Tuesday, March 31, 2009

Field Trials Begin

Over the last 48 hours we finished the final coats of stain and acrylic covering, and i finished the last of the wiring of panels 5 and 6.

We then began harnessing the UMC's to the shelf inside the enclosure, as seen below:

Next came drilling holes for the screws on the face panels. we set the panels in one by one and marked them, took them off and drilled, reset the panels and put all the screws in. way too many, frankly. here is what it looks like underneath now, as we have not yet re-attached the doors:

Finally, we attached the heads to the UMCs (i labeled each head by X-Y, x=UMC channel, Y=which of the four heads it goes on, for a total of 25).

We brought it upstairs to the studio - there is really no room whatsoever left, but i shofted some things around to get it in here. As of now, i am happy to report it is pwoered, connected to the computer, and all controls are working perfectly. will start creating an ableton set to complement it immediately and give it a good thorough exercise over the next week.

here it is in all its glory:

Side view:

We still need to:
* buff it and shine it one last time
* re-attach the doors
* drill a hole or figure out a scheme for the cables coming from the USB hub
* attach the main volume knob somehow
* optionally re-add the small green LED by the channel volume knobs
* find and attach slider knobs (8mm head)

and maybe some other things i can't think of now, but overall this thing is SIIIIIICK!!!

thank you to Dismal/Steve for the Colossus name, Ken at Hale Micro for helping out and answering so many questions, my great wife Jeannette for helping and supporting at every stage, Jnet's dad Howard for the carpentry experience and precision that i never would have come close to, and even to Ethan for setting me on this course (whether you realized it or not). This whole thing started from me wanting to take apart a computer keyboard and wire the letter buttons to other controls that could be used to control Ableton. his first comment was, thats way harder than it sounds, you should try something else. wallah, here is the something else.

Gonna break to play with this thing for a few days, but soon enough i will get started on a set of other controllers in a similar vein (but considerably smaller) to sell and/or give to the equulei compatriots.

If anyone wants more detail on part numbers, etc, feel free to drop me a line any time.

Saturday, March 28, 2009

Gear Porn!!

okay, one more picture just for the drooling aspect. put most panels on even though we aren't ready to actually attach, just to see what it looked like:

good night all. busy day tomorrow working this thing.

Finishing Touches

A couple structural things remained, which were completed this past week.

the doors on the back to gain access to the UMCs, underside of the panels for replacement, wiring, USB hub, etc:

and then some handholds to carry the beast. extra wood placed inside because the outer wall is really just a thin layer of oak ply, and then braced it against the frame itself. they should be pretty sturdy:

above you can also see inside and below the handholds is a small shelf that was put in to accomodate the UMCs. we will bolt them onto that shelf with 4-40 screws and a small standoff which have already been purchased, just waiting for final touches to be finished.

Finally, we finished sanding and beveling edges, and then began staining it. pictured below is after the first coat, we will do at least oen more, and then acrylic goes on that to make it smooth and shiny:

note in the pic above the enclosure is lying on its back.

on the technical front, i have encountered a critical problem and have reached out to Ken @ hale micro to see whats up. last night i tested panel 3 - its the main one with joystick, faders, and main volume knob. i tested it fine on all fronts and even played with ableton briefly. this AM i was showing jnet's dad how it worked, powered it up, and after initially turning on fine when the computer cranked on - it blinked out. since then it has not come on when the computer is attached, but strangely when i plug it just to the USB hub, it powers on fine. no idea whats going on and until i have some advice from ken, not wanting to try plugging it all into one of the extra couple UMCs i have for fear of another one getting fried, if thats what even happened. maybe i am just trippin on some stupid setting, but all the other ones seem to be working great now.

UPDATE: adding schematic below for Ken to check out

Wednesday, March 25, 2009

Panels and Enclosure completed

I have finished wiring and testing out two panels now - in the pic below you may be able to see the bottom row of green/red buttons lit up - they llok great. you can also see the clear difference between the first one i fucked up (on left) and the one i wired correctly (on right):

i am running both through a USB hub that will be velcroed into the bottom of the enclosure to grab data from all 6 devices in the end. its not visible in the pic above.

and i also solved the button issue mentioned below. I had a dip switch setting wrong that was interpreting two buttons as if they were analog pots. done deal. the LED issue remains, but on the second one that i wired up correctly and without the LEDs, everything is running perfectly. i may detach them from the first and just attach to power. they will be always on. interesting but not what i had hoped for.

