GA40 Tone Control Question/Using 5879 in Marshall Topology

[Ed_Chambley]

I have not had the opportunity to play a Gibson GA-40.  Oh the horror, I know an amp I have not played.  Anyway, it looks as if the tone is in series with the cathode cap of the 6V6.  Not sure how it works Someone?

Anyway, I am trying to determine how to add a 5879 Preamp to a KT66 JTM45 I have, but have concerns of it preceding the tonestack.  I have read (Merlin) that pentodes are not good to use prior to a tonestack, but I have used them and do not understand.  The EF86 driving the tonestack in the Dr Z Route 66 works very well.

The chassis is a 4 holer currently with a split cathode 12Ax7.  Anyone know of a way, without adding another tube, to get the 5879 to have dual inputs like the split cathode so it is still possible to mix the inputs?  Just hoping here.  I can go back to a bright cap, but I am just checking if anyone has ever done anything like this and what was the outcome?

[sluckey]

The tone control on that schematic is a simple treble bleed type connected between V4 pin 6 plate and ground. It has nothing to do with the 6V6 cathode cap.

I'm not exactly clear about your 5859 question. But if you just want to add a 5879 preamp to an existing JTM45, then build the preamp and connect the output through a 220K to the .02µF input cap of the PI. You also need to put a 220K between the treble wiper and the .02µF input cap of the PI. These two 220Ks will mix the two preamps together just like a typical Fender blackface amp. (You may prefer 470K mixing resistors but I like 220Ks)

[Ed_Chambley]

The tone control on that schematic is a simple treble bleed type connected between V4 pin 6 plate and ground. It has nothing to do with the 6V6 cathode cap.

I'm not exactly clear about your 5859 question. But if you just want to add a 5879 preamp to an existing JTM45, then build the preamp and connect the output through a 220K to the .02µF input cap of the PI. You also need to put a 220K between the treble wiper and the .02µF input cap of the PI. These two 220Ks will mix the two preamps together just like a typical Fender blackface amp. (You may prefer 470K mixing resistors but I like 220Ks)

I see that tone control now.  Wiper is simply connected to ground.  Thanks.

What I am planning to do is replace the existing preamp, not add.  I played an amp that has 2 5879's in V1 and V2 with a TMB stack.  4 hole Marshall with 2 different voicings.  One brighter than the other, but neither very bright and both usable alone.

So the tube lineup is 5879,5879, 12Ax7 PI, and 2 KT66.  The builder would not share what he has done.  I would like to try the same or similar with a 5879 for a Normal and a EF86 for a Bright and be able to mix them.  I do not know how the TMB tonestack is driven which is what has me confused and he swore no MOSFET CF.

How is it possible?

[tubenit]

I am trying to determine how to add a 5879 Preamp to a KT66 JTM45 I have

See if you can find some ideas from this schematic to use for your JTM45.

5879 into a CF into a tone stack will work.   You can even use a 5879 into a mosfet CF into a tone stack if you want.  That works well.
You can have the 5879 into the CF be in the V1 gain position ............. OR use it in the second gain stage position.

OR have 5879 be V1 and the tone stack be after the cathode follower in V2.

With respect, Tubenit

[tubenit]

And another idea ........... NOT drawn correctly really but you get the idea hopefully of caps on a rotary switch.

I attached the SCH file if someone wants to redraw it correctly

with respect, Tubenit

[octal]

I built a GA-40 lite (-the trem on ch 2.) I configured the second channel as a bright channel by using a very small cathode bypass cap. I'll have to check the value and get back to you.

Please note the "backwards wired" 1m volume controls and the 470K mixing resistors in the GA-40 conspire with the Miller capacitance of the PI tube to make for a lot of built in treble rolloff.


So, just installing a 5879 in a Marshall will probably not duplicate the tone of a GA-40 unless you deliberately do something to kill more treble.


hope that helps.


Nathan

[sluckey]

The builder would not share what he has done.

Stingy builder!

Looks like tubenit has you covered.

[Willabe]

Please note the "backwards wired" 1m volume controls and the 470K mixing resistors in the GA-40 conspire with the Miller capacitance of the PI tube to make for a lot of built in treble rolloff.

Same as a Fender tweed 5E3 Deluxe and 5G9 Tremolux.

[Ed_Chambley]

The builder would not share what he has done.

Stingy builder!

Looks like tubenit has you covered.

Yes, I got a great idea from Tubenit and I do not believe the amp does not have MOSFET.  When I asked him that he quickly became defensive and gave me the ole "It is ALL tube" like a tonestack cares.

I have never done a MOSFET CF, but I can simply replace the normal 12AX7 with 5879 and the Bright 12Ax7 with the EF86 and use a MOSFET CF.  Time to learn something new.

[Ed_Chambley]

I built a GA-40 lite (-the trem on ch 2.) I configured the second channel as a bright channel by using a very small cathode bypass cap. I'll have to check the value and get back to you.

