Not sure if my Plexi build has problems

Discussion in 'Shock Brother's DIY Amps' started by itsGiusto, Oct 26, 2019.

  1. Nickfl

    Nickfl Tele-Afflicted

    Posts:
    1,623
    Joined:
    May 24, 2016
    Location:
    Florida

    I'd definitely move the jumpers to the front side of the board if that is what it takes to make you comfortable shortening the other wires (I prefer to avoid putting anything on the back of the board myself). Leaving the connections to the pots and sockets as long as they are is just asking for trouble. Leaving those leads so long adds up to several feet of extra wire to act as antennas inside the amp and it also leaves so much wire tangled up that you really can't do anything effective with regard to lead dress that would minimize their interaction.

    IME the problems you are having is spot on for a high frequency parasitic oscillation. The volume dropping out and pumping like a tremolo can be due to the oscillation causing your voltages to sag and this can then cause the parasitic feedback loop to drop in and out as the signal level drops below the threshold where it occurs. Also the fact that your NFB pot is effecting it, and that it is worse at higher treble tone control settings, all of which shows it is related to higher gain at higher frequency.

    I would guess that the issue is occurring somewhere in the wires running from the board to the tube sockets, but its hard to have any idea of where specifically. You could try to find it by chop sticking, but with it being a intermittent problem that only occurs with a lot of signal that probably isn't going to work.

    Does the problem get better or worse if you switch between split or common cathode for v1? I've heard people occasionally have some issues with the common cathode version.
     
    dan40, BigDaddy23 and itsGiusto like this.
  2. itsGiusto

    itsGiusto Tele-Meister

    Posts:
    274
    Joined:
    Dec 31, 2017
    Location:
    Massachusetts
    I don't recall if it's worth split or common cathode. I think all the clips I recorded were common, I think.

    What is chop-sticking?

    Also, when people refer to "lead dress" or "proper lead dress", what do they mean exactly? What particular guidelines should I be following?
     
  3. King Fan

    King Fan Friend of Leo's

    Posts:
    4,112
    Joined:
    Jan 1, 2013
    Location:
    Salt Lake City
    Chop-sticking: Tapping connections / moving wires using a non-conductive wooden chopstick (and only one hand) while the amp is on and making noise. Nick is saying that too many long wires will (among other things) make it hard to move the 'right ones.'

    Lead dress is a series of things that make wires short, direct, and neat while taking steps to cross current over signal at right angles, avoid parallel runs of wires that broadcast RF next to wires that can pick up RF, etc.

    I'm no electrical engineer, so I often just try to copy the way the pros do it. Both Ceriatone factory builds and Marshall originals are nice examples.

    upload_2019-10-28_11-5-36.png

    upload_2019-10-28_11-7-18.png
     
    Last edited: Oct 28, 2019
    dan40, Nickfl and Paul-T like this.
  4. BigDaddy23

    BigDaddy23 Tele-Meister

    Age:
    45
    Posts:
    264
    Joined:
    Jan 4, 2019
    Location:
    Australia
    @bikescene 's build is a great example of good lead dress in a Marshall style amp (hope he doesn't mind me linking his thread!) https://www.tdpri.com/threads/50w-plexi-inspired-circuit-build-progress.984709/#post-9394549

    @Nickfl and @King Fan have nailed it - Really try to shorten your leads as much as possible. The signal coupling between adjacent wires is something to look for. I had a similar type issue in my Princeton under heavy OD and was able to get rid of it by re-routing and positioning wires identified in the chopsticking process.

    You are gonna have a very versatile amp there - great build!
     
    itsGiusto likes this.
  5. bikescene

    bikescene TDPRI Member

    Posts:
    9
    Joined:
    Aug 30, 2017
    Location:
    New York
    I hope you get your noise issues sorted. Tube rectifier with the switchable lead option seems very cool.

    I referenced the 50W Metro Amp build guide when I was building my amp, and based a lot of the wiring layout on that guide. It’s hosted on the Valvestorm site now.

    I did a google image search of guy shots of boutique builds from Germino, Metro, Ceriatone, etc.
     
