MOSFET source follower question

[Alan0354]

It's actually not as simnple as that. You should have in mind that in order for the MOSFET follower to "copy" the clipping behaviour of the tube CF it needs some changes which are described in detail here:

Some nuances of using mosfets as followers in tube guitar amplifiers

Also here:

http://www.amtelectronics.com/downlo...e_follower.pdf

I read through the first article today. My observation on the waveform is exactly like the pictures of the cathode follower in the first article. I do not agree with some of the stuff in the second article.

Question is:

1) Is is so obvious that the the cathode resistor (100K R3) run out of steam to put down because of the impedance of the tone stack. The output waveform become a slanted down line. Why people keep using 100K for more? When I saw the condition where the plate of the driving tube goes low and the cathode follower stuck because it got turned off, I immediately change the 100K to 68K and improve the wave form quite a bit!!!! Why writing all these article, taking pictures of the wave form. Is there any magic I missed not having the clipped slanted bottom?

2) Also, when the plate of the driving tube tries to swing up, the cathode follower can run into 0V or slightly +ve Vgk and grid start conducting like a vacuum diode. That prevent the 100K plate resistor from pulling up as it run out of steam also!!!! Are you sure people are not hear all these DEFECTs of the existing circuit where at least the MOSFET can over come the clipping of the top part!!!

Circuit wise, MOSFET or even BJT will work a lot better as they can easily drive more current and don't turn on like a vacuum diode. When I do my design a week or two ago, I seriously thinking about using SS follower. I decided against it ONLY because of the "ALL TUBE" thingy. You can call it silly, but people do look at all tubes and willing to pay a lot more. try explaining to the blind that you are only doing a follower and does not affect the distortion characteristics of the tube stages.


I don't see why they put D1 and D3 just to recreate the bottoming out when R3 runs out of steam. Am I really missing something or people do it as a ritual like using carbon comp, CTS pots, orange drop caps? Get rid of the two diodes and let the zener diode across the source gate to turn on!!! OR better yet, lower the darn 100K resistor to 68K!!!!

[Gregg]

Those circuits are intended to recreate the "imperfect" behaviour of the tube CF as found in most amps (with 100k cathode resistor). Whether this is good or bad or can be improved or is necessary is another question.
If your observations are different from those in the article maybe you can write to KMG or Kempf and get some comments from them. I guess their reply will be of interest to all of us.
Again I think the best way to figure this out is do listening tests and decide for yourself.

[Alan0354]

Those circuits are intended to recreate the "imperfect" behaviour of the tube CF as found in most amps (with 100k cathode resistor). Whether this is good or bad or can be improved or is necessary is another question.
If your observations are different from those in the article maybe you can write to KMG or Kempf and get some comments from them. I guess their reply will be of interest to all of us.
Again I think the best way to figure this out is do listening tests and decide for yourself.

I agree with the first article on the waveform, I don't agree with the second article.

The editorial part is just my old fart ranting!!! ha ha!!! It is just my opinion that there is a defect that the signal cannot get the full swing downwards due to the CF stage got turned off. I would fixed that first. The wave form is too asymmetrical. I don't think that was the intention of Marshall. Original 59 Bassman has 100K for driving the big caps, problem is not as pronounced. When people start lowering to 56K, or even 36K, that's when the problem becomes bad.................UNLESS people like the sound of this defect!!!