Cathode Bypass stages for Tube Amp

[deuce42]

Hello, I am wondering if someone could look at the attached diagram of a circuit I have built based upon a design in a commercial boutique amp and offer any advice about the Cathode Bypass values.

By way of background, I would like the preamp circuit to be that of a Marshall 1959 Super and my first diagram is the preamp section of the Marshall.

The second diagram is the circuit I am using but am confused with the Cathode Bypass sections for the first two gain stages. I'm confused because the original marshall 1959 has both a hi and lo input and so the first tube has 250uf/820r on one side for its CB and 0.68uf/2k7 on the other side. In the circuit I am using however, there is only one input and it uses two separate tubes for its first two stages. I am therefore unsure which CB values should be used for tubes 1 and 2 respectively.

On a final note, can anyone offer any advice on whether the caps after the plates in both tubes should be 2n2 or 22n?

I'd be so grateful for any advice.

Thanks very much

[LeftyStrat]

I'll try and give some background, hopefully someone more knowledgable can chime in with corrections.

The earliest Marshalls actually used a shared resistor and bypass cap for the two input channels, typically 820R and 250uf. This was directly lifted from the Bassman. Notably the JTM-45 used this and with guitar it lets too much bass through and can get 'woofy.' Smaller cap values will roll off the bass and appear to tighten or make a more defined the bass response.

Later Marshalls separated the cathodes into a bass channel and a lead channel. 820R/250uf bass, and 2.7k/0.68u lead.

Now some guys liked this, they would jumper the channels and use the two volume controls to dial in a mixture of what sounded best for them. Others always preferred the 2.7k/0.68uf. And so that second input and triode would go unused.

In the eternal quest for More Gain(tm), the "only uses one side" camp decided that rather than have two inputs with one going to waste, you could drop an input and have an extra gain stage, which is what the second diagram looks like. It is a common 'mod' to plexi's for players that wanted more gain.

Other than the extra gain stage mod, a lot is up to personal preference. As a strat player, I always felt the 2.7k/0.68uf was a little extreme, probably better for humbuckers. However, the 250uf didn't suit me either.

I like 820R/5uf or 2.7k/1uf for the input stage. The thing to remember, sometimes less is more, and the smaller the bypass cap, the less bass, but it can actually make the bass response more 'defined'.

I would think the coupling caps would be 22n/0.022uf.

I might be more able to help you dial in what you're looking for if you can give an example of the classic Marshall tones you're looking for. An unmodified plexi is not a high gain amp. It does a beautiful crunch though.

[brewdude]

Hello, I am wondering if someone could look at the attached diagram of a circuit I have built based upon a design in a commercial boutique amp and offer any advice about the Cathode Bypass values.

By way of background, I would like the preamp circuit to be that of a Marshall 1959 Super and my first diagram is the preamp section of the Marshall.

The second diagram is the circuit I am using but am confused with the Cathode Bypass sections for the first two gain stages. I'm confused because the original marshall 1959 has both a hi and lo input and so the first tube has 250uf/820r on one side for its CB and 0.68uf/2k7 on the other side. In the circuit I am using however, there is only one input and it uses two separate tubes for its first two stages. I am therefore unsure which CB values should be used for tubes 1 and 2 respectively.

On a final note, can anyone offer any advice on whether the caps after the plates in both tubes should be 2n2 or 22n?

I'd be so grateful for any advice.

Thanks very much

It looks like you want to change it to two cascaded triodes instead of two parallel triodes. I don't think it has to be two separate tubes.

If you want to build to the second diagram's specs:
It looks like the first stage is 820R/.68uF with a .0022uF coupling cap.
The second stage looks like 2.7K/1uF with a .022uF coupling cap.

[deuce42]

I'll try and give some background, hopefully someone more knowledgable can chime in with corrections.


Other than the extra gain stage mod, a lot is up to personal preference. As a strat player, I always felt the 2.7k/0.68uf was a little extreme, probably better for humbuckers. However, the 250uf didn't suit me either.

I like 820R/5uf or 2.7k/1uf for the input stage. The thing to remember, sometimes less is more, and the smaller the bypass cap, the less bass, but it can actually make the bass response more 'defined'.

I would think the coupling caps would be 22n/0.022uf.

I might be more able to help you dial in what you're looking for if you can give an example of the classic Marshall tones you're looking for. An unmodified plexi is not a high gain amp. It does a beautiful crunch though.

Hey thanks for that Lefty Strat- Guess I will admit how boring I am in wanting, just like every other unadventurous greenhouse variety guitar player to have a Hendrix Band of Gypsies and Wind Cries Mary sound, played with a front pickup strat, and also I'd like a tight and hard, back humbucker, crunch rhythm sound like a Marshall is known for.

