a variac to do the advance test.  I will see if I can find the neon bulb method.  I do have a bag of those.  In the mean time here is an image of the heat burn from the OT on the chassis.


or

https://sites.google.com/site/partquestion/

[newperson]

So I swapped out the OT with a new one today and now this thing's signal is back to proper volumes. 

I have a few questions if someone would like to teach me something about polarity.  When I first put on the transformer it was getting a very loud oscillation.  I messed around with the wiring positions for a bit without any luck, swapped in different tubes one at a time, and finally swapped the secondary wires around (normal = black to ground and yellow to tip of jack).  This stopped the oscillation problem. 

Why would this fix or stop the problem?  Is it possible that the transformer has the wrong color leads on it (unlikely in my mind)?  Is it fine to leave the wires like this?  Why does this not happen if one swaps a speaker polarity?

How hot can tubes get without cause of concern?  I don't think this thing is near a issue but my meter has a temperature probe so sometimes I check out things.  The 6V6 is reaching up to 200-250F with the probe directly touching the glass.

And finally the new OT has a white wire that is unmarked in the factory's specifications.   Does this have something to do with ultra linear?  I have not read much about ultra linear.  It reads a high DC voltage on my meter. 

[PRR]

You need a Guitar Amp forum. There are many. Some are as polite as this crowd, but more focused on amplifiers than pedals.

> swapped the secondary wires around

No, swap the Primary wires. If you have multiple secondary taps 4/8/16, swapping them is messy; primary can usually be swapped no-fuss.

> a very loud oscillation

The amp has a connection from loudspeaker winding to a stage before the power stage, right? In a Silverface Champ, to the cathode of V1b (actually to its bias resistor). This is "Negative Feedback". It makes the amplifier less "raw", more controlled.

Imagine driving blindfolded. Pretty bad. Now watch where you go and make corrections. Much better. Negative feedback. Eye-hand feedback reduces the error.

Now I give you special glasses which flip your vision like a mirror. You do the wrong thing. wobble worse and worse, until you are ping-ponging off the curbs. That's positive feedback.

The OT can reverse polarity, and turn negative feedback into positive feedback. There does not seem to be a universal standard wire-code. In part because in more complicated amps you might return the NFB to some other point in other signal polarity. Generally you connect, power-up with finger on the OFF switch, and if "BUTT BUTT BUTP!" you shut-down and swap-wire.

So it was a bad OT. Do NOT run a Champ with signal withOUT a load! It WILL kick-up high voltages and fry the OT. If you get to the gig and it seems dead, don't strum harder, check the connections first. Load should be within 2:1 of rated load (up to 8 ohms on the usual 4-ohm tap).

[PRR]

Tape-up that mystery lead.

If it is indeed a UL tap, then this OT is not made for a Champ. A Champ is old good-radio parts dressed as a guitar amp, nuttin fancy. UL is about fancy hi-fi. The UL OT will work, fine, great, but without that trace of strain you get with a cheap radio OT.

> How hot can tubes get

Look it up: http://www.mif.pg.gda.pl/homepages/frank/vs.html

Actually, 6V6 does not have a rating. If you measure plate voltage and current and compute plate Watts, and it is under 14 Watts, it is fine. (If you don't measure every amp you build, you are supposed to aim for 12W to cover the occasional extreme tolerance; in DIY we measure what we build and can use the 14W number.) If the dissipation is within limit, the temperature will be fine.

The bigger/beefier tubes like 6550 say 225 or 250 Centigrade. IIRC, this is 482 degrees Fahrenheit. 6V6 _is_ a cheaper tube than 6550, but only in details. Interestingly 6AQ5 (a mini-6V6) also has a 250 deg C rating. Mil-spec 12AX7WA says 165 deg C (329F), but it would be hard to push 12AX7 that hot, and the mil-spec rating may be very conservative.

By the laws of radiation, if you are anywhere near the design dissipation, you will be VERY near the design max temperature. At 6W diss you may hit 150-200 C.

> The 6V6 is reaching up to 200-250F with the probe directly touching the glass.

That's "too low", but you are measuring it wrong. The glass is so thin that almost any probe will cool it significantly where it touches. Use a non-contact IR thermometer. I've seen them under $20. At that price you get wide-angle, you have to nearly-touch the glass or it will read a lot of "cold" room to either side. You will also learn that the glass is hottest where it is closest to the plate, and may be much cooler at top or at base. (One reason 807 6L6GB EL34 7027 and KT88 are built tall... keep the base cool.)

A power tube (not puny 12AX7) should boil spit (but don't: the glass may crack).

It should NOT set fire to a Kleenex set on top. (This may be the real limit: domestic tubes should not burn the house down even if paper or fabric falls inside the radio. Transmitter tubes are made with other glass and seals, and can run much hotter, but nobody puts drapes near kilo-watt transmitters.)

If you put the back of your hand on a full-hot 6550 (around 215 deg C) for 5 seconds, you will have a full 2nd-degree burn (blister) in the shape of the tube, with spots of 3rd-degree (scorch-flesh).

A power tube can run hot enough to bake cookies. Actually it won't hold heat when you drop dough on glass, but it should feel similarly hot at close range.

[newperson]

Thank you for the information and link.  I was just curious about the temp of tubes and could not find the information with Google.  I don't fear the temp it is setting at.  All the voltages are reading fine. 

I will swap the primary and see if the problem stays away.  As for the wire color code, I am reading the data sheet from the manufacture.  It states Red to power and Blue to plate.  In my mind I was thinking a reversed output would simply cause the speaker to push instead of pull, as it does with a DC battery test to find the polarity of a speaker.  Causing a loud oscillation took me by surprise.  As you can tell I am not fully informed about the functions of transformers. 

Anyway thank you again for the timely and informative replies.  Hopefully I will be able to keep the bad OT so I can unwind it to see how it was burned to still work with the low signal.