Tony,
I use the 6DJ8 as a cathode follower, and at one point tried the 7DJ8 (I have 2 of them here) and then for a lack of total clarity on if it is really is equivalent (or ok to use in my 6DJ8 application) I ended up putting a 6DJ8 back in. If the heater voltage is 6.3V then will there be any problems in how the 7DJ8 performs. I assume as the quoted heater current at 7.6V is 300mA it will end up runing at about 250mA.
Thanks
I use the 6DJ8 as a cathode follower, and at one point tried the 7DJ8 (I have 2 of them here) and then for a lack of total clarity on if it is really is equivalent (or ok to use in my 6DJ8 application) I ended up putting a 6DJ8 back in. If the heater voltage is 6.3V then will there be any problems in how the 7DJ8 performs. I assume as the quoted heater current at 7.6V is 300mA it will end up runing at about 250mA.
Thanks
7DJ8 = PCC88 ? It is for serial 300mA heating, not constant 6.3V heating. Initially designed for TV sets where the heating of all tubes were serial connected. ECC85 is a remote cutoff (variable mu) tube, used for AGC. Its gain depends on the negative bias voltage. Not very useful for audio.
Actually you can find pretty decent linear areas, especially CCS loaded. For example, at 210V, 5mA, -3.5V it is very linear. Also pretty good for LTP. It is pretty cheap too. Example is for the PCC85, but the ECC85 curves are similar (not the same though).
Used in a properly engineered circuit actually designed for it, the Russian 6N1P is an excellent audio tube, with very low distortion. They are quiet, non-microphonic, extremely long lived and very reliable.
My Audio Research Corporation VS110 uses five of them; 2 1/2 per channel driving four 6550s per channel. I use the 6N1P-EV types, and they seem to last forever. I do not recall what the rated lifespan is for that version, but after several years of daily operation the characteristics have not drifted from their original "as-received" tests. Pretty impressive🙂
My Audio Research Corporation VS110 uses five of them; 2 1/2 per channel driving four 6550s per channel. I use the 6N1P-EV types, and they seem to last forever. I do not recall what the rated lifespan is for that version, but after several years of daily operation the characteristics have not drifted from their original "as-received" tests. Pretty impressive🙂
Lingwendil,
The 6BQ7/6BQ7A (other than its higher u), has Gm and rp, that sits in the middle between the 6N1P, and the 'loose family' of 6DJ8 types.
Just for a comparison, there are many common JFET types that have very wide variations in their specifications.
So wide that their variability is more than the difference between a 6N1P versus a 6DJ8.
Of course there are other JFETs that do not have that wide of a range for a single type.
After I initially worked with tubes, I was surprised at the very wide range of IDSS for a single registered part number.
Then I want back to tubes.
The 6BQ7/6BQ7A (other than its higher u), has Gm and rp, that sits in the middle between the 6N1P, and the 'loose family' of 6DJ8 types.
Just for a comparison, there are many common JFET types that have very wide variations in their specifications.
So wide that their variability is more than the difference between a 6N1P versus a 6DJ8.
Of course there are other JFETs that do not have that wide of a range for a single type.
After I initially worked with tubes, I was surprised at the very wide range of IDSS for a single registered part number.
Then I want back to tubes.
Last edited:
i hear you, but....but i once had a line preamp brought to me by a client, it had Russian 6H8 marked as 6sN7 running filaments at 5.2vdc still sounding excellent....if you did not like them, toss them out, that simple...
and so i am least worried about plugging in a 7dj8 to a 6dj8 sockets, i tried the 7dj8 in a cascode to drive the 300b, sounds wonderful too...
gm is cathode current dependent and rp plate voltage dependent, they vary all over the place, just look at any tube datasheet to see...
when these are quoted, usually at a single point in the tube curves, they are never constant and never in any linear 1 to 1 correspondence.....
when these are quoted, usually at a single point in the tube curves, they are never constant and never in any linear 1 to 1 correspondence.....
TonyTecson,
Tube substitutions?
Of course Gm varies, rp varies, and u varies with different voltages and different currents, in the exact same tube.
I bet if someone connects the following connections correctly on a 9 pin EL84 socket, to an Octal 7591 socket.
filament 1 to filament 1; filament 2 to filament 2; Plate to Plate; Cathode to Cathode; Screen to Screen; and control grid to control grid.
And . . .
