here's the Mullard datasheet: http://www.mif.pg.gda.pl/homepages/frank/sheets/129/e/EL34.pdf
here's the Phillips datasheet: http://www.mif.pg.gda.pl/homepages/frank/sheets/030/e/EL34.pdf
=================
1. According to Philips (top of page 7), I need to connect grid2 to plate and I'll get a triode.
BUT according to mullard (page3), to get a triode connection, it's grid2 to plate and grid3 to cathode.
so which is right? It is possible as well that I misread the datasheet.
ps. I've seen replies were people have recommended the Mullard way but Philips must know something if they are recommending it differently.
2. I've seen replies where people have suggested a 100ohm resistor between plate and grid2. But I was looking at the mullard and philips datasheets and it seems I only need a piece of wire to connect the two pins.
what is the purpose of the 100ohm resistor?
Thank you for the help
here's the Phillips datasheet: http://www.mif.pg.gda.pl/homepages/frank/sheets/030/e/EL34.pdf
=================
1. According to Philips (top of page 7), I need to connect grid2 to plate and I'll get a triode.
BUT according to mullard (page3), to get a triode connection, it's grid2 to plate and grid3 to cathode.
so which is right? It is possible as well that I misread the datasheet.
ps. I've seen replies were people have recommended the Mullard way but Philips must know something if they are recommending it differently.
2. I've seen replies where people have suggested a 100ohm resistor between plate and grid2. But I was looking at the mullard and philips datasheets and it seems I only need a piece of wire to connect the two pins.
what is the purpose of the 100ohm resistor?
Thank you for the help
Hi jarthel,
You asked:
Actually, they're both correct, and you did not misread anything. The fact is that Philips just didn't mention the suppressor grid (G3) being connected to the cathode as it should be. Either they assumed one would already know this, or there may be versions where it's internally connected.
In any event, to make an EL34 a triode, connect G2 (screen) to the plate. And G3 (suppressor) to the cathode. Very technically, a straight wire between the screen and plate is all that's needed. However, in practice it's very common (and generally preferable) to use a low value resistor of 100 ohms to link the two. Doing so provides a small amount of decoupling as well as current limiting for the screen. No resistor is used between the suppressor and cathode. Just a piece of wire.
Incidentally, the suppressor and cathode should always be connected together whether in triode or pentode mode. By keeping G3 negative, with respect to the anode, it fulfills it's duty of repelling (or suppressing) secondary emission from the plate.
Victor
You asked:
Actually, they're both correct, and you did not misread anything. The fact is that Philips just didn't mention the suppressor grid (G3) being connected to the cathode as it should be. Either they assumed one would already know this, or there may be versions where it's internally connected.
In any event, to make an EL34 a triode, connect G2 (screen) to the plate. And G3 (suppressor) to the cathode. Very technically, a straight wire between the screen and plate is all that's needed. However, in practice it's very common (and generally preferable) to use a low value resistor of 100 ohms to link the two. Doing so provides a small amount of decoupling as well as current limiting for the screen. No resistor is used between the suppressor and cathode. Just a piece of wire.
Incidentally, the suppressor and cathode should always be connected together whether in triode or pentode mode. By keeping G3 negative, with respect to the anode, it fulfills it's duty of repelling (or suppressing) secondary emission from the plate.
Victor
Hi Victor,
Sorry to say so, but I think you have a double misconception here; but never mind, they are very common ...
Neither the decoupling of g2 is wanted to make a triode out of a tetrode or pentode (instead, g2 shall just swing along with anode; effectively adding its area to the anode area - but when it is "decoupled" (even a small bit - think UL!), it just ain´t triode mode anymore), nor does a typical 100 ohms resistor between anode and g2 do any current limiting whatsoever (and for what reason to boot with? Certainly not because g2 dissipation limits would require) - just do the very simple math and you will see.
The *only* purpose of this resistor is to act as a (screen-) grid stopper. Hardly needed when the anode connection is available at a base pin as with usual audio pentodes; but can make sense with top-capped tubes, like 807 and such.
Tom
Sorry to say so, but I think you have a double misconception here; but never mind, they are very common ...
