cathode follower switch??

[yunger]

being rather new to this stuff, i'm curious about the overdrive effect of a cathode follower. i do like the sound of circuits that use them (e.g. plexi, top boost, blah, blah). my current circuit has 2 preamp stages preceding a tone stack, which can sound fizzy and harsh when driven too hard...not my thing. i do like the sound before it gets too nasty though. i think i read that a cf can compress the fizz and smooth the tone.
my question is, would it be worth the effort to have the option to switch in/out a cf between the 2nd stage and tone stack for different feels/tones??? or based on you're experience, should i just "breadboard" a cf and make a decision before i start building the chassis? i realize that the switch would pop so it would be mounted under or in the chassis to be switched when the amp is off.

[tubeswell]

A quick snapshot of some key points (courtesy of Merlin's book):

The main benefits of a cathode follower are:

- Low output impedance (making the signal virtually impervious to the otherwise potential tone-shunting effects of heavy downstream 'loads' on the signal - like what happens in a TMB tonestack)
- Increased bandwidth (more highs and lows)
- Very linear signal input/output (little distortion)

Plus if they are DC-coupled to a prior inverting gain stage (as in a 5F6A DC-coupled pair) the loss of grid current from the inverting stage into the CF cathode softens the onset of grid current limiting and creates 2nd harmonic distortion, making for smooth overdrive. Reducing the CF cathode resistor in a DC-coupled pair increases the compression.

[MLee]

Actually what Mr.tubeswell says is correct if it's a text book. But we experience in a guitar amplifier is not common with an audio application.
1.We drive the stage such heavily.
2.The load is very unique.

1.You can see the difference when you measure the output waveform.
With CF, (+) side (of the waveform) easily achieves to the +B level, and (-) side won't achieve to GND level. The shape will be a bit round. Without CF, (+) is very fast to distort, and (-) achieves GND level. Both side distorts like a square wave. The shape is like a transistor, so fizzy.

2.If it is Marshall type TONE stack, the load is a bit unique for the output.
With CF, the output impedance is very low, so TONE stack itself won't affect the frequency response or gain of the output.
Without CF, the TONE stack (mainly 33k 0.022u) will be the load for the output, so >100Hz range decreases 6dB gain, and <100Hz 3dB. In other words, the lower frequency sounds like enhanced. This will cause less smoother sound. If Fender type stack, 100k 0.022u will be a load for all frequency, so the output simply decreases 3dB gain.

I don't think you need the CF switch as a function, but I think it's very important for you to try the difference. You can fix the circuit after you find the better. Cheers

[MWJB]

"i think i read that a cf can compress the fizz and smooth the tone." Simply adding CF stage, with the current number of preamp stages, is unlikely to achieve this. CF stages often impart a raspy, crunchy character, compared to a regular plate driven stage. Fender went with it primarily to stop the tone stack from loading the signal & to limit gain.

It might be useful to post your schem, there may be other relevant factors.

[mooreamps]

Actually what Mr.tubeswell says is correct if it's a text book. But we experience in a guitar amplifier is not common with an audio application.
1.We drive the stage such heavily.
2.The load is very unique.

Absoluty correct. For a more correct application of electrical theroy for guitar amps, I would point folks to study the TUT series published by LondonPower.


-g

[yunger]

wow, i forgot about this post. unfortunately, i haven't got around to testing w/ a cf yet. i guess i should have clarified that i was specifically referring to dc coupling to the previous stage. we'll give it the ole ear test soon enough.

[Enzo]

Hi MLee, welcome to the forum.

Just wanted to let you know that the book Tubeswell referred to is about designing guitar amplifier circuits specifically, not high fidelity amplifiers..

[pdf64]

ac coupled cf
The Valve Wizard -Cathode Follower

dc coupled cf
The Valve Wizard
Note the bootstrapping section at the bottom

However, the 'fizzy and harsh' tone may be more due to blocking distortion, see
Blocking Distortion
pre-amp section at the bottom of the page.

