Getting a better OD sound from a 5F2a build

Discussion in 'Amp Tech Center' started by sirnoel, Jul 28, 2021.

  1. sirnoel

    sirnoel Tele-Meister

    Posts:
    157
    Joined:
    Feb 21, 2012
    Location:
    Chicago
    Hey all you circuit junkies. I have been slowly working on putting together an amp for a friend who is a harmonica player. He had played his green bullet through a champ circuit I built and loved it. I thought I might built him something a little more flexible and harp focused. I chose a 5F2a for the added tone control and bought one of those 15 watt upgraded OT from Classic Tones before they closed.

    I build these by depopulating old radio amp chassis and wiring the schematic point to point inside. It's a bit of a brain teaser but it's yielded some very sweet sounding and stable amps. So far this Harp-amp ain't one of them.

    It's quite prone to feedback, without having much of that singing, crunchy midrange distortion that sounds good with an amplified harmonica. When you plug a guitar in it sounds nice at lower volumes and has pretty decent headroom, but once you turn it past half way the distortion you get is not good sounding. It's choked, there is no sustain and it cuts out noisily like a gated fuzz.

    I've tried a number of different tubes, I've played with the voltage on a variac, I put the NFB resistor on a potentiometer so I can set it anywhere between 10k and 60k. I've tried .1, .022 and .01 coupling caps. Nothing makes this sound better.

    So...

    What would you do next? Does that stepped up OT compromise the overdrive?
     
  2. 2L man

    2L man Tele-Holic

    Age:
    61
    Posts:
    693
    Joined:
    Nov 23, 2020
    Location:
    Finland
    Have you "drawn" the loadline. Universal Loadline Calculator is great. Amp idling without signal what is carhode resistor value and voltage over it? What is the tube type, its anode and screen voltages against ground and OT impedance? You can input data yourself or post those and I or someone else do it for you.
     
  3. Lynxtrap

    Lynxtrap Tele-Afflicted

    Posts:
    1,120
    Joined:
    Feb 29, 2016
    Location:
    EU
    I would check all voltages, and the bias to see if the output tube is happy with the 470R.
     
    2L man likes this.
  4. Junior Little

    Junior Little Tele-Meister

    Posts:
    310
    Joined:
    Jul 26, 2018
    Location:
    Rural Illinois
    You want crunchy blues harp? Build one of these. Your harp playing friend will thank you.
    I've built three and they all sound fantastic!
     
    sirnoel likes this.
  5. King Fan

    King Fan Poster Extraordinaire Ad Free Member

    Posts:
    6,848
    Joined:
    Jan 1, 2013
    Location:
    Salt Lake City
    Is it better or worse with NFB lifted? PFB isn’t always a squeal, it can be some other bad sound.

    Is your build all 100% stock except for the big OT? I’d make it so.

    Voltages? Do you have a scope? If not, an audio signal trace, using a second amp, can be helpful.

    Edit: Audio trace suggestions I copied years ago from @FenderLover . No real need for a signal generator IIRC, I think I used a looper playing a simple chord.

    You’ll need a second amp. Remove the plug from a guitar cable and solder a 0.1uF cap on the center lead. The other lead of the cap is your audio probe. Connect the cable's shield to the chassis of the amp under test with an alligator cable.

    Plug the other end of the cable into your second amp with its volume as high or low as necessary and start to follow the input signal on your test amp from stage to stage. Since you have sound, you will listen for when things start going south. With respectable volume settings on the test amp, it should be clear as a bell as you follow it through.
     
    Last edited: Jul 29, 2021
    andrewRneumann and sirnoel like this.
  6. jimgchord

    jimgchord Tele-Afflicted

    Age:
    48
    Posts:
    1,638
    Joined:
    Apr 3, 2018
    Location:
    victor,ny
    Ive built a few of these for harp guys, they prefer variable nfb pot and a 12ay7 in v1.
     
    sirnoel likes this.
  7. Lowerleftcoast

    Lowerleftcoast Friend of Leo's

    Posts:
    4,226
    Joined:
    Dec 30, 2019
    Location:
    california
    The OT should be fine. Are you using the red and blue primary 8K wires of the OT?
    Are you describing *blocking distortion*? The typical *fix* is to use a large 470K to 1M control grid stopper on the power tube.
    Squealing feedback? Is this with the guitar or only with the mic?

    Just because you are building point to point does not mean the lead dress is acceptable. Show us some pictures of what is going on.

    You may want to scrub off some high range frequencies. Harp players don't use them for the most part. Loosing the highs can mean less in terms of feedback.
     
  8. sirnoel

    sirnoel Tele-Meister

    Posts:
    157
    Joined:
    Feb 21, 2012
    Location:
    Chicago
    Thanks! That is not something I know about and am very excited to spend the day doing.
     
