Signal tracing through a jcm800 6V6

Discussion in 'Shock Brother's DIY Amps' started by Tom Kamphuys, Aug 1, 2020.

  1. Tom Kamphuys

    Tom Kamphuys Tele-Meister

    Age:
    40
    Posts:
    471
    Joined:
    Sep 18, 2018
    Location:
    The Netherlands
    Here are some photos of signals.

    Input signal 1kHz, ~200mV pp

    IMG_20200801_231414154.jpg
    First stage output: clean, but slightly triangular. Any idea why? Slew-limiting distortion? Also there is a slight wobble visible on the rising flank, just above 0. Any idea why?

    IMG_20200801_231442685.jpg
    Cold clipper output: only with gain at 8 a flat top becomes visible.

    IMG_20200801_231908394.jpg
    Warm stage: gain at 2-3 we start to get flat spots. Because of the absence of a bypass cap, the warm biasing is not as warm as one might think.

    IMG_20200801_232033400.jpg
    Output of cathode follower: gain at 2-3. Hypothesis: With the gain at 9 the mild cold clipper distortion is swamped by cathode follower distortion.

    Disclaimer: I'm doing this from memory. I should have taken notes...
     
    sds1 and andrewRneumann like this.
  2. String Tree

    String Tree Doctor of Teleocity Ad Free Member

    Posts:
    15,244
    Joined:
    Dec 8, 2010
    Location:
    Up North
    6V6 tube for the output?
    With all due respect, why not EL34's?
     
  3. andrewRneumann

    andrewRneumann Tele-Meister

    Posts:
    291
    Joined:
    Mar 22, 2020
    Location:
    Cincinnati, OH, USA
    Thanks—this is interesting. These were all done with the signal injected at the high input jack?
     
  4. Tom Kamphuys

    Tom Kamphuys Tele-Meister

    Age:
    40
    Posts:
    471
    Joined:
    Sep 18, 2018
    Location:
    The Netherlands
    Yes.
     
  5. sds1

    sds1 Tele-Afflicted

    Age:
    43
    Posts:
    1,432
    Joined:
    May 4, 2017
    Location:
    Orlando, FL, USA
    String Tree likes this.
  6. String Tree

    String Tree Doctor of Teleocity Ad Free Member

    Posts:
    15,244
    Joined:
    Dec 8, 2010
    Location:
    Up North
    Got it!
    I have a '80 JCM800 that I had a tech give an Angus Young Mod.
    KT66's in the back end. Lower Power Output not spoken here! :)
    Plays like a caged Animal.

    Best of luck on your Build.
    I hope it gives you what you are looking for.

    -ST
     
  7. andrewRneumann

    andrewRneumann Tele-Meister

    Posts:
    291
    Joined:
    Mar 22, 2020
    Location:
    Cincinnati, OH, USA
    What do you make of this? Sharp bends equals high frequencies.
    5B7F3F8D-055A-46D6-9608-91C115C4C20D.jpeg
    Ever considered temporarily disabling the Merlin mods on the CF just to see if any discernible difference?
     
  8. Tom Kamphuys

    Tom Kamphuys Tele-Meister

    Age:
    40
    Posts:
    471
    Joined:
    Sep 18, 2018
    Location:
    The Netherlands
    I've no idea...

    Not really. I kinda like how the amp looks. I think it looks nice 'n' clean now. I know no one is ever gonna see it, but I would know... I might have to, in the name of science.
     
    andrewRneumann likes this.
  9. andrewRneumann

    andrewRneumann Tele-Meister

    Posts:
    291
    Joined:
    Mar 22, 2020
    Location:
    Cincinnati, OH, USA
    Hey if you are happy with it, then don’t go changing it. Before experimenting, maybe someone could theorize why that arc protection could cause problems. I have to go back and read that chapter.
     
  10. andrewRneumann

    andrewRneumann Tele-Meister

    Posts:
    291
    Joined:
    Mar 22, 2020
    Location:
    Cincinnati, OH, USA
    On the CF trace, do you have the output signal inverted?
     
  11. Tom Kamphuys

    Tom Kamphuys Tele-Meister

    Age:
    40
    Posts:
    471
    Joined:
    Sep 18, 2018
    Location:
    The Netherlands
    I have never inverted the signal (with the probes and/or scoop).
     
  12. Tom Kamphuys

    Tom Kamphuys Tele-Meister

    Age:
    40
    Posts:
    471
    Joined:
    Sep 18, 2018
    Location:
    The Netherlands
    I was a bit worried. Why would tubes introduce the sharp bends? Everything is gradual with tubes, right? There is a protective diode on the cathode follower though...

    Then I read this on Merlin's site:
    The resistor is included to prevent any switching noise that might be caused if the cathode follower were overdriven (in a hifi circuit the resistor would be omitted).

    Well, it's constantly overdriven...

    So I did the unthinkable and desoldered the diode.

