From old tube manuals
*100 Watt output: Ep = 800, Eg2=400, Eg1= -39, Rpp = 11.000 ohm
*70 Watt output: Ep =500, Eg2=400, Eg1= -36, Rpp = 4000 ohm
*60 Watt output Ep=500, Eg2 UL-connected, Eg1 = -44,5 Rpp = 7000 ohm
*55 Watt output Ep=425, Eg2 = ? Eg1=-38V Rpp = 3400 ohm
*45 Watt output Ep=375, Eg2= ? Eg1=-32V Rpp= 2800 ohm
But is it possible to acheive 100 Watt output (and 800 volt at plate) with current production 6CA7-EL34-KT77 ?
*100 Watt output: Ep = 800, Eg2=400, Eg1= -39, Rpp = 11.000 ohm
*70 Watt output: Ep =500, Eg2=400, Eg1= -36, Rpp = 4000 ohm
*60 Watt output Ep=500, Eg2 UL-connected, Eg1 = -44,5 Rpp = 7000 ohm
*55 Watt output Ep=425, Eg2 = ? Eg1=-38V Rpp = 3400 ohm
*45 Watt output Ep=375, Eg2= ? Eg1=-32V Rpp= 2800 ohm
But is it possible to acheive 100 Watt output (and 800 volt at plate) with current production 6CA7-EL34-KT77 ?
Last edited:
I would think it is possible, it is proven by Roger Modjeski with his RM-10. It uses two EL84 in push-pull pr. channel and gives 30-35W. As I remember, he used 650Va and 325Vg2 and also very high anode load (12-16.000 Ohm?).
I don’t think anybody recommends it anymore. Even back in the day, amplifiers designed like that had reliability issues. Todays EL34’s probably don’t stand a chance.
One reason that plate voltages over 450-500V were rarely used is that it's at that point that you exceed to voltage rating of a single electrolytic B+ filter capacitor, so you have to put capacitors in series, raising build cost and complexity. One of the exceptions was Bogen. They did run high plate voltages and screens about half the plate voltage in a number of their designs.
Why?
The 800V spec is so close to disaster that you MUST provide a regulated supply. In tube days, this regulator would be like another amplifier. It may have made sense in a transmitter as a driver for an even bigger stage, where regulated screen-feed might be needed for a 10KW final.
60W-70W/pair can be done using Marshall voltages and loadings.
Asking 100W/pair is like asking my mini-van to carry a ton of rocks. It's a real strain, short life.
Even today, EL34 are really cheap. Use four or more.
Suppose the output transformer that you use has the following characteristic:
Insertion loss of 1 dB (0.5 dB in the primary, and 0.5 dB in the secondary; due to the primary and secondary DCRs).
Then, with 100 Watt signal from the EL34 tube plates . . .
The 8 Ohm tap will have 79.4 watts out.
Design for 100 Watts, and get 80 Watts out.
Just my opinion.
Is this amplifier for Woodstock III, or to drive an MBL Radialstrahler speaker?
Insertion loss of 1 dB (0.5 dB in the primary, and 0.5 dB in the secondary; due to the primary and secondary DCRs).
Then, with 100 Watt signal from the EL34 tube plates . . .
The 8 Ohm tap will have 79.4 watts out.
Design for 100 Watts, and get 80 Watts out.
Just my opinion.
Is this amplifier for Woodstock III, or to drive an MBL Radialstrahler speaker?
100 watts (per channel even) for Woodstock IIII? That’s a little optimistic. Like the 100 watts from a couple of EL34. The 100 watt transformer is probably better than a dB of loss - half a dB or less is more like it. But you would need the full 800 volts to do it with two EL34’s. That would be approaching or even above 1000 at light load and things can go south in a hurry. A regulator at that power/voltage level would be more difficult than the amplifier itself.
What ultimately limits the power you can get are the power and output transformers. It’s where the expense is. And regardless of which tubes you use and how many you put in parallel you will only get so many watts out of a given transformer set. Choose the iron to be able to pass the power you need/want, then choose tubes that can drive this safely. If you wanted 100 watts with just two tubes there are better ones to use than EL34’s. Ones that can do it with 500-700 volts, and an output trafo that’s easier to wind than 11k a-a. If you go to four or even six tubes your list of possibilities gets much longer, and it can be done on lower voltage.