  6. corliss1

    corliss1 Friend of Leo's

    Posts:
    3,437
    Joined:
    Sep 13, 2008
    Location:
    Lansing, MI
    I also don't believe in underboard wires, both for ease of maintenance, and when I need to replace something the wire doesn't fall into oblivion.

    I also always preach the "built stock first" mantra. Everyone comes in with "I want to build amp _____ but would it be better to start with some mods right from the start?" In my opinion no. Yeah, it may take some extra effort to mod later, but you'll find out two important things:

    -if you really *need* to mod it
    and
    -what the stock amp sounds like so you have a base reference.

    You'll also have a pretty looking (and hopefully pretty functioning amp) without any of the odd behaviors that mods can introduce.

    I would say you've gotta get the layout cleaned up before doing any actual troubleshooting. It's not worth it to try and chase down weird sounds and behavior when there's the potential for oscillation and crosstalk like what has already been mentioned. If it's neat and tidy and you still have issues, then you can start looking at things to fix.

    I had a Marshall.......something, I don't remember, doesn't matter. It had been modded for higher gain, and I had to get to the underside of the board as part of a repair. The input wires were held down with a rubber cement like substance and when I put it all back together I had to get them in EXACTLY the same spot or else it just howled like crazy. So even small changes matter. When chasing down weird things like this you'll find moving a wire a quarter of an inch might make the symptom better, or it might make it worse. Even if it makes it worse, that's a clue. With the current state of the amp you aren't able to do that type of testing and science and get reliable results.
     
    D'tar, King Fan and David Barnett like this.
  7. itsGiusto

    itsGiusto Tele-Meister

    Posts:
    274
    Joined:
    Dec 31, 2017
    Location:
    Massachusetts
    I understand and respect that. And if I had more time to build amps, more money, and more space to keep them, I'd just build stock amps. The purpose of me putting these mods in is so that I can get more iconic amp sounds, so I can hear what they all sound like, and learn about tones. Stock, the amp is a 67 plexi. With the tube rectifier, the amp is basically identical to a JTM50. With the lead stack, split cathode, and swapped PPI coupling caps, the amp is close to a early 70s plexi, etc.

    Also, I was thinking of building this stock first, but Nik from Certiatone strongly encouraged me to plan out my mods beforehand (not that I want to sound like I'm placing the blame on his head... He just inspired me a bit). I know the mods are causing issues for me now, but in theory they could have ended up with a much nicer build. I was able to re-lay-out the board to accomadate for various cap differences, etc, as opposed to what I would have likely done if I did the mods afterward, which would have been to just have all the mod components floating at the switches.

    That's interesting. IDK, maybe I was spoiled by having some other easier amp builds and mods in the past, but for other amps I've worked on, like tweed, blackface, and Vox amps, they just weren't very finicky! I could route the wires every which way and it'd sound fine. I guess the Marshall is just a very different beast? But why is that? The Tweed and Vox are capable of getting some gain, like the Marshall.

    I'll definitely work on cleaning up the build, and post back here once done.
     
  8. corliss1

    corliss1 Friend of Leo's

    Posts:
    3,437
    Joined:
    Sep 13, 2008
    Location:
    Lansing, MI
    The short answer is gain. Way more signal going through the wires means way more sensitivity in the layout. Sure, a tweed amp has gain when you crank it, but not nearly the same amount as a Marshall. Try and get a clean sine wave at the output of any Marshall and it's just not possible :D
     
  9. King Fan

    King Fan Friend of Leo's

    Posts:
    4,112
    Joined:
    Jan 1, 2013
    Location:
    Salt Lake City
    I don't think folks are down on your mods. In fact it seems like several of them are getting a big thumbs up. Nik's key word there may be "plan." Many experts suggest you drill new holes, mount extra turrets, and plan future components and wiring. But, at least on any build more complex than we've done before, experts also suggest *building* stock first and then, maybe one at a time, adding in your planned mods. The biggest of big reasons? The troubleshooting one.

    Also I don't think folks mean to criticize what you've done so far. For sure you'd be tackling a lot of work to take down the existing wiring. And bringing it back stock, getting it working, and adding a mod at a time would be nearly as much work as building a new amp. It's a tough road, but I'd think of it as a way to navigate back to the main highway and get in a free look at the beautiful scenery. Building amps is fun. Having one that doesn't work is not fun -- don't ask me how I know.
     
