SF Twin Reverb power transformer: how hot is too hot?

[asfi]

I've got a 1972 Twin Reverb chassis on the bench. It's been re-capped and "blackfaced" and also has 1 ohm bias measurement resistors fitted. It's on the bench because one of the 6L6's failed in a bad way, toasting the bias sense resistor and also the hum balance resistors.

After repairing the damage, I've got a ground buzz issue I'm sorting out, and after a couple of hours I notice that the power transformer is running HOT. Too hot to grasp for more than a few seconds, and this is with no signal passing through the amp.

PT details: J022756 (code 606-2-03, so 1972), DC resistance measurements red to centre tap 12.0 and 12.9 ohms; bias tap to centre tap 4.2 ohms. Heater winding to plate winding, open circuit. Resistances measured with transformer disconnected. In-circuit I get about 450 VDC after the rectifiers, and that's with 122 volt mains.

The 6L6's are biased for approximately 20 watts idle.

Is there a way of determining if the transformer is faulty?

[Enzo]

Monitor your mains current draw. Is the amp drawing excessively? Is it working real hard at something that doesn't show?

[The Dude]

Anything is possible, but I doubt a transformer problem. Are all tubes biased the same? In other words, is it possible you checked one cathode and there's a different one that's drawing extra current? By "hum balance resistors" are you referring to the bias circuit, or are you talking about the pair of 100 ohm resistors for the filament artificial center tap? How much total current is the amp drawing at idle? Also, check to see that the amp is fused correctly. I've had guys throw a 20 amp car fuse in there after a failure to see if the amp will come back on.

[olddawg]

I've got a 1972 Twin Reverb chassis on the bench. It's been re-capped and "blackfaced" and also has 1 ohm bias measurement resistors fitted. It's on the bench because one of the 6L6's failed in a bad way, toasting the bias sense resistor and also the hum balance resistors.

After repairing the damage, I've got a ground buzz issue I'm sorting out, and after a couple of hours I notice that the power transformer is running HOT. Too hot to grasp for more than a few seconds, and this is with no signal passing through the amp.

PT details: J022756 (code 606-2-03, so 1972), DC resistance measurements red to centre tap 12.0 and 12.9 ohms; bias tap to centre tap 4.2 ohms. Heater winding to plate winding, open circuit. Resistances measured with transformer disconnected. In-circuit I get about 450 VDC after the rectifiers, and that's with 122 volt mains.

The 6L6's are biased for approximately 20 watts idle.

Is there a way of determining if the transformer is faulty?

Float the secondaries. Is it still hot? Is it drawing too much heater current? Are the filter caps leaking? Rectifiers ok? Could be lots of stuff stressing the transformer. If the transformer is hot with no load I might suspect it.

[asfi]

Belatedly to reply:

1) All secondary windings disconnected, I see primary current of 0.31 A. (For comparison, I measured the no load current of a brand-new Hammond 290EX [no, not the same transformer, but similar] and it was 0.28 A.)

2) Adding in the heater winding, the primary current is 0.54 A, and the heater current is 6.08 A.

3) Connecting the centre tap (to chassis gnd) and the bias tap only, the primary current is 0.56 A....no surprises there.

4) HT windings connected but amp in standby, the primary current is now 0.57 A.

5) And lastly, with the amp fully on (but no signal), the primary current is now 1.07 A, or a total power consumption of around 120 watts. Again, no surprises.

I came across this thread: Power transformer--how hot is too hot?

So I think it's safe to conclude that there isn't anything wrong with the PT. It's just really really warm.

[mozz]

8 preamp tubes?

[asfi]

Heater voltage is about 7% higher than nominal, so I'd expect the draw to be 6.75 A. So almost all of the discrepancy is the pilot light (250 mA).

[nevetslab]

The Fender Twin at nominal power tube bias of around 27-30mA each runs at around 125W AC Mains @ 120VAC. That's around 1.2-1.3A. I have found several Twins, both Silver Face as well as Black Face where their power transformer just heats up under that nominal idle current, where as you describe, you can't keep your hands on the transformer core. I've never found the source to that, but, you might unplug all the tubes and the hum balance resistors and see what the excitation current of the transfomer is. I'll have to look and see if I have that data tucked away in my database for the Fender Twin power xfmr. If it is abnormally high, you could unwire the transformer, remove it from the chassis and see if it still draws abnormal current, now free from the steel chassis that could be contributing to leakage. If you have a hipot tester, that would come in handy, in this regard.

[Shawnobi]

I've got an SF Pro Reverb. Essentially a half power twin. It's an Ultralinear with the ridiculously big tranny. That thing gets hot. Never really bothered me too much too be honest.

