banjobater

Thank you Thorsten. That was the key to the problem. I have spent a lot of time trouble-shooting my core when it was actually working fine. :-[ Looking on the bright side it is working now and I am more familiar with the design of the core and the associated software than I would otherwise have been. :) I'm pretty sure this will be a good thing in the long run. As discussed with stryd_one, when I finish this project I am planning to put together a 'dummies guide' to MIDIO128 in order to document some tips and the pitfalls a complete newbie can avoid. I will run it by him for accuracy before posting it. Anyway the bottom line is that as a result of your post I spent the last half hour watching a row of LED's dancing around to the merry tune of 'Morrison's Jig'. This has been my best MIDIBox day. ;D ;D ;D Thanks again Thorsten.

Thanks Thorsten. It's good news that the bidirectional connection is working. I have already re-uploaded the mios_v1_9f_pic18f452.hex file from the mios_update_v1_9f[1]\mios_update_v1_9f\pic18f452\midi directory and I get the "MIOS V1.9f © 2007 T.Klose" and the 'READY' messages on my LCD when I power up. The next thing I do is to attempt to upload the midio128.syx file from the mk_midio128_syx\mk_midio128_syx\pic18f directory using MIDI-OX. All the 1590 Bytes of data goes out as it appears in the Output Monitor but nothing appears in the Input Monitor and there are 0 Bytes received in the Display Window of the SysEx View and Scratchpad. Also there is no reboot or change visible on the LCD. Am I doing the right thing here? Edited to add: I just went through the .syx file generation process using the instructions in http://www.ucapps.de/howto_tools_mbsyx.html just in case I should not have been using the midio128.syx file in the mk_midio128_syx\mk_midio128_syx\pic18f directory . I successfully modified the midio128.ini file and converted it to a .syx file with a new name. I loaded this file in the MIDI-OX SysEx tool and sent it but again nothing appears in the Input Monitor. Any ideas would be welcome. Thanks; Ciaran.

Update: I bought a gameport-MIDI interface and hooked it up to a desktop PC. I got the upload request immediately after power-up every time. I was also able to display text on my LCD using the MIOS LCD Message facility on MIOS Studio. I tried uploading the MIDIO128 application using Midi-Ox but the core still did not reboot and I did not get anything back on my Input Monitor. I went through the TEST PC1 and TEST PC2 steps on http://www.ucapps.de/howto_debug_midi.html Then I went through TEST IN1 to TEST IN6 and TEST INOUT1. All these tests went through with no problem. I tried uploading the software implemented loopback test .hex file from TEST SW1. This didn't go so well. I got the following response from MIOS Studio. The LCD displayed a constant 'Rebooting MIOS' message. When I powered down and back up and ran MIDI-OX with the virtual MIDI Keyboard I got occasional messages back in the Input Monitor but they didn't correspond to the notes in the Output Monitor. I have reloaded MIOS successfully now. Any ideas, advice or suggestions would be much appreciated. :) Thanks.

OK, I'll look into my other options. Thanks for the rapid responses. :)

Here's a shot of the internals of the case. Maybe you can tell me if there is a PIC in there. :) How would you load new PIC firmware? Via the USB?

Hi Thorsten; Unfortunately there is no name on it. I bought it from one of the many Hong Kong vendors selling them on eBay: http://cgi.ebay.com.au/ws/eBayISAPI.dll?ViewItem&ssPageName=STRK:MEWNX:IT&item=180180954171&_trksid=p3984.cWON.m313.lVI Their photos have been taken down but the interface cable comes with the MIDI IN and MIDI OUT plugs on one end, USB on the other and this device in the middle: It came with nothing except these instructions. Would you recommend going with a better/more reliable interface or to keep trying with this one? I also have an older desktop PC with a joystick port. I tried several shops but could not locate an old-style Soundcard -> MIDI IN/OUT cable to suit it. I have found one locally on the web. Would it be worth trying this approach instead? Thanks; Ciaran.

