New JTM45-based Project

BenTobith

Tele-Meister
Joined
Jan 3, 2020
Posts
231
Location
Austin, Texas
Hi all – I am preparing to start a new build project based on a JTM45.

I already have an 18-watt Marshall build for my "cranked Marshall" sound, so this amp is primarily going to be used as a cleaner amp and with pedals. However, I am going to try and setup the High Treble channel to be a nice, Plexi-like channel for when I want it.

I'm making a few modifications, mostly meant to help either with reducing hum, noise, and hiss or to lower the woofy, boomy, excessive (IMO) low-end and lower mids you can get with a stock JTM45 and humbuckers, especially on the Normal channel. For example:
- "Larry" grounding scheme
- Split cathode for v1
- Adjusted capacitor values throughout for helping tune low end and lower mids
- No bright cap on pots or mix resistor

My layout is linked, and would love feedback on this and any other ideas you might have, based on the above.

DIYLC File: Link
PDF: Link

JTM45 Layout.png
 

2L man

Tele-Afflicted
Joined
Nov 23, 2020
Posts
1,437
Age
62
Location
Finland
To me "Larry grounding" seems operationally very strange and quite complicated as well. It uses chassis for secondary return current and therefore make it an antenna for electromagnetic noise. It also send electromagnetic noise to Mains Safety Earth wire now when even in USA ;) you have "safety earth"- system. Also to instrument cable Cold wire which is wrong when in electronics their roles are "shield" against electromagnetic noise. Just yesterday I was replied it is German "Larry" whose idea it is and that is strange because I have thought German engineering to be very advanced :(

"Star grounding" is simpler to make and no current flow thru chassis so chassis operate better as a Shield against electromagnetic noise and return current "pollute" everything less. One wire for B+1 plate/B+2 screen and then one for each following B+_ stages. First pre amp stage is is connected to Chassis/Safety Earth at input jack and then it "shorts" the noise best possible way there is when three wire Mains and two wire symmetrical instrument cable is used.

If needed now it is very easy to install a Ground Loop Hum Killer-circuit when Common "ground" is connected to Chassis only at amp input.

Here in Finland "ground" was decided to divide to Safety Earth on "primary-side" and Common/zero volt/ reference" on "secondary-side" sometimes after WW2...late 1960s and I have thought "European 220VAC Mains world" was the reason? I now have studied and done electronics work 44 years so it was a big suprice to see Chassis being used for operative current when I began building tube amps two years ago :)
 

telemnemonics

Telefied
Ad Free Member
Joined
Mar 2, 2010
Posts
33,464
Age
62
Location
Maine
From your description of goals and mods it sounds like you don’t really like the JTM45 and would prefer a JMP50.

>Less woofy low end
>More Plexi like since it IS
>Generally less primitive circuit so less noisy
>Can still build with tube rectifier like the first “black flag” JMP50 if you’re concerned that you will miss that, but again, tube rec is part of why you get woofy bass instead of warm fat tighter clear bass.

Many repeat the old chestnut that the JTM has great cleans as if the JMP50 did not, but that’s just not true.
Whatever you gain with a more primitive soft squishy circuit comes at a cost, and you seem to want the things you lose with the older design.
 

BenTobith

Tele-Meister
Joined
Jan 3, 2020
Posts
231
Location
Austin, Texas
From your description of goals and mods it sounds like you don’t really like the JTM45 and would prefer a JMP50.

>Less woofy low end
>More Plexi like since it IS
>Generally less primitive circuit so less noisy
>Can still build with tube rectifier like the first “black flag” JMP50 if you’re concerned that you will miss that, but again, tube rec is part of why you get woofy bass instead of warm fat tighter clear bass.

Many repeat the old chestnut that the JTM has great cleans as if the JMP50 did not, but that’s just not true.
Whatever you gain with a more primitive soft squishy circuit comes at a cost, and you seem to want the things you lose with the older design.
I'm not looking to debate which circuit, etc. It should be clear from the OP, the amp is JTM45 based, but somewhere closer to a JTM50 (not JMP), with a little 1987 sprinkled in.

Plexis and JTM45s have the same issue with normal channel cleans, and it has little to do with rectification and 99.9% to do with coupling and tone stack capacitor values and the insane 330uF cathode bypass on V1.

I like tube rectifier tone and feel, not concerned there.

Mostly looking for feedback on more optimal wiring or grounding and other areas like that. Not looking to be convinced I should be building something else.
 

BenTobith

Tele-Meister
Joined
Jan 3, 2020
Posts
231
Location
Austin, Texas
To me "Larry grounding" seems operationally very strange and quite complicated as well. It uses chassis for secondary return current and therefore make it an antenna for electromagnetic noise. It also send electromagnetic noise to Mains Safety Earth wire now when even in USA ;) you have "safety earth"- system. Also to instrument cable Cold wire which is wrong when in electronics their roles are "shield" against electromagnetic noise. Just yesterday I was replied it is German "Larry" whose idea it is and that is strange because I have thought German engineering to be very advanced :(
It seems strange perhaps, but the results over the years are rather proven.
 

telemnemonics

Telefied
Ad Free Member
Joined
Mar 2, 2010
Posts
33,464
Age
62
Location
Maine
I'm not looking to debate which circuit, etc. It should be clear from the OP, the amp is JTM45 based, but somewhere closer to a JTM50 (not JMP), with a little 1987 sprinkled in.