And Brewdude, unfortunately I already made the PCB for the circuit so need to use both tubes cascaded. I can see the original digram values for on the circuit diagram I am using but feel that the circuit sounds a bit plastic sounding and not very vintage organic.

Cheers

[niversen]

If by plastic you mean fizzy, you could try lowering the cathode follower resistor from 100k to 68k. That can help get a more "organic" overdrive sound. I've used that trick before with success in marshall style preamps. Might not be what you are looking for but it's an easy mod.

[rock_mumbles]

As sort of a rule of thumb (but there are no rules)

If you plan on driving the amp into distortion:

With a cold biased partially bypassed stage 2.7k||0.68uf you can use a 0.022uf coupling cap because the cathode is controlling the low frequency content getting to the next stage.

With a center biased fully bypassed stage 1.5k||22uf or warm biased fully bypassed stage 820 ohm (or so) with a 22uf (or larger) bypass cap you will need to reduce the coupling cap to something like a 0.0022uf cap to control the low frequencies.

The combination of the resistor and cap form an RC high pass filter that are used to control low frequency content.
That can be done with the cathode resistor and cap or it can be done with the coupling cap and following grid reference (resistance to ground)

[LeftyStrat]

Hey thanks for that Lefty Strat- Guess I will admit how boring I am in wanting, just like every other unadventurous greenhouse variety guitar player to have a Hendrix Band of Gypsies and Wind Cries Mary sound, played with a front pickup strat, and also I'd like a tight and hard, back humbucker, crunch rhythm sound like a Marshall is known for.

Not boring at all, I've spent my life (I'm 53) searching for those sounds. I can't help with Band of Gypsies, still trying to figure that out.

But if you're looking for any of the tones from Hendrix at Monterey:

http://www.youtube.com/watch?v=LSytbEwg2mM
http://vimeo.com/19982030
http://www.youtube.com/watch?v=iqvU5CVHHmk

I've attached the schematic for the actual amp he played. Note that he also used a dual showman in this concert.

Specific notes about the amp. This is from my own notes and those of Greg Germino:

Lefty:

"Jimi's Monterey Marshall:
Change the early tone stack to 500pf/33K
27k negative feedback"

Greg:
"The other aspect of the JH amp that is not disclosed but is there for the astute eye to see is the pre-amp and tone stage are decoupled by a 10K resistor and each stage gets 32uf of filtering. This is pretty far from the two stages not being decoupled and sharing 16uf. This tightens things up and drops votage at the pre-amp slightly. It's a feel thing and does affect the sound as well."


Hope this helps. Just realize it is a tone journey to find yourself. I ended up liking a few different variants, but Jimi says 'Look for me, but find yourself.'

Okay I made that last part up.

[brewdude]

In the JH schematic, it appears that V5 and V6 do not have grid resistors. Is this drawn correctly?

[LeftyStrat]

In the JH schematic, it appears that V5 and V6 do not have grid resistors. Is this drawn correctly?

That's most likely a mistake i the schematic.

[surfsup]

If by plastic you mean fizzy, you could try lowering the cathode follower resistor from 100k to 68k. That can help get a more "organic" overdrive sound. I've used that trick before with success in marshall style preamps. Might not be what you are looking for but it's an easy mod.

I second this as well. I did this on a high gain amp of mine (100k || 100k for 50k overall) and it made a very nice sounding difference. I don't know the effect this has on cleans as this amp isn't really a clean amp at all. I do plan to put one of the gain stages on a switch to clean it up some, eventually. I'll find out then.

[roberto]

This was done some years ago by Bogner in the Shiva amp.

Have you tried something similar? Both coupling caps are 2n2, because with three bypassed stages and simple -6dB voltage dividers you need very few bass to avoid fizzyness. With a lower CF resistor (dc coupled to a 100k 820 stage around 300V) you obtain less fizz but also less dinamics.

It's better to remove unwanted bass (not more than 1µF as cathode bypass) and annoying fizz (Rp snubbers on 1st or 2nd stage, or grid stoppers on 2nd or 3rd stage).

[surfsup]

Roberto the amp had three gainstages into the CF:

0.68/2k7 - 0.022
1.0/1k2 - 0.022
1.0/820

[roberto]

Yeah, I know they are three. Where have you read about the 1k2 on second stage? With -6dB voltage dividers and all that gain on lows (ok it's 1µ, but you have a 1k2 value in parallel, so there a small difference between bypassed and unbypassed gain), 22n on 1M load is too big. I suggest 2n2 for both stages.

[Roe]

In the JH schematic, it appears that V5 and V6 do not have grid resistors. Is this drawn correctly?

That's most likely a mistake i the schematic.

nope this is how marshall did things

[Roe]

68-69 superleads had 820R/.68u on v1a. this got changed to 2k7 later on