[daz]

Least techy guy here, but i do have ears and i did spend a couple years + trying to make a marshall style design i built sound killer. If you only knew what i did to it in that effort you might be shocked. At HOW MUCH that is. I tried everything you can imagine and some things you can't. The thing that finally make me say "holy %$#@, thats it !" was removing the CF ! I'm not saying thats gonna do it for you. But for me everything just kinda got right after that. all the things i never felt were close enough to perfect and kept me from having a truly killer tone were rectified completely or close enough so that i ended up with an amp i feel is jaw dropping and better than anything of it's type i've ever played. And it all started by removing the CF and simply driving the tone stack from the plate of the 3rd stage. It made me wonder why anyone even uses a CF. It appears to have been the one thing holding the amp back from being absolutely killer. If i were you i'd simple remove it. It only takes a minute to rewire the amp to run off the previous stage's plate, but you WILL have to then change some other things to optimise the tone for the new circuit. When i removed the CF the amp seemed better, but it wasn't a "holy $%#@" moment....i had to change a nuumber of things in the preamp to compliment this change. Once i did then it became apparent how good a change i had made. This is also why i would suggest NOT to make it switchable. You can optimise the tone for CF or for no CF, but i don't think for both. I may be wrong, but in my amp at least it wouldn't work well because one way would sound great and the other i'd never want to use because it wouldn't sound near as good w/o the rest of the circuit being optimised for that. It would seem you'd need 2 completely seperate gain stages to really make it worthwhile, and even then one way will likely sound much better. So i say no to making it switchable, but a big YES to at least trying it w/o a CF. I thought my amp was pretty killer sounding till i did it. then i realize what "killer" really is.

[wizard333]

What you're hearing, I'd be willing to bet, is just what commonly happens when you build something like you described: two (or more) gain stages then a plate driven tone stack. My advice for a cure is a bleeder cap across the plate resistor of the stage driving your tone stack. Somewhere in the range of .001uf to .022uf, experiment for what suits your circuit and your ears. This is a SUPER common trick used to reduce fizzyness, I use it all the time, most high gain designs use it, many over-use it. The key is finding the value that kills the fiz but doesn't make your amp sound dull.

Much, much easier than adding a switchable CF, and much more effective for the problem you are trying to solve.

You can just rig up some wires with aligator clips on both ends (insulated clips, you dont want to electrocute yourself) and clip in various values of cap til you find what you like. I just do that and hang em over the front of the amp, clipping in different values until I find the one that works best for the circuit.

[yunger]

wizard333
What you're hearing, I'd be willing to bet, is just what commonly happens when you build something like you described: two (or more) gain stages then a plate driven tone stack. My advice for a cure is a bleeder cap across the plate resistor of the stage driving your tone stack. Somewhere in the range of .001uf to .022uf, experiment for what suits your circuit and your ears. This is a SUPER common trick used to reduce fizzyness, I use it all the time, most high gain designs use it, many over-use it. The key is finding the value that kills the fiz but doesn't make your amp sound dull.

Much, much easier than adding a switchable CF, and much more effective for the problem you are trying to solve.

You can just rig up some wires with aligator clips on both ends (insulated clips, you dont want to electrocute yourself) and clip in various values of cap til you find what you like. I just do that and hang em over the front of the amp, clipping in different values until I find the one that works best for the circuit.

i did think about this actually. i tried that with a dual ganged volume pot so that most of the high frequency attenuation went into effect when the tubes started going crazy...that way i could still have those highs with cleaner signal. the only difference was i cut the highs by way of miller effect before the second gain stage.
i still plan on trying out the dc coupled cf because i will always wonder if it was a square wave thing that i didn't like. there has to be something to that because i've heard amps with way more gain in the first 2 stages preceding a dc cpld cf and they sound awesome to my ears. we'll see.

