Effects of previous stage on CF

[Reeltarded]

So what I understand from the reading is that since the CF is sucking current off that previous stage it.. umm.. Ok, I don't understand it at all. Truth be told I probably don't need to understand.

Please, could you suggest a couple of useful cathode values for the stage previous to a follower? Is the reason for 820r in a Marshall because of available current from that stage? How high a value could I go on a stage there? Are there any limits that should be observed?

I want to add a second follower after the 3rd stage, the typical 820R being the next stage, or 4th in this instance. It will take a bit of teardown, so I kind of want to have a plan before I start yanking the board.

I was looking at scopes of what the follower does and it made me think I would like an inverted signal that already has the other phase peaks whacked off the top before it arrives at the second.

[martin manning]

What have you been reading?
What amp are you talking about?
Why do you want to put a second CF in it?
Why there?

[Reeltarded]

Reading all kinds of stuff.

Marshall, just before the 820r CF TS.

I want to have compression on the other phase. Squishier.

[Firestorm]

Merlin talks about this, kind of like it was an accident of design in the 5F6A (and thus, in Marshalls). Something about B+ being a little too high and Rk being a little too low, so the tube starts to conduct before the signal swings positive. Haven't looked at it in a while. But I think the point was more compression on one side of the wave, versus the other, hence lovely harmonic distortion. But I would think if you went after the effect for the other phase, you might risk undoing it. I guess it would matter where in the signal path you tried to do it again. Interesting, but over my head.

[LeftyStrat]

I've always been intrigued with this ever since reading Merlins account. I'm not sure how critical the previous stage's cathode resistor is, since Merlin claims it happens in Vox designs also, and they use 1k5 with no bypass cap.

Way back when I first saw the Trainwreck schematic, I actually experimented with using the 10k un-bypassed cathode in place of the 820R in a JTM45. It seemed to increase sustain, but the amp lost a lot of the crunch associated with Marshalls. I think I remember someone on here commenting that got the same result when they tried it.

However, if you're already getting the classic Marshall crunch from somewhere in the circuit you may want to try:

v4a 100k/10k
v4b 0/100k CF

Might be fun to put 820R in series with a 50k pot to tune it. The SLO uses 39k.

[Reeltarded]

I still want to keep the asymetric peaks. 820 is bp .68u before 56k CF, so I am thinking unbypassed 3.3k with 100k on the stage+cf right before that last stage.

Just an experiment for now.

[roberto]

Look at simulations on my sticky thread.
You can see the effect of various configurations of various CF and previous stages.

For certains modern hi-gain sounds I use 220k 820R (bypassed) to a 47k CF but you can use whatever you like. Just keep the tube below its limits.

[Firestorm]

Look at simulations on my sticky thread.
You can see the effect of various configurations of various CF and previous stages.

For certains modern hi-gain sounds I use 220k 820R (bypassed) to a 47k CF but you can use whatever you like. Just keep the tube below its limits.

,
I haven't taken enough time to study what Roberto posted, but it looks like it's all there. You can see the flattening of one side versus the other in the CF.

Sorry, Roberto, I didn't offer you more respect. You might have out-merlined Merlin.

[roberto]

we covered different aspects.

Merlin talked more about different CF resistors values.
In my thread i talked more about different interactions with the previous dc coupled stage.

[Reeltarded]

Ok! I think I will take a run at 1k-3k range with 100k CF just to see if it's worth the thought.

Ya know.. I have an unused triode.

[roberto]

There was a nice idea rumbling on the forum time ago:
use the unused triode as a bias tremolo.

[Reeltarded]

Noooooooooo! Not in this one.