Glacial Communications

It’s close enough to the end of the month, so I might as well recap my SDIY (lack of) progress for March.

I’ve unfortunately gotten very little done building-wise. I’ve soldered a handful of parts onto the Supercontroller PCB, and I did actually complete this simple dual LFO/Joystick controller from start to finish.

Most of my hold up has been “being busy with other things” combined with the fact that most of these projects are now stuck in front panel hell. After learning that I can get reasonably cheap aluminum cut to size, I’ve started heading down that path. It’s been an interesting learning experience thus far, but I don’t have much to show for it yet- especially since I didn’t have a drill press until last Thursday. As of yesterday, I now have what should be a Frac-sized drilled blank panel, which I then found out doesn’t fit my enclosure (a telecom rack frame that I scored for free). I guess tonight’s experiment will be to try drilling Euro-sized/spaced holes and see if that works- my measurements seem to indicate it should. I did also widen out the holes on one of the panels the enclosure came with(Yay Unibit!), so that’s at least something I might be able to get working soon, once I find a circuit to fit in it – maybe a little LFO or something.

So, I guess the “No new projects” rule stays in effect for April. Hopefully something gets finished then. I’d really like to say I have a modular synth, even if it’s only one module…

I finally got around to uploading my RPM Challenge Album, Radioactive Decay – You can find it over here, in easy to use Opentape format.

There’s also a short non-RPM Challenge track on there as well, an outtake from the forthcoming Glacial DIY tape, since today is a somewhat terrible anniversary.

Well, since issuing my March Manifesto, I think I’ve managed to break both rules so far. Excellent! Bear in mind, the parts were semi-unobtanium for cheap (CD4006’s and a 74C922  – fifty cents apiece!), and the new project is something that isn’t going to happen any time soon (an Arduinome – I ordered the shield PCB, and I have a MAX7219. That’s going to be the extent of its progress for quite a while. Must. Clear. Module. Backlog. First.)

I did make some progress on the USB audio interface, but at the moment, it’s only usable as an incredibly expensive  USB-powered LED. Not Good. I’ve painstakingly continuity-tested all the traces on the PCB to pins on the PCM2902, and those all check out, but there may be a bridge or something along those lines, or I fried the chip.  But so far, it’s getting power from USB, the LDO voltage regulator is working, and maybe the IC is doing something? I’m seeing Vccci/2 on the Vcom pin, but the board isn’t getting recognized by either of the computers I’ve attached it to. Frustrating.

In a worst-case scenario, I do have another PCB (cost me just as much to get two as it would have one), and another PCM2902 (yay pairs of samples!). So I can either build a second one, try  the other chip in this one, or some strange combination.

In preparation for whatever I decide to do, I was looking around the Curious Inventor site at some of the surface-mount soldering tutorials, when I found this. I then ended up running out to my local fabric store and finding a slightly different model (a Darice Craft Heat Tool, which is from an Ohio-based company, and the unit is made in Taiwan, in case anyone is looking and wants to know such things.) I haven’t tried using it to solder anything yet, but I did desolder some surface mount parts from a scrap PCB, and it worked really nicely.

I also attempted to get some aluminum panels for various projects, but failed – the posted hours for my local Metal Supermarket didn’t match reality. I will be going back on Friday.

And lastly, I came up with some more ideas I’m not implementing any time soon – a smallish “Analog Computer” module (say 4 op-amp’s worth, think this but way simpler), and a rather crazy Lunetta panel (not sure if I’ll fully do it the way I envisioned – maybe if I can find an ALU chip (74181 or the like) for cheap…)

In my last post, I mentioned that I had a good idea shortly after declaring “no new projects for a little while.”

Doesn’t mean I can’t talk about it. And hey, maybe one of you readers will take up the torch and design the schematic, PCB and/or firmware for me, so that when I clear my project backlog, it’s ready to go!

