I'm measuring plate voltage on a PP amp with EL84, and when turning on Vak jumps up to 380V before falling and hovering around 300V. According to the datasheet, 300V is the maximum value.
I'm concerned that the tubes might be quickly worn out, or start arcing. However, when searching around, it seems others are (for example) using plate voltages in excess of 400V on other kinds of tubes with the same voltage ratings (12AX7 for instance) 😱
I understand that some brands might be more forgiving, but aren't maximum spec maximum for a reason? Is it common to drive tubes so hard? What rules of thumb are there, and when should I worry?
I'm concerned that the tubes might be quickly worn out, or start arcing. However, when searching around, it seems others are (for example) using plate voltages in excess of 400V on other kinds of tubes with the same voltage ratings (12AX7 for instance) 😱
I understand that some brands might be more forgiving, but aren't maximum spec maximum for a reason? Is it common to drive tubes so hard? What rules of thumb are there, and when should I worry?
I would not worry.
The maximum anode (plate) voltage is 550 volts as long as the valve is not drawing any current. You see 380 volts when the valves are not drawing current at first switch on. As they heat up and draw current then that pulls your HT (B+) down to 300 volts. So all within range.
Better to be aware of the dissipation each EL84 has. (Va - Vk x kA cathode current) Do not exceed 14 watts. Some amplifiers do push this and the closer they get to the 14 watts the shorter the life.
You are correct, some types are more tolerant of abuse than others though.
The maximum anode (plate) voltage is 550 volts as long as the valve is not drawing any current. You see 380 volts when the valves are not drawing current at first switch on. As they heat up and draw current then that pulls your HT (B+) down to 300 volts. So all within range.
Better to be aware of the dissipation each EL84 has. (Va - Vk x kA cathode current) Do not exceed 14 watts. Some amplifiers do push this and the closer they get to the 14 watts the shorter the life.
You are correct, some types are more tolerant of abuse than others though.
Last edited:
If the electrolytic capacitors in the B+ are rated at 400V or more, you have nothing to worry about.
After the amplifier is warmed up, and you turn the volume up on a large music signal, each plate in the push pull is going to swing at least far as +/- 200V (each plate starts at +300V, then one plate goes down by 200V to +100V, while the other plate goes up by 200V to +500V; then they swap the up and down directions).
The plates will go from +100V to +500V on the signal peaks.
400V peak into a 10K plate to plate primary is 8 Watts power.
400V peak into a 8k plate to plate primary is 10 Watts power.
300V at 40mA in the plate is 12 Watts (the maximum plate dissipation).
300V at 6.6mA in the screen is 2 Watts (the maximum screen dissipation).
The filament is 6.3V at 0.76A. That is 4.79 Watts.
The total heat is 18.79 Watts in that little glass envelope.
You may want to run the plate and screen dissipation at less than 14 Watts total.
Some tube grids might warp. The EL84 is reasonably linear, but not if the grids warp.
Your Mileage May Vary.
Have fun listening.
After the amplifier is warmed up, and you turn the volume up on a large music signal, each plate in the push pull is going to swing at least far as +/- 200V (each plate starts at +300V, then one plate goes down by 200V to +100V, while the other plate goes up by 200V to +500V; then they swap the up and down directions).
The plates will go from +100V to +500V on the signal peaks.
400V peak into a 10K plate to plate primary is 8 Watts power.
400V peak into a 8k plate to plate primary is 10 Watts power.
300V at 40mA in the plate is 12 Watts (the maximum plate dissipation).
300V at 6.6mA in the screen is 2 Watts (the maximum screen dissipation).
The filament is 6.3V at 0.76A. That is 4.79 Watts.
The total heat is 18.79 Watts in that little glass envelope.
You may want to run the plate and screen dissipation at less than 14 Watts total.
Some tube grids might warp. The EL84 is reasonably linear, but not if the grids warp.
Your Mileage May Vary.
Have fun listening.
Last edited:
If a tube/valve is run above its DESIGNED operating conditions (maximum ) of course it shortens its life .
They say "toffs are careless " in the UK meaning well I can afford to do this or that as I am well off .
