Earthquaker Devices - Rainbow Machine

culturejam · Post by **culturejam** » 12 May 2012, 02:03

Over at madbean's forum, jaidee pointed out that the JetCity youtube demo of the Rainbow Machine includes a decent gut shot. So I grabbed a couple screenshots. As several of us suspected, it has a Spin FV-1 at the core. I tried my best to composite multiple frames into one cohesive image. It's close but not perfect. The video uses the old "Ken Burns Effect" to pan over the still image with some slight zoom effect, so there's no way to get it exactly right. Here's the video:

However, there is no *visible* EEPROM from which a custom program could be loaded. Based on what I can hear in the demos of this pedal, there is almost certainly custom programming going on. I just don't see how it could be done otherwise, unless one of the internal pitch shift programs is being used with some very creative feedbacking setup.

Another possibility, albeit far less likely, is that EQD had Spin load their program onto the internal ROM in the FV-1. This, however, requires a very large purchase of chips, as mentioned here by Frank from OCT Distribution (Spin's distributor in USA):
http://www.spinsemi.com/forum/viewtopic.php?t=131

I think maybe there is an SMD EEPROM on the bottom of the board. I also don't see the crystal oscillator, so maybe that's on the bottom as well. And the crystal is NOT an optional component.

Or maybe the EEPROM is hiding behind that big-ass regulator. Anybody got one of these who can open it up?

KindaFuzzy · Post by **KindaFuzzy** » 12 May 2012, 12:47

I love the crazy zelda stuff this thing does, too bad it probably is some programed chip, it'd bee a cool build.

juansolo · Post by **juansolo** » 13 May 2012, 08:16

I have one... Possibly the most pointless thing in the world ever. But made of purest awesome.

Dirk_Hendrik · Post by **Dirk_Hendrik** » 13 May 2012, 09:33

Juan,
If you have one, are you able to make a photo of the bottom of the PCB? That might answer quite a few questions raised on where the software the thing runs is located.

juansolo · Post by **juansolo** » 13 May 2012, 19:14

Incomming! took some higher res ones of the top also.

Dirk_Hendrik · Post by **Dirk_Hendrik** » 13 May 2012, 19:19

Juan laddie,
you rule!

Nocentelli · Post by **Nocentelli** » 13 May 2012, 19:28

juansolo wrote:... Possibly the most pointless thing in the world ever.

It's certainly weird sounding, it reminds me of the the old Boss PS2 delay pitch-shifter... that was equally useless.

culturejam · Post by **culturejam** » 13 May 2012, 19:32

Well damn. I sure don't see an EEPROM anywhere. So that must mean that the FV-1 itself has the code loaded, which would be pretty expensive.

OR

It is also possible that the Rainbow Machine is doing it's funky magic with one of the default programs and some seriously wacky feedbacking mechanism. That would also make sense with the limited pitch shifting range, which is identical to the range in both of the default FV-1 pitch shift programs. The FV-1 itself is capable of generating more than an octave up and down if programmed to do so, and the default shifter programs are limited to a 4th below root and a 3rd above. And that's *exactly* the range of the Rainbow.

Another thing: I still don't see the crystal oscillator. So I'm thinking the CD4049 is set up as a clock pulse generator:

Pin 9/10 is supposed to have the crystal hooked up there. I see pin 9 runs down to the CD4049, so I guess that's part of the equation. And the Tracking pot ties in to the 4049 as well, so the clock rate must be adjustable. Pretty cool. The more I look at this, the more I'm thinking it uses one of the default FV-1 programs. Gonna review it some more.

juansolo · Post by **juansolo** » 13 May 2012, 19:53

Nocentelli wrote:
juansolo wrote:... Possibly the most pointless thing in the world ever.
It's certainly weird sounding, it reminds me of the the old Boss PS2 delay pitch-shifter... that was equally useless.

It can actually do some useful stuff to be fair, this is a really good overview of the pedal:

culturejam · Post by **culturejam** » 13 May 2012, 20:02

Just had a thought:

The FV-1 has two inputs and two outputs (stereo). I wonder if part of the "magic" is feeding back from output 1 to input 2 and related shenanigans?

