Gonkulator / 1496 ring modulator

[moosapotamus]

Rather than bump up the previous thread on the Gonkulator (which didn't even have "Gonkulator" in the subject - I almost didn't bother to read it), I thought I'd try to keep the fire burning on this one with some much better, much higher resolution pics...

Component side

Solder side[/b]

Yes, it does look like a lot of work to trace. But, to make it even tougher, I don't think it's a single-sided PCB. I think there is another layer of traces that is hidden. For example, look at the pic of the solder side and you'll see three pads in the lower left corner that have no traces connected to them.

Flip to the pic of the component side and you'll see the components in question in the upper left corner... two glass bodied diodes and a small dark yellow capacitor, next to the three pads for upper most pot, the "suck" pot.

I can't see any associated traces with these pads. Even if I hold the board up to a light bulb, I can't seem to see any "shadow" traces. But, using the continuity tester in my DMM, I've confirmed that those three pads are all connected to each other as well as to lugs 1 and 2 of the "suck" pot.

If you look around on the pic of the solder side, you'll find a few more of these seemingly unconnected pads. I've tried hunting around with the continuity tester to try to find connections for some of them, but have been unsuccessful, so far.

I don't have time to dive head first into this one at the moment. Making a clone would be kind of anti-climactic for me since I already have the pedal. And, I've got a bunch of other stuff already in the works that I'd like to finish up.

But, I know there are some folks out there who are interested in this one. So, I hope the pics are helpful. I'd be glad to try to answer any questions about it that I can, too.

~ Charlie

[toneman]

Wellllllllllll,
that's a ground plane on top. 4sure.
usually pcb have even #of layers.
i really doubt Gonkpeople forked out the $$ 4 a 4layer pcb.
Follow all the pos and neg voltaged first.
U know where thry go on the opamps, 1496, etc.
Do the known stuff first, then fill in the blanks.
What R the values and labels 4 the pots?
What does the LED do?
What's the whete connector go to?
inquiring minds etc etc....
tone

[Paul Perry (Frostwave)]

Just in passing, in another thread on PCB layout recently, I mentioned "target" crosshair & circle PCB patterns around connections to groundplanes. If you look at some of the caps connecting to the groundplane (component side) you can see this.
I'm voting for double sided too, but anything is possible! Does the Gromnk have much bleedthrough of the local oscillator?

[moosapotamus]

You guys are good! Yeah, most of those pads that look unconnected from the solder side are connected to that ground plane on top. But, some of the pins on the ICs are simply NC. So yeah, definately double-sided.

My Gonk is very quiet, very little osc bleed.

Controls:
Suck = distortion output level
Smear = ring mod output level
Gunk = distortion gain
Heave = overall output level



~ Charlie

[Dan N]

Thanks Charlie and Paul!

Well, here's what I can see of the circuit board layout:

http://users.rio.com/senorris/junk/Gonkstart.gif

There are some connections on the top of the board I can't see. Plus, I'm lost up by the pots and down by the switch.

Maybe tonight I can start playing with a schematic unless someone else wants to go at it.

[ExpAnonColin]

Well, here's what I can see of the circuit board layout:

http://users.rio.com/senorris/junk/Gonkstart.gif

Man, your layouts are gorgeous, what do you use?

-Colin

[Dan N]

Thanks Colin. I use Photoshop. On my computer I can paste the traces down onto the parts and when I hit a pad or trace with the magic wand it outlines the entire trace. I love Photoshop!

Dan

[niftydog]

"target" crosshair & circle PCB patterns around connections to groundplanes.

"thermal relief" connections to allow some movement of the pad without disturbing the PCB plane.

I used them on my bassballs build anywhere I needed to connect to the ground plane.

I'm betting you'll find that all of the unconnected pads have continuity between them.  Ground plane for sure!

[ExpAnonColin]

Well, here's what I can see of the circuit board layout:

For that other mystery IC, here's a pinout:

[moosapotamus]

I'm betting you'll find that all of the unconnected pads have continuity between them.  Ground plane for sure!

Yeah, definately a ground plane. But, not all of the 'unconnected' pads have continuity (within the ground plane). Some of the IC's have pins that are really not connected to anything.

There are some connections on the top of the board I can't see. Plus, I'm lost up by the pots and down by the switch.

I'll try to fill in some of the blanks when I get a chance.

~ Charlie

[moosapotamus]

There are some connections on the top of the board I can't see. Plus, I'm lost up by the pots and down by the switch.

I think I've got most of the blanks filled in...

Gonk Rough Layout

Let me know if anything looks questionable. Turns out that there are some surface traces within the ground plane that I have shown as traces crossing over each other. And, there are a couple of caps that look like the the other 47n ones (473), but have no markings on them (that I can see, anyway). All the diodes look like 1n914.

So... who's working on a scheme? :mrgreen:

~ Charlie

[RobB]

Wow, you’re a man on a mission!  

Spotted something.  
In the epicentre of  C12, Q5 and Q6 there seems to also be a link between 4k7 and 100k.

You could de solder one side of the mystery caps by applying heat on the track side and gently prying that side of the component up with a small jeweller’s screw driver.  To get it back in you can clean the excess solder off the component with a clean solder tip.  The hole can be cleaned by applying heat on one side then using a solder sucker on the other side.  That’s if you want to be sure about the values.  

Also the 47n cap behind Q5 looks kind of large for that value, but then it could be a different type.  

Good luck.

[Dan N]

Boy Charlie, your additions were a great help! Here is signal into and out of the 1496 (as far as I can see):
http://users.rio.com/senorris/junk/gonkschstart.gif

I'll keep playing with it tomorrow.

Dan

[moosapotamus]

Lookin' good, Dan!
I'll take a look at Rob's comments next chance, too.

~ Charlie

[Marcos - Munky]

Looks cool. What software do you use to draw the schematic?

[moosapotamus]

In the epicentre of C12, Q5 and Q6 there seems to also be a link between 4k7 and 100k.

Good eye, Rob. I caught quite a few more surface traces myself, and added them to rough layout...
Gonk Rough Layout

Here's a cool see-through pic that let's them surface traces stand out...
Gonk Board See-Through

~ Charlie

[Dan N]

Wow! That x-ray shot is cool!

Here's my best shot at a schematic. I sh*t canned the flip flops. Some of the Vb points may be bias for fets that are gone? I'm just don't know.

I wonder what the sound would be tapped straight from the center lug of the "smear" pot.

Anyway...
http://users.rio.com/senorris/junk/gonkschemaybe%3f.gif

Marcos, no real program, just cutting and pasteing and old DOD delay schematic.

[Marcos - Munky]

Cool. I asked this because I saw the same kind of graphics in DOD and other schematics, and I think it was be cool if some software can generate these graphics. Thanks for the schematic, and the x-ray pic is really cool.

[Peter Snowberg]

Great job on the reverse engineering. Now who will be the first to make a build of it?

Take care,
-Peter

[moosapotamus]

Nice work, Dan!

I wonder what the sound would be tapped straight from the center lug of the "smear" pot.

Yup. Or, maybe get rid of the clipping diodes for a basic 'clean blend' control.

How about a varaible frequency control? Anyone see where that could be added?

~ Charlie