  9. sirnoel

    sirnoel Tele-Meister

    Posts:
    157
    Joined:
    Feb 21, 2012
    Location:
    Chicago
    Yes. I should have mentioned that I did some bias work. I have a salvaged PT with a higher than spec B+ so I found a 680 that measured around 700 was putting me in the right spot. But I should retest as I have swapped out some components since then.
     
  10. sirnoel

    sirnoel Tele-Meister

    Posts:
    157
    Joined:
    Feb 21, 2012
    Location:
    Chicago
    Yes on the 8K primary
    I think I will try the grid stopper, thanks. I've not dealt with blocking distortion before.
    I will upload some pictures soon, always good to get your work checked.
    I think that loosing some highs is a good idea, the tone knob is completely unusable above 10 o'clock. I've been playing around with different coupling caps, any other suggestions on taking the highs out?
     
  11. timfred

    timfred Tele-Meister

    Posts:
    155
    Joined:
    Mar 14, 2016
    Location:
    Northern CA
    As @Lowerleftcoast mentioned, big grid stoppers (around 1M) on either the power tubes or the second (driver) triode. Or a small (hundreds of pF) plate bypass capacitor on the driver triode.

    But I'd start with a 50K grid stopper on the power tube and/or a 50-100K grid stopper on the driver triode to control any blocking distortion or oscillation. You may also want to experiment with eliminating one or both of any cathode bypass capacitors you have on the tridoes.
     
    Last edited: Jul 29, 2021
    Lowerleftcoast likes this.
  12. Lowerleftcoast

    Lowerleftcoast Friend of Leo's

    Posts:
    4,226
    Joined:
    Dec 30, 2019
    Location:
    california
    There are a couple of typical places to scrub off highs, the coupling caps won't do it.

    You can try different values in the tone circuit, to get rid of some highs.

    Marshall is known for placing a small cap from plate to cathode on preamp tubes. They also are known to parallel a cap on plate resistors. You can try different values to taste. Marshall tries to get stability using ~100pF or so. Ime a harp amp would start at ten times that to lose highs.

    Fender used a low pass filter before the PI. On the 5F2A probably just before the 6V6 220K grid leak resistor would be workable. Ime 330K resistor with a .0015uF cap might work, but again let your ears be your guide.

    Harp amps are different beasts. For a Harp amp, I would anticipate having some issues with the input resistors. Chances are the 1M is not enough and the 68K could be increased. I would try 3M to 5M and 68K to 330K, maybe more. I would be trying different mu preamp tubes as well. Probably I'd be trying 150K or higher plate resistors and corresponding cathode resistors. IIRC, Gibson and Rickenbacker used up to 470K plate resistors on their SE amps. Anyway, let your ears be your guide.

    Edit: You might also try a cap parallel with the NFB resistance to get rid of some highs. Maybe 470 to 1000pF?

    @King Fan mentioned positive feedback as a possible cause of the distortion, so check that out.
     
    Last edited: Jul 29, 2021
  13. 2L man

    2L man Tele-Holic

    Age:
    61
    Posts:
    693
    Joined:
    Nov 23, 2020
    Location:
    Finland
    Universal Loadline Calculator is quite simple to use. When you have messured your cathode biased amplifier plate and cathode voltages against ground the plate/anode voltage you use in Calculator needs to be same amount less what you did measure from cathode because tube databases use voltage between anode and cathode.

    Screen voltage is good to have also been measued because it effects anode current, which then change everything else, just like physical amplifier.

    When bias is inceased happens most interesting feature of the tube amplifier. Bias current thru primary coil stores energy to output transformer. On loadline we see how loadline rises and same time the max voltage on right end of loadline moves to the right. When single ended A-class biased amp operates it is possible to measure almost double the anode voltage sine wave on tube anode. When amp is overdriven the distorted signal peak to peak voltage can be more than double what B+1 is.

    When "headroom" is set there appear also boxes for distortion estimates. Calculator headroom is a result of simulated drive signal on control grid and it is assumed symmetrical and not distorted. About 40V less than anode voltage is ok start value for max drive but it needs to be test that green area does not pass grid lines on left side or meet zero current on right side.

    For single end power stage starting low and inceasing headroom, first there usually comes little but mostly 2nd harmonic. This is if (and more or less when) output does not sweep equally to both side of "operating point". There are few variables which effect there but only HiFi builders pay attention to them :) but we guitarist just like 2nd "warm harmonics", even seeing it on computer screen ;)

    Then when headroom is still increased 2nd harmonics increase more rapid if other end meets "the end" sooner and flattens. When headroom is still increased the ratio between 2nd and 3rd harmonics turn more towards 3rd when output begins to flatten on both ends.