Neither the decoupling of g2 is wanted to make a triode out of a tetrode or pentode (instead, g2 shall just swing along with anode; effectively adding its area to the anode area - but when it is "decoupled" (even a small bit - think UL!), it just ain´t triode mode anymore), nor does a typical 100 ohms resistor between anode and g2 do any current limiting whatsoever (and for what reason to boot with? Certainly not because g2 dissipation limits would require) - just do the very simple math and you will see.
The *only* purpose of this resistor is to act as a (screen-) grid stopper. Hardly needed when the anode connection is available at a base pin as with usual audio pentodes; but can make sense with top-capped tubes, like 807 and such.
Tom
Hello Tom,
I read with interest your response and do agree with your words to a large part. Unfortunately, I am not an experienced writer and may sometimes use incorrect syntax to convey what I really mean. Decoupled may have been a poor choice of terms. But I do fully understand the concepts of this circuitry.
If you consider how a stopper resistor works, the main thing it does is to slightly desensitize the element by decoupling it from the circuit and lowering the Q to prevent oscillation. Decoupling can be a subtle event, or more direct as in branch-out circuits. It's really a matter of degree. At least that's how the term is used here in the U.S. Prehaps Europeans have a different concept of the meaning.
As far as the current limiting goes I agree that, pragmatically, it is of little value. Especially since the screens are at anode potential (not higher as with UL operation) dissapation should not be an issue. But strictly speaking, by virtue of the fact the resistor was physically and electrically in the circuit, there would be some current limiting. Albeit only a tiny amount.
And btw, I like your icon. I'm a cat person too.
JARTHEL,
You do not need to use this resistor. But you can if you want to. Many designers and builders use it anyway as a preventive just in case. It's up to you.
Victor
I read with interest your response and do agree with your words to a large part. Unfortunately, I am not an experienced writer and may sometimes use incorrect syntax to convey what I really mean. Decoupled may have been a poor choice of terms. But I do fully understand the concepts of this circuitry.
If you consider how a stopper resistor works, the main thing it does is to slightly desensitize the element by decoupling it from the circuit and lowering the Q to prevent oscillation. Decoupling can be a subtle event, or more direct as in branch-out circuits. It's really a matter of degree. At least that's how the term is used here in the U.S. Prehaps Europeans have a different concept of the meaning.
As far as the current limiting goes I agree that, pragmatically, it is of little value. Especially since the screens are at anode potential (not higher as with UL operation) dissapation should not be an issue. But strictly speaking, by virtue of the fact the resistor was physically and electrically in the circuit, there would be some current limiting. Albeit only a tiny amount.
And btw, I like your icon. I'm a cat person too.
JARTHEL,
You do not need to use this resistor. But you can if you want to. Many designers and builders use it anyway as a preventive just in case. It's up to you.
Victor
Hi Jarthel,
Right, it is not needed when you want to get triode characteristics from a pentode or beam power tube.
But it may help preventing parasitic oscillations in certain cases.
In this sense, a resistor of the usual 100 ohms value does not hurt, but definitely does not serve the purpose of "partly decoupling" nor "current limiting".
Tom
jarthel said:
Right, it is not needed when you want to get triode characteristics from a pentode or beam power tube.
But it may help preventing parasitic oscillations in certain cases.
In this sense, a resistor of the usual 100 ohms value does not hurt, but definitely does not serve the purpose of "partly decoupling" nor "current limiting".
Tom
IF by connecting the screen to the plate the dissapation of the screen is exceeded, the tube's life will be decreased considerably. In such cases the resistor should be scaled to keep/prevent excessive screen dissapation at full/high power levels.
It seems that the current crop of non-NOS EL34/6CA7 tubes do not have the same screen dissapation capabilities as the NOS did. For that matter, none of the import tubes (eg 6550, 6L6GC, etc.) appear to really equal the NOS tubes in that respect.
_-_-bear
It seems that the current crop of non-NOS EL34/6CA7 tubes do not have the same screen dissapation capabilities as the NOS did. For that matter, none of the import tubes (eg 6550, 6L6GC, etc.) appear to really equal the NOS tubes in that respect.
