Another MidiboxSID is born. (has now been born)

[286]

I've been planing my midibox for about two months now. But since I've been busy with other stuff too I haven't been able to get started. Until today!

I love what you guys are doing with mb6582s and superdupermidiboxes and sure it would be fun to build something like that someday. But for now it's a mono sid with a simple control surface.

So time to get building:

First mission was to scavenge the SID-chip. Nothing strange here. I knew that it was a 6581 and it wasn't hard-soldered so no problems getting it out of the C64. I also took the powerled. While planing I decided against using the powerplug from the C64 and going with a panelmount one so I didn't take that. The reason for this is that the MIDI-sockets and RCA jacks are going to be panelmount because I want them to be more "trip-on-the-wire-proof". I also don't like to drill square holes. :)

Next mission was to build the PSU. After carefully reading the schematics for the optimized PSU for a last time I started assembling it. Now building on a perfboard with solder islands turned out to be harder than I first thought. Geting the components on the board was easy but connecting them to eachother was a bit trickier. I used the connecting solder blobs technique and it felt a bit risky with the possible cold joints but it turned out fine. I also used wire for the longer connections.

Apart form one bad solder connection progress was slow but smooth. The end result was quite nice and tidy. (Except the ugly underside, I might actually go for stripboard for the CS later on, blobbing just isn't my style.) My caps are good for 35V so they are a bit big but otherwise it's quite compact.

End result: 14V DC = 14.4V DC, 5V DC = 5.3V DC. I guess that will drop closer to x.0 with some load.

To be continued....

Btw crimping the sil-connectors with just a pair of pliers is quite tedious and I've only done four so far. ;)

[Smithy]

Btw crimping the sil-connectors with just a pair of pliers is quite tedious and I've only done four so far. ;)

I bought a Pro's Kit crimping tool for SILs from here.

Cheapest i could find on ebay, and the seller was really nice too.

Smash used sell those ones too but not sure if he has some now, youd have to ask.

[nebula]

While planing I decided against using the powerplug from the C64 and going with a panelmount one so I didn't take that. The reason for this is that the MIDI-sockets and RCA jacks are going to be panelmount because I want them to be more "trip-on-the-wire-proof".

Panel mount is quick 'n easy.  Plus, DIN sockets usually have weird lead spacing and can be hard to fit into some prototype boards.  But the nice thing about board mount is that you can just pull the board out of the enclosure when you want to work on it, without having to unscrew all your panel mount jacks or disconnect them.

I also don't like to drill square holes. :)

Just pick up some square drill bits!

building on a perfboard with solder islands turned out to be harder than I first thought. Geting the components on the board was easy but connecting them to eachother was a bit trickier. I used the connecting solder blobs technique and it felt a bit risky with the possible cold joints but it turned out fine. I also used wire for the longer connections.

It's a matter of preference.  There are a lot of different foil patterns on prototyping boards.  I usually choose either the "pad per hole" style or the "breadboard" style (with little sets of 4 pins that connect together, plus power rails, similar to a breadboard), although stripboard has its uses.  The perfboard with no foil is the cheapest, and sometimes the easiest, but you will always need to manually make every connection.  It is also the easiest to cut to size.

When possible you should always try to make a physical connection before soldering.  Counting on the molten solder to bridge a connection will work, but it is usually less reliable.  On perfboard, try to make your component lead reach the component you have to connect to, and fish it underneath, maybe making a little hook with some needlenose pliers.  This is not a rule, it's just good practice.  Don't undo anything you have done now, it's probably fine... you can usually spot the cold solder joints because they will either be dull, discoloured, or have an irregular shape - good solder joints look like liquid.

End result: 14V DC = 14.4V DC, 5V DC = 5.3V DC. I guess that will drop closer to x.0 with some load.

Yes it should, a little.

To be continued....

Can't wait!

Btw crimping the sil-connectors with just a pair of pliers is quite tedious and I've only done four so far. ;)

Actual crimpers are a must for those little buggers.  As well as few extra terminals to practice on and ruin.