    When testing push pull power stage loadline there is no 2nd harmonics when it gets cancelled on OT.

    On AB class push pull loadline there is a point where loadline begins to rise steeper and it is where other tube quit driving current and other "see" the full impedance.
     
    Last edited: Jul 29, 2021
  14. sirnoel

    sirnoel Tele-Meister

    Posts:
    157
    Joined:
    Feb 21, 2012
    Location:
    Chicago
    This is super helpful. A few questions.
    -I find the Tone Knob very confusing. I know there is a .005 cap and a .0005 cap, If I wanted a more usable sweep for Harp which should go up and which should go down? My understanding is that they are more tied to the frequency cut off point than how much they cut or boost.
    -I'm a bit muddled on the 1M resistor at the input, because the signal coming from a harp mic is so much more than from a guitar pickup wouldn't it be helpful to lower the resistance so that more of the signal goes to ground like in the 2nd input on a fender amp which sees 136K to ground rather than the 1M?
    -My NFB is a 10k resistor in series with a 50K pot set up as a variable resistor. If I put .001 uf cap in series with the 10k will that work?

    Thanks for weighing in on this.
     
  15. dan40

    dan40 Friend of Leo's

    Posts:
    2,296
    Joined:
    Aug 19, 2015
    Location:
    Richmond Va
    All great suggestion so far but I would try King Fan's suggestion of unsoldering the nfb wire from the output jack first. If this simple test cures the issue, all you need to do is resolder the nfb wire and swap the two wires on the OT primary. Quick and easy test that may save a lot of time troubleshooting.

    Edit: You may prefer the tone with the nfb disconnected and if so, simply insulate the end of that wire and tuck it away under the board. By removing the nfb circuit temporarily, you can determine whether or not your circuit is experiencing positive feedback.
     
    Lowerleftcoast likes this.
  16. Lowerleftcoast

    Lowerleftcoast Friend of Leo's

    Posts:
    4,226
    Joined:
    Dec 30, 2019
    Location:
    california
    Yeah. it is interactive with the volume pot. The higher the volume the less the 500pF treble bleed works. You might try lowering that500pF cap to none or 100pF or 250pF but if the volume is always up it won't make a difference. It would be the same as none. The corner frequencies of the .005uF looks like it should handle harp OK.
    I see what your angle is. With my suggestion, I am trying to match the high impedance mic by using a multi meg resistor.
    It would be easy enough to clip in caps across either or both to give it a listen.

    I agree with dan40 and King Fan check out the NFB (possible PFB) first.
     
  17. sirnoel

    sirnoel Tele-Meister

    Posts:
    157
    Joined:
    Feb 21, 2012
    Location:
    Chicago
    Thanks, I should have mentioned that I have already chased this particular gremlin, so I know that the primaries are in the right place. It howled like a banshee when I first wired her up and cut the NFB (it was on a switch initially) so I switched them then. I have the NFB on a variable resistor that can range from 10k-60k now. I have 4 and 8 ohm taps so having some variation seems right, it also seems to help with tailoring your feedback to gain ratio.
     
    dan40 likes this.
  18. sirnoel

    sirnoel Tele-Meister

    Posts:
    157
    Joined:
    Feb 21, 2012
    Location:
    Chicago
    _I do know the primaries are correct as I already switched them (see above).
    _I swapped in a 100pf cap for the 500pf in the tone circuit. It definitely helps. I tried cutting it out entirely and it felt like the sweep stopped at halfway. I might swap out the 1M linear pot for a 1M audio, I wouldn't mind if the treble boost was only in the last 10% of the sweep. Does that make sense?
    -Adding the 1Meg grid stopper definitely solved my distortion problem. Sounds great right now. I might swap in a few smaller resistors, instead just to make sure I'm not overkilling it.
    -From there I'm going to start tailoring the EQ by alligator clipping in some different resistors into the input section and trying out a few capacitors with the NFB resistors.

    Thanks for everyones expected advice.
     
    Lynxtrap and Lowerleftcoast like this.
  19. Lynxtrap

    Lynxtrap Tele-Afflicted

    Posts:
    1,120
    Joined:
    Feb 29, 2016
    Location:
    EU
    Great!
    A screen resistor on the output tube wouldn't hurt either.
     
  20. jimgchord

    jimgchord Tele-Afflicted

    Age:
    48
    Posts:
    1,638
    Joined:
    Apr 3, 2018
    Location:
    victor,ny
    These amps suffer from way too much gain in the preamp stage
     
IMPORTANT: Treat everyone here with respect, no matter how difficult!
No sex, drug, political, religion or hate discussion permitted here.