ultra

hey cimo, i think smash is selling the bourns encoders. i don't know if it's the same footprint but i'll check it out. thanks :)

hi, i've started the design work of an ableton live control surface. it's pretty heavily based on sasha's controller (http://www.midibox.org/forum/index.php/topic,10662.0.html). edit: this didn't show up right for me, link to the full image: http://www.sub-version.net/diy/cs/livecs1.gif this is the initial design, and i'm sure i'll be changing some things. each track has 8 rotary encoders and 6 buttons. their main functionality should be pretty self-explanatory. there's no feedback from live in this design (to make things easier), so when you move an encoder on track 3 for example, there will be code to make live display that track so you can see your parameters. also, each track has a button that looks like live's plug-in icon. clicking that will give the encoders a new set of control change numbers, making it an "instrument mode" that gives you 8 knobs to control whatever is on that track. there's an led to indicate that this option is selected and another led to show the track is selected. one problem i've found with this is there doesn't seem to be a way to assign midi to the track view and clip overview, so i don't know how i can make the instrument or eq automatically show up. the 9 knobs at the top-right are analog pots that change channels with whatever track is selected, in case a rotary encoder just doesn't work right for the instrument. the big black square is a touchpad mouse found here: http://www.ergonomictouchpad.com/ i like the fact that no mouse buttons are required. i want to put this on an integrated usb hub and put the remaining usb ports on the back of the CS case so i can plug in whatever midi controller i'm using at the time. perhaps i can use midibox usb and have everything work off a single usb connection. i'm probably missing some essential controls, but there's space available. the idea is to make a simple controller and not one that does everything... the mouse can take care of that. the lack of a crossfader is intentional and i highly doubt i'll add one. any ideas on what i'd need to add? ultra

does anybody have an eagle part for the alps EC12E series encoders? thanks.

ok sasha, you asked for it. :) what is SEL? just to show the track is active? what is AUX? why 3 sends? i only see A and B per track in live. that's all i have for now. i'm gonna ask more questions soon. thanks!

i suppose you haven't documented anywhere how this works yet? as in, what the buttons are, how the shift buttons work, etc. you're probably using live in a different way than i am, and i recognize a lot of the buttons, but i'm not sure what some of them are there for. i'm not asking you to do this if you haven't, but eventually i'll get all the info out of you in bits and pieces anyway. i decided to go with all encoders because i don't mind looking at the screen, and i really want to keep mine fairly basic, hence the touchpad mouse so i can get at things not on the CS. i figure i'll have some code that changes live to show the track when i move an encoder so i can see what the value is. not to hijack your thread or anything, i'll create my own one of these days :).

sasha, this is great. i think i'm going to do a very similar layout but with all encoders (and of course, much more basic looking :).

they don't have to be a certain way, but i always put the tolerance band either on the bottom or right to make troubleshooting easier.

nils that 13mm tact is the kind that doesn't have a snap-on cap, right?

every part has an identifier and a list will be on the wiki. as a bonus i'll even include a BOM to mouser :P. as stryd said, the resistors aren't polarized :). as far as pcb availability, all i can tell you is there will be an initial group buy. after that i have no idea.

has anybody found an encoder and tactile switch solution where they can be mounted to the same pcb and not have to make any height adjustments? i'm starting on an ableton live control surface with only encoders/switches and i'd like them to all be mounted to the same pcb. thanks.

yeah soon. :)

i've started the documentation for the pcb if anybody's interested in a description of the headers: http://www.sub-version.net/diy/ultracore

thanks for your comments sasha :). usually people don't like posting about it ahead of time because it invites a lot of feature requests, but in this case the feature requests have helped out a lot. i hope to order the prototypes this week and get a group buy going once i can verify everything is absolutely correct. in the meantime i'll get some documentation made and try to get some pricing info :).

it's not pointless, it's a simple tweak that i didn't think of. i'll rotate it and then move it down a couple of clicks, just above R1. the diameter of the cap tends to be slightly smaller than the footprint anyway, but there will be enough space for more. that will help get it out of the way of the power headers. i really have no problem making small usability changes. in fact, this weekend is the time to speak up if you can think of anything else (just no major layout changes please). glad you guys like it. i'm hoping to order the pcbs by early next week and test them asap. i have a mouser order (almost) ready to go to fill it with parts. then the fun begins :). edit: i've already made the change: http://www.sub-version.net/diy/seq/images/core.gif

hopefully it's not too close to the edge now. passed DRC ok.