What ultimately limits the power you can get are the power and output transformers. It’s where the expense is. And regardless of which tubes you use and how many you put in parallel you will only get so many watts out of a given transformer set. Choose the iron to be able to pass the power you need/want, then choose tubes that can drive this safely. If you wanted 100 watts with just two tubes there are better ones to use than EL34’s. Ones that can do it with 500-700 volts, and an output trafo that’s easier to wind than 11k a-a. If you go to four or even six tubes your list of possibilities gets much longer, and it can be done on lower voltage.
You are correct. Not really Woodstock III.
But a 100 Watt amplifier into a midrange horn or tweeter horn that have 104dB sensitivity would put out 124dB.
Stay far away from that, or you might lose your hearing.
But a 100 Watt amplifier into a midrange horn or tweeter horn that have 104dB sensitivity would put out 124dB.
Stay far away from that, or you might lose your hearing.
Why stop there? HK257 (has a later part-number 4E27 widely used and readily available):
315 Watts push-pull Audio without grid current, at only 1500V.
Attachments
“Zero driving power” - ha ha. All tubes do that. The input impedance is capacitive, and the only loss is in the bias network/grid leak - and the plate load of the previous stage. Well, I’m sure there is a small real part to the input admittance (due to grid-plate coupling) but I’m sure you can pretty much ignore it in the overall power budget.
100 watts would be about right for a 2” compression driver by itself, active crossover. If I were doing this and only wanted two tubes per channel I’d use 26DQ5’s, 21LG6 or 26LW6‘s because they are what I have - and can handle 700 volt plate because they have an anode cap not a pin on the socket. At that lower B+ it would need 5 to 6k a-a which is common not an insane 11k. Would make one hell of a tweeter amp. 100 watt HF, 350 mid, and 800 bass (per side) probably wouldn’t do Woodstock III either, but would do any festival I might actually get a crack at doing. The only EL 34’s there might be in one of the muso’s amps.
It would put out around 114 dB of real useable output. Loudness is average power, not peak. Still loud, and if you add the woofer and midrange output it will be in the 120’s, maybe approaching 130. 30 dB less at 100 feet, which is a good average distance.
100 watts would be about right for a 2” compression driver by itself, active crossover. If I were doing this and only wanted two tubes per channel I’d use 26DQ5’s, 21LG6 or 26LW6‘s because they are what I have - and can handle 700 volt plate because they have an anode cap not a pin on the socket. At that lower B+ it would need 5 to 6k a-a which is common not an insane 11k. Would make one hell of a tweeter amp. 100 watt HF, 350 mid, and 800 bass (per side) probably wouldn’t do Woodstock III either, but would do any festival I might actually get a crack at doing. The only EL 34’s there might be in one of the muso’s amps.
A 100 Watt amplifier that can put out 100 Watts rms sine wave . . .
If it can put out the same peak to peak voltage on a square wave, puts out 200Watts.
Any amplifier ought to be able to put out square waves of the same peak voltage as their sine wave.
Of course, music playback systems do not sound good that way.
The power supply will sag after some milliseconds, sooner or later something else is going to give, etc.
But a guitar amplifier that is driven into 100% overload (square waves at the same peak voltage)
is just what some want to do.
Your mileage may vary.
If it can put out the same peak to peak voltage on a square wave, puts out 200Watts.
Any amplifier ought to be able to put out square waves of the same peak voltage as their sine wave.
Of course, music playback systems do not sound good that way.
The power supply will sag after some milliseconds, sooner or later something else is going to give, etc.
But a guitar amplifier that is driven into 100% overload (square waves at the same peak voltage)
is just what some want to do.
Your mileage may vary.
For the most part as stated
80 to 90 watts is more reasonably assumption.
and it will be a royal tube fryer no matter how good the design.
Makes me think of MI amplifiers from the 70's / 80's made by Traynor and MusicMan
They ran ridiculous plate voltage for around 80 to 90 watts from 6CA7 pairs.
