Harmonic Percolator (George Giblet's schem) build report

[JOHNO]

heres a vid of my perc it doesnt sound any thing like dougs and it sounds really misbiased through this amp,i'll have to fiddle with it some more. The circuit is still on the breadboard and all controls cranked right up. The amp is a homebrew thing I modded from an old 100w valve PA and is crystal clean without the effect plugged into it. Recorded into a rode mic straight into imovie. http://www.youtube.com/watch?v=NxBn1PDoGbI Please dont comment on my guitar playing.

[John Lyons]

Comments on doug's video clip and build.

Sounds nice and refined, sustains well and has few artifacts or nasties.
The camera is compressing the poop out of the dynamic range though!
Hear the back ground noise come up between loud sections?
Also notice how the level doesn't change when the volume is turned up.
Part of this is the amp compressing/distorting more and the camera's
limiter sucking that up

Another thing to keep in mind is that the amp sounds great by itself and has
a slight break up with the Brick in bypass. So some of that sustain and dirt is
coming from the amp for sure. Not a bad thing at all. Just pointing it out.

The build sounds very nice for sure.

Geez Doug, why don't you use a shittier amp! 

john

[alex frias]

I would like to make John's words as mine, about Doug's vid.

I also noticed a very compressed overall sound, an artifact very common when using the audio from camera's mic.

Well, the guitar/amp sound is very cool indeed, so we hope for the best when the Brick is activated. And it happens!

The thing, when cranked up, shows much less wheeziness than my version. A high point. Even that I like the wheezzo in my sound, I think we all here are trying to catch the specific HP-1 magic. So I think Doug is closer than me in this aspect. I didn't notice the sub octave effect, but the acoustic feedback was dumbs up. Infinite sustain!!!

I bought the last 10 2N404A I've found as NOS in an old electronics store, downtown Rio de Janeiro. So I will test the Doug's theory about the Q1's hfe. I found some BSX 20 Si NPN with very low hfe, so they will be the second try, as Q2. Soon I will receive more 2N404A and 2N3565 by mail and new tests will be done. Now you all know I'm into that obsessive compulasive search. 

Doug, did you keep the Q1's collector resistor as 22K?

One mor thing: I thought the wheeziness was an artifact linked to the more gained the circuit was working and it could be controlable by the Harmonic's or guitar pots. But I think it's possible to have a more cranked fuzz sound with less zingness by using a good balanced trannies.

[DougH]

Thanks, John. The camera audio is far from perfect, but given that it's really fairly accurate to what I hear in my ears at head level. The biggest issue is the dynamic range due to limiting etc and I can live with that. The room reverb was a little exaggerated but it's easier to pick it up with the camera than it is to set up two mics. In fact, I'll probably be sticking with the camera from now on since it's so darn easy to use. It's much easier than firing up the computer and messing with the mic and mixer and etc.

I don't agree that there's much "amp sound" in this clip at all. I can try it with a couple other amps to see, but the amp is dialed up very clean with single coil pups, and doesn't sound compressed or dirty IMO when the box is bypassed.

Thanks for the comments.

[DougH]

Doug, did you keep the Q1's collector resistor as 22K?

Yes.

Two ways to control "wheez" IMO- raise Q1 Rc from 22k to 100k (91k), or use higher hfe xsistor for Q1.

Thanks for your comments on the clip.

[John Lyons]

Doug
Maybe what I should have said is that the amp seems to be easily boosted into overdrive.
(It sounds great by the way) The same way boosters drive the amp into overdrive.
Played through a fender type amp I'm sure the Brick sounds very different and the
added level would show up as a level boost not compression/overdrive. (Naturally)

Is there a limiter or sensitivity selection for the on camera mic?
The quality of the room and amp are ok. Just the slow deep compression gets distracting.

Sorry for getting off topic.

john

[DougH]

I see what you mean by the 'breathing' of the camera mic, John. I don't think there are any settings to control that but I will check. I might record it again with the computer and sync it to some video and see how that turns out.

