NorthernLightX Posted June 27, 2007 Report Share Posted June 27, 2007 Hi all,strangeness: when I try to run MBSID V2 beta (8580 version) or MBSID V1.74685 (8580 w/o CS version) with my CrystalFontz 20x4 LCD attached but no further Controls attached, MIOS keeps rebooting. When loading the SID testtone V1 app it works fine (produces sound and displays on the LCD). LCD interconnections also appear fine because the text on the LCD is always loud and clear. Also, MIOS is stable (displays READY.) if I upload it and then do not load an app. This is what I see with MBSIDV2 loaded:<Blocks on line 1 and 3>------MIOS V1.9e© 2007 T.Klose------MIDIboxSID V2 BETA1Lauching CS------<reboot>If I leave the LCD disconnected, the app initializes and I can play with the MIDI keyboard in MIOS Studio. Strangeness ???I also tried MBSID V1.74685, then the app message is slightly different of course:------MIDIboxSID V1.74685CS not enabled!------Slight difference is I can't get the SID to make any sound with V1.74685 with LCD not attached, and in the the MIOS Studio MIDI monitor I keep getting Sysex message F0 00 00 7E 40 00 01 F7 on regular intervals, which to my knowlegde indicates a reboot.I have one 1 Core attached to 1 SID, both boards in V2 revision. I also have LTC connected, no MAX stuffed of course. LTC is only used for LEDs right now, no MIDI ports attached. I use my homebrew PSU. I have the LCD connected with a cable with only 12 wires, the 4 spare pins on J15 of the core are left blank. The spare pins on the LCD are connected to ground.I am completely baffled. What can be the problem? Quote Link to comment Share on other sites More sharing options...
MTE Posted June 27, 2007 Report Share Posted June 27, 2007 Maybe wrong LCD-Wiring....did you connect the D0-D3 at the LCD to the ground (VSS) of the LCD ? Quote Link to comment Share on other sites More sharing options...
NorthernLightX Posted June 27, 2007 Author Report Share Posted June 27, 2007 Maybe wrong LCD-Wiring....did you connect the D0-D3 at the LCD to the ground (VSS) of the LCD ?yes I did, i connected pin 7 through 10 to pin 1 of the LCD. I don't think the wiring is the problem because the LCD cable is soldered to the Core, and there is a female SIL connector to the end. I soldered a SIL header to my LCD so I could swap it if necessary. As I stated, It does work with the testtone app and there are no weird characters or anything. Quote Link to comment Share on other sites More sharing options...
TK. Posted June 27, 2007 Report Share Posted June 27, 2007 I would need some more background informations.For example: do you know, why the testtone application was not working yesterday? How do you upload the application? With the latest MIOS Studio release v7.5?Are there any error messages during the upload?And an additional test could be interesting: how does the application behave, when the unused data pins D0-D3 are not connected to ground, but left open?Best Regards, Thorsten. Quote Link to comment Share on other sites More sharing options...
NorthernLightX Posted June 27, 2007 Author Report Share Posted June 27, 2007 I would need some more background informations.For example: do you know, why the testtone application was not working yesterday? How do you upload the application? With the latest MIOS Studio release v7.5?Are there any error messages during the upload?And an additional test could be interesting: how does the application behave, when the unused data pins D0-D3 are not connected to ground, but left open?Best Regards, Thorsten.Hi,I cannot be sure why the testtone was not working yesterday, but I did re-upload MIOS 1.9e a few times today. Maybe my MIOS upload was corrupted yesterday. Also, there might have been a problem with the audio connector yesterday.I use MIOS studio 7.5, downloaded it a few days ago. Had some trouble getting JAR file to run but succeeded at last. My MIDI interface is a Phonic Firefly 302 which I bought also a few days ago.During the upload, if I do a clean MIOS and the app thereafter, there are no error messages. I have had error messages in the past while uploading an application when I did not do a clean MIOS reinstall. I also have seen a message "this error is expected, please ignore" in the past, but not during my last tests with clean MIOS as decribed.Do you mean the datapins on the LCD or on the Core? On the Core, they're open, on the LCD there grounded right now.Could my problem with the SIDV2 application have anything to do with the ControlSurface which I don't have yet? There are special versions of the application for 1.7xxx when CS is absent.I will also test with a different LCD tomorrow. Quote Link to comment Share on other sites More sharing options...