(Quick answer) there is no " magic" tube when you run it at max plate, max load.
It will kill the tube quicker. doesnt matter if it was made in 1940 or 1980 or 1990 etc etc.
My MusicMan amplifier around 90 watts clean 120 clipping
murdered tubes whatever you put in there.
MesaBoogie tested tubes for reliability and sold them in color coded matched pairs.
Those tested and matched ( Chinese) tubes actually survived.
countless fancy pancy vintage European and American vintage tubes.
Which were a lot easier to find in the 90's
And a 90/100 watt EL34 amplifier killed them all.
My Traynor had a fan on the power transformer and also blew air
over the tubes....dont matter. Tube fryer
I would check numerous marshall or musicman amplifiers.
You can do it, but tube life is greatly reduced.
Marshall did not push plate voltage, but the amplifiers are designed
to be run into clipping/
As with any reputable manufacturer learned, specially musician amplifiers.
They tend to use ridiculous high value screen resistors.
since a EL34 or 6CA7 screen pulls a lot of current when clipping
Beam tetrodes do the same , but much more reasonable current
You are way better off using much more moderate, plate voltage
and using the common 100 watt designs with a Quad of 6L6GC
For single pair of tubes be better to look at 6550 or KT88 pair
and again dont push plate voltage to max.
Be realistic and shoot for 65 to 85 watts.
Getting sick and tired of the blame game on " modern tubes"
Ironically the " tube fryers" as well called them.
The 90 watt MusicMan amps destroyed tubes in the 1980s ,90s and so on.
its just a dumb idea to push a tube so high.
Even the " Magical" russian tubes died.
Again, Actually the notorious EL34 that survived abuse
was a Mesa Boogie factory tube.
I used them both in my Traynor and MusicMan amplifier
....they were Chinese tubes....oh know!! how could it be.
80 to 90 watts is more reasonably assumption.
and it will be a royal tube fryer no matter how good the design.
Makes me think of MI amplifiers from the 70's / 80's made by Traynor and MusicMan
They ran ridiculous plate voltage for around 80 to 90 watts from 6CA7 pairs.
(Quick answer) there is no " magic" tube when you run it at max plate, max load.
It will kill the tube quicker. doesnt matter if it was made in 1940 or 1980 or 1990 etc etc.
My MusicMan amplifier around 90 watts clean 120 clipping
murdered tubes whatever you put in there.
MesaBoogie tested tubes for reliability and sold them in color coded matched pairs.
Those tested and matched ( Chinese) tubes actually survived.
countless fancy pancy vintage European and American vintage tubes.
Which were a lot easier to find in the 90's
And a 90/100 watt EL34 amplifier killed them all.
My Traynor had a fan on the power transformer and also blew air
over the tubes....dont matter. Tube fryer
I would check numerous marshall or musicman amplifiers.
You can do it, but tube life is greatly reduced.
Marshall did not push plate voltage, but the amplifiers are designed
to be run into clipping/
As with any reputable manufacturer learned, specially musician amplifiers.
They tend to use ridiculous high value screen resistors.
since a EL34 or 6CA7 screen pulls a lot of current when clipping
Beam tetrodes do the same , but much more reasonable current
You are way better off using much more moderate, plate voltage
and using the common 100 watt designs with a Quad of 6L6GC
For single pair of tubes be better to look at 6550 or KT88 pair
and again dont push plate voltage to max.
Be realistic and shoot for 65 to 85 watts.
Getting sick and tired of the blame game on " modern tubes"
Ironically the " tube fryers" as well called them.
The 90 watt MusicMan amps destroyed tubes in the 1980s ,90s and so on.
its just a dumb idea to push a tube so high.
Even the " Magical" russian tubes died.
Again, Actually the notorious EL34 that survived abuse
was a Mesa Boogie factory tube.
I used them both in my Traynor and MusicMan amplifier
....they were Chinese tubes....oh know!! how could it be.
Fascinated with lots of power?
