tim_pugh Posted March 15, 2011 Report Share Posted March 15, 2011 Hi, After a lengthy hiatus I've got round to finishing the construction of my mb6582 and have a few problems. I have 6 sids plugged in, 2 each into cores 1, 2 & 3. I have uploaded mb6582 hex directly to them all by switching jumper on J11 (had a problem cloning but this is now working fine). I can get sound out of the 3 SID outputs. After going to the manual to learn how to play it, it doesn't seem to be acting right which either means I'm not using it right or it needs troubleshooting. When i boot up I get E002|PInt Ld Chn.1 Sid 1|Lead Patch. This looks fine. Then when I press the various 5 buttons following the manual I get odd results. Pressing the left button to get to ensemble root takes me to a screen with Envelope E1 Settings & switches the SID engine to sid 4. I press the menu and it takes me to the Lead Engine page, but when I choose any of the options it seems to take me to places which don't match up. I press menu again and it takes me back to the main page. I think I have some bad wiring somewhere but have no clue where to start to look. Many other seemingly odd things happen when I press certain buttons. Can anyone offer some hints on where to start to look to troubleshoot. I am a novice at electronics by the way and completely new to using midibox/mios/mbsid, hence my poor description of the problem. I would be very grateful for any help at all as I am a bit flummoxed. Thanks Tim Quote Link to comment Share on other sites More sharing options...
Wilba Posted March 16, 2011 Report Share Posted March 16, 2011 You may have shorts on the far left DOUT module (74HC595) causing the switch matrix to not work, buttons being pressed are thought to be other buttons being pressed. So check there first, then check solder joints on all the 74HC165 and check orientation of the resistor networks. Quote Link to comment Share on other sites More sharing options...
tim_pugh Posted April 5, 2011 Author Report Share Posted April 5, 2011 You may have shorts on the far left DOUT module (74HC595) causing the switch matrix to not work, buttons being pressed are thought to be other buttons being pressed. So check there first, then check solder joints on all the 74HC165 and check orientation of the resistor networks. Hi Wilba Thanks for the reply, sorry for my late response. I posted then had to move house so have no internet access at the moment. I will have a look at your suggestions. Sorry to be ignorant, but how would I go about checking orientation of resistor networks ? I'm fairly new to electronics. Office is in dissaray so won't be able to look at for a while unfourtunately. Thanks again Tim Quote Link to comment Share on other sites More sharing options...
Janis1279 Posted April 5, 2011 Report Share Posted April 5, 2011 Hi About resistor networks right orientation you can read in the point 12 : These are 5 or 6 pin common bus resistor networks, in which one pin is “common†and is connected to all the other pins via resistors. The dot indicates pin 1 (the common pin), which goes in the square pad (left-side) directly above/below the component identifier (eg. “R31â€). Do not get a 6-pin resistor network which is three isolated resistors connected to three pairs of pins. The correct one is Mouser part 264-10K-RC. If buying elsewhere, I believe if given a choice between parts with an “A†or “Bâ€, the “A†means “common bus†and “B†means “isolatedâ€. (See here: http://au.mouser.com/catalog/636/619.pdf) From the MB6582 parts list : http://www.midibox.org/dokuwiki/doku.php?id=wilba_mb_6582_parts_list It's easy to check the resistor network's common pin orientation with the multimeter , too . In the 20Kohm measuring scale. Regards, Janis Quote Link to comment Share on other sites More sharing options...
orange_hand Posted June 11, 2011 Report Share Posted June 11, 2011 Hi, I have got exactly the same issue. Some buttons don't react as expected. Especially the menu button jumps into different other sub menus ... very annoying ... Here is a picture of my CS PCB: Also I have recognised that in the Bassline mode while a sequence is played the LED's 4 to 6 glow always together, rather than separate. Before I soldered the components to the CS PCB I tested all LED's, diods and tact switches. After I noticed the fault, I check the following things: - orientation of the diods and LED's -> all as expected - the function of the diods and LED's -> all ok - the cables from the base pcb to the CS pcb -> all ok - all soldering points -> no shorts To exclude a fault in the base PCB I connected the control surface to another MB6582, and it behaved in the same faulty way. Therefore I can exclude that something on my base PCB is faulty. I installed on all PIC's the MB6582 setup, without any errors. I don't know what else I can check ... Any help or recommendation is appreciated ! Thanks orange Quote Link to comment Share on other sites More sharing options...
