cathode follower....why?

[iknowjohnny]

I know it's supposed to be less lossy, but is that the only reason to use it? The reason i ask is that i tried driving the tone stack from V2A's plate and bypassing the CF and the amp has a ton of volume still and i think it may sound better. (2204 style amp but not cold bias 2nd stage) And if the only reason is to keep gain losses to a minimum, why is it almost always used in marshalls which have a ton of gain to begin with and usually even drop some with voltage dividers?

I'm not sure what else i should change that might be better with a plate driven stack, so if anyone has suggestions i'm all ears.

[Jana]

Marshalls use the cathode follower because that is what the 59 Bassman had and that is the amp that Marshall copied.

[Andy Le Blanc]

tradition is wonderful thing

each type of gain stage has a characteristic impedance

grounded cathode, grounded grid, grounded plate....

[tubeswell]

A DC coupled CF stage in a 5F6A/JTM45 has unusual harmonic distortion when pushed hard. Its to do with the way the CF cathode 'steals' current from the other stage's plate. Merlin's got a big explanation of this in his book.

[ampgeek]

What, generally, are the impedances for each gain stage style?

When we look at a long tail PI, we see a common cathode and a grounded grid driving, effectively, the same control grid circuit for each power tube in a push-pull topology.

Does the typically seen ~20% different plate load resistor difference between each half attempt to make a better match for each type? Seems like that is more a gain difference gig though.

Wondering how this might play into an observed 2X difference in non-quiescent power tube current draw in a pair of well matched EL34s. I have never seen that in similar 6L6, 5881, 7025 equiped LTPI plans.

The 2X's current draw follows the common cathode/82K plate socket when flip-flopping the tubes so it doesn't appear to be a tube response difference.

Maybe that imbalance is normal for the higher Gm'd EL34s?

Probably should consider "rotating the tires" every 1 million mA-seconds to balance out tube wear!

Thanks much,
Dave O.

[martin manning]

At the risk of becoming an accessory to a thread jacking, both LTPI outputs are at the plates, so the impedance looking back into each will be be similar, differing by about the same amount as the plate load resistors. Are the plate load resistor values what they are supposed to be? Pull the PI tube and measure them. What about the PI tube itself? It may not be balanced under static or dynamic conditions... Pull the power tubes and measure the signal at each output.

More to the original question, the CF's lower output impedance is better suited to driving a lossy tone stack. I imagine this would be better than taking the loss and then having to pull the signal back up again.

MPM

[ampgeek]

Gulp...got a little carried away in my excitement upon seeing a discussion of exactly the topic I was studying at the moment. Please forgive me for veering away from the original question.

But...since the gain stage type and its characteristic impedance is germane to IKJ's question, how would you "rate" each?

Grounded cathode = high?
Grounded grid = med?
Grounded plate = low?

Again, sorry for that slight diversion!
Dave O.

[martin manning]

Zin can be important too, depending upon how the stage is used. Zin/Zout would go like this, I guess:

Common Cathode high/high
Common Plate (Cathode Follower) high/low
Common Grid low/high

[iknowjohnny]

More to the original question, the CF's lower output impedance is better suited to driving a lossy tone stack. I imagine this would be better than taking the loss and then having to pull the signal back up again.

That was why i asked, because even tho i always call my amp "hi gain", it's really not. It's about as gainy as a peavey classic 30's disto channel. Yet the fact is i was able to convert it to a plate driven tone stack w/o losing any volume or anything. The typical marshal MV design seems to have more gain all along the signal path than it needs to drive the PA to ear splitting levels even w/o the CF. Thats why it seems to makes so little sense that marshal is the one that uised it while fender, the one who could probably take advantage of it more doesn't. Anyways, i need to play with it more, but i think it sounds better like this. Again, andone with thoughts about what i might want to change to accomodate this new change? (similar to a 2204 with a regular 2nd stage with a 1.5k cathode R)

[Structo]

I understand the reason for the different values of the PI plate resistors in an effort to balance the output but not all amps offset the two resistor values.

Take the Rocket amp for example, it uses two 100K plate resistors on the plates.
Yet when I measure the voltage there they are usually within two or three volts of each other.

The Rocket also uses a DC coupled cathode follower.

Johnny did you notice any tone difference between the plate loaded and CF tone stack?

Merlin has good articles on the AC and DC coupled cathode follower circuits.

http://www.freewebs.com/valvewizard/accf.html

[Andy Le Blanc]

Id take a look the harmonic spectrum as it develops with increasing clip
for both gain structures. The relative even/odd should show a difference.
Are you trying for crunch tone or are you looking for more clean tone?

That follower arrangement gives more opportunities of control.
You can also see the evolution towards an inverter.

Imbalance in an inverter limits the bandwidth and efficiency but...
it encourages harmonic distortion and color in an inverter that limits even
order products.

Thats the danger with increased feedback in a circuit stage, you'll begin to
loose the color.

[iknowjohnny]

Johnny did you notice any tone difference between the plate loaded and CF tone stack?

yeah. Like i said, i think i like the plate fed one, tho i must admit i have only played with it a few minutes. Once i get around to playing it some more i will be more sure. But i wanted a tone that would lean a bit more towards fender in the way the gain felt....less compressed, more dynamic due tio the ability to use LESS gain w/o becoming lifeless. Marshall tone seems not to work as well at low gain levels to my ear. At least THIS particular amp anyways. I think it did that to a degree but i need to do 2 things...play it more and see what things need to change to accomodate the plate fed stack best.

[iknowjohnny]

To update the last post, i played and tweaked it some tonite and i'm really digging this. what i was trying to accomplish by eliminating the CF in favor of a plate driven stack was to get a more fenderish kind of dynamics instead of the more compressed marshall tone. yet i want to retain certain aspects. I think i succeeded quite nicely so far, tho this requires starting from the beginning again in tweaking the amp ! At my age I could possibly die before i finalize this thing ! LOL! But it seems to be what i probably needed all along because there are several aspects of the tone that are doing more of what i wanted all along and could never seem to change no matter how many tweaks i did before. theres definately something quite different to a plate driven stack compared to a CF. I wasn't sure how much at first, but after a few tweaks i'm finding i am able to use values and things that before didn't work well and now work great. there is potential to change things to sound and feel very different than before. Fun stuff !