EL84 g2/g1 u = 19; rp = 40k
7591 g2/g1 u = 16.8; rp = 29k
Then perhaps the voltages and currents can be adjusted to make both tubes have the following:
g2/g1 u = 18; rp = 34.5k
If they can be adjusted to be equal u and rp, I bet it will be at completely different voltages and completely different currents from one tube to the other tube.
Question: Does that make one tube a substitute for the other tube?
Or, try that experiment with 2 dissimilar triodes.
Same thing, same thoughts.
When I do tube rolling (not very often), I generally make sure the tube characteristics are very similar.
Otherwise, if I hear a difference in the measurements and/or the "sound of the amp" I will not be able to determine the cause of any difference.
That is because . . .
Changing the voltages and currents on one tube, to try and make dissimilar tubes "act" the same, makes me ask the question:
Is it the tube, or is it the different parts I used for one tube, versus the parts I used for the other tube?
Thanks for reading!
Tube substitutions?
Of course Gm varies, rp varies, and u varies with different voltages and different currents, in the exact same tube.
I bet if someone connects the following connections correctly on a 9 pin EL84 socket, to an Octal 7591 socket.
filament 1 to filament 1; filament 2 to filament 2; Plate to Plate; Cathode to Cathode; Screen to Screen; and control grid to control grid.
And . . .
EL84 g2/g1 u = 19; rp = 40k
7591 g2/g1 u = 16.8; rp = 29k
Then perhaps the voltages and currents can be adjusted to make both tubes have the following:
g2/g1 u = 18; rp = 34.5k
If they can be adjusted to be equal u and rp, I bet it will be at completely different voltages and completely different currents from one tube to the other tube.
Question: Does that make one tube a substitute for the other tube?
Or, try that experiment with 2 dissimilar triodes.
Same thing, same thoughts.
When I do tube rolling (not very often), I generally make sure the tube characteristics are very similar.
Otherwise, if I hear a difference in the measurements and/or the "sound of the amp" I will not be able to determine the cause of any difference.
That is because . . .
Changing the voltages and currents on one tube, to try and make dissimilar tubes "act" the same, makes me ask the question:
Is it the tube, or is it the different parts I used for one tube, versus the parts I used for the other tube?
Thanks for reading!
i am merely telling what is see in the tube manuals, i suspect that characteristics mentioned there are sweet spots that users can follow and that manufacturers went out of their way to find good spots in their tube's characteristics......all things are in the tube manuals for anyone to see...
i have inserted the 6cg7 in a circuit meant for the 6dj8 and found the sound did not differ much...
i have inserted the 6cg7 in a circuit meant for the 6dj8 and found the sound did not differ much...
we are talking small signal triodes here, power tubes are a different ball of whacks....
now why would i even consider substituting 7591 with an el84? i will never do that....
TonyTecson,
When you inserted the 6CG7 into your 6DJ8 circuit . . .
Wasn't the gain reduced by at least 3 dB?
Was there any negative feedback in that 6DJ8 circuit (local: cathode, plate to grid), Or was there any global negative feedback around that triode and at least one stage before or after the 6DJ8 circuit (often times, negative feedback is essentially a performance 'equalizer').
Thanks!
The following is just my opinions;
(Perhaps someone else will Post a comment about them) . . .
The relationships of u, rp, Gm, and negative feedback effects . . . is the same for small signal triodes as it is for medium power triodes, and also for high power output triodes (even Eimac's 1500 Watt triode has the same relationships as the 12AX7).
Thanks!
When you inserted the 6CG7 into your 6DJ8 circuit . . .
Wasn't the gain reduced by at least 3 dB?
Was there any negative feedback in that 6DJ8 circuit (local: cathode, plate to grid), Or was there any global negative feedback around that triode and at least one stage before or after the 6DJ8 circuit (often times, negative feedback is essentially a performance 'equalizer').
Thanks!
The following is just my opinions;
(Perhaps someone else will Post a comment about them) . . .
The relationships of u, rp, Gm, and negative feedback effects . . . is the same for small signal triodes as it is for medium power triodes, and also for high power output triodes (even Eimac's 1500 Watt triode has the same relationships as the 12AX7).
Thanks!
Last edited:
6N23P is the generally accepted Russian equivalent of the ECC88. It's been said earlier in the thread that it's close but not equivalent. While I'm sure that's probably the case, I've used them in place of ECC88 for several years, they sound good and cause no issue.