Hawkeye Posted June 11, 2011 Report Share Posted June 11, 2011 Hi, Could you post higher-res pictures of the front and backside of the CS PCB? Best regards, Peter Quote Link to comment Share on other sites More sharing options...
Hawkeye Posted June 11, 2011 Report Share Posted June 11, 2011 (edited) Have you insulated the LCD display as described in the photo tutorial? If not, it can cause shorts on the CS. Edited June 11, 2011 by Hawkeye Quote Link to comment Share on other sites More sharing options...
orange_hand Posted June 12, 2011 Report Share Posted June 12, 2011 Hi, I have insulated the Display of course and actually I use some washers in between, because I need some space for the Window. Please find attached some pictures of the backside of the CS: Please note that the screws on the board don't create a short. I have removed them prior to the testing and they also have a nylon wisher underneath: I also think the encoders work fine and modify the expected parameters... Do I need to remove the IC's close to the network resistors when I measure them ? I will use a 20K scale to measure them. Cheers orange Quote Link to comment Share on other sites More sharing options...
Hawkeye Posted June 12, 2011 Report Share Posted June 12, 2011 (edited) Mhm, its hard to tell, maybe some other folks have opinions on this... in the third picture, some joints do not seem to be soldered/tinned... Also for stability reasons, i would solder the stabilizers of the rotary encoders to the CS PCB. You said that you have tested this CS with another baseboard, that works fine with another CS, therefore the problem must be on this CS (Sherlock Holmes :-)), not the resistor networks (which i assumed in last mail). You can always try cleaning flux with alcohol and a toothbrush, sometimes it hides very tiny bridges. Best regards, Peter Edited June 12, 2011 by Hawkeye Quote Link to comment Share on other sites More sharing options...
orange_hand Posted June 12, 2011 Report Share Posted June 12, 2011 Hi Peter, I have re-soldered a couple of joints, but still the same issue. I wanted to measure the network resistors and connected them in the following way: 1 to 2 -> 4,14 K 1 to 3 -> 4,14 K 1 to 4 -> 4,14 K 1 to 5 -> 4,14 K 1 to 6 -> 9,91 K This is just one example (R39). All IC's are still in their sockets and I am not sure if this makes a difference for the measurement. And yes, there could also be some issues on the base PCB, as I couldn't test the other cs on my board, as the plugs are different... Cheers orange Quote Link to comment Share on other sites More sharing options...
Hawkeye Posted June 12, 2011 Report Share Posted June 12, 2011 (edited) Hmm.. interesting, when I measure R30 (can´t reach any other with my probe pins), I´ve got 10K between pins 1-5, also 10K between 2 and 3 for example and 20K between first and last pin, with ICs still in socket and CS not connected... Did you use the resistor networks included in SmashTVs base kit package or buy your own? Let´s wait for some more enlightened folks to give their opinion :-) Edited June 12, 2011 by Hawkeye Quote Link to comment Share on other sites More sharing options...
orange_hand Posted June 12, 2011 Report Share Posted June 12, 2011 Hi, yes, I used the once from smashTV. These are the only resistors that I used without checking them before I soldered them on the board, all the others were checked beforehand. It is difficult to describe the behaviour of the CS, but I have noticed that: a) the menu button does sometimes reacts like the SID 1/2 switch in the right top corner b) the function button 4 sometimes reacts like menu button c) when I run a bassline using the sequencer the LED 4,5 and 6 light up together, whilst all the other light up separately. When I switch to another row, the same thing happens d) the waveform (osc section left, middle position) -> the pulse waveform LED does never light, the switch does the right thing. e) the meter switch works fine, but the meter LED does not light up. : There are so many functions not working properly and I can't believe that I have done so many things wrong !??!?! I used a 10x loupe and checked all joints, there is no short to identify. Cheers orange Quote Link to comment Share on other sites More sharing options...
orange_hand Posted June 12, 2011 Report Share Posted June 12, 2011 (edited) Hi, I have checked issue c), the 3 LED's that always light up at the same time. I have measured a short on the CS PCB on pin 4 to 6 of JD8. I wonder how this is possible, especially as the related tracks directly go from the LED's column to the JD8 connector ? I am pretty sure this is the reason why the LED's 4 to 6 light up together... cheers orange Edited June 12, 2011 by orange_hand Quote Link to comment Share on other sites More sharing options...
