Morning everyone 🙂
I hope that there is somebody in here who can provide me with a knowledge of the input impedance at the EL84 cathode follower stage in the the circuit shown below:
The reason I ask is that my friend and I are going to try a little with a small signal pentode gain stage, which has too high output impedance to see directly into the preamp. And since we have an interest for circuits that is uncommon, it would be exciting to try this cathode follower stage with EL84 out 🙂
I would just be grateful if someone could tell me or calculate the input impedance on this cathode follower stage.. 🙂🙂
Best from
Henrik Vingborg
I hope that there is somebody in here who can provide me with a knowledge of the input impedance at the EL84 cathode follower stage in the the circuit shown below:
An externally hosted image should be here but it was not working when we last tested it.
The reason I ask is that my friend and I are going to try a little with a small signal pentode gain stage, which has too high output impedance to see directly into the preamp. And since we have an interest for circuits that is uncommon, it would be exciting to try this cathode follower stage with EL84 out 🙂
I would just be grateful if someone could tell me or calculate the input impedance on this cathode follower stage.. 🙂🙂
Best from
Henrik Vingborg
The grid input impedance is pretty much infinite. The signal goes to R6 (100k) to C2 (220nF) which (at this impedance) shunts signal to ground, so the input impedance is 100k.
Vingborg, did you mean you mean the input impedance of the circuit, which is 100 kOhm for the reason benb explained, or the input impedance of teh actual EL84 stage seem looking into C3, ie without the triode stage?
To my knowledge, input impedance for cathode followers are so huge, you can ignore it. The reason being the grid leak resistor (R9) getting bootstrapped (voltage swing on each end of the resistor is nearly similar). This increases the input impedance considerably. So, unless your pentode gain stage output impedance is in the Megaohm range, you can be sure the cathode follower won't be a considerable load. Sorry can't help on an exact value..
To calculate the input impedance of the EL84 connected as a triode cathode follower requires knowing the plate resistance of the valve;a figure I do not have at hand.However it can be implied that the input impedance will be greater than the 470k ohm grid resistor.I would consider the resistance of the combination of R12 through to R14 a bit low and it would be better if the total were raised 4 to 5k ohm.The working load impedance as seen by the cathode follower should be at least 3 times higher than the cathode resistance and I would consider removing R15 .
Rin = Rg / (1-A) + (ra||Rk)
Where A is the cathode follower gain, Rg is the grid leak, ra is the internal anode resistance, Rk is the cathode load resistor.
In this case the input resistance would be a couple of megohms into a high-impedance load, but less when heavily loaded by headphones.
Where A is the cathode follower gain, Rg is the grid leak, ra is the internal anode resistance, Rk is the cathode load resistor.
In this case the input resistance would be a couple of megohms into a high-impedance load, but less when heavily loaded by headphones.
Last edited:
Hello again 🙂
Thanks for quick reply..
I am thinking of only the cathode follower stage with the EL84 tube and all the yellow resistors!
I have a EF86 pentode gain stage which requires at least 500K load - Something my preamp dosen't have (47K)
That's why I'm going to use a cathode follower and I really like it all if I could use this follower circuit with EL84, and my pentode stage could see more than 500K.. Or if possible : Increase the input impedance on the EL84 stage to say 1M?
Can I increase R9 to 1M or is it not that simple? 😛
Best again from
Henrik Vingborg
Thanks for quick reply..
I am thinking of only the cathode follower stage with the EL84 tube and all the yellow resistors!
I have a EF86 pentode gain stage which requires at least 500K load - Something my preamp dosen't have (47K)
That's why I'm going to use a cathode follower and I really like it all if I could use this follower circuit with EL84, and my pentode stage could see more than 500K.. Or if possible : Increase the input impedance on the EL84 stage to say 1M?
Can I increase R9 to 1M or is it not that simple? 😛
Best again from
Henrik Vingborg
Gain of cathode follower itself is A = mu / (mu + 1). mu of EL84 triode connection is 19.5 from datasheet so A = 19.5 / (19.5 + 1) = 0.951.
This gain will decrease to A' by divided with cathode output inpedance and 330//330 plus 1.8k, also 3.3k load paralled with them.
A' = A * (330//330 + 1800)//3300 / ( 100 + (330//330 + 1800)//3300) = 0.88
Thus Rin = Rg / (1-A) as Merlinb wrote,
Rin = 470k / (1-0.88) = 3.9meg ohm
If you wish to increase Rin, you can increase R9 to 1meg and you get 8.3meg Rin.
But I think 3.9meg is still high enough for this circuit.
This gain will decrease to A' by divided with cathode output inpedance and 330//330 plus 1.8k, also 3.3k load paralled with them.