Try this: https://worldradiohistory.com/UK/Bernards-And-Babani/Bernards/BP12 HiFi Guitar and PA Design.pdf
Babani Hi-Fi, P.A. Guitar and Discotheque amplifier design handbook by B.B. Babani , Babani Press, London
Try this: https://worldradiohistory.com/UK/Bernards-And-Babani/Bernards/BP12 HiFi Guitar and PA Design.pdf
Babani Hi-Fi, P.A. Guitar and Discotheque amplifier design handbook by B.B. Babani , Babani Press, London
You know, the requests here about making 100 watts out of a pair of (formerly) $15 EL34’s remind me of the same broken record we listen to constantly on the solid state side - “Can I make a 200 watt per channel amplifier out of a couple of 2N3055’s? Please show me a schematic I can build”. Results are the same, except things go south a lot faster with transistors. Eventually the discussion comes around to “it takes three or four pair of MJ15024 per side” - but everything else you need is exactly the same. Or they end up settling for 30 watts per channel using those 2N3055’s - after seeing the sticker shock on transformers and 100 volt 22000 uF caps.
Fact of life now -EL34’s are going to go up many times the rate of inflation. If you learn how and why things work you can make a proper decision on where to put your money - and you’ll find that cramming too much power through a pair of EL34’s provides a poor return. When the goal is a working amplifier. The Tubelab experiments were usually done with tubes George bought for a buck apiece. If they turn red-face, not much is lost and some knowledge of where safe operation ends is gained.
Fact of life now -EL34’s are going to go up many times the rate of inflation. If you learn how and why things work you can make a proper decision on where to put your money - and you’ll find that cramming too much power through a pair of EL34’s provides a poor return. When the goal is a working amplifier. The Tubelab experiments were usually done with tubes George bought for a buck apiece. If they turn red-face, not much is lost and some knowledge of where safe operation ends is gained.
4 EL34's in a parallel push-pull making 100-120W will last longer than 2 pairs of EL34s in a 100W push-pull amp.
If more power is desired best solution is pentode mode with regulated supply. Then one can use different forms of feedback. I prefer moderate cathode feedback (`10-15% max) and no loop fbk.
If more power is desired best solution is pentode mode with regulated supply. Then one can use different forms of feedback. I prefer moderate cathode feedback (`10-15% max) and no loop fbk.
RF power tubes are routinely run into heavy grid current and real grid power.
I guess you rarely see this in Audio, but there are examples. Original McIntosh. Original and second Ampeg SVT. Fender 300. A bunch of battery-types. The vast majority of radio transmitter modulators over 50W. See 833. That's 29 Real Watts before we consider capacitance. Note that the grid is driven 140V positive, and sucks like crazy (about 2k-1k equivalent resistance).
Last edited:
Well, AB2/B2 operation….. not what we would normally do for low distortion audio. I guess they were bragging about the fact that the 4E27 could do the 235 watts and stay in in AB1. If you stay in AB1 nothing takes any grid power. You don’t have to go anywhere near AB2 to beat a pair of EL34 to death. Might be something to try with some expendable tubes - maybe some vertical amp types that have curves for positive g1, and may have been designed with that in mind.
Well, there is always parasitic capacitance that draws power from the control grid circuit even in AB1, especially using parallel tubes, and increases with frequency. It is very small though.
Grid to plate capacitance does couple a bit of energy into the plate circuit. Grid to cathode is strictly displacement current in AB1 - energy to charge/discharge it comes from outside (as in it taxes the previous stage, but doesn’t dissipate power in itself). It’s tiny in tubes. Can be HUGE in switching mosfets (they make drivers capable of 10 amps for the purpose).
Its all just semantics but I did think it was funny (since it’s not doing something any KT88 wouldn’t).
Its all just semantics but I did think it was funny (since it’s not doing something any KT88 wouldn’t).
The problem with running 800V plate voltage into a transformer coupled load is that under maximum power you have up to 1600V on the octal socket, and that is simply playing with fire given the filament is only 1 pin away from the plate on an EL34. A bit of dust and moisture and the thing arcs over and carbon tracks, destroying your amp. When using output tubes >400V it's safer to use a tube with a top cap instead.
Last edited: