Hollow-State Hard-Clipper

Discussion in 'Shock Brother's DIY Amps' started by mrriggs, Jun 28, 2020.

  1. mrriggs

    mrriggs Tele-Meister

    Age:
    41
    Posts:
    229
    Joined:
    May 24, 2020
    Location:
    Vancouver, WA
    I've got boxes of old TV-tubes and find it fun to come up with uses for them. There's like two dozen 6AL5 dual-diodes in there which got me thinking about building a hard-clipping distortion stage using tube diodes. I've also got this old Bogen PA amp chassis that I picked up at a flea market twenty years ago. To make this a really fun and challenging project, I stipulated that I had to use only the transformers, pots and tube sockets that were already in the chassis, and everything else had to be parts that I had on-hand, no buying parts for the project.

    Unfortunately, the 7-pin tube socket had been robbed out of the chassis so no 6AL5 this time. I did, however, have a 9-pin 6FM8 which has two diodes like the 6AL5 AND a high-mu triode for a gain stage. A 6N2P takes up the other 9-pin socket and does the duty of pre-amp and cold-clipper.

    The output transformer has a really low plate-to-plate impedance, perfect for a TV sweep tube like 6DQ6B. This chassis has Novar sockets and I have some 6GT5's which are the guts of a 6DQ6B shoved into a Novar tube. That leaves two more Novar sockets. I dug through the datasheets for the Novar tube in the box and came up with the 6GF7 which has one high-mu triode and one power triode. With two 6GF7's I could use the two high-mu triodes for a long-tail phase inverter. That leaves the two power triodes for another wacky idea I've been wanting to try... screen regulators hooked to a "Volume" knob on the front of the amp to adjust when the power tubes break up.

    After a bit of trial and error, this is the circuit I've settled on.

    [​IMG]

    The hard-clipper is before the cold-clipper and is configured to clip sooner. The tube-diodes can be switched off which will allow the cold-clipper to be overdriven. With the gain turned down and the "Power" knob [screen voltage] turned down you can easily overdrive the power tubes. With the Power knob on zero it's only putting out 400 mW of clean power.

    Just for fun, I recorded a short sound sample of each stage of distortion. Can you identify which one is the hard-clipping, cold-clipping and power-amp?

    http://www.gofastforless.com/junk/HSHCclip1.mp3

    http://www.gofastforless.com/junk/HSHCclip2.mp3

    http://www.gofastforless.com/junk/HSHCclip3.mp3
     
    Last edited: Jun 28, 2020
    LetItGrowTone and FenderLover like this.
  2. Mexitele Blues

    Mexitele Blues Tele-Afflicted

    Age:
    43
    Posts:
    1,725
    Joined:
    Mar 7, 2018
    Location:
    Westminster, CO
    I'll give it a try.

    1 - Hard clipping
    2 - Power amp
    3 - Cold clipper

    There is a lot happening in your schematic. Can you explain in layman's terms what's going on with your screen regulators?
     
    Mongo Park likes this.
  3. mrriggs

    mrriggs Tele-Meister

    Age:
    41
    Posts:
    229
    Joined:
    May 24, 2020
    Location:
    Vancouver, WA
    The pot and resistors make up a simple voltage divider to set the voltage. The regulator tubes are cathode-followers to do all the heavy lifting. You could hook the plates of the regulators directly to B+ to make it a straight forward voltage regulator to get maximum power. I chose to exploit a characteristic of sweep tubes, the low screen voltage, and hooked the plate of the regulator to the plate of the power tube. This way it will act as a voltage regulator only until the power amp is pushed hard. When the power tube plate voltage drops to the regulator voltage then the regulator can no longer regulate and the screen voltage will drop with the plate voltage. Basically, switching the pentodes to triode mode of operation.
     
    LetItGrowTone and Mexitele Blues like this.
  4. sds1

    sds1 Tele-Afflicted

    Age:
    44
    Posts:
    1,762
    Joined:
    May 4, 2017
    Location:
    Orlando, FL, USA
    I'll go: cold, power, hard
     
  5. Mexitele Blues

    Mexitele Blues Tele-Afflicted

    Age:
    43
    Posts:
    1,725
    Joined:
    Mar 7, 2018
    Location:
    Westminster, CO
    Thanks, I will have to read this a few times once i get home.
     
  6. mrriggs

    mrriggs Tele-Meister

    Age:
    41
    Posts:
    229
    Joined:
    May 24, 2020
    Location:
    Vancouver, WA
    You got it.

    I didn't adjust the tone controls between samples but probably should have. For the cold-clipper to sound decent, you really need to pull out some bass.

    That is doable with the tonestack in this amp. Setting both the Bass/Mid and Treble knobs to 12-o-clock gives a pretty flat response. Turning the Bass/Mid knob counterclockwise cuts bass, turning it clockwise cuts mids. The Treble knob will cut or add treble.
     
    LetItGrowTone and Mongo Park like this.
  7. mrriggs

    mrriggs Tele-Meister

    Age:
    41
    Posts:
    229
    Joined:
    May 24, 2020
    Location:
    Vancouver, WA
    I made a change to the screen regulator circuit this morning. Instead of hooking the lower end of the Power pot to ground, it was moved to a negative voltage supply to get the screen voltage to go down further.

    [​IMG]

    The power tube overdrive is now AMAZING when the screens are turned down. It starts clipping at 0.1 Watts. Very softly at first, and progressively hardens up as it's pushed further. At full square wave output it's still only 0.85 Watts which is a very pleasant volume for playing when the kids are home.
     
    Last edited: Jul 1, 2020
    LetItGrowTone likes this.
  8. mrriggs

    mrriggs Tele-Meister

    Age:
    41
    Posts:
    229
    Joined:
    May 24, 2020
    Location:
    Vancouver, WA
    Made a couple more tweaks to the screen regulator circuit.

    [​IMG]

    One of the 6GF7 tubes shorted out and fried the Power potentiometer. A 470k grid resistor was added to prevent that happening again. Also pulled out a couple of the resistors in the voltage divider to simplify it and get a little more range for the screen regulator.
     
    LetItGrowTone likes this.
  9. mrriggs

    mrriggs Tele-Meister

    Age:
    41
    Posts:
    229
    Joined:
    May 24, 2020
    Location:
    Vancouver, WA
    I absolutely love the sound of this amp but the bass/mid control is very touchy. All of the big changes happen between 0-1 and 9-10. The entire schematic was designed around the pots in this Bogen chassis because they have a unique long "D" shaft. The tone stack makes use of the extra-odd 2M-L w/1M tap pot. It works but, like I said, it's touchy.

    What it really needs is an S-taper pot. A 1M-S pot would plug right in and work with no other changes to the circuit but, believe it or not, I wasn't able to source a 1M S-taper pot with the correct long "D" shaft. I did buy a 2.5M pot that was advertised as an S-taper. However, when it showed up, it turned out to be a linear-taper. Womp, womp.

    Alright. What do you do when "they" don't make what you need? Make it yourself!

    I found a 100K-L pot in the pot drawer, cracked it open, then scraped the carbon off of the center of the track which turned it into a 220K-S pot. Sure, that's not the 1M I initially wanted but the circuit can be revised to make it work.

    [​IMG]

    That was the easy part. It was a standard 1/4" shaft pot so it'll need a long "D" extension to work in this Bogen chassis.

    First step, make a "D" shaft out of 1/4" rod. Only need about an inch but I did extra in case it'll be needed again in the future.

    [​IMG]

    Next, machine the tip of the pot shaft to a "D" to mate with the extension.

    [​IMG]

    Finally, clamp the two together with a split collar. Notice, a relief had to be turned in the collar to clear the flange on the back of the knob. Shown next to one of the original Bogen pots.

    [​IMG]
     
  10. Mongo Park

    Mongo Park Tele-Afflicted

    Posts:
    1,155
    Joined:
    Nov 28, 2010
    Location:
    Winnipeg
    Yikes man you are at a way higher level of DIY self containment. Beyond my understanding but enjoying the read.
    Always nice to see amp makers who bring other skills into play in their projects.
    Cheers Ron.
     
