The Vesperado Mod thread for the JCM2000 TSL-100: VERSIO DJENTER

[Vesperado]

I will be updating this post from time to time with the latest developments (see post #85), so revisit as oft as you need. If there are any questions, post them here in this thread.

After reading some comments from some unhappy TSL owners out there, I am providing a public service to address the amp's shortcomings in this thread, free of charge. I feel that this amp was rushed into production without the proper R&D, for had it undergone the appropriate testing it would have been doubtless a huge success. Mine is a 2001, corrected and improved to meet what I envision a Marshall to have been for the turn of the Millenium. Wherever there is room for several mod options, it will be given so as to give the user full access to the amp's hidden potential--and there is good potential for all types of users here.

Mainboard Issue 20, corrections, improvements, and suggestions:

- R31 should be left alone, unless you are after JCM800 tones, then an 820R gets you there. Or put it on a switch (remember to switch it in Standbay Power only). I don't like it as it compresses everything and that's just not my bag, baby.
- The treble peaker cascade is something I stumbled upon out of dumb luck, Peakers I are on their respective channel's board already, Peaker IV may be 470p for a JCM800 spec. These two do not effect Clean, only when Peaker V is engaged does Clean become effected.
- I have cleaned up excess Trebles and Bass in the preamp. N.B. Treble bleeds and boosts are EVERYTHING with Marshall circuits.
- Filament filtering has been improved and made ruggeized with raising the Bridge Rectifier off the board a few cm.
- R68/77 bias pair stock values are way too hot. Unless you enjoy burning through tubes, or enjoy power-stage clipping, cool the engines, try lowering plate dissipation rates, it affords more clarity and dynamics, believe it or not. I have 2W metox types in these positions and the amp's bias remains solid.
- C46 should be above a 2kV rating. Check the component first, a 6kV rating was fitted on mine from the factory.
- Further improvements which are not annotated on the .pdf: cleaning up the socket pins V5-8 by removing all flux residue and utilizing silver-tin solder, drilling air-motes between Pins 4/5 to isolate them better, reinforcing traces for higher current handling, relocating R8/78 to socket V5, etc. See my previous threads for more information.
- Upgrading R71 with a choke is preferrable. 3H is good, 10H better.
- Upgrading diodes to FRED-types, same.
- Upgrading the Power Transformer; preferrable.
- Upgrading the Output tranny; maybe.
- Desire to run at half power? Utilize a matched pair of EL34 in V5/8 ONLY, and ensure you load a 16ohm speaker (cab) off an 8ohm tap if you do this, or an 8ohm speaker (cab) off a 4ohm tap. Adjust bias pots as necessary. If Plate dissipation is too hot, take care of R68/77 as outlined above and try 'er again.

Crunch Channel PCB corrections, improvements, and suggestions:

- Between C14/VR8 can be fitted a 2k2 or larger resistor to tame some of the picky highs over the Clean channel. Access to B25K parts is grim, try a B50k with a 50k resistor across it.

Clean PCB corrections, improvements, and suggestions:

- Bright cap, C3, can be either 56p for S-Coils, or even 68p if they are inherently bright. This is indeed applicable to speakers as well. These values provide for the chime wanting with the stock value.
- L2 chokes the Bass freqs of your axe, not to mention low freq dynamics! It is unwelcomed. I used copper wire to link this spot.
- C1 is entirely unwelcomed. Having this here along with Main C9 just deprives you of that Marshall "kerang", spank, and pick attack you hear so much of (think Plexi). Not only so, but robs pedals equally of their mojo.
- The rest is the spec for my "Super Glass" voicing. N.B. This is not voiced to be a pedal platform channel. You might have success keeping it stock and removing Mainboard C24. Nevertheless, my Super Glass works well with Fuzz, ymmv.
- Once relocating the clean inject on the preamp, there will be (3) warm and chimey, beautiful stages of gain for that channel, and headroom for days.

 

[TheLoudness!!]

Great info!!

They should have called the tone shift button the "ass" switch. Mainly because I always thought it made the amp sound like ass when pushed in lol

I think they need a choke and resonance control. I might tinker with mine at a later date.

 

[Vesperado]

Little variations in values impact everything. Try your best to match what I have. Many months of research have gone into this, and if a value is not on the money then it will likely not fair well. I claim no liability for anyone who deviates from my mod tone-wise (nor do I claim liability for you proceeding forward with my mod). If you want to hear the TSL at its finest, stick to my values.