I always stick to design parameters --yes you have military/government versions etc which will stand higher voltages etc but running a tube.valve over its designed and manufactured tolerances will shorten its life.
Nowadays many ignore this wanting that extra bit of gain/drive ,anyway have a read of and look at the graphs in this simple explanation of their characteristics -
https://www.bvwtm.org.uk/articles/Valves and Valve Testing.pdf
They say "toffs are careless " in the UK meaning well I can afford to do this or that as I am well off .
I always stick to design parameters --yes you have military/government versions etc which will stand higher voltages etc but running a tube.valve over its designed and manufactured tolerances will shorten its life.
Nowadays many ignore this wanting that extra bit of gain/drive ,anyway have a read of and look at the graphs in this simple explanation of their characteristics -
https://www.bvwtm.org.uk/articles/Valves and Valve Testing.pdf
Thank you for the input!
The voltages mentioned are at 220V AC input, which is what the device is rated for. However, where I live, the mains are ~247V. I'm wondering if a 12% increase in B+ will be detrimental to the tubes. Filament voltage I can control with a few power resistors, but B+ is another matter..
B+ caps are rated 450V. What I understand from this, then, is that higher voltages will affect tube life (how much?) - however, it's the current that kills.
@duncan2 Thank you for the link, I've only glanced at it so far, but seems informative and well put together 🙂
The voltages mentioned are at 220V AC input, which is what the device is rated for. However, where I live, the mains are ~247V. I'm wondering if a 12% increase in B+ will be detrimental to the tubes. Filament voltage I can control with a few power resistors, but B+ is another matter..
B+ caps are rated 450V. What I understand from this, then, is that higher voltages will affect tube life (how much?) - however, it's the current that kills.
@duncan2 Thank you for the link, I've only glanced at it so far, but seems informative and well put together 🙂
Guitar amps routinly run EL84s up near 350V. They don't last as long, sure, but they generally give a few thousand hours all the same. 300V is not so bad.
Hey, this company claims to be running 6BQ5's at 700 volts for 35 watts per channel, although I can't find any info mentioning that number today.
The most I have run through these types of tubes is 430 volts on the plates with 300 volts on the screens. Only JJ EL84's or old 6BQ5's would put up with that abuse though.
Excess plate voltage is generally not what kills a tube. It's usually excess plate power dissipation, or worse excess screen grid dissipation often caused by excess screen grid voltage.
EL84 max power for HiFi Amp
Music Reference RM-10 Vacuum Tube Amplifer
The most I have run through these types of tubes is 430 volts on the plates with 300 volts on the screens. Only JJ EL84's or old 6BQ5's would put up with that abuse though.
Excess plate voltage is generally not what kills a tube. It's usually excess plate power dissipation, or worse excess screen grid dissipation often caused by excess screen grid voltage.
EL84 max power for HiFi Amp
Music Reference RM-10 Vacuum Tube Amplifer
Don’t some tubes have more of a tendency for the bias to run away on elevated plate supplies? That’s what I would be afraid of - the IQ being in dissipation spec initially and going red plate after several hours (when you’re not looking, of course).
The voltage rating of electrolytic capacitors that are employed for B+ filtering, has nothing to do with the life of an output tube.
B+ electrolytic caps that are exposed to more voltage than their maximum voltage rating will experience:
Shorter life time
Possible failure
It depends on many things: time of exposure, ambient temperature, surge voltage rating of the cap if any, and the material & build quality of the capacitor.
If you take the RPM limiter out of a 1987 Volkswagen Jetta, you can rev the engine to 8,500 RPM . . .
The Life of Your Engine May Vary.
B+ electrolytic caps that are exposed to more voltage than their maximum voltage rating will experience:
Shorter life time
Possible failure
It depends on many things: time of exposure, ambient temperature, surge voltage rating of the cap if any, and the material & build quality of the capacitor.
If you take the RPM limiter out of a 1987 Volkswagen Jetta, you can rev the engine to 8,500 RPM . . .
The Life of Your Engine May Vary.
Last edited:
I looked at some original manufacturers EL84 data sheets, and some modern EL84 data sheets.