  11. NSB_Chris

    NSB_Chris Tele-Holic

    Age:
    56
    Posts:
    514
    Joined:
    Sep 2, 2017
    Location:
    New Smyrna Beach Florida
    This is really cool. Tough for us mortals with all the NOS tubes. Took me some time to figure out what was going on with your hard clipper. Makes me want to try it!

    Can you explain the rational for the rectifier. Not sure if you would call that a full wave "double bridge" or "stacked bridge" or something???
     
    tubegeek likes this.
  12. mrriggs

    mrriggs Tele-Meister

    Age:
    41
    Posts:
    229
    Joined:
    May 24, 2020
    Location:
    Vancouver, WA
    It's a full-wave doubler. There are certainly doubler schemes that look simpler on paper but despite how complicated this one looks it's actually only four parts; two bridge rectifiers and two capacitors.

    [​IMG]
     
    LetItGrowTone likes this.
  13. NSB_Chris

    NSB_Chris Tele-Holic

    Age:
    56
    Posts:
    514
    Joined:
    Sep 2, 2017
    Location:
    New Smyrna Beach Florida
    Interesting! Had to look up the Cockcroft-Walton full wave series multiplier to understand how it works. Not intuitive but it seems that the first bridge rectifies and then the AC is added back on top of the rectified DC through a coupling capacitor and then rectified again.

    I assume you did that because your PT secondaries were too low. You use the first stage for the power tube plate and screens which is where most of the current will come in. You also use this tap for the clipping diodes because you need to reduce that to a very low voltage. Then you use the second stage for the preamp tubes and the PI where the current demand is less. Am I close?

    Are you doing this to keep the voltage on the power tubes low to get earlier power tube distortion? Then using a higher starting B+ for the preamp to allow filtering without dropping the voltage too low? So in effect, being able to dial in the power tube voltages independently of the preamp stages.
     
  14. mrriggs

    mrriggs Tele-Meister

    Age:
    41
    Posts:
    229
    Joined:
    May 24, 2020
    Location:
    Vancouver, WA
    Yeah, the doubler was necessary because the transformer only had a singe low-voltage output with no center tap. It'd easily supply enough current to drive the power tubes through the doubler. The Bogen originally doubled the output to drive four 7868 power tubes!

    I used the un-doubled output for the power tubes because they are TV "sweep" tubes which want lower-voltage/higher-current than output tubes normally used in guitar amps.

    If the screen regulators were hooked directly to B+ instead of the plates of the power tubes then this thing would put out about 20 Watts of clean power, even at the un-doubled voltage. I'll probably switch it over to do that since all I have to do to lower it again is turn the knob down. With the "Power" knob turned all the way down it'll start to break up at around 100 mW.
     
    LetItGrowTone likes this.
  15. mrriggs

    mrriggs Tele-Meister

    Age:
    41
    Posts:
    229
    Joined:
    May 24, 2020
    Location:
    Vancouver, WA
    Modified another pot today. While I have the amp open to replace the Bass/Mid pot I wanted to revise the Master Volume control as well. Where it's at now it doesn't turn the volume down all the way, it's like it only goes down to "3". Picked up a 2W 250k-A pot to go in place of the plate resistor on the cold-clipper. Found one with an extra long shaft so it was much easier to modify than the other pot.

    Mill a flat in the shaft, then cut off the extra.

    [​IMG]

    [​IMG]
     
    LetItGrowTone likes this.
  16. mrriggs

    mrriggs Tele-Meister

    Age:
    41
    Posts:
    229
    Joined:
    May 24, 2020
    Location:
    Vancouver, WA
    A revised schematic showing the changes to the tone-stack, master volume and screen regulators.

    [​IMG]

    With the screen regulators configured this way it isn't necessary to use a separate triode for each screen. The two screen resistors could be tied to the cathode of a single power tube. Heck, a paralleled 12AU7 could probably be used instead of a power tube.
     