TK. Posted June 27, 2007 Report Share Posted June 27, 2007 The message "this error is expected, please ignore" should appear whenever you are trying to upload MIOS. It's a failsave mechanism. Just ignore it (as it says)Don't know why a MIOS update should be required. With "in the past" - do you mean in the last days with the new PICs, or weeks ago? So long there isn't an error message, the code upload should be fine.datapins: I mean the data pins of the LCD which are currently clamped to ground (or vss, or Vs, or 0V - it's all the same).It would be interesting for me, what happens, if the unusued LCD data pins are open instead - this is just a test which is important to gain experiences with different LCDsD0-D3 of the core are not connected. Later you will use D2 and D3 of the core for the CAN Rx/Tx lines, but I guess you've already connected at least the 1k Pull-Up resistor between D3 and Vdd like shown in http://www.ucapps.de/midibox_sid/mbsid_v2_communication.pdf ?If not I cannot guarantee for anything - don't know how ECAN behaves if it receives random data values.I'm also not sure what happens when no control surface is connected. It's unfortunately impossible for me to test this, as I'm using Wilba's PCB. An unconnected DIN shouldn't be a problem, as nobody has reported an issue in the past. It is important of course, that the 10k pull-up (R9) at J9:SI of the core module is connected, but I guess your core module is (hopefully) completely stuffed? I mean: even the two pull-up resistors for the IIC bus? (I just mention this, as somebody else forgot this and missed to mention this detail...)Best Regards, Thorsten. Quote Link to comment Share on other sites More sharing options...
NorthernLightX Posted June 28, 2007 Author Report Share Posted June 28, 2007 I use a fully stuffed V2 Core, if the pull-up resitors were in the schematic back then, they are mounted now. There is no bankstick connected yet.I have nothing connected to the CAN bus as I am only using one Core a the moment. Should I solder the resistor and the diode and connect them both to +5v?LCD-wise, will it suffice to disconnect the ground-wire from the 4 LCD pins, or do they need to be disconnected from eachother too?With in the past I meant the last few days, a week at most.What will suffice as a control surface? Do I need the Full monty like Wilba's, or are a few buttons and an encoder enough? I would like to do the tests for you but I am nowhere near designing my CS yet, wanted to wait for the final version to be released. Will post the testresults from my last post toningt.Thanks, Alex. Quote Link to comment Share on other sites More sharing options...
Wilba Posted June 28, 2007 Report Share Posted June 28, 2007 I would suggest first connecting the diode and pullup resistor, as per diagram, as if you were going to connect to other PICs, this would be my prime candidate for "the thing which is different in your setup, and thus why yours doesn't work" ;DAlso, I noticed my displays preferred the four unused data pins to be open (not grounded) to make 4-bit mode work, but then I also observed the status byte being returned with nibbles reversed, so I had to abandon 4-bit mode and connect them with 8-bit data. This is probably irrelevant to your problem. If communication with the display goes into timeout, then it just won't be used anymore (display "freezes"), but the sound engine and control surface (buttons/LEDs) continue to work. Quote Link to comment Share on other sites More sharing options...
Jegge Posted June 28, 2007 Report Share Posted June 28, 2007 Hi, I also experienced something similiar: Never ending reboots and sysex messages (only with version V1.74685, I did not try out V2 up to now, V1.7303b worked more or less). The mistake I made was leaving out the 100N caps located on the bottom side of the LTC module. After having soldered these in, it suddenly worked.I suppose that, the more peripherials are connected, the more likely a missing cap will lead to errors (read: corrupt signals), so I guess this triggered some kind of watchdog... CU Jegge Quote Link to comment Share on other sites More sharing options...
NorthernLightX Posted June 28, 2007 Author Report Share Posted June 28, 2007 Quick reply: I just tested the LCD with the 4 unused datapins still clamped together but not clamped to Vss and the behaviour is the same. I also checked another LCD (powertip 16x2) in both confugurations with the same results.@Jegge: all 100nF caps are in place on the bottom of the PCBs.I am now going to solder the CAN bus things, hope it helps :) Quote Link to comment Share on other sites More sharing options...
NorthernLightX Posted June 28, 2007 Author Report Share Posted June 28, 2007 Soldered the components for the CAN bus (1K resistor and 1N4148 diode) together, now MIOS doesn't boot at all, only the blocks appear on the LCD ??? Quote Link to comment Share on other sites More sharing options...
TK. Posted June 28, 2007 Report Share Posted June 28, 2007 Are you sure, that this isn't a general problem with your core module? A bad connection, or a short somewhere?Do you still see the upload request? Are you still able to download another application?Best Regards, Thorsten. Quote Link to comment Share on other sites More sharing options...