Further Update: As a result of SmashTV spending a couple of hours with me on chat I was able to confirm I am getting that single LED flicker on power-up (TEST OUT1). I also see a flicker on the input LED on my USB-MIDI interface. TEST PC2: I'm able to get data back on a loopback set-up but the received bytes is much less than the sent. It looks like I am having problems with my USB interface. This may be the reason I don't see the upload request (on start-up) in my In Monitor on Midi-Ox or MiosStudio. I have tried updating the USB-MIDI interface driver as Thorsten suggests in the MIDI Interface Troubleshooting page but the windows message says I have the best one. I also lowered the Low Level Output Buffer size and increased the output delay in the MIDI-Ox SysEx configuration menu as suggested with no success. Has anyone else had issues with these USB interfaces who might be able to point me in the right direction? Thanks. :)

Thanks nILS. I'll do some more reading and then keep plugging away until I see some sign of life. :)

Oh dear. :-[ :-[ :-[ I believe I may have spent the last 5 days troubleshooting a problem that didn't exist. Maybe the reason I wasn't receiving the MIOS upload request was that the PIC already had MIOS loaded..... Doh! I was following Thorston's Core testing segment and when he said "The next step is to upload MIOS" I followed that procedure. I just went back to SmashTV's website and confirmed that the PIC comes "Programmed with the bootstrap loader and MIOS". Oh well, at least I am now intimately familiar with the Core PCB layout. :) Thanks very much for the response nILS. Not only have you saved me more days of pointless troubleshooting: but now I know how to test the polarity of a diode! :D This newbie owes you ein weissbier!

Update: Moved ahead to TEST CORE7: - don't know how to test the polarity of a diode but the direction was marked on the SmashTV PCB so I'm pretty sure that's OK. TEST CORE8: R8 and R7=220 Ohm - checked and OK TEST CORE9: crosslink between the core module and PC is correct TEST CORE10: not applicable. Not getting any messages TEST CORE11: 2200uF Cap looks to be soldered in fine. TEST OUT1: No Good - not getting any flickering of the LED TEST OUT2: Didn't try but disconnected PSU on laptop and still no upload request or LED flickering. TEST PC1: check the MIDI-Ox configuration -downloaded MIDI-OX and routing looks OK TEST PC2: didn't get it working with MIDI-OX SysEx view but managed to receive MIDISync timestamp and other data sent when the loopback was connnected so my USB-MIDI interface appears to be OK. I also got a flickering on an LED connected to the MIDI out pins. I should have mentioned that: I am getting the 'READY.' message on my LCD so some of the PIC and Core appears to be working OK. I installed the 10 MHz crystal as supplied in the SmashTV kit - TEST CORE1 -looks fine. I have gone through the schematic and continuity and resistance checked all the available soldered connections (TEST CORE3) -all tested fine. I have checked the Vdd and ground connections of the PCB as per the schematics (TEST CORE4 and TEST CORE5) - all fine. I'm not sure what is meant by the TEST PROG1: 'use the verify function of IC-Prog/P18 ' but since I bought the PIC with bootstrap loader from SmashTV it's probably not applicable. Am I correct in saying that there is no point in going on to testing the MIDI-In or the Opto-coupler if I am not getting the upload request or getting a flicker on an LED connected to my MIDI Out port? Thanks in advance to anyone who can spare a few minutes to help out a hopeless newbie. ;D

Hi. Apologies in advance for the newbie questions. I am currently working my way through the MIDI Interface Troubleshooting page on uCApps. I have a USB-MIDI interface and I am not seeing an upload request or getting a flicker on an LED connected to my MIDI Out port. I am stuck at the point of testing the polarity of my MIDI plugs (TEST CORE6) using this diagram: I have a few questions/confirmations: 1) Should the lower pin on J12 be M+ instead of M-? 2) The MIDI in and MIDI out refer to the female MIDI sockets and not the male plugs right? 3) Am I supposed to be testing the polarity of the plugs or the sockets? 4) Should I be testing with my USB plugged into my PC and the plugs in the sockets or not? 5) I am looking for positive DC voltage of pin M+ with respect to M- right? Apologies again for all the questions but I have been stuck at this point for 4 days now. :(

Good point DragonMaster. I didn't consider cutting the optocoupler pins and de-soldering individually as an option if it fails. I'll pick up a new optocoupler to have as a back-up but I'll leave the current one in until it gives me trouble. Thanks everyone for the advice.

I've finished my DIN and DOUT boards and gone back to review the core. I've tested my voltages and they seem to be right according to uCApps and the opto-coupler appears in the right orientation according to the circuit diagram. Based on Wilba's comments above I would prefer to install the socket if it is possible to remove the IC with damaging it or the board. Is there a good technique for desoldering a multi-pin IC or socket? When I tried I was working one pin at a time and each pin seemed to re-solder with residual solder every time. I couldn't make any progress and was adding a lot of heat to the IC. Any suggestions?

Thanks for that guys. :) I'll leave it in and see how it runs.

The good news is that I've just finished soldeing my first core module. :) The bad news is that I have a component left over. :( Spot the newbie - I soldered in the opto-coupler instead of the IC-Socket. :-[ I'm not having much joy de-soldering it with my pump or braid and I'm afraid I will overheat or crack it. Should I keep trying or is it best to leave it where it is in the hope it will work?

