Why are my tubes running so hot?

[joshdfrazier]

Just finished another ax84 50w build. Having some issues with redplating tubes and low output. Here's a quick link to the schematic: http://www.ax84.com/static/corepoweramp ... ematic.pdf

I wanted the amp to run 6L6, so I switched the grid resistors to 470r. Also, for L1 (choke) I use a 100r 10watt resistor, instead of the 120/25w. It's what I had on hand. Nearly everything else besides the tone stack, follows the schematic exactly. The bias circuit is unchanged.

Currently my bias voltage is -30. Plates are at 396, screens at 396 and idle current draw of each tube is only 27ma. Right now the tubes are very hot. If I bias them any hotter, they will redplate. Most 6l6's see voltages well above 400 and draw much more current, yet remain relatively cool. Any idea what gives? Is my bias circuit totally unfit for 6l6 utilization?

[Stevem]

If your biasing for class A then your plate voltage is 30 volts too high, your screen is over 100 volts too high, the cathode resistor needs to be 250 ohms and the OT needs to be 8 to 9K!

[joshdfrazier]

Stevem, this is a push pull fixed bias amp! The irons used are the ones recommended by the schematic.

[cbass]

Im confused Who mentioned class A?
What OT are you using?

[joshdfrazier]

OT is hammond 1650N. 4.3k primary. Not a class A amp, I think Steve misread something.

[cbass]

Well the I guess it begs the question Hvae you tried different output tubes?And how are you determing that the tubes are drawing 27mA apiece?

[joshdfrazier]

Yes, I've swapped the tubes. I have a pair of JJ's that I can get around 38-40ma of current from, but this is still not even 60% dissipation at idle, and still, running extremely hot. Measuring current across the 1 ohm resistor on the cathodes.

[cbass]

Hmm Make sure you don't have any leaky caps In the PI. SOmetimes the PI tube its self can create some funky problems

And make sure that one ohm resistor reads one ohm

[VacuumVoodoo]

"Right now the tubes are very hot."
This does not say anything about how hot they really are.
It's not unusual to measure glass temperatures around 200°C.

[LOUDthud]

You may have an ultrasonic oscillation. Does R11 get hot with no signal and speaker connected? Do the tubes seem hot with the PI tube removed?

[joshdfrazier]

i'm aware that tubes get very hot under normal operating conditions, however my issue is that the tubes should not be on the verge of redplating at 27ma. maybe i should rephrase and say "extremely, abnormally hot".

r11 does not get hot. removing the pi tube doesn't affect the power tubes heat dissipation.

[pdf64]

my issue is that the tubes should not be on the verge of redplating at 27ma. maybe i should rephrase and say "extremely, abnormally hot".

By hot, do you mean calculated plate dissipation, the measured temperature of the glass envelope, or maybe even an infra red thermometer measurement of the plate?
Or that the plates appear to be visibly redplating under static conditions, even though the calculated static plate dissipation is only ~10 watts?
Pete

[Stevem]

Like Loudthud posted if the amp is oscillating at a frequency above what the speaker can reproduce than the output stage could be running hard even before you pump signal thru it.
If you hook up a meter set for ac volts across the speaker and read anything more than 12 mv of normal amp stuff (60 and 120hz noise) than the amp is singing all on its own!

Sorry, read your post very quick and though that the low 120 ohm resistor was on the cathode of the outputs.

[rp]

How's it sound? If it sounds good, dynamic with good treble extension no mud and not dark that's a good sign of no ultra sonic oscillation. If you have a good sized PT and it gets very hot just idling that's a bad sign. Those are pretty casual tests but just something to start with.

[Tillydog]

How are you determining the bias current?

Did you use the 1 ohm resistors in the schematic? If so, treble check that they are the correct value!