The Thomas Henry Mega Percussive Synthesizer

[Scott Stites]

[Editors Note: Circuit boards now available at the electro-music.com store: http://electro-music.com/catalog/product_info.php/products_id/104 --mosc 16/Apr/2009]

I guess I better extricate this from the Clangora thread before it derails that fine conversation. I hope I don't sound like I'm hyping this thing unnecessarily, but I truly am buzzed by this design. This one has been as hard to keep quiet about as the Mankato Filter. In fact, I often think about this as the Mankato of drum voices, it's that good. I'll start at the top:

This summer (summer 2007), Thomas quietly developed a drum voice that just sat me back in my seat. I'm not sure I've ever seen him work so hard on a single design - I'm sure he has, but in my experience breadboard testing things, I'm certain this one took more work than I've experienced with any other project. He literally designed this thing from the ground up, and we very thoroughly rung it out.

In the Clangora thread, Thomas mentioned how versatile it was (and it is extremely versatile). On top of that, the sound of this drum voice for me is simply stunning. A good deal of that is due to an innovation Thomas threw in there; it's something that I've never seen on any other drum voice. It was an idea that he picked up from an interview with Roger Powell years ago (it was Roger Powell, wasn't it Thomas? I'm kinda fuzzy here). It has to do with the impact circuit - I swear, it literally sounds like someone is striking this think with a real stick/mallet/hand/sledgehammer (depending on how it's tuned).

Just to give a rundown of the elements: the voice has three oscillators, a noise source, a balanced modulator (that can be unbalanced as well), three envelope generators, two VCAs, a noise source, a LP/BP switchable resonant VCF, and a versatile mixing section with send/receive loops. This one is a blast to tweak as it plays, BTW. It's the only drum voice I've played that can dissolve from a cowbell to a landing alien craft in a very non-seventies, non-disco-era Simmons way (though it can do that, too, if that's your thing).

Fortunately, Thomas designed it, so it is a very elegantly designed circuit (translation, it will fit on a single PCB). There are a lot of controls, so it won't be a small panel.

Right now, the target for the project is as the next electro-music PCB series, in the same vein as the Klee project. We're working on drafting a certain man from Nambucca Heads to crank out a PCB, and we're going to Klee team it to make sure what you get will be the best quality PCB we can offer. The documentation part already is very well done - can't beat those Thomas Henry schematics! I'm hoping to avoid the whole reservation process which is really a pain, but this is all in the prelim stage so far.

Expect samples.

Cheerio,
Scott

[Alienation]

Hey, hey, hey something else to play.

Sounds great, you guys are getting better and better.

[Scott Stites]

Here's a list of controls - I'm not sure if these are all the correct labels yet, but they at least describe the function. I post this now, for all of those that like to design and envision panels:

Inputs:
Trigger Input
Shell CV
Shell/Clank Return
Impact Return
Snares Return

Outputs:
Synth Out
Line Out
Shell Send
Impact Send
Snares Send

Potentiometers:
Trigger Input Sensitivity
Shell CV Range
Shell Sweep
Shell Pitch
VCO Pitch (for Ring modulator)
VCO Depth (for Ring modulator)
Ring Modulator Balance
Shell/Clank EG Decay
Impact EG Decay
Impact Sweep
Impact Pitch
Noise Filter EG Decay
Noise Filter EG Sweep
Noise Filter Center Frequency
Noise Filter Resonance
Shell/Clank Mix
Impact Mix
Snares Mix
Line Volume

Switches:
Filter LP/BP Select

Indicators:
Trigger Status LED

[Scott Stites]

Here's the economy of design part: six ICs.

3 LM13700
1 CD4046
1 XR2206
1 TL074

Yeah

I'll let Thomas describe the impact circuit, if/when he wants to.

Cheers,
Scott

[Photon]

This sounds *really* cool. I will definitely be saving a spot next to my clangora for this one.
Is this circuit meant to be an updated and improved source of clangerous sounds (ie. a clangora for th' 21st century)?
It looks like it may also do drum sounds also (?). Please do tell...

Thanks again Thomas and Scott. You guys are the best!

