Gristleizer PCB group buy? Who's interested?

[Taylor]

It's a quad and a dual opamp. So TL072 or 4558 should be fine for the dual. TL074 or LM324 for the quad. MAX1044 for the bipolar power supply.

There's no vactrol in this project, it uses a JFET to control amplitude/filtering. The JFET is a 2n3819.

The spec'ed diodes are 1n34A, which are available at Mouser, but Roy said that any diode should do, so if you have hundreds of 1n4148/1n914 as I do, they should work and are cheaper.

The parts list will be pretty much the same as this, barring the above notes:

http://gaussmarkov.net/layouts/gristle/gristle-parts.txt

[Derringer]

awesome dude

thx

does the jfet need to be a 2n3819 or will others work ?

[Taylor]

According to Gaussmarkov, any JFET should work:

http://gaussmarkov.net/wordpress/circuits/gristleizer/

But, this guy says the opposite:

http://www.thelongafternoon.com/?p=437

So I don't know. I'd spring for the right one, since it's cheap and available.

[Rectangular]

I don't know who "that guy" is, but I read his blog post on the gristleizer. is he a hobbyist or is he one of these guys thats manufacturing/selling them ?  basically I'm just trying to figure out where his level of understanding is. I seriously doubt that the 2N3819 is the only one that works, I'm guessing his either got the pinout wrong (if he's a beginner) or that he didn't rig up an FET tester and find one that matches the 2n3819.  in either case, I don't see popular electronics releasing a hobby effects pedal schematic if it could only use one transistor on the planet. they were smart guys, they understood that substitution is an essential part of DIY.

[Derringer]

According to Gaussmarkov, any JFET should work:

http://gaussmarkov.net/wordpress/circuits/gristleizer/

But, this guy says the opposite:

http://www.thelongafternoon.com/?p=437

So I don't know. I'd spring for the right one, since it's cheap and available.

agreed

[Taylor]

I don't know who "that guy" is, but I read his blog post on the gristleizer. is he a hobbyist or is he one of these guys thats manufacturing/selling them ?  basically I'm just trying to figure out where his level of understanding is. I seriously doubt that the 2N3819 is the only one that works, I'm guessing his either got the pinout wrong (if he's a beginner) or that he didn't rig up an FET tester and find one that matches the 2n3819.  in either case, I don't see popular electronics releasing a hobby effects pedal schematic if it could only use one transistor on the planet. they were smart guys, they understood that substitution is an essential part of DIY.

I agree with you, and I don't know who that guy is, somebody just pointed that out to me at another forum. But I don't really think it matters except for people who only buy electronics at Radio Shack (in fact you can be sure that that guy is a noob if he buys all his parts at Radio Shack).

[Rectangular]

(in fact you can be sure that that guy is a noob if he buys all his parts at Radio Shack).

true enough !

[Taylor]

OK, so the PCB is nearly done. It will fit comfortably in a 1590BB. I still need to implement some of the little additions people have mentioned through this thread. I recently bought RG's book on PCB layout, and I have tried to implement as many of his design rules as I could.

I'm a little unsure of how to implement the bipolar power supply. According to RG's page, using the stock charge pump schematic from the datasheet will lead to switching noise if using stereo jack switching for the battery. He suggests an alternate way to to do this, but I have to admit that I'm confused by it. The audio input signal goes on the sleeve of the input jack?   That doesn't make sense to me, and I'm not sure how to separate audio and power ground.

Josh, it looks to me like you used the stereo switching schematic for your power supply (I can see the 2 resistors and one transistor on your daughterboard) but it doesn't look like you're using stereo jacks.

Anyone have some thoughts on this?

[jacobyjd]

Josh, it looks to me like you used the stereo switching schematic for your power supply (I can see the 2 resistors and one transistor on your daughterboard) but it doesn't look like you're using stereo jacks.

Anyone have some thoughts on this?

Personally, I don't use batteries in my DIY effects. Those extra parts you're seeing are actually the components needed for the flashing rate LED.

RG's charge pump stereo switching schematic works well (it uses a transistor to switch the ground). I'd say you should incorporate it, but make sure it's laid out in such a way that it could be ignored if a battery is not used. The power supply gets MUCH simpler without the jack-switching circuitry (basically the IC and 2 caps).

[Derringer]

+1

I don't use batteries either

[Taylor]

Yeah, I don't use batteries either, but I figured a lot of people would want to have that ability. I'm trying to facilitate the needs of as many people as I can here - kind of a juggling act.

