EL-84 Specs ?

[rreiser]

I'm curious what are the max specs an EL-84 can run in a guitar amp and not wear out quickly ? What is typical in the industry ? Max plate voltage and milliamps per tube ( bias ) ? Thanks, Russ

[RobRoyMcCoy]

You can run some robust EL84's at up to 450V on the anode but the idle plate current must be low to allow for this.

If you have a lower plate voltage you can run more current through them at idle. Also the reflected load, in conjunction with the electrical conditions on all of the electrodes (anode, screen, cathode, grid) will determine if the operation of the tube is safe once AC signal is applied and amplified.

Unfortunately you need to take into account all of these factors to determine if tubes are running safely.

I typically measure the idle AC and DC voltages and current on all of the electrodes (obviously some will be zero at idle), the same measurements at maximum clean headroom and the same at maximum power.

This way you can measure if the screens (for instance) are over-dissipating at maximum power or even at clipping.

As a basic rule of thumb at 250VDC on the anode and 200VDC on the screens an EL84 can safely handle up to 50mA of quiescent current and with the right load will run in class A. As the voltage increases the current should be reduced to compensate, taking into account that as the tube is biased into class AB it will cut-off for a portion of its duty cycle and will therefore "cool off" for a short period of its operating cycle. I suspect I am going into places you really didn't intend me to go so I will just shut up now.:o
Rob

[rreiser]

Hi Rob , Thanks for the info. I had two new EL-84 tubes fail recently and the plate voltage was 330 and the ma per tube was about 38 ma. The seller of the tubes feels that the tubes dissipation ( whats dissipation ? ) was too high and that the amp probably caused the problem. He is replacing the two tubes so I have no beef with him but I am trying to understand why I never had other NOS tubes fail in this amp. There was no plate glow on the tubes. I have no technical knowledge so that's why I'm trying to get some more data.

[RobRoyMcCoy]

The seller is correct unfortunately.

At 330VDC on the anode and with 38mA plate current that works out to be 12.5W of plate dissipation per tube. The maximum specified is 12W. Unless the amplifer is class A the plate current (and moreso the screen current) will rise as more and more signal is applied. The anode voltage will remain somewhat constant therefore the dissipation in the tube will rise further. Note the 38mA is probably cathode current but it is close enough.

For an EL84 to run at 100% dissipation safely it should be in class A and around 250VDC on the anode maximum. Aiken amps has a really good description on his website.

If your plate voltage is 330VDC it it highly likely that your amp is a class AB design and the ideal safe plate current would be between 19mA and 26mA.

Dissipation is simply how much heat the plate can cope with safely. An EL84 plate can operate normally with 12W of heat passing through the plate without failing. Red-plating is when the plate has more heat than it can safely handle. The heat is caused by the electron flow through the tube.

Rob

[groutfulone]

And yet my amp runs El84's at 350vdc on the plates at 50ma with no problems. I have run Mullards, 50's Mazdas, Dutch Phillips, Sylvania's, RCA's, and new TAD's like this. No problems at all.

It doesn't make sense to me, because it is over 100% dissipation. I used to worry, but it has been fine for over two years.

[Jade]

And yet my amp runs El84's at 350vdc on the plates at 50ma with no problems. I have run Mullards, 50's Mazdas, Dutch Phillips, Sylvania's, RCA's, and new TAD's like this. No problems at all.

It doesn't make sense to me, because it is over 100% dissipation. I used to worry, but it has been fine for over two years.

I've seen amps with el84's, and other tubes that have been run this way with the idle dissipation above what "Should" be the tubes limit.

IMO, the real killer in these situations, is Screen Current. The plate structure of any tube is it's largest element, and pretty hardy, the screens on the other hand, are small, and very fragile in comparison. Increasing the value of the screen grid resistor, or the power supply dropping resistor in the screen circuit COULD alleviate enough of the stress on on the OP's tubes to let it run in the condition it's in.

The best thing to do though would be simply to re bias the amp to get the idle current down to a more reasonable level. To my ears, most amps sound much better biased down around the 60-65% mark. Many times when you go above that, the tone just get's kind of mushy, and undefined.

Jade

[rreiser]

This is great data . I did have the power reduced to 310 on the plate and 33 ma per tube as the builder said he is running his amps cooler these days. My amp is two years old.. It seems to sound fine and I put in a pair of Siemens E84L tubes that used to go into plate glow ( so I never used them) and now they work fine in the amp. I guess the point that hung me up is why my other tubes didn't fail but the new one's did. Any ways, I'm glad the seller was accurate and not falsely blaming the amp voltages when it could be poor tube quality.

[David Robrecht]

Just to chime in here, Russ' amp is a Class A-ish design and cathode biased. You do need to subtract the cathode voltage from the measured plate voltage to determine the effective plate voltage the tube is "seeing". I don't see that addressed here. If the current reading was determined from the drop across the cathode resistor there is going to be some screen current included in that measurement. His bias condition "may" be closer to acceptable than we think because we don't have all the info.
Dave

[RobRoyMcCoy]

True. If we drop the voltage by say 20v for the cathode (a lot I know) and allow say 4mA of screen current (SAME) you are left with 310*.034=10.5W so if it is running in or close to Class A it could be OK.

Careful measurement of all electrodes at idle and maximum clean power I say would reveal if it was OK.
:aok

[MartyStrat54]

I think a lot of amps that use current production tubes are always running on thin ice. EL84's by nature get very hot. After all, they are in small envelopes and are usually placed closely together (like my DSL401's. I have two). The earlier version of this amp ran some ungodly bias and it got hot enough to flow solder off of the bridge rectifier. It really makes me wonder why an amp would be designed this way. Granted, my 1999 401 sounded so much better than my 2005 and 2006 models.

Sometimes I wish they still made the 7189. It was a big EL84 and could handle 400 volts on the plate. That's what I would run in my 401's. I love NOS tubes and I have like 400 12AX7 type tubes among others, but they want $80 a piece for 7189. $640 to tube two amps! Wow. I guess I'll stick with the hot EL84's.

MARTY:dude

[JJman]

Seems to me that most all guitar amps designed for el84s are designed to idle OVER the max "recommended" wattage. EL84s get no respect in that department.

Could be the whole "rock and roll" thing.:dude

[rockon1]

My Palomino V32 runs at 13 watts idle dissapation . This is substracting the cathode voltage. It eats most tubes quickly. Ive settled on EL84M's. Perhaps not the most toneful but at least they havent failed and redplated on me like the JJ's,Sovtek 6BQ5's,unfortunately RCA's . One day maybe I'll swap out the cathode resistor but as long as it works with the M's probably not. Bob

[groutfulone]

Is the screen grid voltage just pin 9 to ground using my DMM and the screen grid resistor is the r connected to pin 9?