Peter

[Scott Stites]

Bear in mind I've never operated a Clangora, but knowing what a Clangora is, I'd say this is certainly a different beast. This design can produces bass, snare, cowbell, toms, wonderful solid metallic sounds, and so forth. I'd say it would be a good partner to the Clangora and the UD-1 as well. I intend to have at least a couple of these. With that impact circuit, you could almost call this one "deep drum". Once you start tweaking it for other sounds, the synthetic drums it'll create are much more varied than the pyeeeooow species of electronic drum. In certain ways, it's almost like an exotic noise box disguised as a drum voice, which becomes apparent once the EG times are stretched out, and you start twisting knobs.

Cheerio,
Scott

[bdu]

Wow, that's getting an awful lot out of very little! non-ic parts count?

[krisp14u]

[elektro80]

Oh yeah, there will be no more dreams of an SDS-3 for me. This seems like the kind of wall padding I would love to have in my studio. Bring it on!

[Scott Stites]

I'd love to have this for wall paper


@bdu:

96 resistors (that's counting all the panel mount pots, there aren't any trimpots)
34 caps
14 transistors (7 NPN, 7 PNP)
7 diodes

Cheerios,
Scott

[tomcat]

My XR2206's are jumping in their drawer

[elektro80]

Oh shit, seems like I would want several channels of TH MPS excellence. Will 8 do?

[Funky40]

[elektro80]

[Thomas Henry]

Well, I'm delighted to learn this may be a popular project. As Scott mentioned, it was one of the most difficult I've ever undertaken. I put in a full two weeks of twelve hour days each (no kidding) on the pencil and paper design alone. I had set myself a goal of making it as versatile as possible, but simultaneously trying to keep it as elegant as possible. Those two criteria weren't very compatible in this case, but after some prodding from Scott, I finally overcame the difficulties. And then came the breadboarding! After just a bit of scrambling to suck in some slack, I was able to conclude with a design which accounts for any chip-to-chip variation. I'm glad I didn't give up (which almost happened!

I remember taking a walk in the woods every single day during this design, just to ponder whatever the current difficulty was that needed to be overcome. When it was all over, I said to myself that I had worked so hard on it, that I simply had to sell it. But now that the sweat is behind me, I decided to release it to the EM crowd. Mosc has provided such a fine service here---in which I'm having a great time---that I wanted to pay him back with a bit of fun thing.

So, this is for EM patrons. Please don't copy what'll be coming in the ensuing weeks, but instead link back to EM directly so others will find the horse's mouth.

And a big thanks to Scott for coordinating this effort. It'll be a lot of work for him, but maybe it'll help him take his mind off of Sam Brownback (his hero) withdrawing from the presidential race.

To help keep the momentum of interest going, I've attached the block diagram for you to ponder.

And, oh, it doesn't really replace the Clangora although there is a large intersection of sounds. Instead of the open/closed logic, you'll find an impact generator that'll blow you away. And yes, I first learned of this technique in a Roger Powell article back in the 80's. If you listen to my "Morpheus" composition, you'll hear it there, but that was a $2000 Moog modular doing the work. The MPS does the whole shebang for $50, and more!

Thanks again for the interest; that's what keeps me going...

Thomas Henry

---

block.pdf
Description:	MPS Block Diagram in line art/pdf format.	Download (listen)
Filename:	block.pdf
Filesize:	24.67 KB
Downloaded:	3089 Time(s)

---

[Dan Lavin]

Thomas,

This is just awesome! I really felt guilty dismantling my Blacet Noise Voice and my TH Snare+ to make my 2 SN Voices, but now I can replace them with something better! Yea! Thanks so much for this!

Of course, now we all curse you as this is another project for the queue!

[loss1234]

cant wait!!!
guess i need to put in an order for another xr2206 as i have already dedicated the other one to the xrvco i am starting this week.

thanks

[The Alison Project]

Damn, I will catch up on all these projects one day I swear. Looking forward to retirement hahaha, 30-35 years to go.

[bugbrand]

Very interesting indeed!

The implementation of 3 envelopes seems like it'll be supremely great.

And, with the impact - am I maybe right to think i could try such ideas (short thuds with a sqr wave setting) on the UD1. Mmmm.