Lead PCB corrections, improvements, and suggestions:

- VR4, for you early EVH heads, should work (I have not tried it) provided you power the amp off a Variac ~90 VAC or so, and dime the B50K, so they say. Update, Main R3 should get a small cap across it to enhance the highs above 8kHz. This is imperative to reach Eddy.
- The Deep Boost reduction and the Presence value corrections I have gotten from Matt over at Atomiumamps. Hats off to him!
- Replace all aluminum caps with Tantalums in the presence circuit.
- IC4: Roll to your hearts contentment. Higher slew rates equate to good bandwidth and fast responses to your palm-mutting.
- IC2/3: NJM2121D are the top of the line replacements. Good luck and happy hunting!
- R25/26 have been tuned with the modifications done to the Reverb/Mainboard PCB. If you choose to leave the Reverb as is, but prefer longer decay on the Clean channel (my value, that is), you can make R25 mirror R26 for even decays on all three channels.
- C23 is unwelcomed as it deprives the drives of 'verb. Remove it unless you find it to be too much, then reduce the value instead.
- C36/37 handle high voltages off the preamp. 63V, a typo for 630V? N.B. If you use the XLR out, install 630VAC types, as they will provide better audio resolution than the 400VAC type. I put 630V in mine.
- There is a workaround to reassign the Clean's Presence/Deep over to the Crunch (OD1), which I have outlined in another thread. I haven't tried it, but it should work. This might be preferrable for someone who could dispense with their Clean channel, not I, for my Super Glass voicing is a dream with my G&L ASAT Classic; it requires that circuit, and spanks and chimes oh so good!
- All "B22K" pots are no more than B20K on the boards. B25K is what you need (good luck finding any). Acquire a B50k and trow a 50k resistor across it to get an A50k.

Reverb PCB corrections, improvements, and suggestions:

- R1 is optimal at 100k for the reverb unit which came stock with the head. This may need to be readjusted upon upgrading the tank. My board came fitted with a 220k.
- C5 dido.
- Remove R4 and ZD1/2; they are unwanted neither were they fitted with the third revision (Issue 3).
- C13 may be removed or reduced. This is the global treble bleed for the reverb.
- C12 may be reduced for more treble freqq. pass-through for the reverb.
- The IC should be first removed and fitted with a socket before installing the NJM2121D, as they are now scarce.
- Mainboard TR3 cuts Reverb audio signal whenever a channel is switched over. Removing this switching FET will allow for that signal to "echo" until fully dissipated. Option is availabe to those who prefer this.
- The physical tank enclosure is exposed to radio interference underneath, and ought to be closed via aluminum tape or screening, then be bonded to the chasis to provide it a connection to ground. Otherwise, you get noisy reverb everytime you engage it.
- Replace the RCA cable with a quality replacement.
- Replace the Link (Mainboard R43) with copper wire.
- N.B. whenever transporting the Head, remove the tank and disconnect the springs to prolong their longevity. This goes for storing in humid environments as well. These springs can no longer be acquired through ordinary channels, unfortunately.

Rear PCB corrections, improvements, and suggestions:

- R19 utilizes the 4ohm tap off the OT. The 16ohm tap may be employed if so desired. The NFB needs some feedback to work properly, so do not dip under 20k or so.
- R22/23 greatly benefits from an higher rated resistor, as this helps combat heat generated from the rectifier circuit. I found some 10W wirewounds with long leads, raising them off the board at an angle to dissipate the heat towards the mainboard, where the socket plate has vent holes which I drilled in it to provide exhaust. Tin-silver solder is preferrable here, too.
- IC2 is notibly important. Better performance is preferrable.
- Remove zener diodes 3-6. They are entirely unwelcomed.
- The remaining resistor values have been corrected for unity gain. We must thank Matt over at Atomiumamps for those values. Remember, if it carries signal, make it a Metal Film type.
- Lead C24 and Main C30 need to be Tantalums. You can save some doe and build C30 with (2) 22u, the PCB will accomodate them.

Thoughts:
Overall the TSL is limited in terms of functionality, but still quite capable to cover all the Marshall tones imaginable, Plexi onward. There are just enough knobs and buttons to do the trick.

For those needing a pedalboard-friendly Foot Controller, check out ebay, there are some Four Botton-types floating around. You could rehouse them in a Hammond 1590BX2--they even have painted ones!