[Taylor]

Ok, so on to the flashing LED. Here's the idea:

No, I didn't see that until now. I hadn't read the comments.

"From IC1 pin 6 I have another
transistor rigged as a follower with a LED plus load in its collector to +9V and its emitter grounded. 6k8 between its base and IC1 pin 6. This gives a flashing indication of the rate. Crude but maybe useful."

I'll implement that.

I drew a schem of what I think this means:



So this should be an NPN transistor, right? Like 2n3904? Do I have the idea right?

[jacobyjd]

Yupper. I used a 2n5088, but any normal NPN should be ok. I placed mine close to the power supply for my final setup. However, I had no ticking problems whatsoever when testing it with long leads on my breadboard.

[Skruffyhound]

Yeah, the same device is used on Ricks "Vibracaster" for a rate LED, he's got a 47k on the ground leg as well, but maybe thats just peculiar to that circuit.

Nice picture

[frequencycentral]

Yeah, the same device is used on Ricks "Vibracaster" for a rate LED, he's got a 47k on the ground leg as well, but maybe thats just peculiar to that circuit.

Nice picture

It never totally cured the ticking though, so I ended up not using it.  


EDIT: ......but then again, that was before I used a 3.3uF cap to reduce the ticking in the circuit, so maybe it should work in the Vibracaster now - hmmm need to try it now! BTW, I 'stole' that little device from a Roland 100m Modular synthesiser schematic, so it's certainly not a unique approach.

[Roy]

Sorry for silence - I've been away. I half thought I might get thread replies in email, but no.

Opamps - 741's are old hat. There are much better modern opamps, as people have said.  I'd use duals. TL082 would do in a +/-9 environment (and that's pretty old now too). LM6132 I'm fond of these days. Rail-to-rail, fast, low power. Not as cheap as some, but if you're just making a few of something it doesn't matter much what it costs. Time is money too.

Given new opamps I'd get the input impedance up to 1M or more. Also there are places where the 100nF coupling caps form an HPF that rolls off too high. They need rooting out. Also in the LFO the integrator capacitor could be reduced by an order of magnitude, which reduces ticking currents. Just increase the resistor driving the integrator by the same factor.

Any old FET WILL do. I use J112 or J113 these days. But remember that the Vgs(off) is all over the shop with JFETs, even within the same batch. Use the bias control to get round this variation. J112, for example, is somewhere between 1 and 5 volts. Such precision and repeatability!

A LED to mirror the LFO is good, like Chris Carter added. A LED to show effect in/out is also good. Millenium Bypass is the answer to that. A cracking design!

I started the 2009 reissue in Feb this year thinking I'd just revamp it for modern stuff. Op-amps on a single 9V rail would be much cleaner than two batteries or switching negative converters. But, as with many things I start, it grew into more than I'd expected. It is now sufficiently different to justify having both. I redid the filter with three bandpass sections because I've been playing around with voice formants. It sounds really nice on rock organ/gritty Hammond tones. The deal with the LFO is that I think the original is a bit of a bugger to play along with. You have to let it take the lead on timing. You follow it, it doesn't follow you. What I wanted was an LFO with sync from an envelope follower. And that led me into AVR because I've got most of the software components written already for another project. All the waveshaping stuff and that cumbersome rotary switch get subsumed into software.

You're right - it does need to be future-proofed. The object code as a free download on the web in several places at the least. I'm not going to put the source code out there - that's my job and I need to earn money out of it.

I'm not ready to publish on the 2009 reissue anyway, so don't hold your breath. Full steam ahead on the PCB's I would say - and could I have one or two please? I'll build it up and pot comments.

I'm STILL waiting for a Gristleizer from Endangered Audio. Ho Hum.


Roy

[Taylor]

Thanks for the comments Roy. I'm not really sure what to do to change the input impedance, is this just a matter of changing the resistor to ground at the input?

I will of course send you some PCBs when they're complete.

[Roy]

Yes, it's just component value changes, so no need to change the PCB. Same with the choice of op-amps.

Roy

[Derringer]

Jacobyjd ,

did you use a DP4W switch as per gaussmarkov's design?

where did you get it from?

I've never worked with a rotary switch before.

thx

[jacobyjd]

Jacobyjd ,

did you use a DP4W switch as per gaussmarkov's design?

where did you get it from?

I've never worked with a rotary switch before.

thx

I actually used a 2p6t switch, so I have 2 unused positions. however, you can get a 3p4t switches pretty readily--you'll just leave 1 pole unused.