Please note the "backwards wired" 1m volume controls and the 470K mixing resistors in the GA-40 conspire with the Miller capacitance of the PI tube to make for a lot of built in treble rolloff.


So, just installing a 5879 in a Marshall will probably not duplicate the tone of a GA-40 unless you deliberately do something to kill more treble.


hope that helps.


Nathan

Thanks, but I really not trying to build a GA40, I just knew the amp uses the 5879 in V1 and I referenced it for this reason.  Also the original GA40 Schematic gives me the voltages I need to shoot for.

I was going through a bunch of tubes I just got and there are 3 sleeves of RCA 5879 new.  I did not know RCA even made the tube, I thought Sylvaina made the majority of the 5879.  These definitely look different than the Sylvania's I have.

Time to use some of them.  I put one in the Dual EL34 SE amp and I really like the tone.

[tubenit]

When Geezer and I worked on the Tweed Overdrive Special,  we just used a 5E3 type tone stack.  And I thought it worked just fine and was quite happy with it.  If it is supposed to not work .............. I didn't get that memo?
 

I typically do a pseudo 5E3 that you can see on the D'Mars ODS overdrive tone stack. Pretty similar really.

With respect, Tubenit

[Ed_Chambley]

When Geezer and I worked on the Tweed Overdrive Special,  we just used a 5E3 type tone stack.  And I thought it worked just fine and was quite happy with it.  If it is supposed to not work .............. I didn't get that memo?
 

I typically do a pseudo 5E3 that you can see on the D'Mars ODS overdrive tone stack. Pretty similar really.

With respect, Tubenit

I agree, it is just Merlin's preamp book speaks of the tone of a pentode and he makes a statement that their use is better for tone when used later in the circuit.  For overdrive such as you have.  I have quite a few amps with a pentode in V1, but the amps either have cap switching (Ac15) or a 12Ax7 driven TMB except for one amp which is the Route 66 type which has Treble and Bass plate driven EF86.

I see you have a plate driven tone control from the 5879.  What are your thoughts on driving a Marshall type TMB using it?  Have you ever seen a Cathode driven tonestack using a Pentode?

I am trying not to hack up the chassis, but I am losing 2 triodes if I replace V1 and V2 with a pentode.  I am thinking to do what I want I will have to add a 9 pin socket or a MOSFET.  That is if I plan to keep 2 Volumes, TMB and Presence.  Otherwise I will simply lose 1 triode if I only use the one 5879.  This still will make it a 1 channel amp, but the original JTM45 had a shared cathode.

I guess I could do this first or bite the bullet and add a tube socket.  I am just really wondering what having an EF86 and 5879 dual channel KT66 would sound like.  The dual 5879 I played was very responsive, cleaned up great and would really get nasty.

[tubenit]

Just a couple of thoughts for you.  PLEASE hang a question mark over them because I do.

I'm thinking one channel of EF86 and then another channel of 5879 may not offer enough of a difference to be that useful?  And I could be wrong about that? 

As best as I am understanding it ............. I am thinking a 5879 with a mosfet cathode follower with tone stack may be your better answer.

With respect, Tubenit

[tubenit]

I thought I had drawn up something at one time with two 5879 gain stages.  This was drawn around 7/2013.

I built one amp where I tried about 5 different wirings for the preamp section.  I might have tried this when I was experimenting with that
but can't remember for sure?  I drew up a bunch of stuff and kept trying different designs out to see what I liked.

I've attached the SCH schematic and layout if anyone is interested?

With respect, Tubenit

[DummyLoad]

The tone control on that schematic is a simple treble bleed type connected between V4 pin 6 plate and ground. It has nothing to do with the 6V6 cathode cap.

I'm not exactly clear about your 5859 question. But if you just want to add a 5879 preamp to an existing JTM45, then build the preamp and connect the output through a 220K to the .02µF input cap of the PI. You also need to put a 220K between the treble wiper and the .02µF input cap of the PI. These two 220Ks will mix the two preamps together just like a typical Fender blackface amp. (You may prefer 470K mixing resistors but I like 220Ks)

I see that tone control now.  Wiper is simply connected to ground.  Thanks.

What I am planning to do is replace the existing preamp, not add.  I played an amp that has 2 5879's in V1 and V2 with a TMB stack.  4 hole Marshall with 2 different voicings.  One brighter than the other, but neither very bright and both usable alone.

So the tube lineup is 5879,5879, 12Ax7 PI, and 2 KT66.  The builder would not share what he has done.  I would like to try the same or similar with a 5879 for a Normal and a EF86 for a Bright and be able to mix them.  I do not know how the TMB tonestack is driven which is what has me confused and he swore no MOSFET CF.

How is it possible?

something active is hidden? maybe something like in the attached schema?

--pete

[Ed_Chambley]

Just a couple of thoughts for you.  PLEASE hang a question mark over them because I do.