[wizard333]

I'm assuming that using a dual ganged pot you had a bleeder cap on the pot dumping high end directly to ground; i.e. grid to cathode? Or were you going grid to plate? Not sure which part of the "miller effect" you were using. What I've found about using that method vs. going across the plate resistor is that it makes the pick attack and high end dull and tends to kill touch sensitivity. A lot of amps do it and invariably I find the affect on touch and pick attack tremendously annoying. I had a thread on here a while ago discussing why going across the plate resistor yields such different results vs a direct cap to ground, and the best possible explanation put forth, iirc, was by Merlin who suggested the difference might have to do with changing the inter-stage capacitance. Whatever the technical reason, its obvious to the ears and fingers.

I'd also tend to go across the last stage, I think a smaller cap can be used at that point.

In any case, direct cap to ground = bad juju, cap across plate resistor = fuzz gone without bad juju. I've been able to do builds with a screaming amount of gain on tap doing that without any obnoxious fizz, and without any bad affects on pick attack or touch.

In terms of keeping highs with lower gain settings, you can always turn your treble up, its what I do. How often are you reaching back during a song and rotating your gain knob?

I'm curious to hear your results with the CF in there, I can't think of a reason it would make a difference. Not sure how much gain you are using but I've seen higher gain Marshalls where, even with the CF, they have bleeders in the circuit to prevent fizz. Same thing on some Mesas. Both companies tend to over-do it on the bleeders IMO, but if you just remove them all its fizz city.

[Tonewood]

...i had to change a nuumber of things in the preamp to compliment this change. Once i did then it became apparent how good a change i had made. .

What were some of these things?

Thanks

[Chuck H]

I had a thread on here a while ago discussing why going across the plate resistor yields such different results vs a direct cap to ground, and the best possible explanation put forth, iirc, was by Merlin who suggested the difference might have to do with changing the inter-stage capacitance. Whatever the technical reason, its obvious to the ears and fingers.

I'd also tend to go across the last stage, I think a smaller cap can be used at that point.

In any case, direct cap to ground = bad juju, cap across plate resistor = fuzz gone without bad juju. I've been able to do builds with a screaming amount of gain on tap doing that without any obnoxious fizz, and without any bad affects on pick attack or touch.

I never got this either. How is a cap that goes to ground via a really big cap any different from a cap that goes directly to ground post coupling cap???

I hate bleeder circuits in all their forms. But if you've ever tried to build without them you know that the glassy highs can be annoying. This, for me, is true of vintage amps as well. Cranked up Marshalls can sound great, except for that little glassy grit on the tippy top. The right speakers will squash that but you then lose any ability to get good top end with a clean tone.

My least hated top end bleeder is a very low pf cap (3pf to 10pf depending) between the grid and plate pins of a preamp triode. I've also used conjunctive filters with a comparably small cap value (1500pf) to smooth out voltage spikes appearant in output clipping. These are the two least obnoxious bleeder circuits I've used.

Chuck

[daz]

the thing that finally worked for me in removing the fizzies are grid stoppers. i always hated them because they always seemed to knock out too much high end and the tone always suffered similar to a bleeder cap. then when i changed the circuit for the umteenth time a while back, that circuit somehow seemed to benefit greatly from them. i always used them to help with oscillation and blocking problems, but the by product was a dull tone. Till now that is. And the benefit is an end to that harsh or fizzy high end. I since haven't had a need to remove highs at all.

What were some of these things?

You name it. I pretty much changed most every value in the first stages including filtering, plus adding the grid stoppers i mentioned above. But heres the kicker....since i last posted above about how much i like the tone with the CF removed, turns out it wasn't that at all. It was the things i did to compensate for the new plate driven stack. then it hit me when changing some of the things in the first stages in effeor to make it even better than anything i changed made this tone that i was really digging go away. it then hit me it probably were those changes i made to compensate for the CF removal that were what sounded so good, not the removal of the CF. I then went back to a CF stage and sure enough the great tone is still there and even better. crazy shite huh? Amp design is like a puzzle with infinite pieces, so it never ends. But every time i go thru something like this i learn new thing, and this one taught me a heck of a lot including a lot about filtering, which is a lot more important to tone than some will tell you.