Cathode follower distortion results

[ashok]

I tried to simulate three schemes of cathode followers using a 6DJ8 and +/- 35 volt supplies.
Grid goes to ground through a large resistor.
The three schemes are :
1. 12K resistor from cathode to -35 volts
2. A CCCS of 1mA in between cathode and -35 V.
3. A CCS of 1mA in the cathode and a bootstrapped anode using a MOSFET (IRF840).

The distortion I get in % is shown in the chart. There doesn't seem to be much difference between them and the CCS scheme has the lowest 2nd harmonic distortion.

Might be that my load of 68K is high.
Can anyone simulate this with their Spice program and see if they come up with a similar result ?

Has anyone compared the sound of all three schemes? The CCS scheme sounds good as compared to a direct connection to the power amp. Any improvement (?) or change in sound with the bootstrapped scheme.
Thanks,
Cheers,
Ashok.

 

[EC8010]

You really haven't got enough HT. If you use a CCS, you can get away with the -35V supply, but you really ought to have 90-100V across the 6DJ8. Similarly, 1mA is a very low current for this valve, >5mA would be better. Given that a cathode follower has 100% negative feedback, it is capable of vanishingly low distortion provided that the valve is optimally biased. You could use a voltage multiplier to produce the positive supply. With a 90V HT and a CCS load down to -35V, I would expect to measure figures an order of magnitude better than your simulation.

 

[ashok]

Hmmm.....................

Hi EC,
I did post earlier on a different thread about my problems with the 6922 (Electro Harmonix) in this circuit. It refused to bias properly.
http://www.diyaudio.com/forums/showthread.php?s=&threadid=39025&highlight=

The Sovtek 6DJ8 works. I tried to simulate higher current (6mA) and 90 volts at the anode. It shows very little improvement.
Now all that's simulation and it could be just bull with respect to actual figures.
Can you guys try it out with your simulators and see what you get ?
Today several variables are out of gear.
I'm really loaded down with work ( non profitable !).
It raining very heavily.
The humidity is almost 100%
The circuit simulation is bugging me .
There is a hum on one channel ( visible on the scope )
AND I'm beginning to doubt that the cathode follower is almost like a straight wire !

AND Murphy is all around !
Cheers,
Ashok

 

[fdegrove]

Hi,

I did post earlier on a different thread about my problems with the 6922 (Electro Harmonix) in this circuit. It refused to bias properly.

Allow me to doubt that this is typical of this tube.
It should bias fine unless there's something wrong with the one you used.

The Sovtek 6DJ8 works. I tried to simulate higher current (6mA) and 90 volts at the anode. It shows very little improvement.

I doubt it's a true 6DJ8, more likely a 6N1-P which is not the same tube even though that's what many vendors sell it for.

The CF you're using is probably the buffer of a gainclone circuit, it should work just fine as is and IMHO it's probably the most obvious choice within the confines of the GC concept bar, perhaps the CCS version, a 2SK170 FET in the tail of the CF should do.

As EC8010 pointed out, you shouldn't expect miracles from a CF running at such low voltages.
The only thing that can save it from performing only soso is to run sufficient current through it, otherwise there's ;ittle point in having it there in the first place.


Cheers,