    LetItGrowTone likes this.
  17. mrriggs

    mrriggs Tele-Meister

    Age:
    41
    Posts:
    229
    Joined:
    May 24, 2020
    Location:
    Vancouver, WA
    On Saturday I was able to fire up the soldering iron and get everything switched over. The changes to the tone-stack are even better than I anticipated and the extra power output really beefed up the bottom end. Unfortunately, there is some "motorboating". It did that before if I dimed all the knobs. Never investigated it because I never played with all the knobs dimed. It seems a bit more prone now so I can't ignore it anymore.

    With the chassis on the workbench, plugged into a different cab, it was motorboating constantly. Found a 60kHz oscillation coming out of the cold-clipper, no doubt due to the new plate resistor (master volume pot) being located eight inches from the tube. Adding a 220pF snubber cap from plate to B+ eliminated the oscillation but had no effect on the motorboating.

    After hours of probing and trying different things I checked the impedance of the speaker cab I was using. It's a 2x12 cab with the speakers wired in parallel so it should be 4 Ohms, right? Nope, 16 Ohms. At some point in the past I rewired the speakers in series and totally forgot about it. Switched the output transformer to the 16 Ohm tap and, wouldn't you know it, the motorboating is gone.

    What's interesting, though, is that with the impedance mismatch it had WAY more negative feedback. Like pushing the phase-inverter to under-unity levels of negative feedback. This output transformer has an independent "70-Volt" winding that I'm using for only the negative feedback. I just assumed that this would give consistent negative feedback regardless of the speaker impedance setting.

    I'll need to experiment with this some more to get a better grasp on how it all interacts.
     
    LetItGrowTone likes this.
  18. mrriggs

    mrriggs Tele-Meister

    Age:
    41
    Posts:
    229
    Joined:
    May 24, 2020
    Location:
    Vancouver, WA
    Many more hours of probing and experimenting, I'd say it's all ironed out. The weirdness was traced to the inverting half of the phase-splitter which was oscillating. A 120kOhm grid stopper made this thing rock stable.

    I miscalculated the snubber cap on the cold-clipper plate and it was attenuating well into the audible range. Luckily, with the grid stopper on the phase-inverter, this cap wasn't necessary so I pulled it out.

    While I had it on the bench I took some shots of the tube diode "hard" clipper in action. The clipper was in the 3 Vp-p setting. Scope was hooked to the speaker output with a resistive dummy load. Power and Volume knobs on ten.

    First a clean signal.
    [​IMG]

    Turn up the Gain.
    [​IMG]

    Even more gain.
    [​IMG]

    As you can see, calling this a hard-clipper is a misnomer. The tube diodes are as effective at clipping the signal as drum brakes were at stopping 60's American automobiles; little more than a suggestion. Notice that the amplitude continues to creep up as you push the tube diodes harder.
     
    LetItGrowTone likes this.
  19. NSB_Chris

    NSB_Chris Tele-Holic

    Age:
    56
    Posts:
    514
    Joined:
    Sep 2, 2017
    Location:
    New Smyrna Beach Florida
    Very cool! I wondered how soft the tube diodes would clip. Love the drum brake analogy.
    How does it look when you adjust for asymmetric clipping? Or adjust so that you get only one lobe clipped. Brilliant feature!
     
  20. mrriggs

    mrriggs Tele-Meister

    Age:
    41
    Posts:
    229
    Joined:
    May 24, 2020
    Location:
    Vancouver, WA
    I didn't get any shots of it with the Asym control off center but it looks just like you'd imagine, rounding off one side before the other.

    The Asym control was added as a curiosity. You hear a lot about symmetrical vs asymmetrical clipping, I wanted to hear the difference for myself with the ability to change it on-the-fly. It was a real eye-opener and I know now that I prefer the "cleaner" sound of symmetrical clipping. Shifting it off-center definitely makes it "woollier".

    If I was to build this amp again, knowing what I know now, I wouldn't bother with the second coupling cap, adjustable bias, and ridiculously inefficient low-voltage supply. Instead, I'd just use a couple AA batteries for the positive and negative voltages. It wouldn't really draw any current from the batteries so they should last for many years.

    [​IMG]
     
    Bendyha likes this.
IMPORTANT: Treat everyone here with respect, no matter how difficult!
No sex, drug, political, religion or hate discussion permitted here.