A' = A * (330//330 + 1800)//3300 / ( 100 + (330//330 + 1800)//3300) = 0.88
Thus Rin = Rg / (1-A) as Merlinb wrote,
Rin = 470k / (1-0.88) = 3.9meg ohm
If you wish to increase Rin, you can increase R9 to 1meg and you get 8.3meg Rin.
But I think 3.9meg is still high enough for this circuit.
You could use an EL84 follower to bootstrap the anode load resistor of the EF86. As others have said, use a larger value for the cathode resistor and the output ground leak resistor. A cathode follower is a gain stage with 100% negative feedback so the usual rule 'make it linear, then add feedback' still applies. If an EL84 needs, say, 5k anode load as an output stage then it needs around 5k (or more) as a cathode load when used as a follower.
As others have said, that resistor is bootstrapped so its effective value is multiplied. This is like Miller effect in reverse.
Hello folks 🙂
This was some useful answers, I must say!
3,9 meg ohm is more than plenty for my pentode stage, I'm sure!!
I thank you all very much🙂
Best from Henrik
This was some useful answers, I must say!
3,9 meg ohm is more than plenty for my pentode stage, I'm sure!!
I thank you all very much🙂
Best from Henrik
Hej CF,
If you just connect the EL84 as pentode you will get much better performance than as triode connection. You can skip C3 and R9 and DC connect the to stages. Remember to adjust R14.
Hilsen
Karsten
If you just connect the EL84 as pentode you will get much better performance than as triode connection. You can skip C3 and R9 and DC connect the to stages. Remember to adjust R14.
Hilsen
Karsten
Using the EL84 as a pentode follower, and DC coupling the stages, are two different things. To use the EL84 as a pentode you need to bootstrap g2 from the cathode.
Heey country friend 🙂
How to connect the EL84 as a pentode?
I thought it allready was that in this EL84 circuit. G3 connects internal to the cathode - G2 is not connected directly to the anode, but with a resistor ect..
Taken the R9 away will cause the change in the input impedance of the CF circuit (I believe)
I plan to put this EL84 CF stage to my EF86 gain stage and I would really like it if I could leave C3 and R10 out, because I want to reach a signal path as short as possible!!🙂 Yes?😀
Hilsen Henrik
How to connect the EL84 as a pentode?
I thought it allready was that in this EL84 circuit. G3 connects internal to the cathode - G2 is not connected directly to the anode, but with a resistor ect..
Taken the R9 away will cause the change in the input impedance of the CF circuit (I believe)
I plan to put this EL84 CF stage to my EF86 gain stage and I would really like it if I could leave C3 and R10 out, because I want to reach a signal path as short as possible!!🙂 Yes?😀
Hilsen Henrik
To make a pentode follower you need to bootstrap the screen from the cathode. Usually not worth doing. The screen feed resistor then appears in parallel with the cathode load and loads the output. This partly negates the increased transconductance of a pentode follower.Vingborg said:
Morning folks 🙂
DF96: Bootstrap? I don't understand that term, and I can't translate it by google.. Can you explane it for me😉
Merlinb: That looks interesting the schematic you posted, can you tell me the diode role instead of R9 470k? Sorry I'm not an engineer or something🙂
And how about the R10 100ohm gridstopper.. Can I leave it out, so the EL84 control grid connects directly to EF86 plate?
Best regards from
Henrik🙂
DF96: Bootstrap? I don't understand that term, and I can't translate it by google.. Can you explane it for me😉
Merlinb: That looks interesting the schematic you posted, can you tell me the diode role instead of R9 470k? Sorry I'm not an engineer or something🙂
And how about the R10 100ohm gridstopper.. Can I leave it out, so the EL84 control grid connects directly to EF86 plate?
Best regards from
Henrik🙂
The diode stops the grid from being pulled hundreds of volts above the cathode when power is first applied. It stops arcing inside the valve. During normal operation the diode is reverse biased, off, and does not affect the audio.
No you cannot leave it out; it protects against oscillation.
You need to put a highish value resistor from the supply rail to g2, and a capacitor (probably electrolytic) from g2 to the cathode. This means that the cathode-g2 voltage is fixed so the EL84 acts as a pentode.Vingborg said:
A low value resistor in series with C4/C5 (circuit in post 14) may help stability too.
The zener version will work down to DC, but add a little noise. It will limit the maximum voltage swing to whatever quiescent voltage is dropped across the g2 feed resistor. As it steals some current from this resistor, the resistor will need to be smaller in value and so load the output more.
So if you need pentode follower action down to DC use a zener. Otherwise use a capacitor and accept that at lower frequencies the action will change from pentode follower to triode follower.
Or just use it as a triode follower. Good enough for most purposes.
So if you need pentode follower action down to DC use a zener. Otherwise use a capacitor and accept that at lower frequencies the action will change from pentode follower to triode follower.
Or just use it as a triode follower. Good enough for most purposes.
- Status
- Not open for further replies.