Jtm30 - The Right Way To Remove The Clipping Diodes.

Discussion in 'The Workbench' started by DesolationBlvd, Sep 16, 2017.

  1. DesolationBlvd

    DesolationBlvd Active Member

    Joined:
    Sep 16, 2017
    Messages:
    68
    Likes Received:
    101
    EDIT: See post #6 for details. Original post follows...

    So I recently got a JTM30. I can't leave good enough alone. I replaced the 1N4148s in the clipping section with LEDs, and I really like the overall character of the drive now. It plays much better with pedals, but the bass response is now rather flabby even with the bass knob all the way down. Where should I go from here?

    Hmm, if I'm reading the schematic right, R114 and C113 seem to form a low-pass at 720 Hz. That sounds really low, and might match up with what I'm hearing. Is it safe to bypass this?

    I may have melted away a solder pad, because the initial result was like the one described in this thread: http://www.marshallforum.com/threads/marshall-jtm-30-i-removed-the-clipping-diodes.63517/ , and while gimmeagig seemed to like the sound of the clipping diodes simply removed, I did not. I jumpered D101 to C103 and got the boost channel back.


    Also, how do I get the main board in/out with that plastic rail in place? I really struggled, before finally getting it out. Now I couldn't get it back in, so I removed the little plastic rail thing stuck to the board. Works fine, but I'm worried about support of the board.
     
    Last edited: Oct 21, 2017
  2. ampmadscientist

    ampmadscientist Well-Known Member

    Joined:
    Jun 26, 2014
    Messages:
    16,422
    Likes Received:
    10,332
    Location:
    Bio-Isolation Lock Down
    (another one of Marshall's retarded schematics, which is very difficult and confusing to follow)
    I would not rely on the schematic as "accurate," it was obviously intended to confuse....either that, or the person who drew it was mentally ill.

    http://drtube.com/schematics/marshall/jtm3-61-02.pdf
    http://drtube.com/schematics/marshall/jtm3-60-02.pdf

    WHEN you remove the clipping diodes, the level of the audio is going to increase several times...
    Therefore, the next preamp stage is going to be overloaded with input signal...

    The preamp stage was not designed for that high level signal (which you are now feeding into it)....
    AND the over-loading is causing the bass (low frequency) to become very distorted. It's compressing the low frequency.

    Diodes:
    CLIP the signal...which means that the signal cannot go above a certain level. Once that level is reached, it can increase no more.

    BUT without the diodes: there is no limit to the audio signal. You have removed the limiting from the audio path.

    CIRCUIT BOARD
    is very easily damaged by inexperienced soldering. This requires professional soldering equipment and the experience to use it.


    If I were crazy enough to modify this amp: (you already know that I am crazy enough)

    A. I would start by creating a complete read-able schematic....

    B. I would not attempt to modify this amp without test equipment (such as a scope).
     
    Last edited: Sep 17, 2017
  3. DesolationBlvd

    DesolationBlvd Active Member

    Joined:
    Sep 16, 2017
    Messages:
    68
    Likes Received:
    101
    Didn't see this post until I was done with more tinkering. Wish I could change the title of this thread...Maybe "JTM30: The Quest for Tone?"

    I already replaced R114 with a jumper and removed C113. Low pass going into V101B is gone, and so is that blanket over the sound. I'm so much closer to the sound I want. From 2 to 6 on the gain knob, I have most of the Marshall sounds in my head at my fingertips. Setting the gain to 8 and up is unusable though, sounds like I'm overloading it at that point.

    The clipping diodes are now two red LEDs on each side. Much more natural sounding than the "distortion pedal" sound from before.
     
  4. slide222

    slide222 Well-Known Member

    Joined:
    Apr 5, 2013
    Messages:
    2,682
    Likes Received:
    1,733
    Location:
    cambridgeshire uk
    my jtm60 has a switch to turn clipping on or off , it was modded before I got it with this switch and a fan under the power tubes pulling the heat out and I much prefere it with the clipping switch in the off position
    and it gives it more tube distortion than fizzy

    and last year I sent it to marshall (1 hours drive from me) to fit new el34's and bias and they checked out the mods to make sure they were safe ,as they refuse to work on modded amps unless they check them over first
     
  5. DesolationBlvd

    DesolationBlvd Active Member

    Joined:
    Sep 16, 2017
    Messages:
    68
    Likes Received:
    101
    ^ I took a look at the JTM60 / JCM600 schematic. It seems the preamp is pretty different from the JTM30. The clipping diodes go straight to ground on the 60, not on the 30.