I hear no difference between film and ceramic, ymmv. There are many places in the signal path where I have MLCC and they sound great. One thing to be said about cap value building, for the signal path (as well as power supplies) limit these to parallel, not series in event you need to build (and you will with my mod). Otherwise if you put caps in series your ears will pick up on the "lag". No good. Everything else, however, (bleeds, switching, etc.) does not make much difference whether series or parallel.

For speaker cabs, mine is a 1960A JCM900, with one caveat (but a nice and welcomed caveat): one side is 1" deeper than the other. Someone at the factory went over one inch, and made the cab anyways What this means is, my 1960A is oversized. The thunder it produces is amazing because of this. Thought I'd share that lil' slip with you guys

I hope my mod is a worhty one to join the Marshall hall of fame.

Keep 'em valves aglowing gents!

 

[Rodical]

Thanks for all the info... Alot of info to read through...Bought mine in 2003... Mine has been a really good purchase... Running through 1960TV cab and 1960bv...

 

[Vesperado]

I think it would be good to talk about tone options for those who would like to try out different combinations on their amps.

You can discover which combo suits you best before closing her up, or wire in a separate enclosure that can piggyback off the Reverb Tank (mind your space and temps!).

For Bluesbreaker mids, replace the 22n mids cap with a 20n on OD1 and/or OD2, also defeat the additional slope resistor and additional highs cap via links to make the Tone Shift switch perform as a "Mids Shift". Simply replace the additional mids cap with a 220n film (22n + 220n = 20n). Ofcourse, if you like the Tone Shift (I know I do!) you can switch between the two values for Lead C13 and Crunch C10, say a 20n for a 2n in parallel. Leaving the 20n selected when engaging the Tone Shift will produce an altered tone from that with the 22n, mind you.

Crunch PCB "Mids Shift" option:
R9 > Link
C8 > Link
C9 22n > 220n film

Lead PCB "Mids Shift" option:
R15 > Link
C14 > Link
C12 22n > 220n film

Back to the tone talk.

I'd say a 3-way toggle is all you need for each peaker, the trick is narrowing down your preferred values and getting the peaking cascade to work from there. Be sure that you reserve an OFF position for the last two peakers, unless you wish to dispense with the classic Marshall topology; it's your choice.

Stomps become useless above a pair of peakers. If you desire more gain with the trio, then employ a linear boost/EQ, or get a gain doubler valve socket kit if you prefer to keep it fully analog. Ofcourse, you could add in a fourth with some further modification as outlined in a post above, but in all reality three is plenty at SPL.

 

[TheLoudness!!]

As far as the speaker jack goes, what ohms are safe to run?? All I have is an 8 ohm cab.

I'm asking about this because of a speaker ground mod I read about. On here most likely??

 

[Vesperado]

If all you got is 8ohm, your fine. I have already noted the 16ohm grounding fix on the Rear PCB PDF in the thread above, which is a good idea if you plan on using a 16ohm load. It's also a good idea to check your cabs just to make sure.

However, should you elect for V5/8 operation only, you must adjust for (1) load, and (2) plate dissipation. There are tutorials floating around online to fill in the gaps.

 

[Vesperado]

Updates:

Playing on the amp more, I tweeked the values some to get even better tone at SPL.

- Main C17 > 796p 630V MLCC
- Main C18 > 1n6 630V MLCC
- Main R14/15/17/34 > 1k5 1/2 watt
- Main C13 > .8u 63V pol. (.76u also works well)
- Main BR1 > BR106
- Lead VR2 bleeder = 168p (182p for bright axes)
- Crunch VR2 bleeder = 168p (182p for bright axes)
- Crunch C9 > 26.3n (makes ~12n with Tone Shift)

Observations:

- Crunch Tone Shift mids have been shifted down a few Hertz. Sounds so much better to my ears, and dishes out those lovely chorused, old-school metal vibes. Throw a delay into the loop and let your neighbors know how much your amp is better than theirs
- Main BR1 crapped out on me, so I replaced it with a better spec'd one. Mouser sent one smaller than that which I ordered, and it fits the PCB outline perfectly (raised off the board, of course), so call me skippy. Let me mention that those of you who want to run into an auto-tranny ought to upgrade your bridge rectos for increased current draw. Probably why mine crapped out...
- Still need to test out Stage-four cathode bypass...
- A quick note on the loops; you need to strongly consider a high-grade opamp as well as upgrading the mixers located on the Lead PCB. It's essential for good tone. It could be the difference between lush, ambient "Turn it up!" and noisey/farty "AHHH, get this out of my signal!!!". Also, don't forget those metal films!

Once I finish up this weekend, I'll post the updated drawings!