In addition, the entire enclosure is done save for staining it, which i think we will do this weekend. and i have connected all buttons and pots, ready to be wired as soon as i can:

Sunday, March 22, 2009

Panel Work

I have finally completed all of the first panel, which is equivalent to one full UMC32 circuit board. Here it is below:

You will notice i failed to put two of the small knobs on - this is due to discovery of an issue that i did not realize until near the end, and at the moment i am not willing to back out and redo a bunch of soldering - i just really want to move on to the next panel and fix the mistake there. the issue arose despite my best planning for which controls would go on which of the four heads of the UMC. the last one was gooing to have 6 analog pots and 2 digital buttons.... unfortunately the UMC DIP switch settings for the last head only allow for 0, 2, 4, or 8 analogs... so i did not even wire up the last two pots. thus the lack of knob, which will be an obvious sign to me when playing it that nothing is attached to those controls. i'll fix it later. for the next identical panel i am spreading the analogs out more.

Other issue is a soldered and wired up the volume knobs to the LEDs before realizing my original design, and much more sense from a space perspective, is that they should be swapped with the pan knobs, smaller all black with vertical white line seen at the bottom of the panel. the other panels will correct for this as well.

And i am still having the LED issue - the two volume knobs are acting like log pots and giving me fits when trying to fine adjust volume of a track.... such is life. if the next one does it too, i may just unlink the LEDs from the volume and give them power. it was a good thought, although i admit to stealing it from a youtube video DIY project.

Another thing i discovered - although i intentionally waited to attach the knobs until each pot was confirmed good (because they won't come back off without serious risk to the panel), it has a price. the leverage needed to squeeze them onto the shaft is gone with the crazy wire mess in the back, trying not to touch wires and solder joints, etc. the result is some knobs i can't get all the way on, so they have slight variations in elevation from the base of the panel. not sure if its obvious in these pics.

Only other issue i have seen thus far is a couple of the illuminated buttons are acting strange in ableton - not sure if it is maybe a setting i have in ableton rather than the buttons... problem is they are assignable, ableton recognizes them, but when i assign them to things like mute or stop, they only operate once (mute track, but will not unmute, or vice versa). after i click manually with mouse, they will perform the operation once, and not again. not sure what that means.

Overall though pretty good success - all controls responding and i assigned the entire thing to a new set in ableton just to test it all out. hopefully some of the bugs go with it soon.

Here is a profile pic just to give an idea of the crazy mess of wires on the underside. it would probably look better were i more professional about it:

Starting on the next panel ASAP, and after getting the hang of it and discovering a few tricks, its moving faster. No pics but jnet's dad is done with the rear doors, the corner trim, and the shelf for the UMC's to attach to.

Friday, March 20, 2009

Box Almost Done

We spent the majority of today finishing assembly on the box and beginning to size for the plexiglass panels. There are lots of notes on this, but here is how it looks right now, from the side and including the trim we are putting around all the corners, ala oak cabinets style:

And a brief view from the back side:

Finally got a view of the plexiglass panels, and they rock. initially they are covered with some protective film, so you will see in many pics i've left it on except a couple to start attaching all the controls. They do the laser cuts with the film still on. Here are the panels that will eventually go on the device:

In these views, I have attached all the potentiometers, but no knobs are attached yet. I want to plug in and test each before i attach them, because in many cases they are next to impossible to remove once shoved over the shafts.

Here we set up all 6 panels after getting the trim around the perimeter of them that is set to be flush with the panels:

And finally, after taking it home and spending over two hours (much longer than i thought), i have one head of one UMC plugged in and running successfully (so, only 23 more to go), including the LED signal from volume knob:

The LED attachment works but is causing weird issues.... i wonder if i have the wrong resistor, but the knobs they attach to are acting almost more like log pots than linear. will investigate further.

Time permitting, we will finish the trim, the shelf across the bottom to attach the UMC's to, and even the doors in the back. then its all soldering until dawn.


Wednesday, March 18, 2009

The Face Panels

Face panels were completed, paid for, and received yesterday. J-net's dad bumped them up against the frame as pictured below and they appear to fit perfectly, so we are a go for trying to finish it this weekend!!!! many pics hopefully to follow.