    As for my situation with the gain knob, I originally switching the diodes increased the forward voltage and could be feeding too much into the feedback loop. Debunked...

    I decided to remove D101 - D104 and C103 completely. Removing the low pass filter yielded by C113 and R114 made the difference. The schematic for the boost channel does look a lot like the 2203 preamp once those clipping diodes are gone. Gain above 8 results in an unusable squeal. Originally thought to reduce the gain knob's range, but I realized that there was way too much gain in the first place. I brought preamp component values to 2203 stock except where noted: V101A cathode cap to 1u (instead of .68u), V101A cathode resistor R111 to 2.7k, V101B cathode resistor R115 to 4.7k (instead of 10k), coupling caps C111 and C112 to 22n, and then replaced R25 on the jack board with the components formerly known as R114 and C113 to put the original high-pass in. Just so happens that they are 470k and 470u respectively.

    EDIT TO ADD: Switched R115 back to 10k. Had a 10k resistor kicking around. Still farty. Replacing C4 with jumper. It was 220n in series with the coupling cap.

    Now I need an attenuator, since it's so loud now, the slightest twitch of the master volume results in blasting my apartment full of that JCM800 roar.
     
    Last edited: Sep 24, 2017
  6. DesolationBlvd

    DesolationBlvd Active Member

    Joined:
    Sep 16, 2017
    Messages:
    68
    Likes Received:
    101
    Bumping with the almost-final result.

    I was still not fully happy with the bass response. The Marshall tone stack was meant to be fed by a cathode follower, not right off the plate. I first reduced the output impedance as much as I could. V2B, the stage feeding the tone stack, went from 220k to 100k on the plate, and from 2.2k to 820r with a 25u bypass capacitor. I still had to set the bass to around 2 or 3, so I chose to switch the bass pot to 100k audio to shift the range accordingly. I notice Mojotone ended up having to feed a Marshall tone stack from a plate in both their Studio One and 18-watt TMB kits. Instead of bypassing the cathode resistor of the stage feeding the tone stack, they went with a 47k slope resistor in the tone stack. EDIT: Went to .68u on V2B and put the bass pot back. Seems to have tightened the sound up.

    During the modding process, I had the opportunity to try a 2203x. That amp seemed to have more gain on the highest settings than my modded JTM30, so I decided to reduce the resistor going into the preamp gain to 100k when I came back home to it. However, it was almost "farty" past 8 or 9 on the preamp gain when I hit it with a Duncan Distorition, so I raised the resistor to 220k. I also switched to a 220pF bright cap on this channel, from the stock-on-this-amp 100pF. Just enough brightness when I dial back the gain.

    I am very happy with the end result. The boost channel has the JCM800 character I craved and takes pedals very well. I might try switching V2A (clean input) to the 1959/1987 bright channel values (2.7k / .68u on the cathode, 2.2n coupling after it) the next time I have a soldering iron out. EDIT: Made V2A changes and am even happier.

    Side note: A .68u can be very closely approximated by putting a 1u and 2.2u in series. Just a heads-up since most of the .68u caps I've seen were too big for this tight little PCB.

    Final list of changes:
    D101, D102, D103, D104, C103 (diode clipping circuit): removed
    R114, C113 (low-pass after clipping circuit): removed, with jumper over R114


    R111 (V1A cathode): 2.7k
    C110 (V1A cathode bypass): .68u

    C111 (coupling cap after V1A): 22n
    R25 (resistor before boost channel gain): 220k, with 470p capacitor piggybacked on it
    C3 (boost channel bright cap): 220p
    R115 (V1B cathode): 10k

    C112 (coupling cap after V1B): 22n
    VR3 (boost channel volume): Add 470p bright cap (between input and wiper)
    R119 (V2B plate): 100k
    R118 (V2B cathode): 820r
    V2B cathode bypass capacitor: .68u (there is an empty spot next to R118 for it)

    I know the JTM30 clean channel is pretty well regarded. I think I went astray when I increased the boost channel coupling caps and then tried to compensate for the increased bass further down the circuit. If I wanted to preserve the clean channel, I would start with just the underlined changes.

    Also, the boost channel volume is no longer used after this. It is after V1B, where a fixed voltage divider with a bright cap was in the 2203 preamp circuit. Setting the boost channel volume right between 7 and 8 gets the voltage divider closest to the original 2203 spec. I have a feeling the clipping diodes were put in to make the boost channel volume usable in the first place, allowing the player to balance the volume between the clean and boost channels.
     