 

[Vesperado]

I have the amp back in its case. The drive channels are now fully accessable, whereas before the 1k8 cathode resistors did not fare so well with the normal modes. I still prefer them over the 1k5 for the Tone Shift modes, anyhow. But now, she's up and running "across the board".

I settled on a .76u V1A cathode bypass (OD2), and a 2n2 for Peaker II for the drives, as I still need to make a separate enclosure to house all the switches for the big job. The only pedals which seem useable through the amp now are my Strange Master Treble Booster clone and a TS9 clone, oh, and ofcourse my improved Maxon DS-830; basically your old-school "classic" stomps. I don't use dirt over the Clean channel.

The fourth-stage cathode bypass works, and can provide two options for the user: 1) to tighten up the bass globally, and, 2) to fuzz up the bottom freqs globally, depending on which size you install. A 1u works great for both channels as far as tightening goes. I tried a 3u3, which added some low-mid flub, a 22u, which added some high-low flub, a 47u, which added good low flub, and a 100u, which made for good flatulance. Methought a 34u was a neat trick to get some Angus-style tones. This trick doesn't get djent, but a sort-of full stack rumble. You could also leave the stock value in for Main C27 to emulate nearly the same effect.

 

[Vesperado]

I'm revamping the Reverb for the drives. There are limitations with the circuit that I'm working to get around. Stayed tuned.

As for the switching enclosure, I'm ditching it. I should be able to get away with installing (4) mini toggles between the top plate and the OT. I think I'll leave the first Peakers fixed at 470p and 985p, and will only toggle 1720p/2150p/OFF for Peaker II, and 2150p/2n7/OFF for III. This covers my bases. The other switch is reserved for the coupling cap for the drives, 2n2/4n7, and the final one for a Stage IV bypass cap...or perhaps a 4PDT for +1n across the four coupling caps in the preamp to get some more extension.

I have the updated drawings nearly ready, just need to verify my tweaks. Just a few more days and they should be up...

 

[Kitchensinkelectronics]

Hello iam new to this forum. Ive done quite a lot of fixes and mods to my tsl122 and just recently found your mods (have followed atomium amps mod page for the tsl100 and implemented his mods) Will look in to your mods
The amp is already a a lot better, now changing marginal gains. I have a few questions I need help with.

1. I’ve installed a bias drift mod kit and got the bias drift issues working for a couple of months but now the last two tubes won’t bias over 50mA. Do you think it’s tube related or do you have other hints were to look for when troubleshooting?
2. Iam looking to replace coupling caps with sozo caps but they only go up to 500V and the PI CC:s call for 630V. Would 500 be fine or should I get Mallory 630V rated instead?
All help appreciated!

 

[Vesperado]

Hello iam new to this forum. Ive done quite a lot of fixes and mods to my tsl122 and just recently found your mods (have followed atomium amps mod page for the tsl100 and implemented his mods) Will look in to your mods
The amp is already a a lot better, now changing marginal gains. I have a few questions I need help with.

1. I’ve installed a bias drift mod kit and got the bias drift issues working for a couple of months but now the last two tubes won’t bias over 50mA. Do you think it’s tube related or do you have other hints were to look for when troubleshooting?
2. Iam looking to replace coupling caps with sozo caps but they only go up to 500V and the PI CC:s call for 630V. Would 500 be fine or should I get Mallory 630V rated instead?
All help appreciated!

(1) Issue 20 needs a complete rework, which see this thread: https://marshallforum.com/threads/t...t-entailed-their-changes.114845/#post-2293924
(2) Mallory 630V, or what I just got in: Nichicon QXP. Panasonic 800V RF types are also recommended, provided they fit. Ensure that you verify the physical dimensions so that they will fit the board. Anything Metal Polypro made in Japan will be a cut above the rest imo.

I started with Atomium mods, then I began to tinker more with the amp while in search for my tone. I am waiting on a few more parts to arrive. Once I verify the changes, I will publish the data. Otherwise, Matt has been a big help, hats off to him!

 

[Kitchensinkelectronics]

Thanks for your reply! Just realised I’ve already did two of your mods on the preamp board(C1&L1) . Really appreciated! Before the amp what’s useless without a pedal cranked in to it. I will be following your progress! Nice work!

Just do clarify do you think a 500v sozo would be a bad idea?

 

[Vesperado]

Thanks for your reply! Just realised I’ve already did two of your mods on the preamp board(C1&L1) . Really appreciated! Before the amp what’s useless without a pedal cranked in to it. I will be following your progress! Nice work!

