I just finished assembling the Shruti and everything seems to work fine (buttons, menus, pots)… except I can’t get it to see the midi input. Voltatges are ok on all the test points… the jumper is set on midi
What could it be? Broken opto coupler? Did I break something else? What could I do to troubleshoot the problem?
forget it! The problem was my midi keyboard… attached it to another one and now it works! YEAHHHH I’m the happiest man on earth!
i won’t be envious for too long, mine’s on the way too
Getting accustomed with the interface, soon I’ll upload some sounds!
I have to figure out how to dump the preset sounds on the computer, in case I want to get back to them sooner or later (there’s a couple of them I quite like). Though of course I want to create my own sounds, this little device has quite some stuff to play with, and my fingers are already itching!
Anyway, the problem with the keyboard was quite a weird one, so it might be worth telling you about it: I have a Midiman Oxygen8, I had the midi wire come out directly from it and into the Shruti, which would give no signs of incoming midi data. Then I attached it to the Blofeld Keyboard just to see, and it worked. So then I reattached the Shruti to the Oxygen8, but this time passing through the Korg DSS-1, (which I have ususally attached to the output of the Oxygen8). And this worked as well. I guess something in the way the Oxygen8 outputs midi doesn’t go well with the Shruti…
I’m puzzled as to this one… Weak signal? In any case, more likely to be a hardware problem than a firmware problem as the DSS-1 is likely to just pass the data to through, rather than decode it, reformat it and resend it
Same problem here. My M-Audio Axiom 25 somehow does not work with the Shruti-1. My cheap midi keyboard does. I will check further Shruti-hardware-connections the next days.
Bitbluster: Is your “Midi out to USB” turned on or off? The axiom is weird in how it handles routing and if it is on, then you wont get anything on the midi out unless you route it back to itself in software (idiotic scheme)
Rumpelfilter: Do you have a scope by any chance? This could be an optocoupler gain issue coupled with a low output from the oxygen. I had this problem on a project once and going to a different brand opto which had a higher gain fixed it. When probing the output of the opto on a scope, the data had these very rounded tops vs the input which had a very pronounced square-ish appearance
I’m sorry for that… I tested it with 4 MIDI devices (JX8P, FA-66 interface, access virus and the MIDI output circuit of another Shruti-1 proto) and did not have any problem… The problem seems to be specific to those USB powered/controller keyboards. I don’t think there’s another optocupler you can swap in unfortunately - the ones which are likely to solve a weak signal problem are the active ones, which need a power rail and are able to amplify the signal - and the one used in the Shruti-1 is passive, so it doesn’t have the right pinout…
That might be the problem, unfortunately I don’t have an scope to test it.
pichenettes: don’t worry, the oxygen is a piece of crap, as most M-Audios, so it’s not really your fault and anyway I wouldn’t attach the Shruti directly on the Oxygen anyway. I have to chain all the hardware in a row (only have 2 Midi outs and have to drive 6 device, some of which use several midichannels). Guess I’ll have to buy a 4x4 Midi interface sooner or later…
@altitude: thanks for your advice. but i paid attention to the “midi to usb” or “midi out” issue. my midibox sid runs fine with the axiom. it uses a 6n138 opto coupler. my lsdj-mc2 gameboy midi interface also uses the 6n138 and there are no problems.
i remember that there also has been a strange behavior of the shruti while hooked up to ableton via one of my midi devices. i think it was the axiom in “midi out” mode. the shruti did not play every note. but the “midi out” mode works great with my midibox sid.
i will try to use a wall adapter for the axiom, later.
@pichenettes: don´t worry. problems are there to be solved
What I’ve put in the next synth is a 6N137, which is digital and has a darlington stage… I’m sad there’s this bad side of “early revisions” hardware - the good side is the raw sound and the use of a rare chip. Those are two things I’ll probably never do again. One bit of info related to that: the next synth has a compatibility board for the CEM3379 - this is what I used for testing/debugging the digital section before I attempted to design the analog filter section - so if you want to enjoy a sound close to the original Shruti-1 with all the revised stuff, it’ll be possible.
@BB: that was the next thing I was going to recommend.
pichenettes: that’s great news! I have a couple of friends that are quite envious of my Shruti-1 and would be really interested in a similar device!
I had the MIDI in issue as well with my Analog Factory Experience USB-Keyboard and Roland PCM-M1…
However, I have a working fix for the problem:
The optocoupler can be replaced with a CNY17-1. It is pin-compatible so just desolder the 4N28 and put in the CNY17-1.
My Shruti now works with the weak output of the above keyboards as well!
Thanks a lot for sharing this fix! Any particular characteristic of the CNY17-1 that makes it work better? What gave you the hint that it would work better?
actually I don’t really know why it works better…
I tried it because it had solved a similar problem for me some time ago, when I was experimenting with EZ-KIT and MIDI, so I thought I’d give it a shot…
Many people online claim that the CNY17-1 isn’t capable of doing MIDI at all, but it sure worked for me. My theory (which maybe total nonsense) is that the higher forward voltage of the diode in the CNY17-1 according to the datasheet, may help providing steeper edges with a signal already “rounded” or noisy…??
As you might have seen from the stuff detailed in this thread I’m pretty sure I need me a new optocoupler as well. I’m a bit new at electronics (although learning a lot from the Shruti-1!), so I just wanted to double check - I’m in the UK, so it looks like Farnell is my best bet for sourcing these parts. Is this the opto I’m looking for?
I confess to not understanding a lot of the technical specifications, but it looks like quite a different beast to the 4N28, which I assume is this one (although the picture is wrong)
Have I got the right product, and is it considered a “safe” replacement?
@ElectroDruid. It looks right but I’ll let TD confirm. There are two versions of the CNY17-1 - one with pin 6 connected and the other one not. I don’t know which one TD used.
Also, don’t bother looking at the images on the Farnell pages - sometimes they use generic ones and it looks wrong (on the 4N28 page it shows up as a 8 DIP package).
@TD - is this the CNY17-1 I’m looking for?
@pichenettes - Yeah, I saw that the image for the 4N28 was wrong, but I’m guessing that the actual part is the same thing that you shipped with the kit… What I’m thinking is that if the components are so cheap and I’m likely to have to pay a bit of postage anyway, I might as well buy a couple of 4N28s and a couple of CNY17-1s whilst I’m ordering, just because they’re nice things to have as spares.
might be worth trying different brands. Vishay, fairchild and liteon all make that part