12 Watts Plate
2 Watts Screen
So Triode Wire it and dissipate 14 Watts if you want (I would not).
How do you know the proportions of plate and screen dissipation, unless you measure both plate and screen currents, times their different voltages.
Or, dissipate 14 Watts in the plate for Pentode or UL operation (I would not).
And 2 Watts in the Screen, or what ???
Your Tube Life May Vary.
At 300V and 12 Watts plate the current is 40mA.
At 300V and 14 Watts plate the current is 47mA.
The signal power dB difference of 40 mA versus 47 mA is 1.4 dB.
An old Hi Fi rule:
if you are purchasing an amplifier to replace your old amplifier, simply because you want more power (and no other reason), then be sure to purchase an amplifier that has at least 3dB more power.
1.4dB? . . . 3dB?
This is the Hi Fi thread, not a Guitar Amp thread.
Guitar Amplifiers follow different rules (and more often break more rules).
The above are just my opinions.
12 Watts Plate
2 Watts Screen
So Triode Wire it and dissipate 14 Watts if you want (I would not).
How do you know the proportions of plate and screen dissipation, unless you measure both plate and screen currents, times their different voltages.
Or, dissipate 14 Watts in the plate for Pentode or UL operation (I would not).
And 2 Watts in the Screen, or what ???
Your Tube Life May Vary.
At 300V and 12 Watts plate the current is 40mA.
At 300V and 14 Watts plate the current is 47mA.
The signal power dB difference of 40 mA versus 47 mA is 1.4 dB.
An old Hi Fi rule:
if you are purchasing an amplifier to replace your old amplifier, simply because you want more power (and no other reason), then be sure to purchase an amplifier that has at least 3dB more power.
1.4dB? . . . 3dB?
This is the Hi Fi thread, not a Guitar Amp thread.
Guitar Amplifiers follow different rules (and more often break more rules).
The above are just my opinions.
Last edited:
Thank you for your feedback. I'll build an autotransformer and play it safe.
On a side note, talking about all of this - I see solid state rectifiers with tube sockets to use as replacements. They state they will increase the B+ of about 10-20%. This, combined with instant on, wouldn't that be tough for the tubes?
On a side note, talking about all of this - I see solid state rectifiers with tube sockets to use as replacements. They state they will increase the B+ of about 10-20%. This, combined with instant on, wouldn't that be tough for the tubes?
And the capacitors .
I have several original commercial conversions that made use of an IO tube base that included high wattage series resistors , this also applied to some old CRT televisions with series heaters although it seems now this is ignored by many .
I have several original commercial conversions that made use of an IO tube base that included high wattage series resistors , this also applied to some old CRT televisions with series heaters although it seems now this is ignored by many .
> wouldn't that be tough for the tubes?
Bah. That's why we socket them. EL84 is about the *cheapest* power tube ever made. Use 'em, abuse 'em, replace them. But as said, even dedicated EL84 abusers rarely go broke from replacement costs.
And know your rating systems. 300V 12W would be "design centre", a vague middle of the road target. Later US Design Max ratings were 330V and 14W.
And know that ratings are also Marketing tools. If you want a "big" EL84, they would $ell you the 7189, same thing rated for 400V-440V. Was it really different inside? Or was it maybe just a better day in the EL84 factory? (Sadly true 7189 are vanished, but guitar-market EL84 come close, because SO many g-amps over-volted their EL84s. A short life but a merry life, like their users.
Bah. That's why we socket them. EL84 is about the *cheapest* power tube ever made. Use 'em, abuse 'em, replace them. But as said, even dedicated EL84 abusers rarely go broke from replacement costs.
And know your rating systems. 300V 12W would be "design centre", a vague middle of the road target. Later US Design Max ratings were 330V and 14W.
And know that ratings are also Marketing tools. If you want a "big" EL84, they would $ell you the 7189, same thing rated for 400V-440V. Was it really different inside? Or was it maybe just a better day in the EL84 factory? (Sadly true 7189 are vanished, but guitar-market EL84 come close, because SO many g-amps over-volted their EL84s. A short life but a merry life, like their users.