    Last edited: Oct 23, 2018
  7. gimmeagig

    gimmeagig New Member

    Joined:
    Sep 3, 2013
    Messages:
    26
    Likes Received:
    1
    I am not an electronics expert so to mod my JTM30 I have to take it to my repair man. I don't want to spend more that the amp is worth. I had my repair guy dis able the clipping diodes a while ago and I was pretty happy with the result except that now the boost channel is really dark sounding and I have to crank the treble on it so much that the clean channel ends up being way too bright. So I can't channel switch.
    I just found this thread here and I'm curious about replacing the clipping diodes with (red) LED. I guess different colors would give different results. I'm not after a super heavy over the top distortion. Guys I like are Robben Ford, John Mayer, Richie Blackmore.
    In this thread are so many options and I don't understand them. Can you guys help me out?
    If I was just going to do one thing to my amp. Which would be the best bang for the buck?
    The LEDs instead of the clipping diodes?
     
  8. DesolationBlvd

    DesolationBlvd Active Member

    Joined:
    Sep 16, 2017
    Messages:
    68
    Likes Received:
    101
    Bumping now that I'm back from completely converting it to 2204 preamp. This is all on top of my previous changes.

    I disabled the clean channel completely (obliterated the trace going into V2A's grid), so I could use V2A as a cathode follower. A wire from V2B's plate to V2A's grid, then replacing V2A's plate resistor (R117) with a jumper and V2A's cathode resistor (R116) with a 100K, moving W102 to where C114 was. C115 is gone too.

    I removed V2B's bypass cap. I originally used it to decrease output impedance to make the tone stack more responsive, but it's no longer needed. The cathode follower has a far lower output impedance than going off the plate ever will, even bypassed.

    The next step was to permanently set the second-to-third-stage attenuator instead of using the boost channel "volume" pot for that. L101 between V1B's output and CON102 wire 4 was replaced with 470K/470p, and the jumper between CON105 wire 1 and V2B grid is gone too. The other end of CON102 wire 4 goes to the end closest to V2B's grid, and a 470K resistor replaces the jumper. CON105 is gone and replaced with a jumper, so the new 470K resistor can go to ground. With that, the attenuator is complete.

    I removed RL1, the relay, since this is a single channel amp now. I could have routed W102 to W2 like normal (though a longer wire is a must now), but that would completely bypass the FX loop and reverb. I connected W102 to the "boost volume" pot and jumped CON5 wire 1 to W2 to keep the FX loop and reverb in.

    At this point it's basically a 4010 but 2x10 and minus the presence knob, and I'm over the moon with it. As a bonus, the master volume is less twitchy. I'd still like to get the FX loop much less hot, the reverb louder, and possibly be able to use the footswitch jack as the low sensitivity input, but I'm fine as it is right now.

    EDIT: Tried to make the low sensitivity input and failed. I had a nasty squeal which was coming from the reverb tank. I decided to bypass the FX loop and reverb altogether. Now the volume's twitchy again, though.

    ...Can't really say much about this now that my JTM30's a single channel amp, but I'll try to help.

    The biggest thing darkening it is the low pass after the clipping diodes. First try removing C113 and using a jumper instead of R114.

    Next, the LED color matters. I'd go with a red LED - and just one on each side (a jumper replacing the other spot). The two 1N4148s gave a forward voltage of 1.3V, while one red LED is closer to 2V. Closer to the blue end of the spectrum and the forward voltages get higher and higher - might as well be nothing at all.

    To brighten the tone, I'd also look at adjusting V1A's cathode resistor and bypass cap (C110 and R111). Stock is 1.5K and 22u, letting all the bass through. I'd change it to the traditional Marshall values of 2.7K and .68u to cut that heavy bass. I did have a hard time finding a .68u cap that would fit.
     
    Last edited: Oct 21, 2018
  9. RickyLee

    RickyLee Well-Known Member VIP Member

    Joined:
    May 22, 2010
    Messages:
    9,815
    Likes Received:
    3,255
    Location:
    Riverside CA U.S.A.
    I am just curious to know if you are a Sweet fan? I am sure there might not be anyone here even wondering what might have me thinking this way. Actually, you might be scratching your head over this question as well lol.

