Jose Plexi mods

[Spencerleehorton]

Hi all,

Armed now with a great source of knowledge and working my way through this book:

https://www.valvewizard.co.uk/Book1.html

I’m going to attempt to mod my other plexi 1959 50W I’ve built to try out some mods and see how it sounds.

I’ve added another first gain stage into the plexi and having two input jacks, one for standard plexi and the other for the extra gain stage.
The other jack sockets will be replaced with 1M gain pot and 1M master volume, as I’m going to remove the PPIMV pot and put back the two 220k resistors.

I’ve added the gain stage and so far the plexi sounds great at half gain but I’ve not changed the PPIMV to mst vol between tone stack and PI.
When I turn up extra gain stage past half way I get squeals and wah wah overloads the signal?
Could this be the PPIMV?
I’ll add the mst vol tomorrow and see if it improves.

Will also try mst vol pre tone stack to see how it sounds.

 

[Spencerleehorton]

This is how it sounds so far

 

[Spencerleehorton]

Have put in 1M master volume pot and put two 20v diodes back to back across it on outside legs as per Jose mst vol mod.
Gain pot gets to about 3/4 now before the dreaded whistle!!!???
I suspect EMF and more screened cable need, so will go through and tap components and see if I can find the noise.
Suspect the volume pots need to be screened.

 

[Spencerleehorton]

The main aim once I get the high gain going and the normal plexi side all sorted it to put in some relays and two masters to switch between the channels.

Here is a Jose schematic I’m working off but will be fairly different but I will try and alter and post.

 

[dai h.]

IMO the original layout (5F6-A) isn't very good, but one probably has to be fair and consider it was a bass amp and not some high gain thing for guitar. (Consider for example the Soldano SLO100 which seems an evolutionary offshoot which has the preamp tubes very close to the front panel which helps to shorten lead lengths. In a Marshall, the high impedance (pickup) signal goes from the input all the way across the chassis to the grid, then the output goes all the way across to the 1M volume, then (yet again) goes all the way back across the chassis to the next tube's grid. Also the ridiculous(?) long lengths of wire (okay maybe not for a bass amp) esp. to the pots and to the preamp tubes. Big opportunity for inconsistent performance depending on where they are physically. (I once tried using a shielded cable (using the shield to connect signal and not connected to ground) to connect some of those wires around the input/pots area (to see what happened by getting a lot of capacitive coupling) and it sounded weird (I don't quite remember exactly but I think it was some odd fuzzy mosquito sound IIRC)

I think once you start modding for higher gain you have to take more measures like protecting the signal from interacting with other signals that instability and noise (using distance, shielded cable, shielding (you can achieve shielding effect by laying wires tight to the chassis--wires close by will interact less if you put them on a ground plane)). The high gain mods where you see parts mounted to the pins (or quite close) or very short lengths of wire to a PCB make sense to me. I think you want to try to keep loop areas small and carefully manage the relative orientation (and distance) of the heaters to the grid(s) (assuming AC heaters). Also (my novice understanding) of things like caps added to try to quell oscillation: if the performance with them meets your goal they are ok (i.e. the ones that are there are part of the overall design), but they could result in less highs when you have the gain maxed out (I could kind of see that with something like a plate to grid cap as output goes up) so maybe good (conventional) layout takes precedence than trying to add (what some term) "band-aids" after the fact.

Anyway, there seem to be tons and tons of resources nowdays on the Jose stuff (and some sort of resurgence in popularity and appeal) but (possibly?) something to keep in mind is that not everybody seemed to be enamoured with the mods (and to be fair there were apparently differences). (To my recollection) Terry Kilgore (ex. DLR) checked out the Jose-modded Marshalls DLR had (apparently he had a number of them modded by Jose seemingly believing they were part of EdVH's "secret sauce") but TK was apparently not impressed with the amps.

Some Jose mod-related links (maybe you've seen them already?) :







https://www.rig-talk.com/forum/threads/sharing-josé-arredondo-amps-guts-one-exclusive-owned-by-david-lee-roth.318437/

 

[Spencerleehorton]

I’ve also been reading up on valve wizard (Merlin’s) ideal gnd scheme and will make the changes needed.

Found Lyals explanation very helpful.

 

[Spencerleehorton]

Thanks for that resource!! That’s great, I’ve found it hard finding stuff through the other forums and tbh I like this forum and wanted to try and build similar info here but have it more schematic based and get rid of the myths!!
I did noticed on both the Jose schematics that I’ve got now that there is no grid resistor on the extra gain stage, and it keeps the 68k into the next stage.
So have read that you are meant to have a grid resistor of at least 10k but the Jose way doesn’t seem to agree with that?