    With:
    IMG_20200802_203213569.jpg

    Without:
    IMG_20200802_203102691.jpg

    No difference.
     
  13. tubegeek

    tubegeek Friend of Leo's

    Age:
    59
    Posts:
    2,495
    Joined:
    Jan 31, 2020
    Location:
    Brooklyn, NY
    Is the cathode follower DC biased or does it have a coupling cap feeding it? What's the bias voltage on the CF?
     
  14. Tom Kamphuys

    Tom Kamphuys Tele-Meister

    Age:
    40
    Posts:
    471
    Joined:
    Sep 18, 2018
    Location:
    The Netherlands
    DC coupled cathode follower. Bias is ~167V.

    Here is the schematic. Basically a (robrob) jcm800.

    MP2104_Master_Volume_6V6_Schematic.png

    And here's the voltage chart:
    image%3A38890.png
     
    Last edited: Aug 3, 2020
  15. Tom Kamphuys

    Tom Kamphuys Tele-Meister

    Age:
    40
    Posts:
    471
    Joined:
    Sep 18, 2018
    Location:
    The Netherlands
    Adding a 10k resistor in parallel to the 6.8k dropping resistor does not change the 'sharp bends'. Neither does bypassing the Merlin grid stopper before the cathode follower.

    Adding a 57k resistor in parallel to the 100k cf cathode resistor (eff. 36k) does solve the sharp bends. As discussed at the Valvewizard site, it also distorts more.

    Gain 3 (100k cat res would have shown sharp bends):
    IMG_20200803_155605779.jpg

    Gain 10
    IMG_20200803_160514599.jpg

    Also weird, but no sharp bends.

    I still don't understand what's going on though.

    The wave looks like a dotted line. I've seen this before. Not always. Could that be a high frequency signal/oscillation that is only visible/present with high gain?
     
  16. Tom Kamphuys

    Tom Kamphuys Tele-Meister

    Age:
    40
    Posts:
    471
    Joined:
    Sep 18, 2018
    Location:
    The Netherlands
    Kinda agrees with Merlin:
    "[This] is why this circuit is so often used in high gain amps- it can warm up a signal that already contains too many high-order harmonics, and return a rather fuzzy sound to a rich, creamy distortion tone."
     
  17. andrewRneumann

    andrewRneumann Tele-Meister

    Posts:
    291
    Joined:
    Mar 22, 2020
    Location:
    Cincinnati, OH, USA
    I’m pretty much out of ideas. Your original traces all made sense to me except the phase seemed to be either shifted 180 or inverted, and the weird angle of the clipping on the CF output. I agree with your analysis that the cathode follower (actually, the combination of it and the prior gain stage) is clipping well before the “cold clipper”. If you were to bypass the cold clipper cathode, then you might actually have some real cold clipping, but that would just push the next stages farther into overdrive so there’s really no big advantage. Maybe someone else with a similar build and scope could provide some comparisons?
     
  18. mrriggs

    mrriggs TDPRI Member

    Age:
    40
    Posts:
    46
    Joined:
    May 24, 2020
    Location:
    Vancouver, WA
    Nope. That is a myth perpetuated by tube lore. Just because the player perceives the sound as smoother does not mean that it shows up on an oscilloscope as curvy smooth lines.

    Look at the trace in my avatar. That is the output of a 5F1 laid over the input signal. The bottom of the waveform is chopped off perfectly square.

    The dotted lines you see are likely just the switching circuit of your oscilloscope. With the rare exception, an oscilloscope only has a single electron gun so it can only draw one line at a time.

    A dual trace scope either draws one channel then the other alternately or switches back and forth very quickly drawing a little of each. The alternate sweep works well on fast signals (MHz) but can be too flickery with slower (audio) signals. The 'chop' method works better for slower signals. It's normally unnoticeable because of the persistance of the screen (and our eyes) and because the chop frequency usually doesn't sync with the frequency that you are testing.

    Some scopes will have a switch or button to allow you to switch between alternate or chopped traces.

    If you do happen to notice the chopping [and don't have an ALT-CHOP switch] then you can change your test frequency slightly so that it doesn't sync, turn off the channel that you aren't currently looking at, or simply ignore it.
     
  19. andrewRneumann

    andrewRneumann Tele-Meister

    Posts:
    291
    Joined:
    Mar 22, 2020
    Location:
    Cincinnati, OH, USA
    Is there actually anything wrong with the sound of the amp since you moved the MV to pre-PI?
     
  20. Tom Kamphuys

    Tom Kamphuys Tele-Meister

    Age:
    40
    Posts:
    471
    Joined:
    Sep 18, 2018
    Location:
    The Netherlands
    Sounds ok to me, but don't have a lot experience with all kinds of amps. Still quite a lot of highs, but I think that is to be expected with all the bright caps.

    The problem is that I don't understand what is going on...
     
IMPORTANT: Treat everyone here with respect, no matter how difficult!
No sex, drug, political, religion or hate discussion permitted here.