[nevetslab]

Having just finished servicing the two Fender SF Twin Reverbs, I found their AC mains current was running around 1.4A/145W @ 120VAC. After leaving one of them on for about an hour, the power transformer core was as you described...you couldn't keep your hands on it. Just the outside core temperature was around 51 deg C. When I get back to the shop tomorrow, I'll leave one on for a few hours, and see what it is then. If I really wanted to dig into it, I could set up to measure the temperature rise of the windings and see just how hot this is getting internally. I know they are Class B insulation system (130 deg C), and pretty sure there's a thermal fuse inside. I don't know if Fender ever had used thermal breakers within the windings...those will re-set after the unit cools down, which takes well over an hour if using 30 deg C differential breakers.

[nevetslab]

I taped a thermocouple with white grease to the PT core, and at 81 deg C I shut it down, knowing the inside temperature is considerably hotter. Now, granted I had the thermocouple taped on the side nearest the power tubes, unable to get my hands into the front cavity with any success fumbling with gaffer's tape and slipping in the thermocouple. The amp was running at idle in Operate mode (not S/B mode). AC mains was 120VAC, 1.5A/143W. Power tubes were in the 33-35mA range, with plate voltage at 495VDC.

I had intended on letting this amp run for around 4 hrs, but upon seeing it at 81 deg C in an hour and a half, I got cold feet and shut it off. DEFINITELY hot enough to burn your hands. I suppose I should at least do this one more time, moving the thermocouple to the opposite side, if I can successfully get it taped down, so there's not the local heat source next to it from the four power tubes. There's barely a 1/2" gap between the 6L6's (V9 & V10) and the power xfmr.

[Jazz P Bass]

See how hot it gets if you crank back the bias.

That amp shouldn't be pulling 143 watts at idle.

[nevetslab]

This one must be a SF 135W Twin Reverb, since the plate supply voltage is in the 500VDC range. The SF 100W Twin is around 445VDC plate supply voltage. Of course CBS Fender altered the bias circuit so you can't adjust the bias level without tweaking a resistor. Bias Balance is what they give you, and it was never modified to have an added bias LEVEL pot (or change the bias BALANCE to bias LEVEL). Well, I still have both in the shop, and can get them back up on the bench to revise before they go back to inventory.

[Shawnobi]

This one must be a SF 135W Twin Reverb, since the plate supply voltage is in the 500VDC range. The SF 100W Twin is around 445VDC plate supply voltage. Of course CBS Fender altered the bias circuit so you can't adjust the bias level without tweaking a resistor. Bias Balance is what they give you, and it was never modified to have an added bias LEVEL pot (or change the bias BALANCE to bias LEVEL). Well, I still have both in the shop, and can get them back up on the bench to revise before they go back to inventory.

Yeah, as I said my UL gets too hot to touch. Have it running for hours at rehearsal and never worry about it. Though our new room has a nice big fan that I stick behind it. I think it's just the way they are.

[nevetslab]

Well, one of the two SF Twins was the 135W model. I had them mixed up, though the 100W model WAS running 143W/1.5A @ 120V. I pulled the chassis again on that, lowered the source resistor from the first filter stage of the bias circuit, raising the neg bias voltage, and adjusted the plate current to 32-33mA, Plate voltage 444VDC, AC Mains pulling 1.3A/126W @ 120VAC.

The 135W SF Twin was also pulling 148W @ 120VAC, it having the larger power xfmr, and the Ultralinear O/T (has screen taps). I lowered the plate current on it to 27mA, 500VDC plate, puling 1.34A/130W @ 120VAC.

I just re-started the 100W Twin running at idle, with the thermocouple this time inserted between the copper flux band and the core, and no longer being influenced by the heat of the power tubes. That may have been contributing to the temp, as when I had turned the amp off, the thermocouple environment dropped quite fast...faster than I recall a transformer core drop (though it's been a good 15 yrs or more since I've done any of these measurements). This amp has been running for 40 minutes, started at 28 deg C, and now up to 48 deg C and climbing. Our Guitar Dept staff will be coming by in another hour or two to pick up all the gear that's been serviced (preventative maintenance), so I'll have further data coming.

Update: After 75 minutes running the 100W SF Twin @ idle, xfmr core temp reached 60 deg C. Guitar Dept staff showed up early to pick up the gear, so that's as far as I got. I haven't plotted it, but I'd say the temp rise slope is lower from dropping the plate current, reflected in lower AC mains current flowing thru the P/T. Regardless, as Shawnobi indicated, his amps has been running reliably. I've replaced many Re-issue Deluxe Reverb power xfmrs, but never a Twin Reverb power xfmr. Just one of these things that make ya go hmmmmmm..............?