Plexis and JTM45s have the same issue with normal channel cleans, and it has little to do with rectification and 99.9% to do with coupling and tone stack capacitor values and the insane 330uF cathode bypass on V1.

I like tube rectifier tone and feel, not concerned there.

Mostly looking for feedback on more optimal wiring or grounding and other areas like that. Not looking to be convinced I should be building something else.
Fair enough, I’ve owned most of the versions and weighed in on that.
Forgot the first 50 was still JTM.
Had a cool hand bent chassis Park 45/50 as well that was brutally loud clean and tight.
Could be we are thinking of two different things where you identify excess bass and I identify loose indistinct bass, two different causes and my complaint can’t be fixed with a cap swap.
Carry on!
 

dan40

Friend of Leo's
Joined
Aug 19, 2015
Posts
2,731
Location
Richmond Va
The Larry Grounding method has worked well for me in a couple of my first Plexi style builds. It was very popular on the Metroamp Forum years ago and a few guys still use it in their builds. My only gripe with it was having to drill so many holes and the ground runs be so long. The filament centertap ground at the input is extremely long and I just felt that it made the lead dress look a bit messy whenever I built with this scheme. These days I have switched over to the single ground buss scheme for all of my builds including the higher gain 4 and 5 stage circuits. It's a much cleaner, simpler scheme and the amps are always dead quiet.
 

2L man

Tele-Afflicted
Joined
Nov 23, 2020
Posts
1,437
Age
62
Location
Finland
Because Mains Safety Earth wire has some resistance it is stupid to use part of it (read chassis) to cause more noise there. And then there is instrument in the end, behind a symmetrical cable which half of the signal is usually connected to Chassis. It is selfish approach because it will cause some noise to other band member amplifiers. Perhaps it was fine when mains did not have Safety Earth but nowadays it is not. It would surprice me if it is still accepted in USA "electric regulations" but no-one has not confirmed it?
 

dan40

Friend of Leo's
Joined
Aug 19, 2015
Posts
2,731
Location
Richmond Va
Perhaps it was fine when mains did not have Safety Earth but nowadays it is not. It would surprice me if it is still accepted in USA "electric regulations" but no-one has not confirmed it?

I doubt Larry's scheme would meet electrical code but it has been used by many builders with good success. This scheme was developed by a German amp builder named Larry Grohmann, owner of Larry Amplification. He builds some beautiful amps and does work for many pro touring acts.
 

BenTobith

Tele-Meister
Joined
Jan 3, 2020
Posts
231
Location
Austin, Texas
The Larry Grounding method has worked well for me in a couple of my first Plexi style builds. It was very popular on the Metroamp Forum years ago and a few guys still use it in their builds. My only gripe with it was having to drill so many holes and the ground runs be so long. The filament centertap ground at the input is extremely long and I just felt that it made the lead dress look a bit messy whenever I built with this scheme. These days I have switched over to the single ground buss scheme for all of my builds including the higher gain 4 and 5 stage circuits. It's a much cleaner, simpler scheme and the amps are always dead quiet.
This is how I did my 18 watt TMB build. Buss wire on board with all grounds leading to the input jack ground lug. Very quiet from a grounding perspective.
 

BenTobith

Tele-Meister
Joined
Jan 3, 2020
Posts
231
Location
Austin, Texas
I've updated my layout a bit. Going to try a single ground near input approach. Please let me know if you spot any issues? Note that the pink striped wires are RG-316 (shielded).

JTM45 Layout Alternative.png
 

Attachments

  • JTM45 Layout Alternative.pdf
    276 KB · Views: 11

2L man

Tele-Afflicted
Joined
Nov 23, 2020
Posts
1,437
Age
62
Location
Finland
Looks good!

I believe filament elevating has good effect vs. its small incressed complexity. Read recent Bleed Circuit thread where are ideas to use bleed circuit to filament lift.

I would install 1 ohm biasing resistors to both power tube cathodes! I have also began installing bias adjust which has also a balance potentiometer.
 
Last edited:

dan40

Friend of Leo's
Joined
Aug 19, 2015
Posts
2,731
Location
Richmond Va
If you want to cut back on those long ground wire runs, I like to mount a single 18awg ground buss wire to the chassis just above the board and land all of the ground wires to it it. The first ones would be the filter cap grounds, filament CT wire and PT CT wires and the rest of the grounds from the pots and board would land in order as you get closer to the input jack end of the buss. To mount the buss, I attach a couple of the 3 terminal tag strips to chassis and clip off the grounded center tab. I will then solder the buss wire to the isolated lugs on the tag strips and attach the end of the buss to the chassis near the jacks with double lockwashers and a lock nut. The 1 ohmers on the power tube sockets for biasing will get grounded right near the sockets unless I'm using tip jacks to create bias test points on the rear panel.
 

