Power transformer- how hot is too hot?

[octal]

I am using a mystery transformer in an amp build of mine. After about an hour idle time, I've got about 135deg (F) on the bell of the transformer and 145deg on the core lams, measured with an IR thermometer.  I found an old thread where PRR said that 140deg was A-OK. At what temp SHOULD I be concerned?


N.B. this is a vintage transformer, so it may not have the benefits of newfangled improvements in winding insulation temp resistance and such.

[SILVERGUN]

Here's an article I remember reading from the MM site which I take with a grain of salt because they are obviously trying to sell transformers.
BUT, a hot transformer IS an angry xfmr and you really cant predict how long it will last.
If this is a stage amp, I'd replace it without question.....if it's the amp you use to play for your brother in law in the basement once a month, I'd ride it out.

http://www.mercurymagnetics.com/pages/mainframe.htm

Also, check and post your filament voltage....if you are overloading that winding your voltage will be much lower than the rated 6.3VAC

[HotBluePlates]

Corrected MM Link

[octal]

Thanks! I'm familiar with that MM piece, it came up when I was googling transformer temperatures-  but I found contradictory info in an old post from PRR- that's why I'm asking.
Since PRR didn't make any absurd claims like "hot power transformers are bad for your tone" I'm more inclined to believe him. :-)
Intuitively, this transformer feels much hotter than I'm used to experiencing in my builds. The amp IS going to a friend who plays out and does kind of long sets, so I'm worried.


Nathan

[jazbo8]

Do you have a surface thermometer or IR Laser thermometer? The finger-on-the-transformer "measurement" is not a good judge for the "safe" operating temperature. Hot water out of the faucet is typically 120F, while "normal" operating temperature for a power transformer can easily be 140-150F, and it is not considered "hot" until ~175F, with ~200F being the safe limit. Don't think you want to put your hand/finger on that...

[octal]

From the post I stared the thread with- "145 deg F on the core lams, measured with an IR thermometer"
So, yes, I do have one. :-)

What is MM's deal? They're saying "temperatures above 120 are considered hot." I guess they want to sell their goods.
You're saying not to worry until 175-200. That's quite a spread.

[jazbo8]

Doh! Skipped over it, anyway, try taking a some reading during the performance every hour or so (if possible) to see if the temperature rise - if it stablize at ~140-150F, then it should be fine.

[octal]

Thanks for your advice, will do.

[vibrolax]

The power transformer in my 1967 Vibrolux Reverb gets about that hot.  I was also uncomfortable with it being that way.  So I mounted a surplus 120V boxer fan on a homemade bracket that slips over the bottom rear panel of the cabinet.  The fan points up at the octal tubes and power transformer, and keeps everything very close to ambient temperature no matter how long or hard I play it.  It mounts/unmounts without tools in a few seconds, so I can easily move it between my Princeton and Vibrolux.  The bracket was made from a few scraps of steel furnace duct mounting straps.  It plugs into the amp's accessory outlet.  The straps are a little long for the Princeton, but it holds fine.

I am absolutely sure that there is nothing electrically wrong with either amp.
Jon

[trobbins]

Transformer may take longer than an hour to stabilise, so that advise is important.

Are you measuring those temps on all similar parts of the transformer - or were those points easier to access.

Are you able to test the transformer when there is no secondary side loading?   Perhaps you are operating the transformer primary near its maximim voltage.

[octal]

My first theory was that I was overtaxing the filament winding with 2X EL34s (3amps!)  and 3 12AXX preamp tubes. So, I floated the two EL34 filament terminals, wired them in series and powered them off of a 12V external SMPS. After an hour, the PT was still measuring 145deg, though it seemed to take longer to heat up. As a side note, I don't think EL34 filaments are designed to be wired in series... the filament voltages of the two tubes were a bit unequal.

I'm afraid to run the amp with no load on the secondaries per TROBBINS test. The B+ (unloaded) is a bit over the max rating of the first filter cap.

I'm thinking I'm just drawing too much current from the B+ winding. This is a cathode biased amp. I wonder if I convert to fixed bias, and dial back the idle current a good bit if I can get the temps at idle down to a manageable level.

[sluckey]

I like Steven Raichlen's method... 1 Mississippi, 2 Mississipi, 3 Mis.... Ouch! That's a hot grill!     

[Willabe]

 

[octal]

EZ-BAKE power transformer! Comes with bonus amplifier!

[Willabe]

Yeah, bake a cake for break time.      

[trobbins]

Disconnect the high voltage secondary side loading at the transformer, not the B+.  That could be as simple  as removing a rectifier valve. Remove the valves to remove heater loading.

[octal]

 So, I disconnected the HV winding and let it sit on for an hour (output tubes were also unplugged.)  Transformer was nice and cool, no need to even take a measurement.  So, I guess I'm just drawing too much current from the HV winding. (Given that I tried powering the output tube filaments with an external supply and the transformer got just as hot, it seems it must be the HV winding)
 

[trobbins]

Can you lower the output stage bias current easily without causing problems like B+ getting too high?

[shooter]

I guess I'm just drawing too much current

Are you comfortable measuring current?  If so it's pretty easy to know your *real* current draw.
   

[octal]

Oh, sure, I'm comfortable measuring current. It's just that since it's a mystery surplus power transformer, I don't know what its current ratings are, so knowing that I'm drawing, say,  100mA from the B+ doesn't tell me too much. Transformer could be rated for 85mA or 150mA.

Backed down the bias by increasing the cathode resistor and the transformer seems much happier, I had temps in the 125 deg range after an hour. I'll need to keep checking and see if the temp stabilizes at that level or keeps climbing.

[sluckey]

since it's a mystery surplus power transformer

And there it is! You gotta be willing to live on the bleeding edge when using mystery transformers.    I know because I do it all the time. 

Get a fan.

[shooter]

Get a fan

Yup, the last mystery one I used, you couldn't get to 3 Mississippi without a doctors visit, it's been in the public domain for 2 years, last I heard

[lkrasner]

A fan will help a LOT, but even proper ventilation will too. I don't know what the enclosure is like, but hot air rises, so if you use that to your advantage, it can create some flow if you put vents so that cool air pulls in the bottom, heats up and flows out the top. A tight enclosure with bad ventilation heats up like crazy