Govmnt_Lacky · Post by **Govmnt_Lacky** » 13 May 2012, 20:17

culturejam wrote: Another thing: I still don't see the crystal oscillator. So I'm thinking the CD4049 is set up as a clock pulse generator:

This looks pretty good. I believe there is another stage in there (as Pin 7 of the 4049 is in use)

Also, the 100K in your pic is actually the Tracking pot. On the mask side, the Tracking pot is labeled 10KB. The actual pot used in the pics is 25KB

Food for thought

ppluis0 · Post by **ppluis0** » 13 May 2012, 21:16

Hi there,

Here`s the link to the datasheet of the chip labeled FV-1: http://www.spinsemi.com/products.html

Cheers,
Jose

chicago_mike · Post by **chicago_mike** » 13 May 2012, 22:32

thats kinda cool.

Jack Deville · Post by **Jack Deville** » 14 May 2012, 00:31

Um...

Who tracked that board?
For serious. Take a close look at the tracking, component placement and massive copper pours.

And also, what's up with the aux. jumper at the top? It appears to be an omitted analog ground connection?

culturejam · Post by **culturejam** » 14 May 2012, 02:04

Jack Deville wrote:Who tracked that board?

DipTrace?

Jack Deville · Post by **Jack Deville** » 14 May 2012, 02:15

I was turned on to that CAD package recently.
I always enjoy learning, so I thought I'd give it a chance!
Well, no sir, I didn't like it.
Clumsy, poor schematic capture, and rudimentary (albeit highly user-friendly) PCB layout section.
It did have a bunch of "Oh-boy!" features and add-ons, like the in-program 3D mockup, but that didn't surpass its shortcomings in my eyes.
As usual, the autorouter was loose and in-efficient, but then again, I didn't put 30hrs into dialing it to my requirements.

Seriously though, have a close look at the tracking/component placement on the PCB.
I'm like: LOLWUT!?

Sweetalk · Post by **Sweetalk** » 15 May 2012, 19:25

I knew it has a FV1!!

. If pin 13 is tied to ground it's internal rom programs , if it's tied to +V, external EEPROM. I can't see it well in the pictures. I think that the Magic it's some kind of feedback and the tracking it's changing the clock and so changes the resolution.

PS: The layout is HORRIBLE.

culturejam · Post by **culturejam** » 15 May 2012, 20:01

All signs are pointing to internal ROM, it seems. I'm impressed with the creativity of this project, I'll have to say.

Sweetalk wrote:I think that the Magic it's some kind of feedback and the tracking it's changing the clock and so changes the resolution.

Yes, it's a feedback monstrosity. And definitely the Tracking control is adjusting the clock speed.

Sweetalk wrote:PS: The layout is HORRIBLE.

Maybe it's not "artwork," but it's functional and not noisy. So I'm going to call it a "good" layout from that perspective. It's not like the Rainbow Machine is part of the Space Shuttle guidance system or anything.

(but I'm biased, because I usually make ugly layouts

)

teej212 · Post by **teej212** » 15 May 2012, 20:54

ive recently put my whammy IV in a feedback looper and have gotten the EXACT same sounds as the rainbow machine. If I were to guess it is just a pitch shifter on the fv-1 which im pretty sure is a pre loaded program on the spin chip.

Sweetalk · Post by **Sweetalk** » 16 May 2012, 00:22

teej212 wrote:ive recently put my whammy IV in a feedback looper and have gotten the EXACT same sounds as the rainbow machine. If I were to guess it is just a pitch shifter on the fv-1 which im pretty sure is a pre loaded program on the spin chip.

The fv1 comes with a +-2 semitones pitch shifter in the internal rom. The pitch control its on the rom program, primary and secondary must be some harmonies volumen. Check if the center lug of primary and secondary pots are tied to pins 21, 20 or 10. Ill try some algorithms to make this!.