BenTobith

Tele-Meister
Joined
Jan 3, 2020
Posts
231
Location
Austin, Texas
Question about switches (I wish I better understood this aspect of amp builds):

Because of the transformer I'm using, I have to twist a pair of 18 awg wires (120v) from the PT to the mains switch, along with one of the wires from the amber neon lamp. That's 3 18 awg wires for one tiny lug. It's obviously a bit messy, even with me trying my best to make it neat.

I'm curious, could I just use DPST switches instead, and split the lamp wire and those twisted, 120v PT wires across two lugs on the same side of a DPST switch? Wouldn't that be the same, "electrically," as being on the same lug of a SPST switch?
 

BenTobith

Tele-Meister
Joined
Jan 3, 2020
Posts
231
Location
Austin, Texas
Looks good!

I believe filament elevating has good effect vs. its small incressed complexity. Read recent Bleed Circuit thread where are ideas to use bleed circuit to filament lift.

I would install 1 ohm biasing resistors to both power tube cathodes! I have also began installing bias adjust which has also a balance potentiometer.
Thanks for the feedback.

The only reason I don't use biasing resistors on the power tubes is I use a TAD Bias Master, so I've no use for them typically.
 

dan40

Friend of Leo's
Joined
Aug 19, 2015
Posts
2,731
Location
Richmond Va
I'm curious, could I just use DPST switches instead, and split the lamp wire and those twisted, 120v PT wires across two lugs on the same side of a DPST switch? Wouldn't that be the same, "electrically," as being on the same lug of a SPST switch?


Sure...just be sure to use a jumper wire between the two lugs you want to use because they are electrically isolated from each other inside of the switch.
 

Wyatt

Tele-Afflicted
Joined
Nov 3, 2004
Posts
1,264
1. You have the Neutral fused...no good. Modern safety wiring dictates Hot to fuse then switch then PT
2. If you need a place to tie in the Neutral, consider switching both Hot and Neutral (its actually a requirement in much of the world)
3. Move the Standby switch after the first filter cap. Or just don't wire or use it. You need the first filter cap filling while the GZ34/5AR4 is ramping up. If you try and fill the cap after the tube is warm, the current in-rush can kill the tube. It's a very well-known failure point on the JTM45 (as well as the Vox AC30CC and a few others)

Personally, I'm a fan of the Hoffman grounding scheme, but each their own.

index.php
 

BenTobith

Tele-Meister
Joined
Jan 3, 2020
Posts
231
Location
Austin, Texas
1. You have the Neutral fused...no good. Modern safety wiring dictates Hot to fuse then switch then PT
2. If you need a place to tie in the Neutral, consider switching both Hot and Neutral (its actually a requirement in much of the world)
3. Move the Standby switch after the first filter cap. Or just don't wire or use it. You need the first filter cap filling while the GZ34/5AR4 is ramping up. If you try and fill the cap after the tube is warm, the current in-rush can kill the tube. It's a very well-known failure point on the JTM45 (as well as the Vox AC30CC and a few others)

Personally, I'm a fan of the Hoffman grounding scheme, but each their own.

index.php
Hey, thanks for the feedback.

I'm mostly following the layout and details provided by Metro Amps/Valvestorm for their JTM45 kits, which I think are fairly historic-based. So of course, some of the old design flaws and quirks are there.

I'm open to correcting what I can, but I need some guidance so I don't screw things up. I'm not enough of an expert to deviate too much without feeling confident in what I'm changing or doing.

Regarding point 1/2, can you clarify how I would do this, based on my existing layout? EDIT: Could the following changes work?:
1. Move hot/black wire from IEC to "tip" of fuse holder
2. Move neutral/white wire from IEC to "side" of fuse holder, where it connects with the PT 0v common/lamp (?)
3. Add wire from "side" of fuse holder to mains switch (?)


Would appreciate tips, advice, etc. on the above from @Wyatt @robrob @2L man and others.

Regarding point 3 (standby), I'm using a 2 watt 100k resistor (ceramic) to help mitigate some of the standby design flaw. If I were to go beyond that and rewire, I'd need some specific instructions. Otherwise, I won't feel comfortable mucking around with it. EDIT: I might just disconnect the Standby altogether. Leave the switch for looks..

Regarding grounding, I'm too far along with the build to alter that at this point. It should be solid.
 
Last edited:

2L man

Tele-Afflicted
Joined
Nov 23, 2020
Posts
1,437
Age
62
Location
Finland
Edit: "Sorry Ben, this following was based to the layout Wyatt posted, not yours!"

It is almost always easier to build everything "right" than it is to fix later. That layout will hum because there are ground loops and current flow thru chassis. Soldering current wire between pot cases is bad idea!

If the B+1 capacitor negative is not big enough to take power tube cathode current resistors, PT Ct and B+3, B+4, etc. current wires just solder insulated something larger close to it which can take few wires and then only two wires need to go to filter capacitor (from this lug and PT CT. Pre amp current is very low and you can use very thin wire.
 
Last edited:




Top