    :D
     
  10. DesolationBlvd

    DesolationBlvd Active Member

    Joined:
    Sep 16, 2017
    Messages:
    68
    Likes Received:
    101
    ^Yup. First album name to come to mind when I needed a username for musical gear forums. Andy Scott inspired me to pick up an ES-335. (Epiphone version, I still don't think I could afford the Gibson version).

    As for my JTM30 conversion to 2204, it still sounded a little off to me, so I corrected some values (R25 going into the preamp volume, restored to 470K, and C3 bright cap, restored to 1n) and looked at the tone stack. I put a 4n7 cap between the middle tone pot wiper and ground like the 2204-based combos do, and moved the cap between the treble pot wiper and master volume (which is not there on the 2204 schematic) to replace C11 going into the XLR out section, putting a jumper to connect the treble pot to the master volume. Now I think I'm done.
     
  11. mickeydg5

    mickeydg5 Well-Known Member

    Joined:
    Sep 30, 2011
    Messages:
    24,016
    Likes Received:
    11,010
    Location:
    US of A
    I am not going into detail but from what I can tell you cannot read schematics well. I stopped at R114 and C113. I hope you get what is aimed for however in the end.

    I am referencing the same schematics as Ampmadscientist. Are you?

    And oh, Marshall has far worse schematics.
     
  12. DesolationBlvd

    DesolationBlvd Active Member

    Joined:
    Sep 16, 2017
    Messages:
    68
    Likes Received:
    101
    Yeah, I only know enough to be dangerous. This thread really was more of a chronicle of my misadventure, in trying to improve an amp that sounded better in the store than taking it home, than anything meant to be informative, and I wish I could change the thread name.

    However, most of my changes were not related to simply removing the clipping diodes. I should have been clearer. (Yes, I was using the schematics from ampmadscientist's post.)
     
  13. mickeydg5

    mickeydg5 Well-Known Member

    Joined:
    Sep 30, 2011
    Messages:
    24,016
    Likes Received:
    11,010
    Location:
    US of A
    I was a bit harsh there. I should have been more careful when reading and writing. Maybe I was on too many boards last night. So I am sorry for that. I reread some and brought up the schematics.
    You removed C113 and R114 from the picture. The 470k was part of the low pass but it also curtails some lower guitar frequency. You bumped up all the frequency passing there prior to V101B. That part of the circuit also works in conjunction with the prior gain control. If the bass is still bothering you a lower value resistor can be tried in stead of the jumper. (If it still applies, read on.)
    It seems you were removing and changing quite a bit of stuff. I did not look at it all but was thinking some may have been just as counterproductive as productive.
    I read further and see you basically have a new front end now. Wow. As I said I hope it is all good in the end.
     
  14. DesolationBlvd

    DesolationBlvd Active Member

    Joined:
    Sep 16, 2017
    Messages:
    68
    Likes Received:
    101
    Yeah, the "can't read a schematic" bit hit me a little hard. When I knew I wasn't going to use the JTM30's extra features (clean channel, reverb, FX loop), I started on the total conversion. I was going back and forth between the JTM30 and 2204/4010 schematics, and the JTM30's physical layout, and making every change to get the JTM30's signal path to match the 4010's (minus the low sensitivity input) from the high sensitivity jack to the master volume.

    And it all paid off. I had a SansAmp Classic into some line-level FX and then my interface as my headphone jamming (and in the future, recording) solution, and though the SansAmp sounds pretty good, the JTM30 XLR-to-1/4" low output sounds that much better. The SansAmp will be on its way out.

    My next project will probably be a standalone reverb unit using the tank that's no longer in my converted JTM30's signal path. That way, I could use it with synths too. My Boss RV-5 is acting up and real spring reverb would be a huge upgrade.
     
    Last edited: Oct 25, 2018
  15. mickeydg5

    mickeydg5 Well-Known Member

    Joined:
    Sep 30, 2011
    Messages:
    24,016
    Likes Received:
    11,010
    Location:
    US of A
    Well do not pay any attention to my rants. Obviously you have knowledge and understanding of what was wanted taking care of it fairly expediently. I am kind of surprised you did away with some or so many of the gadgets in that amplifier. You are a man on a mission. March on.
     
  16. ibmorjamn

    ibmorjamn Well-Known Member

    Joined:
    Nov 3, 2012
    Messages:
    7,266
    Likes Received:
    4,989
    Location:
    North
    Necroposting , 2204 single channel ? Interesting. However did we switch from 5881's to EL34's ?
     

Share This Page

  1. This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
    By continuing to use this site, you are consenting to our use of cookies.
    Dismiss Notice