protofuse Posted October 4, 2009 Report Share Posted October 4, 2009 as suggested by TK, I'm opening this thread in order to aggregate all the tests/troubleshot and solutions I found here and there.- rgb led matrix & common anode matrix & ISRsolved => code was ok, but static keyword involves "non loading" at startup- DIN problem, strange messages received in midiOXsolved => I listened midi link encapsulation, cause I didn't listen at the right place (wrong midi out..)- blinking leda BIG TK's answer not tested yet by me- false detection by 4051 analog muxthis post itself, see lower AND this post tooI'll update this thread as soon as I'd have solutions, questions etc etc.I hope it is ok for all the eventual helpers... ;) Quote Link to comment Share on other sites More sharing options...
protofuse Posted October 4, 2009 Author Report Share Posted October 4, 2009 My last problem deals about AIN/pots.Here is a schematic of the pots wiring of my protodeck (attached piece)I have 2 core.AIN are all AINx4 & are named like that: AINxy (x=core number, y=AINx4 number in the chain)AIN11 is filled with 32 pots.none problem with them ! nothing strange etc.AIN21 is like shown on the schematic:- SR1: 8 pots ok- SR2: 8 pots- SR3: 8 pots- SR4: 6 pots + 2 pins clamped to GNDAIN22 is like shown on the schematic:- SR1: 6 pots + 2 pins clamped to GND- SR2: 6 pots + 2 pins clamped to GND- SR3: 8 pots- SR4: 5 pots + 3 pins clamped to GNDas I have to scan all pots in the chain, my MIOS_AIN_NumberSet(AIN_NUMBER_INPUTS); goes with AIN_NUMBER_INPUTS defined to 64.strange things happens witj AIN21 and AIN22.if I "touch" SR with my finger, or by squeezing corresponding flat ribbon onto them, randomly value appears... on pins correspondingsometimes, when I turn pots on AIN22-SR1 for instance, 0 value is 0 (ok), I turn it ... value increase, but decrease inconsistently and become 0 instead of 7Fthe biggest problem is... it happens sometimes (I guess, when i touch flat ribbons connection to SRs) sometimes all seems very ok. but I need a more reliable deckso questions are:- does my wiring seems to be VERY strange? (flat ribbon for values, red/black 0,25mm wires)- what should I test first? in second ? I mean , a process to track that...I checked my midilink, it works very fine: I mean, pots value from core1 (from AIN11) are VERY stable, reliable etc.I checked 10 times my GND clamps... seems ok.any help would be very appreciated! Quote Link to comment Share on other sites More sharing options...
protofuse Posted October 4, 2009 Author Report Share Posted October 4, 2009 I have to add this.I read http://www.midibox.org/dokuwiki/doku.php?id=troubleshooting#how_can_i_avoid_pot_flickering_jitter, and http://www.midibox.org/users/tor_arne/midibox64_walkthrough/potsbuttons.html and other precious posts.My frontpanel & my box aren't connected to ground.could it really fix that?I mean, I can test it but, does the reason come from bad soldering/wiring? Quote Link to comment Share on other sites More sharing options...
protofuse Posted October 5, 2009 Author Report Share Posted October 5, 2009 apparently, groundind the case isn't a way...I'll recheck all my connections and just hope I won't have to REDO ALL THE POTS wiring ?!!!!nILS and ssp helped me : I would have to check the analog and digital cable proximity... near from AIN and core. and I guess it could be the problem.indeed, "moving cable analog" means "make cables closer from the SR, from the digital line" ..... it should be the cause!I'll test it asap ! Quote Link to comment Share on other sites More sharing options...
protofuse Posted October 6, 2009 Author Report Share Posted October 6, 2009 should I move all my pots cable to these ?http://fr.farnell.com/3m/3517-10/cable-en-nappe-10-voies-par-m/dp/8608571or maybe, only CORE<=>AIN cable..my problem comes only from this cause: proximity between wiggling analog cable from pots & digital cable between CORE<=>AIN Quote Link to comment Share on other sites More sharing options...
nebula Posted October 6, 2009 Report Share Posted October 6, 2009 interesting ... you could try shielding your ribbons yourself using aluminum foil, shrink tubing and a bare "drain wire" connected to ground at one end. it might be cheaper and faster. Quote Link to comment Share on other sites More sharing options...
protofuse Posted October 6, 2009 Author Report Share Posted October 6, 2009 hello nebula,already bought some shielded wire.I'm trying that asap!I'm a little bit despaired by all of that..I tried to do my best in all stepand this unstability problem gives me headacheit is only with (some) pots, I mean: some ain.leds, buttons are very ok ... indeed, it is positive :) Quote Link to comment Share on other sites More sharing options...
noofny Posted October 7, 2009 Report Share Posted October 7, 2009 Here I come again with my obvious general questions! ;)Isolate the issue is your best friend. If you swap the connections from AIN->CORE around does the same problem happen on both cores? This is basically isolating the problem to be a core or ain issue. Then I would suggest if you can, swapping pots between AIN's - this isolates the problem to a single AIN module. Also I would try swapping the SR's around to see if the problem may be a flaky SR (not likely tho). Did you solder the filter caps onto the AIN's ? Some people like myself don't seem to need them but others do, but it can affect things.And don't forget to power down any nearby Deathstars or particle accelerators whilst testing MidiBox... (sorry) 8) Quote Link to comment Share on other sites More sharing options...
protofuse Posted October 8, 2009 Author Report Share Posted October 8, 2009 I changed all the AIN<>CORE cable by shielded one.nothing has really changed.I found one of the 4051 was freaky dead ! easy to isolate. I had doubt so I putted it at the end of the AIN chain and put the value of pots in main.h to 56 in order to let it for later... so now I 'm sure it is a dead component (or almost dead)my PSU seems to be noisy (it is a switchmode PSU)the proximity between pots wire (ribbon between pots and AIN) and PSU gives me big flow of false detections.so I had around 2 problem, but one consequence: freaky flow of false detections.indeed, by puttin the cable far enough from psu, I decrease the data flow false detection by an amount of 10000. really.after that, I found sometimes little false detections, but only between very little range of values. I guess DEADBAND parameter could correct a little bit this. but it isn't made for it, I guess (too)so, one big advice:- buy good PSU as less noisy as possible. (switchmode means noisy, too)- put a big distance between pots wires and PSUtomorrow, I'll try to put a couple of capacitor between 5v and GND at the main supply of the protodeck.we'll see. but I guess my problem is found and almost solved (at least in my head!)big thanks to nils who didn't laught at me too much with my questions and tests ;) Quote Link to comment Share on other sites More sharing options...
protofuse Posted October 8, 2009 Author Report Share Posted October 8, 2009 a nice site that clearly explains bypass capacitor concept:http://www.seattlerobotics.org/Encoder/jun97/basics.html Quote Link to comment Share on other sites More sharing options...
protofuse Posted November 9, 2009 Author Report Share Posted November 9, 2009 (edited) everything works fine now with protodeck.:frantics:=> the famous & damn it is working dance when I'll have higher quality pictures etc, I'll try to be midibox of the week :) I'll apply for that Edited November 9, 2009 by protofuse Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.