How hot should an EL84 run in a class A?

[g1]

Why not try increasing the cathode resistor to reduce the bias and heat. If it has too much of a negative impact on the tone you can put it back the way it was and suggest a fan.

[Randall]

Why not try increasing the cathode resistor to reduce the bias and heat. If it has too much of a negative impact on the tone you can put it back the way it was and suggest a fan.

Yes g-one, but I'd rather know if there is an actual problem that needs to be fixed before I look at re-engineering an amp that I don't know whether it usually runs this hot or not.

[Leo_Gnardo]

Yes g-one, but I'd rather know if there is an actual problem that needs to be fixed before I look at re-engineering an amp that I don't know whether it usually runs this hot or not.

With the Velo's I looked at back in the 20th century, g-one's was the best solution. I simply ran sine wave thru, tried increasing Rk until I saw & heard crossover distortion, then backed down a notch. Still ran hot but not so much & no customer complaints.

Does it have a history of frequently burning up output tubes? If its owner says yes then you better dial down the bias. As I suggested now you can select low Pc EL84's which should let the amp operate cooler with the same Rk in place if you're disinclined to experiment with changing Rk value. Which I don't consider re-engineering, not any more than twisting a bias adjustment pot on a fixed-bias amp. I keep a stock of low Pc JJ's tested & matched by Ruby just for Vox & other similar amps. Typically 16 thru 19, bottom of the scale. Amps that have fixed-bias, use the higher value Pc on them and adjust accordingly. Horses for courses as they say in Old Blighty.

[The Dude]

My $.02

Whether it was designed that way or not, it's not a good thing to be running an EL84 with that amount of dissipation. I'm with g-one on this one. Try increasing the cathode resistor and reign in your bias. Sell the customer on reliability. I can't imagine there's going to be a vast tonal difference unless you go too far the other way.

And, while we're discussing......never hint to a customer that his amp might not sound as good because of a problem you corrected. There is a placebo effect that goes with this. Tell him, for instance, the amp was biased too hot and you corrected it and that his tubes will last longer and be less prone to failure. Tell the truth, but don't set yourself up for a customer who thinks they hear something they don't.

[Dave H]

...I'd rather know if there is an actual problem that needs to be fixed

I don't think there is a problem. It's just the way it was designed. With 350V+ plate voltage and a 120R cathode resistor the plate dissipation will be 15W.

[pdf64]

'it was clearly the best sounding amp in the UK back in the '60s'

Yes, and five decades later I don't think it's been bettered!

For the Trace Elliot, how about dropping the VB+ a little with a sag resistor, eg 100 ohms >10 watts.
Perhaps sell it to the customer as being to simulate a tube rectifier, with a further benefit of keeping the EL84 alive longer.

I prefer this to increasing the cathode bias resistor much beyond say 270 ohms per EL84, as that tends to result in excessive crossover distortion / a more 'reedy' overdriven tone.

[Randall]

OK, I guess I'll try a 150 ohm 5 watt. I'll let you all know what comes of it. I'm finding that in my new digs down here in SW FL, I have no source for parts like I did in Maine. Everything takes a week plus shipping here.

[Leo_Gnardo]

OK, I guess I'll try a 150 ohm 5 watt. I'll let you all know what comes of it. I'm finding that in my new digs down here in SW FL, I have no source for parts like I did in Maine. Everything takes a week plus shipping here.

Gotta stock up Randall! Who knows there might even be an ol' fashioned parts dealer somewhere in the neighborhood, though they've become rare as rocking horse turds. If you run across any ham radio guys ask them. If you see a ham handle on someone's license plate, or extra antennas sticking out of a car, quiz them & maybe you'll get a good answer. FWIW I get a handful of whatever Rk values might come in handy plus whatever else - Mouser offers 2-day delivery at ground rates if you order over $200 - so I "pad" my orders of necessary parts with items that should be in stock just so I don't have to wait wait wait for that box to arrive. Comes in handy sometimes & I look like a hero (if only for a minute ) when I can turn around a repair in short time.

[Steve A.]

I have to believe 500 degrees means you are measuring the internal structure temp through the glass.

Presumably with an infrared thermostat- right? They can be very misleading... I would try a thermocouple placed directly on the tube surface (I'd tape it on with masking tape, not electrical tape!)

Steve Ahola

P.S. 350v is kinda high for EL84's- you might try lowering it a little bit with zeners (a lot of builders like 320vdc but that might be for fixed bias.) I was just checking for Vox schematics with voltages listed and the Cambridge was running the plates at 270vdc with a 120 ohm resistor for two EL84's.

[Randall]

OK, I put a 150R cathode resistor in it, and after noticing my line voltage is at 123 vac I put it on a variac and dialed it down to 120 vac. Now the math comes out out to 12.5 watts after figuring it at 90% of total dissipation. Still quite warm, but not as bad. I'm at the point now of believing this is just a hot running amp. I'm giving it back and suggesting the fan idea. Thanks for the help.

[Randall]

Wow! Here is the final outcome on this little pistol of an amp.

I had it all back together and decided to let it run for a while in my shop before I gave it back, because my gut was nagging at me a bit. When I checked on it later in the dark I could see the EL84s red plating just a bit, with the cathode Voltage up about 10%. After much measuring and gnashing of teeth only to find nothing out of tolerance, I hit upon the answer. With my probe I gently poked the top of the 220uF bias cap, and out oozed a bit of goo! I switched off the amp, and watched as the goo retreated back into the tiny hole in the top.

I pulled the cap and it measured fine, however the plastic wrap around the aluminum can was separated. Another small order to Mouser, and another week to receive said part. Installed today, and now it is running very solidly. It seems the bias cap was leaking and shorting the bias resistor as it warmed up, causing the tubes to draw more and more current, and get even hotter, which in turn cooked the guys tubes as well as open up a filter cap and burn the markings of two 4 watt PS resistors.

I must remember this, as this is the first time this has happened in my experience. Probably more common than I'd realized?

I shall leave the cathode resistor mod, as it is idling solidly at about 12.5 watts now.

[Jazz P Bass]

[QUOTE=Randall;365676]With my probe I gently poked the top of the 220uF bias cap, and out oozed a bit of goo! I switched off the amp, and watched as the goo retreated back into the tiny hole in the top. QUOTE]

A Rat Capacitor.