yes the idea regarding the heatsink is a good one. i'll move some things around. i agree that it would look nicer with the power header in line with the midi i/o, and i had thought of that. i put it down there because i wanted to position it in the middle of the small pcb i'd be stacking on top of it, but i guess i'd want it off to the side anyway. shouldn't hurt if it's hanging over the board and i could put a spacer in the other side anyway (if this makes any sense). these changes can be taken care of pretty easily so i'll do ahead and do it :). edit: check the link again. took about 1 minute :).

i've been working on a control surface pcb for seq but i think you're better off starting from scratch. i've learned eagle recently and the seq control surface pcb is the first i've made. if you're going to rearrange an existing pcb, you'll be basically doing the same thing as if you started it from scratch, only you'll just have to make the schematic first. by the way gold phoenix does pcbs up to 14.5" wide.

hi, the pcb is pretty much complete. the final size is 160 x 80 mm. http://www.sub-version.net/diy/seq/images/core.gif the power and midi headers have changed a lot, and here are some details on that. J1A is the same as J1 on smash's core. J1B breaks the connection and goes out to a switch. simply bridge it if you don't use this. J2A and J2B are identical and are the same as J2 on smash's core. you can use these to power the core and also power something else, or just get vdd/vss out of it if you need it. these should be pretty self-explanatory. the decision to use DIL was to give more options. also, i plan on adding a sub-circuit over the power header (not a pcb that will be released) and DIL will offer more support. on any of these connections, SIL connectors can still be used, even if you use DIL headers. in fact, two SIL connectors side-by-side in a DIL header will be more stable. i hope to be able to get this made and test it out very soon. i'll let you know what happens :).

splint type. if they were red on black, i'd go for 36 of those. grey on black, 100. i'll have to get back to you on a quantity for white on black, sorry. it's going to depend on what my 9090 case looks like and i haven't received it yet. ultra

am i reading this price correctly? about 32 cents USD per knob seems really cheap and i would buy 100 of these if they were grey on black.

you must mean J4, since it's the only headers (besides midi i/o) that go to the IIC modules. it's my understanding that these give access to midi i/o on each IIC PIC, bypassing the optocouplers. could be useful to someone. the additional pcb height is necessary anyway to make room for the connectors on j8/j9, j10, and j11. i've tested the spacing with connectors and what is there is necessary (with a little extra). the lcd headers couldn't be moved up because i need the extra space by the power headers. i plan on adding a sub-circuit that will slip on top of those headers to make a pushbutton power toggler using a tact switch. there's just enough room between those headers for an 8 pin dil socket to be mounted over them. sorry. i don't know anything about this and it's not in my plans.

i've reconsidered on using DIL headers for everything but the midi i/o. the pcb is a little bigger now, but still not bad at all. http://www.sub-version.net/diy/seq/images/core.gif how do you suggest this would look? the best i can think if would be 5 sets of DIL headers that look like this: X+G-X X+X-X each header would be one set of i/o, so DIL can be used off to another board, but SIL headers will fit on it as well (at least mine will). i wouldn't object to doing this since it would be a simple change and wouldn't add much space at all to the pcb. you're gonna keep going forever. i have klee and wogglebug pcbs sitting here that i haven't touched yet. gotta finish the 9090, box-o-trix... oh and this seq thing i have going on. it never ends :)

i'm hoping soon. i think this week i'll be able to have a prototype made. i don't know a price yet but it will be relatively cheap. i would guess not more than $15 or $20 USD, and i think that's a high estimate.

if you're only using one IIC module, then it's already addressed as the first by the grounding on pins 12/13 of your IIC PIC. are you etching this pcb yourself? if not, i'm making a core/4xIIC/8x bankstick pcb and it might end up being cheaper because there will be a run of them available. you only have to use the parts you want. the only problem is i still haven't had a test pcb made. i expect to finish it up this week. http://www.midibox.org/forum/index.php/topic,11057.0.html

apparently 5052 aluminum is for bending.