  10. itsGiusto

    itsGiusto Tele-Meister

    Posts:
    274
    Joined:
    Dec 31, 2017
    Location:
    Massachusetts
    A few questions.

    1. Which wires should I be twisting together, which ones should I be zip-tying together, but not twisting, and which should I be keeping far apart? Some specific ones I want to know about:
    Wires to and from dc caps
    Wires to and from the output transformer
    Wires to and from the power transformer (in particular, should I keep wires from separate taps away from each other, or does it matter?)
    Choke wires
    Any other specific ones to call-out?

    2. Is it important to keep ground wires short and far away from other wires, or do ground wires not really matter much in terms of lead dress? What about wires carrying DC current?

    3. Which wires do you recommend using shielded cable for? I'm currently using it for the wire from the NFB to the presence pot, the wire to and from the effects loop, and the wires running from the input jacks to the board and the board to the first preamp stages.
     
  11. corliss1

    corliss1 Friend of Leo's

    Posts:
    3,437
    Joined:
    Sep 13, 2008
    Location:
    Lansing, MI
    We twist AC signal wire. It's the same reason that phone/ethernet cable is twisted pair - it reduces crosstalk. That's why the wires from the power transformer and the heater wires are always twisted, as those are regular AC signal.

    DC signals we don't care about as much. We do want to make sure wires cross at 90 degrees when possible, and don't want to be sloppy, but our audio signal is not usually going to be present on these except for the short runs from a plate of a tube to a capacitor.

    Ground wires don't seem to be as critical in their lead dress as others, but you do want to make sure you're grounding to the correct point.

    Shielded cable is usually reserved for the signal path, especially early on the amp. It's common to run that right to the input jack and then to the first tube. If we get noise in the signal early, it will only get amplified throughout the rest of the gain stages.
     
    Nickfl likes this.
  12. itsGiusto

    itsGiusto Tele-Meister

    Posts:
    274
    Joined:
    Dec 31, 2017
    Location:
    Massachusetts
    In the Ceriatone images, they're not twisting the output transformer secondary wires, even though they are AC. Should I be twisting those or not? Same goes for the power transformer HT wires going to the tube rectifier.

    Also some AC wires should not be twisted together, right? Couldn't twisting certain ones together (such as signal wires to and from the preamp tubes) cause more crosstalk between individual wires, leading to parasitic oscillation?
     
  13. corliss1

    corliss1 Friend of Leo's

    Posts:
    3,437
    Joined:
    Sep 13, 2008
    Location:
    Lansing, MI
    We don't want to twist signal wires. We do want to twist wires that have the same things on them (heaters, secondaries) so they cancel out.

    Here's a blackface bassman I pulled from an online search, before being serviced:

    screenshot.1.jpg

    so that's what we want to see. Both primary (black) and secondary (red and green) are twisted. The center taps to ground are twisted as well, partly for function partly because it makes it easier to deal with.
     
    itsGiusto likes this.
  14. itsGiusto

    itsGiusto Tele-Meister

    Posts:
    274
    Joined:
    Dec 31, 2017
    Location:
    Massachusetts
    In the Ceriatone build image, they're twisting the signal wires from the PPI coupling caps to the grids of the power tubes. Should I avoid doing that?

    [​IMG]amp_7 by lin ceriatone, on Flickr

    Also, the two wires leading from the power tubes to the OT should not be twisted?
    How about the choke wires? Since they're DC, they should probably not be twisted as well?

    Also, does it matter if the heater wires for the rectifier tube run close to the PT secondary HT wires? What about if either of those run close to the 120v power input wires, or the PT primary wires?
     
  15. itsGiusto

    itsGiusto Tele-Meister

    Posts:
    274
    Joined:
    Dec 31, 2017
    Location:
    Massachusetts
    Does anyone have a recommendation for a brand that makes a multicolored set of stiff wires which will stay where you put them, and also rated up to 600v? I can't seem to find one online that I know is suitable, but I want to use this for trying to do better lead dress.

    Above, people mentioned that solid core is better than stranded for staying where you put it. What about "tinned stranded wire", will that be stiff enough to stay?
     