Good luck with your SID!  I have just completed my MB-6582 and I have pretty much only done stuff so far with a single pair of SIDs.  It FAR exceeded my expectations.  I always loved the C64, and C64 music, but MIDIbox really takes it to a new level.  As you toil over finding parts, making messes, burning yourself, and forking over money for whatever parts you can find while dealing with back orders on the parts you can't find, and having no social life, remember that you are making something open source which is way better for making music than any other commercial SID product, including the Commodore 64 itself.

[286]

Day2. Core day.

Today I started assembling the core. It was straight forward and having smashtvs homepage open during the process was great help. There are many solder joints but they are all pretty easy so it didn't take very long. I was careful soldering the transistor, using a crocodile clamp as heatsink.

After assembly I tested the 5V points and got 0V everywhere. It didn't take long to figure out that I had forgotten about bridging IC3. After that i got about 5.3V. As mentioned yesterday my PSU delivers about 5.3V on the 5V line so thats OK.

To be able to test it I decided to hook up the LCD. Now this was a total pain and the reason for that was that my wirestripper doesn't handle wire as thin as that. I tried a knife, wirecutter, scissors... you name it. After struggling for a long time with getting all 16 wires correctly stripped without accidently cutting one to short I finally gave up. I went to the shop and bought myself a nice new wirestripper. What I couldn't do in like 2hours took about 5 minutes when I got back. My LCD layout is mirrored to the one in the schematics so I took great care in getting everything right.

I hooked it all up and got... a blank LCD. I first blamed one of my homemade dil-connectors (I will hard-solder those connections after building everything anyways but the plier crimped power connectors work for now). But it turned out that it was in fact the pots that just needed adjusting. A few seconds of tweaking and I got excellent light and contrast. Doesn't show up that great in the picture, plus the protective sheet is still on, but it's negative white on blue backlight. I've never seen any word as beautiful as "READY."

[nebula]

Looking good! 

If your DIL connection is good, and you can wiggle it without causing any weirdness, I wouldn't recommend you bother changing it.  You could probably clean up the connections on the LCD though :)

(They often start out messy while you are trying to figure out which conductors are which)

[286]

Day 3 (of building).

I've been busy with other stuff but today I finally had time to assemble the SID-module. It was really straight forward. Only problem was that I didn't have SIL-headers for the audio-connectors but I cut some DIL-headers to get bi-pole SILs :)

Tested it with some crocodile clamps and and more plier made connectors. 11.95V between the 12V points of the IC. Another success!

However now I've run into some trouble. It's time to connect the SID-module with the CORE-module and as mentioned earlier I use the optimized PSU-layout. I've edited the schematics to visualize my problem for you. The "problem" is the 2200uF in point 2.a. I can understand the need for this cap in a multi-SID environment but the 5V line here is already filtered in point 1, and there really isn't anything  in between except from the 100nF cap.

So I basically have three questions. (see attached schematic)

1. Do I need the 2200uF cap in point 2.a even in this single SID environment.

2. Does it have to be as close as possible to J2 on the SID or could I put it in point 2.b? It would be more easy to solder it to the PSU-board.

3. If it has to be as close as possible to J2. How are people solving that, and what do people attach them to? (Pictures would be great.)

[zgba]

I haven't soldered this cap and in my case there's no audible difference with or without it. I doubt it's necessary in 1 SID environment. But if you insist -  in theory it should be as close as possible to avoid ground loops, but it's really hard to solder it under the PCB. Maybe try to use some short insulated wires?

[286]

Day 4.

I finally got everything hooked up. Sent some funny messages to the LCD through the debug interface in MIOS Studio. Then it was time to upload the setup_6581.hex. Now this is where things went from happy sunshine to dark clouds in a few seconds. MIOS Studio started to send the code but then the LCD showed "MIOS rebooting" but nothing happened. I waited for a few minutes but since nothing ever happened I eventually restarted the MIDIBOX.

And now all I get is two blank lines on the LCD. No OS-info. No READY. Only two blank lines and two lines with blocks. I've started reading documents on troubleshooting but I'm really afraid to mess up even more... :(

*EDIT* I may have figured out the problem... I'll let you know.

[Labelwhore]

<=== patiently waits for an update

I'm about to begin work on my own Midibox Sid soon so I'm enjoying this thread, seeing what it takes to get one off the ground. I can't wait to see the finished product.

[286]

Ok I've figured out how I messed up. I've not been able to solve it yet but I'm working on it. All help is appreciated.