_-_-bear
Hi Bear,
Can you please give an actual, real world audio duty operation point example for your "if-clause" with EL34 in triode mode?
By a resistor that is of a value high enough to be able to "limit" (nah, better lets say "lower") g2 current, hence g2 dissipation, significantly, you would have left triode mode by far already.
Tom
bear said:
Can you please give an actual, real world audio duty operation point example for your "if-clause" with EL34 in triode mode?
By a resistor that is of a value high enough to be able to "limit" (nah, better lets say "lower") g2 current, hence g2 dissipation, significantly, you would have left triode mode by far already.
Tom
I know this is old
I know this is an old thread but I wanted to say to convert to pentode you can go to a 1K resistor and then you would take the screen straight to the power supply however distortion will go up. There is a way to go pentode and keep the distortion down, you would need to build a voltage regulated supply for the screen I would say around 200 volts and keep the regulation tight.
I know this is an old thread but I wanted to say to convert to pentode you can go to a 1K resistor and then you would take the screen straight to the power supply however distortion will go up. There is a way to go pentode and keep the distortion down, you would need to build a voltage regulated supply for the screen I would say around 200 volts and keep the regulation tight.
Beam Power/Beam Tetrode, and Pentode tubes:
Given a fixed Screen voltage (300V for example) from a low impedance voltage source (Low Z and low DC Ohms):
The worst case screen dissipation is in Pentode mode, when the Plate swings toward the cathode (as an example when V plate to cathode is 50V), and the screen is at a high voltage.
Check the data sheet graphs and look at the screen current curves.
Just do not operate the amp into clipping 100% of the time. The screen will be at max current 50% of the time (times the fixed 300V).
Next, at the same screen voltage (quiescent voltage, 300V), UL has less screen dissipation because when the plate swings down 250 Volts to 50 Volts, the screen also swings down
(40% x 250 Volts, screen @ 200V). When the plate is at 50V, the screen voltage has partially reduced along with it.
Re-calculate and check according to the UL tap %: 20, 26, 40, 50, 75, etc.
Next, at the same screen voltage (quiescent voltage, 300V), in Triode wired mode, when the plate swings down to 50V, the screen is also 50V too.
Less dissipation on the screen than both Beam Power/Pentode mode, and less than UL mode.
That is why an 807 Beam Power tube has a 300V screen rating, when wired as a Beam Power tube.
But, when triode wired, the screen is rated for up to 400V.
I estimate that for UL taps of 40% or 50%, the screen will work fine at about 340V or 350V, respectively.
A few good calculations can tell you whether you can sleep good at night, knowing your amplifier is designed for reliability, and not trading off power output for early tube failure.
Given a fixed Screen voltage (300V for example) from a low impedance voltage source (Low Z and low DC Ohms):
The worst case screen dissipation is in Pentode mode, when the Plate swings toward the cathode (as an example when V plate to cathode is 50V), and the screen is at a high voltage.
Check the data sheet graphs and look at the screen current curves.
Just do not operate the amp into clipping 100% of the time. The screen will be at max current 50% of the time (times the fixed 300V).
Next, at the same screen voltage (quiescent voltage, 300V), UL has less screen dissipation because when the plate swings down 250 Volts to 50 Volts, the screen also swings down
(40% x 250 Volts, screen @ 200V). When the plate is at 50V, the screen voltage has partially reduced along with it.
Re-calculate and check according to the UL tap %: 20, 26, 40, 50, 75, etc.
Next, at the same screen voltage (quiescent voltage, 300V), in Triode wired mode, when the plate swings down to 50V, the screen is also 50V too.
Less dissipation on the screen than both Beam Power/Pentode mode, and less than UL mode.
That is why an 807 Beam Power tube has a 300V screen rating, when wired as a Beam Power tube.
But, when triode wired, the screen is rated for up to 400V.
I estimate that for UL taps of 40% or 50%, the screen will work fine at about 340V or 350V, respectively.
A few good calculations can tell you whether you can sleep good at night, knowing your amplifier is designed for reliability, and not trading off power output for early tube failure.
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.