I’ve gotten really interested in the Spin Semiconductor FV-1 effects processor IC. It’s got built-in ADCs and DACs (so it’s a one-chip solution unlike some others), and if I’m reading the information on the web correctly, they almost seem to be encouraging the equivalent of a ‘clock bend’. I’m sort of curious as to how slowly it can be clocked – their examples don’t go below 20 kHz, but I wonder how lo-fi one could get it. I should ask on the messageboard…

Here’s kind of the feature set I was imagining:

• Switchable clock rate, between a fixed clock (at the suggested 32768 kHz or maybe better if you’re some kind of audiophile type), and the variable schmitt trigger clock outlined on the website, possibly tweaked to go lower if that’s technically feasible. Hot-switching between modes need not be required- I’d be cool with having to power-cycle it when changing clock mode from fixed to variable.
• The FV-1 on its own supports the use of a Serial EEPROM for additional effects programs. Change this to some kind of bank-switching multiplexing arrangement to allow multiple Serial EEPROMs to be used.
• Front panel/microcontroller. Use something like an ATMega168 (preferably with Arduino-type bootloader so upgrades to the firmware are easy, and include the headers for an FTDI programming cable!)  to control a 2×20 character LCD panel, and set the bank/program numbers. If you need more inputs, outputs or memory, feel free to use something bigger (ATMega644p in a Sanguino arrangement, etc.) Don’t worry about program names for the external banks. “BANK 3, PROGRAM 4″ is fine. Feel free to use whatever menu navigating mechanism you want (buttons, pots, etc.)
• Additional features would be up to you. If you can add a simple mechanism for uploading new effects programs to the Serial EEPROMs (especially from Mac OS X), that would win, but I’ll live with the possibility of using a separate EEPROM writer and removing/inserting the ICs.

A starting point for looking into might be this one, but I think I want mine in a 1U rack form factor – I’ve got too many desktop doodads already, and it’s time to get vertical. This post also has some tantalizing ideas (Arduino as EEPROM writer! Write your DSP code  in Java!) but they’re not released yet.

So, there’s my idea.

I read Bre Pettis and Kio Stark’s Manifesto of the Cult of Done earlier today. While I liked some elements, I  didn’t like others (#5 in particular – I usually can’t even source parts for a project in a week, let alone finish…). It did get me thinking, though. So I came up with my own.

My own, simple, manifesto for the month of March:

• No new projects. Finish all the ones I’ve started. Then I can move on to the next ones when these are Done.
• No buying parts for unstarted projects. Only those needed to finish ones that are started(*).

I’ve got something like six or eight different circuits in various states of construction or debugging. Time to get that number down. Zero would be preferable, but I’ll see where I end up at the end of the month…

The sad thing is, about ten minutes after I declared this to myself, I came up with some more awesome ideas. But they now have to wait. No sooner than April, or maybe even May (I’ve got Blockparty to think about too!) Instead, these new ideas are just getting written down in my “Electronic Project List” in Google Docs. When I’m ready to take on a new project, they’ll be there, waiting for me.

(*) I will declare a small exemption to this – If I order parts from some place where shipping is a total PITA and I’ll be waiting 2+weeks anyway (i.e. Futurlec), I may batch up some other things. I do not currently anticipate having to do this, though.

What did I accomplish building-stuff-wise in February?

• Built this SN76477-based noisemaker for my sweetie.
• Made some small progress on the SN-Voice and MPS. They’ll be done soon. I swears. Also got a TH SuperController PCB, so add that to my Thomas Henry queue
• Started on a Blacet Dual SVF (he was clearancing the PCB. I said “why not?”) Of course, I neglected to get a front panel for it. Gotta rectify that.
• Threw together an incredibly simple triangle wave LFO (it’s an LM567 tone decoder/PLL IC, a cap, a resistor and a pot. Add an LM7809 if powering off 12/15V). It’s only on a breadboard now, but I’ll fully document it in a future post.
• built my own lousy rendition of a Gene Rayburn Microphone, as well as the SparkFun mic preamp.
• Designed and ordered a PCB (two, actually) for the USB audio interface I decided to build – I decided flying wires on a protoboard was just too much opportunity for noise issues. Once those come in (tomorrow?) I guess I’ll have to get started on one.
• Finally received the Gristleizer PCBs I ordered in January. Put one together, but it’s not quite right yet. Waiting to get my account approved on diystompboxes.com to try to get some tech support.
• Attempted some more debugging on the 4069 VCO. fixed one bad connection, and switched out the 4069 with a more appropriate one (4069UB instead of HCF4069). Still no sound. Getting stumped.

I also acquired a number of useful tools this weekend – a mitre box/backsaw combo, a center punch, a clamp, nibbler tool (which should allow me to cut rectangles!), some electronics-quality tweezers, and a set of helping hands.