Hawkeye Posted June 12, 2011 Report Share Posted June 12, 2011 (edited) a few more things... * regarding c) does that short occur with the baseboard connected? if so, can you disconnect the baseboard and see if the short is still there? * regarding e) the meter depends on the patch, it is normal for some patches that nothing is displayed * how many SIDs do you have on the baseboard? are you using a standard c64 psu? could you re-check the voltages with everything connected? perhaps there is a voltage drop that causes strange behaviour... * have you checked for correct placement and orientation of the 74HC595s vs the 74HC165s? * have you checked the correct orientation of the resistor networks? best regards, Peter Edited June 12, 2011 by Hawkeye Quote Link to comment Share on other sites More sharing options...
orange_hand Posted June 12, 2011 Report Share Posted June 12, 2011 Hi Peter, no, the short is clearly on the CS PCB. The network resistors are correctly oriented. I run 2 8580er SID's, the chips are in the right socket with the correct orientation. I don't think that there are voltage drops... First I have to find the short on the CS PCB causing the 3 LED's lighting up together. Is there a documentation which pad (JD1 to JD8) servers which purpose ? It is difficult to follow the tracks on the boards as they use both sides. Cheers orange Quote Link to comment Share on other sites More sharing options...
Janis1279 Posted June 12, 2011 Report Share Posted June 12, 2011 (edited) Hi, From the Wiki pages "Control Surface Wiring These describe the wiring of the control surface PCB. They can be used as a guide for constructing your own control surface. You can freely assign LEDs and switches to the matrix, but for the best performance in the firmware, the Modulation Matrix LEDs should be wired with common cathodes in vertical columns, so that the LED matrix will illuminate one column of 8 LEDs each shift register update. LEDs and switches marked with an asterisk (*) are V2-specific features or just things I added for my own use. Refer to http://www.ucapps.de..._manual_fp.html for more information. BTW I just noticed my CS DOUT wiring PDF is wrong… “LFO Sin†and “LFO Tri†are swapped and “Syncâ€, “CCâ€, “Editâ€, “Play†are in the wrong order (should be “Playâ€, “Editâ€, “CCâ€, “Syncâ€) EDIT 2 - The menu button should be at the intersection of “JD8-D3 & JD5-D3†NOT “JD8-D3 & JD5-D4†http://www.midibox.o...dout_wiring.pdf http://www.midibox.o..._din_wiring.pdf The rotary encoders are connected directly to the first four DIN modules on the base PCB (i.e. JD1 to JD4) as per the normal way of connecting rotary encoders to DIN modules. Just like the LEDs and switches, you can assign them to any pins of DIN modules and configure this in the table in the .asm file. http://www.midibox.o...der_assignments " Regards, Janis Edited June 12, 2011 by Janis1279 Quote Link to comment Share on other sites More sharing options...
orange_hand Posted June 13, 2011 Report Share Posted June 13, 2011 Thanks Janis, The material makes it more transparent to me. I hope I can find the faulty spot. Apart from that my family is already pissed off, because I spent so much time within the last weeks to build the MB6582.... Anyway, I keep you in the loop. Cheers Matt Quote Link to comment Share on other sites More sharing options...
orange_hand Posted June 14, 2011 Report Share Posted June 14, 2011 (edited) Hi, I did some more troubleshooting and figured out that there is a mistake in the din_wiring.pdf (for the switches) or my CS PCB (from SmashTV) is different. a) The menu switch is connected to JD5-D3 instead of JD5-D4 b) The "filter mode" and "filter osc" switches are swapped Also I found a fault on the PCB, which causes the "SID L/R" and "LFO Select" switch not to work. Now I have to find the short on JD8-D3 to JD8-D5, which won't be easy... I would appreciate if somebody could confirm to whether the din_wiring.pdf is not correct or my CS PCB. I also wonder if somebody could post the link to the latest DOut wiring documentation. The one on UCAPS doesn't match the CS PCB from SmashTV, or I got the wrong doc .... Cheers orange Edited June 14, 2011 by orange_hand Quote Link to comment Share on other sites More sharing options...