Sparx; I am planning a similar project using lots of solenoids like the one being discussed here. Am I correct in saying that although the output of the ULN2803 is "inverted", it is still a suitable device for driving a bank of relays used to power solenoids? Thanks; Ciaran

Will do. I've written a good few Standard Operating Procedures/work instructions over the years for engineering tasks so I'm well used to documenting work. The only problem is that it will be written from a newbie/non electronics literate viewpoint. This would make it good from a newbie reading perspective but it may need some proof-reading for accuracy. ::)

No worries Bill. I plan to try out lots of solenoids once I get the DOUT module working with LED's. It could be a bit down the track though because from what I gather I have a few hours/days/months of learning, assembly, programming and de-bugging ahead of me. :o :o :o I've only gone as far as getting my MB module kits & accessories components on order so it make take some time to get to the solenoid trialing stage. I will probably use rotary solenoids with some sort of latching relay for the picking action and a more standard relay/solenoid arrangement for the fret-board action. Am I correct in saying that to allow proper note resonance you need to get your hammer away from the bar being struck immediately after striking, irrespective of whether the MIDI note status is still on or not? Or will you just use stacatto MIDI notes to minimise the duration of the output pulse? It might be an idea to use the solenoid to give momentum to a hammer and use a spring or gravity to return the hammer to a rest position away from the xylophone bar immediately after striking. For a good sound do you need the hammer to strike the xylophone bar at right angles like it does in the video?

;D :D ;D :D ;D :D ;D

Thanks for the advice stryd_one and lylehaze. After doing a bit more reading I believe to start off, I will get away with the DIN, DOUT and CORE kits plus some LED's, 16 core ribbon and spare connectors/pins. Smash don't have any LCD's in stock but from what I read I should be able to pick one up locally. I will need a power supply and some soldering, cutting and crimping tools so I need to do a trip to the local electronics shop anyway. I don't like the sound of that coding task stryd_one. :-[ The last time I did any programming was in 1985 on my Sinclair Spectrum. ;D You are correct lylehaze one note can be played on more than one string but this it is highly unusual to do so on the instrument (tenor banjo) that I will initially be working on, so to start off it will be one note per fret/string combination. There is more scope for playing the same note on two strings on a 5-string banjo which is part of some future phase of this project ::). Your comment has given me some ideas though. One of the reasons for this project is to produce sounds on an acoustic instrument that cannot pysically be produced by a person (e.g flat-picking two non-adjacent strings simultaneously without picking the intermediate one). Using different MIDI channels for different music 'parts' will be a great way to display the capabilities of the system. One quick confirmation/question before I place my order. Would a good starting point be a single CORE module with an PIC ID of 0x00, set up for a common character display LCD and common MIDI baudrate? (i.e. PIC ID of 0000 0000 0000 0000)

Hi all. Let me apologise in advance for incorrect terminology and lack of technical knowledge. I am a complete newbie to microcontrollers etc. :-[ I have a project in mind which involves automating a plucked stringed instrument (a banjo). I want to use a MIDI file (preferably stored locally on a controller) to DO control (via relays) the solenoids used to pluck and press the strings. (Ultimately I like to use DI's to select and start and pause different MIDI files but that can come later.) The idea is that the control system would enable the MIDI file notes to be played in real time to the tempo defined in the MIDI file. I was initially looking into using a PLC (based on advice from a system integrator / PLC programmer I work with) and then leaned more towards a PDA/laptop/PC linked to a MIDI-to-Parallel Converter like the j-Omega Electronics MTP-8 ( http://www.j-omega.co.uk/mtp8.html ). Having come across this site and uCApps.de, I am now leaning towards the Midibox hardware platform as I believe the modular design leaves my options open for expanding the project later. I would also prefer to understand the design of the controller rather than take a 'black box' approach. I am a musician and a mech. eng. with some exposure to PLC's and electronics. I believe I can work out the components from the DO's on but find the Midibox side of things daunting. My question has to do with the best way to learn about Midibox. I learn better in a hands-on environment and generally prefer to have something physical to work with. Would it be easier to learn if I go ahead and buy the components I think I need and try some stuf with LED's connected to the DO's? Or can someone point me to where I can do some basic tutorials so that I buy the right components to begin with? From what I can gather here and on uCApps.de I will (at a minimum) need the Core and DOUT Modules as well as an LCD, some LED's, ribbon cables and IDC sockets. Which is the best approach:buy now or do more research? Thanks. Ciaran