Baah, what to do.. I really would prefer to pop in a couple of power resistors for the filament and then just connect it to my 247v mains. Everything would be so much easier. But I still have no idea how much it would have to say for tube lifetime.
I'm listening at low volumes, so 15W is far away from my consumption. It's PP, but I think the amp itself is 7.5W per channel max output. Not sure how much is lost on the way from the tube to the banana jacks?
I'm listening at low volumes, so 15W is far away from my consumption. It's PP, but I think the amp itself is 7.5W per channel max output. Not sure how much is lost on the way from the tube to the banana jacks?
You have not told us which amplifier it is.
So things to consider,
You can now work out the dissipation of the EL84s in your amplifier.
Measure the voltage on the EL84 cathodes, pin 3, then by ohms law calculate the current each one takes. (Example if the cathode resistor is 270Ω and the voltage is 12 volts, then the current is 44mA. That is 12 divided by 270.)
If you multiply that by the HT voltage minus the cathode voltage you get the power dissipation in Watts. (Example HT is 300 - 12 = 288. 288 x 0.044 = 12.7 watts.) Below 14w is Okay above is pushing it.
Next consider the mains transformer. You will be over running it by 27 volts, it will run hotter and may fail if marginal. A not uncommon problem in the UK on cheap 220 volt transformers.
A bucking transformer is a simple solution to drop the mains down to 220 volts. Search here Bucking Xfmrs That will get the heater voltage and HT voltage in the designed band...
Last I assume the amp runs in class A most of the time. In that case it essentially uses the same power all the time. Reducing the volume does not reduce the load or stress on the valves. They are 'wearing out' at the same rate all the time it is on. Fact is you do need to replace them.
So things to consider,
You can now work out the dissipation of the EL84s in your amplifier.
Measure the voltage on the EL84 cathodes, pin 3, then by ohms law calculate the current each one takes. (Example if the cathode resistor is 270Ω and the voltage is 12 volts, then the current is 44mA. That is 12 divided by 270.)
If you multiply that by the HT voltage minus the cathode voltage you get the power dissipation in Watts. (Example HT is 300 - 12 = 288. 288 x 0.044 = 12.7 watts.) Below 14w is Okay above is pushing it.
Next consider the mains transformer. You will be over running it by 27 volts, it will run hotter and may fail if marginal. A not uncommon problem in the UK on cheap 220 volt transformers.
A bucking transformer is a simple solution to drop the mains down to 220 volts. Search here Bucking Xfmrs That will get the heater voltage and HT voltage in the designed band...
Last I assume the amp runs in class A most of the time. In that case it essentially uses the same power all the time. Reducing the volume does not reduce the load or stress on the valves. They are 'wearing out' at the same rate all the time it is on. Fact is you do need to replace them.
SY's Red Light District Amp with 550 volts on plates of EL84
On SYs site he said he made a version of the Red Light District Amp that ran the plates at 550 volts and got 25 watts output.
SYclotron Audio | The Red Light District, p4
Steve
On SYs site he said he made a version of the Red Light District Amp that ran the plates at 550 volts and got 25 watts output.
SYclotron Audio | The Red Light District, p4
Steve
550Volts ??--- and the guaranteed lifetime in hours ?
No wonder he says---"he likes a lot of ventilation".
No wonder he says---"he likes a lot of ventilation".
Sy said,
"Anecdotally, I just changed my output tubes for the first time in ten years, not because the old ones failed, but just as a matter of routine maintenance."
Steve
"Anecdotally, I just changed my output tubes for the first time in ten years, not because the old ones failed, but just as a matter of routine maintenance."
Steve
Radio stations used to replace the high power RF and High Power modulator tubes, after a specific number of hours.
They could not afford to go off the air during scheduled airtime.
Some of those tubes were sent in to be re-built.
I have some tubes that very well may have been taken out of service, when they had reached the prescribed number hours.
They could not afford to go off the air during scheduled airtime.
Some of those tubes were sent in to be re-built.
I have some tubes that very well may have been taken out of service, when they had reached the prescribed number hours.