orange_hand Posted June 16, 2011 Report Share Posted June 16, 2011 Hi, I have now spent more time to find the fault on the CS PCB, as I needed to solder it. I could manage to fix the faulty interconnection between the front and backside of the board. But I can't find the short on JD8-D3 to JD8-D5, I spent hours with a jewlery's loupe to find something but without success. I am pretty sure that it is a fault somewhere on the board, as I was very precise when I soldered the components on the board. You can't believe how pissed off I am... I have now decided to throw this board away, as I can't manage to find the fault. I will order another CS PCB and I hope there are now issues on the PCB.... What a shit, all the lost time.... Cheers orange Quote Link to comment Share on other sites More sharing options...
Hawkeye Posted June 16, 2011 Report Share Posted June 16, 2011 (edited) Mhm, now this does not sound good... aren´t there any MB6582 owners in your area, maybe someone willing to test your board before you flush it down the toilet? A second pair of eyes sometimes helps... edit: can´t you selectively lighten up the problematic leds without the baseboard attached, for debugging, with an external power source (no 9v block :-))? like this you could "close in" on the short... Edited June 16, 2011 by Hawkeye Quote Link to comment Share on other sites More sharing options...
orange_hand Posted June 16, 2011 Report Share Posted June 16, 2011 Hi Peter, I have already done this. The only thing I will see with this test is that always 3 LED's in the row light up together. This is because there is a short on the GND side on JD8-D3 to D5, but as you know the GND tracks are all over the board. There must be a spot somewhere where these 3 tracks have got a short.... Cheers orange Quote Link to comment Share on other sites More sharing options...
Janis1279 Posted June 16, 2011 Report Share Posted June 16, 2011 Hi, I found the some mistakes and the some differencies in the names of some functions if comparing pdfs with the .asm file, too. You are able to measure and compare similar schematic parts of the CS board with your multimeter too ( in the beep mode, diode measuring mode without powering CS board). Regards, Janis Quote Link to comment Share on other sites More sharing options...
orange_hand Posted June 16, 2011 Report Share Posted June 16, 2011 Hi Janis, yes, that's exactly the way I test the board at the moment. The CS board is disconnected from the base PCB and I used the diode test mode of my multimeter. I found out that just 2 switches are not working (SID L/R and LFO Select). Based on this information I found one faulty connector that I fixed. You know the short at JD8-D3 to D5 has the following effect. If you press one of the buttons which is connected to JD8-D3 to D5 then the firmware is not able to distinguish which one that was. It will take one of three .... In practise: if you press for example select 1, the firmware could interpret it as "shift" as they are on the same connector (JD5-D0). Or the firmware could determine that it was "up" as this button is at the same connector JD5-D0).... The only way I see to find the short is to cut the tracks on the board one after the other in order to find the track where the short is. It could be one spot for all three tracks or 2 different spots.... I have already spent so much time, the whole thing looks great but doesn't work properly.... I am short before smashing the whole thing against the wall. You can not imagine how aggressive this makes me .... Cheers orange Quote Link to comment Share on other sites More sharing options...
Janis1279 Posted June 18, 2011 Report Share Posted June 18, 2011 Hi, Are you know these links in the Wiki : http://www.mb6582.or...PCB_R2_Mono.pdf http://www.mb6582.org/plans/MB-6582_CS_PCB_R2_Color.pdf http://www.mb6582.org/plans/MB-6582_Base_PCB_R2_Mono.pdf http://www.mb6582.org/plans/MB-6582_Base_PCB_R2_Color.pdf p.s. I have not the CS board from Smash ( I slowly made my own ), but many people are succesfuly soldered and debuged it. Regards, Janis Quote Link to comment Share on other sites More sharing options...
orange_hand Posted June 18, 2011 Report Share Posted June 18, 2011 Hi Janis, thanks for the links, I hope this helps... Cheers orange 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.