#58

Cool project. The end result looks really neat!

I especially like how it showcases how easy it is to get something working using platforms like Teensy and Arduino. It really takes away the technical difficulties of for instance USB to allow people to be creative.

You may be able to cut cost by not using a PCB and manually connecting the faders instead.
Similar to this box: http://www.mpc-forums.com/viewtopic.php?f=41&t=140209
To accommodate the cables, the enclosure would need to be a bit higher though. And it wouldn’t be as clean of course.

Coincidentally, I made a very similar controller but with some other design choices.
You can check it here (images) and here (source).
Not sure if it’s okay to post it here? Do I better post this in separate thread?

4 Likes
#59

Does anyone have an extra board for this laying around?

1 Like
#60

Wow, these are gorgeous!

The caps in the original post look like they fit better than the ones from Mammoth. Where are they from?

#61

I’d be interested for sure if 2 others wanna do an OSHpark run

#62

Those were scavenged from some spare toggle switches I had. If I recall those are easier to find in red, black and white.

1 Like
#63

Ps I’m really curious to know how many of these have been built

#64

I’d be in for one!

Anyone want to be the 3rd?

#65

so tempting but I know I wont have time to build it myself

somebody help these guys out!

#66

Has this version been tried yet?

Way cool.

Another idea: 16 sliding pot controlled attenuverters feeding a summing mixer.

Anyway, I’d be game for going in on a group buy of any version.

I also see that maybe some spare/extra boards were previously ordered, wonder if those are spoken for by now?

1 Like
#67

Great!

Yeah I was wondering that too… that’s why I revived this thread.

#68

It has so not been tested.

I had a quick look and didn’t like old me’s routing, so have tidied that up and also added a label as to what each standard is.

Untested because, personally, I don’t need one, and it’s a slightly expensive board to get made owing to size. Switch and jack are where they are because I still don’t have pay-for Eagle.

I’ve stuck it on Github for now - https://github.com/infovore/16n-faderbank-minijack - but am happy to remove if @tehn/@shellfritsch are unhappy. All that’s on Github is the EAGLE files. Similarly, I’m happy to put the other things - code, CAD - there if appropriate. Although that would imply that my hack really needs testing…

2 Likes
#69

Still need to test my i2c hack; was too focused on the Teletype expanders to get around to it. I know that the unit works as a USB MIDI device - I use it regularly. Just not sure if my i2c hack functions. I’ll post when I’ve figured that part out. :wink:

I don’t have any spare boards. I only got two back from OSH - the third was damaged in production. (They were kind enough to credit me for it. Nice folks!!)

The one thing that I did learn is that Tung Oil is a HORRIBLE finish for this kind of thing. It still stinks after all this time of airing out and is also still slightly oily. Sigh. I will be recutting the body when I get my hands on a laser cutter.

1 Like
#70

I love the look of this.

I have pretty much no DIY experience (Landscape all flesh only) but have been dreaming about maybe making a simpler alternative version of this. Essentially turning this into a 16 channel passive attenuator with faders - with the 32 jacks in the convenient space next to them.

To turn all / some of them into CV sources, these could then be fed with a multed CV offset from the modular. Now you have a fader bank that requires no external power and should be pretty simple to DIY. Given that it wouldn’t need the PCB or teensy it should also be pretty cheap.

The first problem I thought of is that if you passive mult an offset then the more destinations it goes to, the weaker it becomes - and as you increase one fader, the others will reduce their output a little. Now if you send the offset through a buffered multiple, I believe this problem should be resolved, or at least really reduced. I tested this with my 4 channel passive mult and I get only a tiny bit of voltage drop when using a buff mult.

Would love to hear your thoughts!

1 Like
#71

I definitely got distracted thinking about what it might take to do a small-ish run of the original thing, with a well-located 3.5mm stereo MIDI jack. It turns out that even for low volumes - like, 10 - my usual European manufacturer will do a board that size for about €35 - it’s at the point where OSH becomes less effective.

I also contemplated the rather neat configuration mechanism the CV.OCD uses - configuring from a webpage via WebMIDI Sysex - so if you wanted to change the CCs on offer, you wouldn’t need to piss around with arduino - but I think that might be a step too far.

Then I went back to work…

3 Likes
#72

Given that it wouldn’t need the PCB … it should also be pretty cheap.

The faders need to mount to something, as do the jacks. A PCB gives you mechanical support for the faders and jack sockets, it connects them all together in a way that’s far more reliable than tons of flying wires - and it would be tons of flying wires - and it can be assembled far faster.

Even for DIY projects: value your time. It’s only ‘cheaper’ if you can put it together as fast, and it’s as reliable.

Bear in mind that the PCB cost from OSHpark is pretty much as expensive as the board gets - OSH are designed as prototyping service; high quality boards that are a little expensive but fast and very low volume. Even going to 10 or 20 you could save by going to a more general-purpose manufacturer.

#73

Thanks for the feedback.

Both faders and jacks could just be panel mounted could they not?

Just to clarify I was thinking about making a one-off prototype rather than a batch and wiring up 16 attenuators seems to require relatively little wiring (Basing this on Doepfer DIY page and the Thonk ATATAT documents)

#74

You can normally panel mount faders with a m2 bolt - make sure you get the size and depth right from the data sheet

1 Like
#75

So I think I’m going to go ahead with my plan on this.

Essentially I will modify the design so that it is 16 passive attenuators. However these will be grouped into 4 sets of 4 and within these groups, the inputs will be normalled to the ‘group master input’ using switching jacks. This will allow faders to be used either as fader control (with a buffered CV offset at the group master input) or individual passive attenuators.

I have sourced alternative faders (http://uk.rs-online.com/web/p/potentiometers/2499418/) which are 50kΩ instead of the original design which was 10kΩ. I believe this will reduce the load on the buffered signal (coming from MI Links with OPA4171 OP amps). Schematic for 1 group is below.

image1 (4)
image1 (4).JPG480×640 60.1 KB

My electronics knowledge is pretty poor but I think this all seems OK. My only concern is the potential strain on the Op amps but I believe this should be OK?

1 Like
#76

I would recommend buffers for that design, otherwise all your variable resistors are in parallel to ground… so if you try to use the normals you’ll have the position of the subsequent sliders affecting the output of the others.

1 Like
#77

My plan is to use Mutable Instrument Links - I’ve demo’d this with the SSF/WMD quad passive attenuator and all seems good. My concern was whether the Op amps in Links would be able to drive the 4 parallel faders ( I don’t know how to do the maths to work this out…) though this was more of a concern with my original design which had all 16x10kΩ faders in parallel together.