 

[Spencerleehorton]

Reading through the valve wizard grounding and it mentions several different gnd schemes but for higher gain it says several star gnds which should be connected in sections?
But only connected to chassis at input?
I was under the impression up to this point that the gnd connections would be connected directly to chassis rather than all isolated either on the bus bar or raise isolated points not connected to chassis?
This maybe obvious to those who know but it’s not clear within the description which way it’s done and I have seen several amps where each section is connected directly to chassis?
If someone could clear this up please as I do understand it could be done either way but I want to do it the best way to reduce noise and for it to be as correct as possible.

it maybe clearer if i make a list of all the gnd points and link them to what section they should goto:

IEC gnd = chassis as close as possible

HT CT gnd
Heater CT gnd
Diode bridge gnd
bias gnd
Speaker Out gnd

V0 cathode gnd (extra gain stage)
V1 Filter cap gnd
V1 cathode gnd
V2 Filter cap gnd
V2 cathode gnd (cathode follower)

Tonestack/master volume section gnd

Phase Inverter section gnd

EL34 screen/ cathode gnd (1R)

Input Jack gnd

 

[Spencerleehorton]

This so far is what I’ve found which is closest to what I’m aiming for!

I have the 5v tap on the mains txf for the relays.
Already have pretty much everything, only thing is now I have an extra gain stage!!!
This seems to sort of follow the broken up star gnd?

 

[Walter66]

Thats a great book and its online for free. I downloaded my copy some time ago.
If one thinks that a replica of an old tube amp, build with modern parts, would sound the same as the original: it will never do.

 

[Spencerleehorton]

here is the ideal gnd layout (b) using a bus bar for high gain.
using this i have made a layout for what im purposing and just want to check its correct!!

 

[Spencerleehorton]

i paid for my hard back version of the book

 

[Spencerleehorton]

This is how im going to do the grounding


im going to leave out the extra gain stage at the moment and i will also include v1b cathode on the 5th gnd section.
at the moment everything is star grounded, so hopefully there should be a significant change in noise

 

[dai h.]

(I'm just a hobbyist and I'm not very technically literate, so take it for what it's worth, but) overall, (for grounding and layout) these seem to be important to think about:

V/R = I (current is key)

and

basic transformer action (loops)

So for V/R = I (as you do with the "1 ohm R on the cathode measurement"). As the term "electronic circuit(s)" implies, there are loops everywhere. DC, AC, or both, and if the loop comes back over a ground, there is a voltage drop, so recognizing where the loops are, how much and what kind of voltage drops are occuring are important to sorting how the grounds should be connected together

As far as basic transformer action, you don't want to physically arrange loops lined up in such a way where you are getting some deleterious interaction (noise, instability, distortion (maybe not for a fuzz or distorting guitar amp for that one)). So for example you wouldn't want to make a big giant physical hoop out of the rectifier to primary cap and back to B+ center tap loop, then put the input stage tube inside of that.

As far as using the chassis or not using the chassis, as far as I can observe you can achieve a workable grounding scheme even if you have more than one connection from circuit ground to chassis (so I don't think the notion of "only have one connection from circuit ground to chassis" is an absolute), but the scheme needs to make sense.

(The following article was helpful for me (besides stuff from here, various internet sources, and things on paper, and also staring at chassis and PCBs and trying to make sense of them)) :

https://ampgarage.com/forum/viewtopic.php?f=19&t=220

There is also an excellent ongoing thread there with comments from R.G. Keen that should help to answer some questions:

https://ampgarage.com/forum/viewtopic.php?f=6&t=37513

 

[Spencerleehorton]

i'd be lying if i said i understood grounding totally but i am experimenting with different placement, so far the star gnd scheme seems to be the best with this Plexi / Jose mod amp but i hope to find a better way!!

 

[Spencerleehorton]

Success today with building the Ceriatone JCM51 800/ plexi although I have only connected the gain channel and not the plexi yet as waiting for the relays to arrive.
Sounds a bit ice picky to me so will look at some of the cathode bias and put in a 100pf cap on v1a from plate to cathode.
I’ve also added a 3 way switch and got 2 x 20v diodes and 2 x MPSA06 trannys with 220k to gnd on middle position as per one of Jose mods.

 

[Spencerleehorton]

getting some nice tones now

 

[john12ax7]

One thing to keep in mind, if you are looking to add extra gain stages to a typical marshall style amp you often need to roll off high end somewhere. Or you end up with a fizzy feedback prone amp when you crank the gain. Look at something like an slo, you have large grid stopper resistors and also capacitors across some of the plates loads. Both decrease high frequencies and help keep things under conttol.

Of course grounding and lead dress matter a lot too. But they can often be insufficient on their own with high gain.

 

[Spencerleehorton]

By putting in the 100pf cap on v1a and changing the cold bias from 10k to 5.6k it’s sounding much better now.
Yes I do want to try putting in another gain stage as I’m having so much fun!!!