NorthernLightX Posted June 28, 2007 Author Report Share Posted June 28, 2007 Oops, big fault from my side, I accidentally soldered the wire to GND (I have the 5v wires for CAN and SID on the same pin, I had to desolder the 5v for the SID first, then made the mistake of choosing the wrong pin to resolder the wires to.I also made a 2 pin SIL connector to the CAN wires for testing, when the CAN wires are connected the Core doesn't reboot anymore :DBut instead launching CS stays on screen. :-\I will upload the new BETA2 application now, to see if that helps. I will also record any Sysex messages and post them.So far, thanks for your help and patience TK ;)I hope I'm not wasting too much of your time. Quote Link to comment Share on other sites More sharing options...
NorthernLightX Posted June 28, 2007 Author Report Share Posted June 28, 2007 I connected the midibox to MIOS studio and I am able to play with the midikeyboard again, so I think this is as far as I got without LCD connected.No Sysex messages exept the upload request at boot.I think the problem is solved.maybe a FAQ marker?"When you connect an LCD, the CAN bus has to be connected too"I am happy again, and am going for the Step A now....new problem topic soon maybe, considering my skills to create problems ;DP.S.forgot to mention that this was all done with Beta1 loaded, I don't think it makes any difference though. Quote Link to comment Share on other sites More sharing options...
TK. Posted June 28, 2007 Report Share Posted June 28, 2007 Great! :)Yes, of course, this input will go into the hardware discussion page that I'm planning to write this weekend.I wasn't aware that ECAN behaves so sensible to an unconnected receive input. However, thats the purpose of the beta test - to find out and document all possible issues.Best Regards, Thorsten. Quote Link to comment Share on other sites More sharing options...
NorthernLightX Posted June 28, 2007 Author Report Share Posted June 28, 2007 Wheee, I'm an official MBSIDV2 Beta tester 8)I think I'm going to order a T-shirt with that tekst ;DOne last question before I continue: I'm considering making a very minimal frontpanel with 2x16 LCD, 4 buttons and 1 encoder. I saw this is supported for SID V1, are there any reasons (in term of functionality) not to do this for V2? I do not plan to make this box either stereo or multi-core, just one core and sid for jamming and testing, for the big works I'll wait for wilba's PCB. Quote Link to comment Share on other sites More sharing options...
TK. Posted June 28, 2007 Report Share Posted June 28, 2007 The SHIFT button is still important, because it's the only way to switch the main encoder to FAST mode (to speed up parameter changes), to change the engine, to init a patch and to send a SysEx dump.I know, there could be an alternative solution, e.g. a well hidden menu page, or a special key combination, just to save this additional button. Just today I spent again thoughts for a way to replace the SHIFT button, but without a conclusion which fits all needs. I'm really sorry for that, but so long I don't see a clear way which is future proven (e.g. the button could get some additional functions in bassline and drum mode), I can only repeat what I'm writing since weeks: please add this button!Best Regards, Thorsten. Quote Link to comment Share on other sites More sharing options...
NorthernLightX Posted June 28, 2007 Author Report Share Posted June 28, 2007 No problem, I will add the button, if I use a DINX1 there is 1 spare pin anyway. If I want to connect the extra encoders I'll have to add a second DIN, but i'd want to make LEDrings for them too so that would require DOUTs as well. I think veroboard will be the way to go for this one.Cool, I'm going to fetch me some buttons this weekend :-* Quote Link to comment Share on other sites More sharing options...
stryd_one Posted June 28, 2007 Report Share Posted June 28, 2007 "I beta tested MBSIDv2 and all I got was this lousy TShirt"(and on the back)"And this bassline, biatch" Quote Link to comment Share on other sites More sharing options...
Wilba Posted June 29, 2007 Report Share Posted June 29, 2007 http://cgi.ebay.com.au/New-8580-Sid-thing-Commodore-64-C64-C128-Amiga-Cap-Hat_W0QQitemZ180134666442QQihZ008 Quote Link to comment Share on other sites More sharing options...
TK. Posted June 29, 2007 Report Share Posted June 29, 2007 but i'd want to make LEDrings for them tooLED Rings are not supported, and it is impossible to add this to the firmware due to conceptional reasonsBest Regards, Thorsten. Quote Link to comment Share on other sites More sharing options...
NorthernLightX Posted June 29, 2007 Author Report Share Posted June 29, 2007 LED Rings are not supported, and it is impossible to add this to the firmware due to conceptional reasonsWhoops, thanks for pointing that out. Makes te project a lot simpler too :) 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.