Sunday, March 15, 2009

The Frame

After a brief visit over to the in-laws house, here is how the frame is shaping up to be:

I have to admit this is substantially cooler and better quality than i ever could have done on my own. with luck, the panels are done in the next few days and we can do final assembly of the box next weekend. then comes the fun part of soldering 190 control points plus LED power, etc.

This thing is getting close!

Saturday, March 14, 2009

Some Pics to Bide Time

Here are a few shots from the Visio drafts of version 4 of Colossus - and i did finally get the call from the plastics place. they quoted me and began work on the panels - likely to be done mid-week.

here is the main panel appearance:

And actual cut positions for the laser:

And a view of the connection scheme to the UMC devices (each color on each panel represents one of four heads per board, and i've tried to combine the analog and digitals together as much as possible since the DIP switches must be set accordingly):

And finally a picture of the frame that jeannette's daad began building:

I believe each is a support, so spaced across three feet, each on holds the edge of one panel below and one panel above, for a total of 6 panels at 12x18". more pics after some work or viewing of it this weekend.

Thursday, March 12, 2009


i finished the basics of the x0xb0x last night and bolted it (sort of) to the case and turned it on. the system works, and most functionality appears fine. i haven't had a chance to test it with MIDI, computer/USB control, etc, but so far so good. i was able to make patterns and it sounded just like a tb303. AWESOME.

some miscellaneous troubles, however, that i will need to work out. the standoffs, screws, and faceplate do not seem to interact well together, evne after following the instructions to a tee. this means that some of the buttons are a little too sunken in, but i am able to trigger them all.

additionally, i can tell the tuning is not correct, which could be a result of not having an oscilloscope to test with when i was building. before i bolt the thing together finally i will get under the circuit board and adjust the trim pots to see if that helps. i also can tell that a couple of the other pots (cutoff and decay, for example) really don't have much effect at all until at least half way through the turn, almost as if they are supposed to be log pots rather than linear, or vice versa.... not sure but i will see if that issue exists in the ladyada tech forums.

finally, i cannot find the screws, if they ever existed, to bolt the plastic case together from the bottom. very annoying, but i can probably find some elsewhere if i have to.

overall very excited though that such a complex project, for being my first foray, actually even turned on, didn't explode, makes noise, and responds to stuff that i do. acid basslines here i come.

For colossus, NEWSFLASH - the plasticare guys are STILL not finished... they have promised (as well as a few times previously) to have the final quote and readiness to start by um... yesterday. i am crossing fingers and praying that they might pull a miracle and have it by this weekend, because:

the frame and wooden enclosure is almost complete! jeannette's dad is building almost the whole thing and says he is ready for the plastic panels... i will go check it out this weekend and take pics, but this is a huuuuge help, as i know NOTHING about woodwork. we drew up a design on a piece of paper last week, and before you knew it, he was half done. retirement has its benefits. ;)

Wednesday, March 4, 2009

x0x and Colossus Progress Report

Been awhile since i posted, so a coupel quick updates. the x0x is coming along nicely, though i keep forgetting to swipe my wife's camera to take some pics of the circuit board and what not. i have finished all the internal components (VCO, VCA, envelope, I/O section, etc) except the sequencer - and planning to tackle that this weekend. all the tests i ran to see if i was doing it right seem to indicate all sections have adequate power, etc. the one thing i didn't have was an oscilloscope, and decided to wing it without it after seeing how much an oscilloscope costed. if i decide to embark on more analog synth projects, i will go and get one.

As for the Colossus, i am STILL waiting on the plastics place to finish my design, though i think its pretty close. i still think they are kind of idiots for taking so long (and they seem to have real issues with working via email - welcome to the 19th century douchebags).... but i settled on a design that was cheap and easy for them. it amounts to 6 panels each 12"x18" and only an 1/8" thick, negating the need for countersinks. there will be a single brace behind each one across the middle, and i think that will be plenty. each panel has approximately 32 control points, so they equate to one UMC each. the case is not going to be all clear, the panels are now black, and the overall case is wood - and i have been toying with different options for making it somewhat "modular" - so if i didn't want to take the whoel gargantuan thing to a show, i could just grab one or two panels or whatever, and wire them to the USB hub.

I'll try and post some drawings of that later. kudos go to my wife and to her dad for being very eager to help build that wood case, as i am stupid about that stuff. we worked on a little prototype a couple weeks ago and i am actually going to keep it and wire another single UMC to it when i get the time between the other larger projects above.