For some goodforsaken stupid reason my cheap USB->MIDI connector has the MIDI-OUT and MIDI-IN mixed. All other devices I've ever used have had MIDI OUT on PC connects to MIDI IN on another device, MIDI IN on PC connects to MIDI OUT on another device. Now I don't know if they just invented their own "lets mix the cables standard" or if someone accidentally mixed the connectors while assembling it but it has gotten me into deep problems.

Anyway the problem is that I tried uploading the sid-application while the connections were crossed. This somehow has corrupted MIOS. I've now switched the cables and now I get  "F0 00 00 7E 40 00 01 F7" every two seconds. This means that the bootloader still is alive and that there is hope! I dont have a pic-burner so thats really good news.

The problem now is that when I try to upload MIOS I get:

Starting upload of mios_v1_9f_pic18f4685.hex
Hex file contains code in MIOS range, forcing reboot!
Received Upload Request
Sending block 00000400-000004FF
Error: Received unexpected Upload Request

After 16 times MIOS gives up just as it's supposed to. Now you can go ahead and call me noob and tell me to RTFM but I really can't figure out how to get MIOS back in. I can't think clearly. It's time for a cup of coffee to clear my mind but if someone who has a clue should read this. Do NOT hesitate to give me some directions. :)

[smashtv]

The problem now is that when I try to upload MIOS I get:

Starting upload of mios_v1_9f_pic18f4685.hex
Hex file contains code in MIOS range, forcing reboot!
Received Upload Request
Sending block 00000400-000004FF
Error: Received unexpected Upload Request[/code]
After 16 times MIOS gives up just as it's supposed to. Now you can go ahead and call me noob and tell me to RTFM but I really can't figure out how to get MIOS back in. I can't think clearly. It's time for a cup of coffee to clear my mind but if someone who has a clue should read this. Do NOT hesitate to give me some directions. :)
This symptom -usually- means the MIDI in on the Core is not working......  
Cause could still be the PC though, until you test to verify (loopback test) that the MIDI the PC says it's sending is actually being sent and making it to the Core......considering the fight the USB->MIDI interface has been giving you a loopback test should be next on the list.
Could also be as simple as the wrong polarity to the midi jack connection @ J13.
What tests pass/fail from http://www.ucapps.de/howto_debug_midi.html?
Best regards
SmashTV

[286]

Testing shows that this may in fact be a case of bad wiring. I will redo my wiring to the MIDI jacks.

*edit* It turns out it was a bad MIDI-jack. HUGE thanks to SmashTV and Nils for helping me out, you guys rock! 

*edit2* Btw while doing all the testing I switched between windows and linux just to rule out any OS settings, and I can now confirm what I have suspected all the time: MIOS Studio works flawlessly in 64-bit Vista.

[bugfight]

...

Best regards

SmashTV

SmashTV?? who is that??

*whack*

[stryd_one]

confirm what I have suspected all the time: MIOS Studio works flawlessly in 64-bit Vista.

Crosspost: Re: MIOS Studio beta7

[286]

Day 5. Let there be sound.

After spending last night trying to figure out why my midi-connection was acting up I woke up really tired. I never got around testing the SID-module last night so today was a big day. A quick breakfast and then it was time to hook the SID-module back up and give it a go:

And the result? It sounds just beautiful! I'm really happy I took the time needed to build the optimized version of the PSU. There really isn't much noise at all. At least I cant hear any with my current speaker setup. I've included a short clip (approx 40 seconds) of my Midibox playing the great demotune Axis of Evil by Zabutom if anyone is interested in how my 6581 sounds. I still need to do more testing to find out the characteristics of my 6581 to see what it can and what it can't do.

This is a major milestone for me.

Next up will be editing the .asm so my LCD uses 4x20. After that it's time to start thinking about control surfaces and enclosures. I've done a lot of planning but after building the modules I may have to rethink some of my design choices for the enclosure.

I also included a picture of what it looks like now. The 14V wire is connected with crocodile clips and it's all mounted on a cardboard sheet from a box of beer cans, but it's starting to look like a Midibox. :)

test2.mp3

test2.mp3

[stryd_one]

Day 5. Let there be sound.

Day 5. Fastest midibox in the west. :D

!yay

[stryd_one]

Oombox went over here....