  16. dan40

    dan40 Tele-Afflicted

    Posts:
    1,355
    Joined:
    Aug 19, 2015
    Location:
    Richmond Va
    This is the wire you want. Valvestorm sells it by the foot and everyone that has used it has been happy with its's performance. Pretinned so it's easy to solder and holds it's shape very well. I like to use the 22awg for everything in the amp except for the ground runs which I use 18awg.

    http://valvestorm.com/Products/Wire

    Most Marshall style kits recommend twisting the wires leading from the PI to the power tube grids and the OT secondary wires. In Fender type kits, you rarely see these wire pairs twisted. Definitely a good idea to twist your AC filament leads, PT secondary leads and AC power wiring. Others may be twisted just for ease of installation and appearance.

    It really helps to study the lead dress of different builders to get an idea of what's important and whats not. As the gain levels are increased in an amp's design, lead dress becomes more an more critical for quiet performance. These instructions to the old Metroamp kits have a bunch of good info on how to properly dress a Marshall circuit.

    http://valvestorm.com/sites/default/files/50_WATT_KIT_INSTRUCTIONS.pdf
     
  17. itsGiusto

    itsGiusto Tele-Meister

    Posts:
    274
    Joined:
    Dec 31, 2017
    Location:
    Massachusetts
    Thanks! But are you sure to twist the OT secondary wires (that would be the wires leading to the 16, 8, and 4 ohm selector, right?) Most pictures I see, including Ceriatone's and the link you sent, show those wires not twisted, but zip-tied together. Did you mean OT primary wires?
     
  18. dan40

    dan40 Tele-Afflicted

    Posts:
    1,355
    Joined:
    Aug 19, 2015
    Location:
    Richmond Va
    Page 14 and 15 of the Metroamp build guide shows the OT secondary wires twisted in a bundle and heading to the selector switch. I do not believe that this was done for noise reasons though, more for ease of routing I would guess. I have done it with a twist and without twisting and never had any issues with noise. Do be careful with the wire leading from the board to the presence pot though. This needs to be run in a certain manner to avoid crosstalk with the tonestack wiring and George gives details for proper routing in the build guide also.
     
    itsGiusto likes this.
  19. King Fan

    King Fan Friend of Leo's

    Posts:
    4,112
    Joined:
    Jan 1, 2013
    Location:
    Salt Lake City
    Don't overthink this. Hey, I'm the guy who put up the Ceriatone pic, but you don't want to build *that* compulsively. Our friends @corliss1 and @dan40 are exactly right -- copy vintage Fender or Marshall neatness, not that 'built by nanobots' look.
     
    corliss1 and dan40 like this.
  20. itsGiusto

    itsGiusto Tele-Meister

    Posts:
    274
    Joined:
    Dec 31, 2017
    Location:
    Massachusetts
    Gah! I just spent weeks basically rebuilding the amp, trying to keep all the leads routed away from each other, and shortening them up. It's still not the neatest, but the leads are a lot shorter and I tried to pay good attention to keeping signal wires away from noisy components, etc. But STILL, I'm getting the same weird buzzy modulating tones, I was getting before. I don't really think it any sounds better at all. It sounds like an octave fuzz, but has weird spitty modulating overtones that modulate around if you hold a note or chord, especially at higher gain settings.

    Here are some trends I'm noticing while playing:

    1) the weird sounds happen more on the normal channel, especially if I turn up the volume between 5 and 8

    2)However, I can get similarly weird tones to happen on the treble channel if I switch to the larger bypass cap on the treble channel.

    3) the weird sounds are more prominent when i dial the NFB pot from 27k to about 75k to 100k.

    I'm feeling so exhausted from this project, and I'm not sure what to do to diagnose it... I guess I should try playing something recorded through a loop pedal while chopsticking some of the wires.
    However, what if the problem is actually an issue with the power supply being faulty and unable to supply clean power while under load? How could I diagnose that? I have a weird suspicion about the power supply, due to this sounding like it's almost like a crossover distortion. And I was reading some weird, really high voltages before on the tube rectifier's heater wires. Maybe that's symptomatic of a malfunctioning power supply?

    I'll post some sounds clips and shots of the chassis later...
     
IMPORTANT: Treat everyone here with respect, no matter how difficult!
No sex, drug, political, religion or hate discussion permitted here.


  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.