Sending SacnMatrix to AIN

[henrygr]

I know. The idea of it- so shoot me. But, to be honest, I am under pressure, and don't have time for UPS to get me DOUT modules grom SmashTV. It's a nine by nine matrix (yes,once agian, Hammond drawbars). Very much discussed on the forum, but with no real conclusion, except for the implemtation of the scan matrix http://www.midibox.org/forum/index.php?topic=2755.0 here.

Here is what I have done thus far. I have split the bus-bars into 81 seperate switches and am happy with the result (only took about three hours, so not so bad.), and have also rendered the switching system "break before make". I have absolutely no idea of the sense that this topic makeshttp://www.midibox.org/forum/index.php?topic=2754.0, about a "switched resistor scheme". I have never seen the schematic, but on reading through it a couple of times, I was left with the impression that all information would be mixed as each wiper shares a common ground.

Before I go on, I can still revert to the original Hammond layout of the bus-bars, so if someone can clear up the second topic I posted, I will go ahead that way.

Otherwise, I will continue with the 81 seperate switches. As they are now break before make, I am playing with a model of crossing the ground via a 7.5K resistor to the AIN to eliminte jitter. At the wiper/busbar contact section, I have run a series of nine resistors from 100R to 10K from switches one to nine. The theory is that a value will be returned based on the differential of voltage between the Vs and the Vd and translated into MIDI. Its a bit messy. But I will keep at it (I am doing it drawbar by drawbar (wiper by wiper) for the sake of quality control).

Any tips, suggestions or obscenities would be appreciated.

WARNING-Anyone else thinking of doing it, don't- it's a last resort from the topic of the first link!!!

[stryd_one]

I've read of this being done by using resistors all of the same value, and at each stop, all the previous resistors are switched into the bus, so the resistance is accumulated, and it works like a pot... Your way should be basically the same in effect, and the digital approach is just as valid... I reckon it just depends on the particular drawbar model you've got, and your level of programming experience, as to which method is most appropriate.

Given that you are actually able to choose whichever of these methods you'd prefer, it comes down to a question of whether the extra work is in software or hardware, by coding the scanmatrix handler or by soldering resistors. I think you already made that decision :)

My tip, suggestion or obscenity would be : Rock the f*@^ on, and don't f$^&in forget to put it in the f!)% wiki  ;D

[audiocommander]

haha! I knew I would find that link again:

http://www.eecs.tufts.edu/~dsculley/RCX/

I hope this helps ;)

Cheers,

Michael

ps: storing my bookmarks on my own sitebar server. Pretty cool if you have more than one browser/computer :D

[henrygr]

Just a quick question. When one leaves an AIN pin ungrounded, and then touches a ground to it during all it confusing jitter, should the final value be "0"?

[henrygr]

Thanks for both your replies.

And an update....

The idea is fairly sound, and works only OK for the moment. The big difficulty is the action of the drawbars, and the atypical American Over-enginerring that does not leave much room for modifications. It's like everything was designed for a task and that task only, whereas with european clasic keyboards, parts were borrowed or reused from other models (for the purpose of budget one can summise), making spare parts both easier to source, and also easier to find information on. All I can compare working with it is to filing the frets on a guitar, where each string has a seperate fret!!

I opted not to do the voltage divider method. The main advantages of it was that the drawbars could be left as make before break and there would be no need for additional grounding of the contacts, as they are never in a state of flux. It is in fact the best way to go. But in considerstion a break before make system had the advantage of being compatable wth pre 1954 model Hammonds (up till then all systems were break before make).

I may change my mind yet. But the short and narrow of it is that there is certainly more than one way to skin this cat. He'd better bloody taste good, that's all i'll say.....

stryd_one, would love to do a Wiki on all things organ, and try to amass all the information here into one concise article. Perhaps over the year I will do this. For the meantime though, got to finish this keyboard as the clock is ticking, and I'm back on the road fairly shortly!!

Hammond fanatics, this link is to the most comprehensive guide I have yet to see on the net about the Hammond. http://theatreorgans.com/hammond/faq/files/faq.txt

Will keep ye all posted,

Mark.

[stryd_one]

Yep, toys first, stories later. Forget the wiki until you've had your fun ;)

[henrygr]

Just realised that I posted this whole thing in the wrong area of te forum. Apologies. :o

[henrygr]

This is working. Not rocket science, bur not the easiest thing in the world to build either Will need to put a cross cap between the Vs and Vd close to the drawbar mechanism to cut down on superflous spillage (old machinery, old contacts). Any suggestion for a value?

[henrygr]

OK. Take it from me and lots of sweat. DO NOT, NOT EVER, TRY TO DO WHAT I HAVE ATTEMPTED.

The drawbars wont take to the modified enclosure, It's like when you fix one error, an other one arrives. The solution is Scan Matrix. I got my DOUT module in the post today, so off I pop.

[stryd_one]

:(

Sorry to hear it, and thx for the advice man. Hope the matrix treats you better!

[henrygr]

I think I have you roped in stryd_one......... ;D

Please. More pity........he he.