Just do clarify do you think a 500v sozo would be a bad idea?

What you can do is run a dmm across the plates to confirm voltages. Keep one hand behind the back while you check that, of course. I know audiophiles prefer higher rated caps over lower rated ones, fwiw, they say that provides for better detail. From a safety standpoint, higher rated caps outlive transient spikes, such as a tube failure Less maintenance, etc.

 

[Kitchensinkelectronics]

Thanks! The amp lives in my studio think I will go with 500v sozo caps and take the risk!

 

[Kitchensinkelectronics]

The plate voltages were around 450v and the bias measured 70mV when the bias trimpot was maxed out (for the last tube pair) the other pot were able to go much higher.

 

[Vesperado]

The project has been delayed a few weeks. Had to order more parts than anticipated (isn't that how it goes?). Apologies to those waiting on new drawings.

Once more I am revisiting the Treble Bleeds. They are close, but still can use a better tune-up. I am happy to say that the new drawings will include a switchable Global SPL Bleed for those who wish to perform loud. Who doesn't? It should take the edge off so the room won't clear. At this moment I am open to suggestions on a good bleeding scheme for the drives...

I am ditching the Bright Caps.

I have a new choke, yippie.

I am revamping the Reverb and the Effects Loop circuits. And on that note, I will experiment with the input opamp NFB loop some more in effort to take advantage of increased output, or rather input, to the parallel mix. I feel that there should be a negative input to ground blocking cap here: CAN ANY OF YOU CONFIRM THIS?

I'm relocating the Clean Channel's second stage inject.

I'm ditching all aluminum cans lurking in the signal path, among some other small improvements.

One fnal note, I found a flaw in my last drawing which will be corrected in the upcoming. Sometimes I am half-awake when making my annotations. I appreciate everyone's patience as I move forward!

 

[Vesperado]

No takers?

WELL I have it all figured out and just got finished confirming those changes. Doing this all without an oscillascope, signal generator, or what have you, just reasoning it out as I go

Amp is an absolute BEAST now...can't imagine what it's gonna sound like once I ditch the old choke...

Standby to standby...

 

[Vesperado]

"File too large" to attach, apparently. It's less than 1.7M. Hmm...

 

[Vesperado]

I did it. Don't know how, but I did it.

Here are the tonal options on tap--ALL GLORIOUS:

(1) Green channel = 4 gain stages
- Normal Tone Stack
- Mid Boosted Tone Stack: disengages Mid control
- Independent Presence
- Selectable Independent Deep
- Global NFB
- Global Treble Bleed for SPL
(2) Amber channel = 6 gain stages
- Normal Tone Stack
- Tone Shifted Tone Stack
- Shared Deep
- Shared Presence
- Global NFB
- Global Treble Bleed for SPL
- Coupler cap select: 2n2 / 4n7
- 1959 mode (Peakers II/III: OFF)
- Peaker II mode: 1720p / 2150p
- Peaker III mode: 2150p / 2n7
- Peaker II + III modes: 4 selections thereof
(3) Red channel = 6 gain stages
- Normal Tone Stack
- Tone Shifted Tone Stack
- Shared Deep
- Shared Presence
- Global NFB
- Global Treble Bleed for SPL
- Coupler cap select: 2n2 / 4n7
- Hot Rod mode (Peaker I: 985p, Peakers II/III: OFF, Bypass cap: .76u)
- Peaker II mode: 1720p / 2150p
- Peaker III mode: 2150p / 2n7
- Peaker II + III modes: 4 selections thereof

...and not counting the V.P.R. which darkens the tone when engaged, I have a total of (4) different base tones for Cleans, (20) options for a Crunch base, and (20) for a Lead tone. Those are independent, btw, and tally to (44) tonal bases in MY AMP !! FORTY-FOUR. The global-controls overall effect the cleans two-fold (2) and the overdrives the same (2). The overdrives share (2) parameters. There are also (2) Parallel Effects Loops, which when ran together split one loop to the Clean, and the other to the Overdrives. But, I have remedied the problem with the loops; that means I have the option to maintain an unity volume between ON/OFF, or boost (you heard correctly) my signal for a solo when engaged. All with ZERO tonal degradation. Reverb is drippy, too.

Can anyone boast FORTY-FOUR ANALOGUE TONAL BASES with their amp? I trow not.

Oh, and the 10 henry choke is outstanding. Thank you for the fast shipping, Mercury Magnetics