[alex frias]

Well, we are talking top level video/audio productions here!!!

Jokes off, I'm here just to endorse Doug's theory about the best Q1's hfe. I've tested several 2N404A's and one "unknown" Ge PNP.
The 2N404A's with hfe around 100 sounded best. It deals with the gain and the quality of distortion and fuzziness (and fizziness in certain cases). 

At the end of the tests the subtleties were scaping from my ears. But I used 2N404A hfe=105 as Q1 and BSX 20 hfe=108 as Q2. The sound and playability was very good, but, even with a deteriorated perception, I think 2N3565 sounds smoother... When I obtain more conclusive answers I will do new soundclips to compare and try to help a little.

[DougH]

The 2n3565 sounded smoother than other xsistors I tried for Q2 too. It's subtle but I can hear it.

[km-r]

damn, i missed, like, three pages of this read.

will somebody put a summary after this epic discussion?

[DougH]

Go ahead and have at it.

[alex frias]

Just to assure Doug's comments about Q1 2N404A hfe~100 I put a utub vid.
Using the same hand drawn scheme I posted titled "Sardonic Percloretor", now renamed "Expresso Tone Machine", with a 2N404A hfe=110 as Q1.
In the first part I used a BSX20 (metal can, for super-mojo!!!) with a hfe=105 for the Q2, then at the second part a 2N3565 with a hfe=203 was used.
The setup used to record thie audio was exact the same as the one used in my old soundclips, just for an easyer comparison.

http://www.youtube.com/watch?v=PWlNgtUCM4M

The only thing I didn't obtain from this circuit (in the soundclips) was that wonderfull acoustic feedback, for an obvious reason: I did all the recording through headphones.
But I tested it using a small Marshall combo and... Yes!!! It feedbacked as it was alive in a very musical way!

For me, this new seting of trannies sounds way better than the others I've been trying, so I really agree with Doug H. when he claims the magic is something about the "current thing", as I saw no significant differences in voltage measures. This new combo certainly is closer to what "I suppose" a good HP-1 sounds.

[DougH]

Sounds nice, Alex! Thanks for making the video.

[alex frias]

After all experimentation my conclusion at this time is the Doug's New Brick design gaves me the most interesting sound and performance close to how I "imagine" the original HP-1 is supposed to behave. And I must to thank Doug Hammond very much because it worthes any effort to get the right components and setings. The smoothest combination was achieved using the New Brick design with a 2N404A hfe=150 and a 2N3565 hfe=205. Several combinations achieved good results too, but always using devices around those hfe's. It's the real zen drive, because balance is the key word here!

Two observations:

1 - It's another beast when played with some volume from a good amplifier, it doesn't need to be that loud. It really comes to life and shows what differs it from the other distortion boxes. Even with medium to high Harmonics (Fuzz on the Brick) control position you can get very good sustain and the fantastically musical feedback.

2 - Using the Hamonics Pot or the Guitars Volume Pot you can control not only the distortion intensity, you can also TUNE the pedal behaviour! It's a beast actually!

Two mod's I've tested with good results: as said before, the Gagan's Easyface input tone control works great with the HP as a pickup matcher or to change some timbre character; using more diodes in series with the cliping one's can give a more "normal" drive sound and a considerable level boost.

Well, as you all can perceive, I'm very impressed!

Doug, thanx again and congratulations for your effort and sharing, your kindness and your very good performances on the demos.
Thanks to all people helping and sharing here too!

[Ry]

I would like to thank Doug as well.  I've been following this thread and finally built one up.  I happened to have a few 2N404's (not 2N404As) and Fairchild 2N3565's.  I found one of the 404s with a gain of 98 and an 3565 with a gain of 250.  It sounds amazing...everything I've ever wanted out of a fuzz circuit (other then the crazy skyripper stuff), the guitar tone and definition is retained, but with grit and low end grunt.

I tried a lot of different transistor combinations and different gains of the 'magic' combination of 404/3565 and found that they resulted in reduced gain or a very thin sounding circuit.