I'm wondering whether it'd be nice to be able to repatch a little within the module perhaps? Any plans to be able to re-route other sources into the VCF, for example?

This is one of the times I slightly wish I was using jack sockets rather than bananas!

[Scott Stites]

Well, well, well. The breadboard the MPS occupied has been sitting on a lonely corner of a shelf in the la-bohr-atory. I thought I'd fire it up last night (I haven't touched it since mid-July, being occupied with the Klee project). I thought it'd be nice to supply a sample, but I ran into a slight problem - the breadboard bristles with so many pots (non of them labeled, of course), I couldn't remember which did what! So, I just went with it.

After sorting out where the trigger input should go, I attached that to a trigger out on the Klee, and it came to life. I also plugged one of the Klee voltage outputs into the shell control input. Then, I just started tweaking pots to see what did what. I got a late start, and it was 'round midnight, so I made a quick, spontaneous sample of it doing....something. I debated whether to post it, but I suppose it sounds like you'd expect on bring up after the build - just discovering (in my case, rediscovering) the control functions. So, the sample doesn't have any of the finely tuned drum voices that one can get out of it, it's just me tweaking the pots around.

It starts out with a very long, un-drum-like shell envelope - the initial sound heard is the shell oscillator tuning being tweaked while the MPS is being triggered by the Klee. Then I tweak the ring oscillator. Then I bring up the impact tone (wherever it was set at the time) and noise. Around this time I bring up a couple of Klee voices. After that I just sort of tweak this and that. I only had about 3 minutes recording time left on the D8, so it cut out just as things were getting interesting - on the fadeout, you can hear some of the groovy cowbellish metallic tones the MPS can generate.

Cheerio,
Scott

---

mega_perc_test.mp3
Description:	Resurrected MPS breadboard waking up.....	Download (listen)
Filename:	mega_perc_test.mp3
Filesize:	2.63 MB
Downloaded:	3651 Time(s)

---

[mono-poly]

[Alienation]

Great sample!! This sounds like a must have.

[Scott Stites]

[kokoon]

this sample is just fantastic!!! could you tell us what exactly was used besides the (one) MPS and (one) klee?

[Scott Stites]

The sample just uses the same voices I've used since I proto-typed the Klee. My bench is at full saturation - when Thomas visited the la-bohr-atory last August, it was in a very unnatural state (comparatively clean and organized). This was due to the wife, mostly. Anyway, once I proto-typed the Klee, I kept the breadboarded Klee there, attached to a number of cables - I was in such a hurry to proto the Klee, I just disconnected those cables and ran cables over the top of that mess. Lot's of them. When I hooked the MPS in the other night, I just plopped that breadboard on top of a cushion of coaxial cables and wire and plugged that in. Yeah, I need to tear down...

But anyway, that explains the voices I've been using now for a month or so - I'm kinda hemmed in.

The low voice is my dual Schmitz VCF - one half of it is resonating, passed through the CGS wave multiplier, then through the other half of the filter, through a VCA into the mixer module.

The high voice is a Schmitz VCO-1 through the Schmitz 2040 transistor filter through a VCA into the mixer module.

Each voice is controlled by an EG that is triggered by the Klee.

The MPS is connected to the mixer module, and the output of the mixer module is going through my Lexicon delay straight into the recorder.

Actually, the two synth voices aren't doing a whole lot - the MPS was making all the drum-like voices and weird tones throughout the sample. In fact, the high voice doesn't even appear until the 1:00 mark, and the low voice was mixed in at 1:10.

Thomas put in an impressive array of controls so that a particular voice can be tweaked in just right - I'd like to concentrate on demonstrating that aspect of it (this sample was just a "hey, it's alive" moment). At the same time, those elements that are used to create the drum voice are merely a knob tweak away - you can reach out, tweak a control and add a little surprise now and then that you wouldn't expect out of a drum voice, other than "pyeeeewwwww" (which of course it will do as well, should Donna Summer show up for a jam session). Anyway, I'd love a panel that had four of these in it.

This weekend will be dedicated to stuffing and testing the production Klee boards, then after that I'll tear down and concentrate on making some more comprehensive samples of the MPS.

Cheers,
Scott