I'm thinking one channel of EF86 and then another channel of 5879 may not offer enough of a difference to be that useful?  And I could be wrong about that? 

As best as I am understanding it ............. I am thinking a 5879 with a mosfet cathode follower with tone stack may be your better answer.

With respect, Tubenit

Consideration noted, however the amp I played had 2 5879.  The normal channel was nice and thick and very clean and this is what got my attention first.  The channels were internally jumper-ed because he had a Master in one of the holes.  He did give the type of master.  He said it was a Orange/Matchless style which is what it looks like Pete has drawn.

Now plugging into the bright channel is where you get the nasty overdriven tone.  The problem with his design is you cannot cleanup if the bright is too high and there is no way to use a AB switcher and he had on channel switching.  So if you dial in too much of the bright channel you are sort if stuck in distortion.  Plus there is too much gain from the bright channel 5879 by itself for my taste preferring the cleaner side of tones.

That is why I considered a different tube.  You are probably right, but also consider guys have been jumping Plexi's to mix a 12Ax7 which is the same tube and getting vastly different tones and that is why I thought 2 different preamp tubes would take the same idea.

[Ed_Chambley]

The tone control on that schematic is a simple treble bleed type connected between V4 pin 6 plate and ground. It has nothing to do with the 6V6 cathode cap.

I'm not exactly clear about your 5859 question. But if you just want to add a 5879 preamp to an existing JTM45, then build the preamp and connect the output through a 220K to the .02µF input cap of the PI. You also need to put a 220K between the treble wiper and the .02µF input cap of the PI. These two 220Ks will mix the two preamps together just like a typical Fender blackface amp. (You may prefer 470K mixing resistors but I like 220Ks)

I see that tone control now.  Wiper is simply connected to ground.  Thanks.

What I am planning to do is replace the existing preamp, not add.  I played an amp that has 2 5879's in V1 and V2 with a TMB stack.  4 hole Marshall with 2 different voicings.  One brighter than the other, but neither very bright and both usable alone.

So the tube lineup is 5879,5879, 12Ax7 PI, and 2 KT66.  The builder would not share what he has done.  I would like to try the same or similar with a 5879 for a Normal and a EF86 for a Bright and be able to mix them.  I do not know how the TMB tonestack is driven which is what has me confused and he swore no MOSFET CF.

How is it possible?

something active is hidden? maybe something like in the attached schema?

--pete

I guess he could have had a tube inside the chassis since he said he had no MOSFET, but I doubt it.  To me it is not difficult to add a tube to a Marshall 50 watt chassis.  I have added one to mod a few amps.  You know if you have another 12Ax7 you will sound just like Slash!

Anyway, thanks for all the input guys.  I opened up the amp last night to see how much room I have to work and there is plenty.  Now I gotta go check out a couple of tubes you have in the drawing that I am not familiar with.

[DummyLoad]

 i'd use depletion FET pair: LND150 one as a self-biased source follower to drive stack and second as grounded source amp for recovery. FETs are powered by B+ rail.

either loop PCB easily adapted to drive TMB stack and hidden. no sand state.

http://riotdevices.com/wp-content/uploads/2015/02/loops.jpg

--pete

[Fresh_Start]

Why can't you drive a tone stack with a pentode?  If you can drive it with a 12AX7 triode, you certainly should be able to drive it with a pentode. Will you like the tone?  Won't know 'til you try it. You like your Route 66, eh?

I swear that I've seen a schematic with a pentode going into a TMB tone stack and then into a PI. I thought it was a Peavey but can't find it. IIRC the PI was a concertina. IOW another gain stage followed by the actual phase inverter. Now I don't know if a 12AX7 concertina will drive KT66 power tubes, but the low gain/high current side of a 12DW7 might. Again, I know I've seen a 12AU7 concertina PI somewhere.  Fixed bias for the concertina and a grid stopper might help get the headroom you need there. See Merlin's site for enhancements to a concertina PI. Also, R,G. Keen has published some work on MOSFETS as power tube drivers...

Guys seem to think the long-tailed PI has more gain. Well, the concertina triode itself has slightly less than unity gain but the stage before it is a perfectly normal gain stage with more gain than an LTPI if I remember right. The downside to the concertina is the potential for nasty distortion when overdriven, Merlin addresses some of that.

One big advantage of using MOSFET (or tube) cathode followerS after the pentodes is that you wouldn't have to worry about mixing resistors before the tone stack. You'd get perfect isolation and greatly reduce the insertion loss from the TMB at the same time.

You know those subminiature dual triodes on the effects loop board make me wonder if that's what's hiding inside your mystery man's circuit - each triode as a cathode follower for one of the pentodes.

Spitballing/daydreaming here: what if one channel uses a pentode and the other parallel 12AX7 triodes Riggs up with a Fat control on one like Tubenit's circuit shown above. That would give you a purpose for the first volume/gain pot you already have and very different tone compared to the pentode.

Hope that helps a bit and isn't pure nonsense,

Chip