Thanks again!

[DougH]

Maybe what I should have said is that the amp seems to be easily boosted into overdrive.
(It sounds great by the way) The same way boosters drive the amp into overdrive.
Played through a fender type amp I'm sure the Brick sounds very different and the
added level would show up as a level boost not compression/overdrive. (Naturally)

Hey John, I tried it with both my Octal Fatness dialed up clean and my clean SS amp (which has excessive headroom btw) and it sounded pretty much the same. Of course the amps all sound slightly different and I couldn't drive the SS amp as hard with the Brick volume control- but overall it sounded very similar.

Just wanted to follow up on that and report since I had tried it.

[DougH]

Ry and Alex,

Thanks for the comments. But you should really thank George, Gus and the many others that contributed to this and the older HP threads. George got the voicing "right" AFAIC and he and Gus both got us all thinking about the DC operating point and what it takes to make it sound "good".

The Brick is not a "design"- it's just the way I happened to build the Harmonic Percolator from all the various schematics that have been posted. I called it the "Brick" because of the way the finish on the box turned out and I wanted to avoid any issues using the name "Harmonic Percolator" which is now trademarked. Only difference is I removed what I considered "band-aids"- input diode and filtering, etc and changed the output pot to 100k. I even dumped my 1n46 diodes and went back to  the 'standard' 1n34a's that everyone else is using.

I don't want people to think the Brick is my design, because it's not. It's just another slight variation on the HP. I feel my only contribution to this has been finding a good target HFE for the 2n404a.

I'm glad you guys like the clip and were able to get your circuits sounding good. This has been a cool thread.

[John Lyons]

"Hey John, I tried it with both my Octal Fatness dialed up clean and my clean SS amp (which has excessive headroom btw) and it sounded pretty much the same. Of course the amps all sound slightly different and I couldn't drive the SS amp as hard with the Brick volume control- but overall it sounded very similar.

Just wanted to follow up on that and report since I had tried it."


Ok, cool , thanks for the follow up. Thats good that it's not as sensitive to amp inputs.
Now I guess I ought to give this one a go now
Thanks for the field testing here!

john

[alex frias]

I would like to express my gratitude to everyone who gave the clues we have to work on!

But... it seems I can't stop yet! Yesterday I was playing with the "Brickolator" and very happy with it. Very cool sounds and lots of sustained mutant timbres, don't forget the very impressive and musical acoustic feedback.  Well ... I was still feeling it was lacking that stronger sub-harmonic-octave-down effect, so beautifull exaggerated on Albinis vid.

Then I gave a break and decided to reorganise the information stuff I have regarding the Harmonic Percolator in my computer HD.

I saw a there set of 3 pics that I completely forgot: they show a configuration very close to our George's PNP-NPN version.

I quickly made a PCB and tried the values I saw on the pics. What intrigued me was that Q2 10K collector-to-ground resistor, i didn't use values above 91K with success until DougH tell me that it has to do more with the trannies hfe matching. I built the circuit exactly as the following image (sorry for reusing the schem, if it hurts anybody, just tell me!):



If you get the same trannies with close hfe's you will get it! It has THE sound and it adds a very discernible sub-octave effect to. As I did few tests with different hfe trannies,  the best I achieve was with that combo shown at the picture. The sub octave is not completely stable, it appears at the note attack, then oscilates while it sustains, in some regions of the scale it's more evident than in others. it's not so evident as in a proper octaver, but much more elegant. I will try to find out a more accurate combination...

[bool]

Hi Alex - what other trannies have you checked?
(of Ge's, I currently only have 2N219 PNP ...and AC141, AC 187, AF239, and a 2SA235 which may be faulty)
Any of these you tested? 2N404 seems hard-to-find in EU. Maybe I havent' looked at the right places. (I suppose that circuit would work "flipped", with a npn ge as well)

I saw you tested a 2n2222 for the Si NPN. Could be that 2N2219 also works?