weird behavior on mbsidx4 with full cs

[tarzan boy]

Greetings,

Tonight was the big night when I plugged all the ICs into their sockets and ran my bad boy for its first time.

First I'll go through what I've already done up until this point -- in order to set aside any misunderstandings of what I might have done wrong:

way back in october I got my "Step 1" mbsid working; LCD, 5 buttons, and a 6581R1 sid making some sounds. Since then, off and on because of travel for work, I've:

-Built up all 3 slave modules

-created a 'optimized C64 power supply'

-programmed successfully each PIC with header and corresponding mios with device IDs 1-3

-programmed each slave PIC with correct type of SID app, i.e. IDx1 and IDx3 for 8580 and IDx2 for 6581

-created a full control surface from an old C16 case, a C64 motherboard as panel, and lots of drilling

  -tested all of the buttons and LEDs for correct wiring

-confirmed all voltages to be correct for each module

    -the 6581's have 12 volts

    -the 8580's have 9 volts

   

Tonight I popped the PICs all in, and since I hadn't messed with the ICs part of the project since december, I put them in in order of what I thought was 1-2-3-4, but in fact I put them in 1-3-4-2 (because I was looking at the Device ID number I had written in Sharpie on the top of each of my slave pics --the IDx0 pic is from Smash so its the only one with a white sticker on it -- so i dropped "2" in thinking it was the 2nd one, then 3, then the 'other' one.)

Then i tested voltages in the SID modules one more time. Everything was gravy, so then i put the sids in, paying careful attention to put the 6581's where the 12volts was, and 8580's where the 9volts was.

So i fired it up-- with the pics in the wrong places, mind you-- this means that instead of

IDx0       IDx1    IDx2   IDx3                     IDx0        IDx2    IDx3    IDx1

master    slave   slave  slave                     master    slave    slave   slave

6581       8580   6581   8580          I had:   6581        8580    6581    8580

which means the slave pics 2&3 had an incorrect model of SID with them.

So without knowing it I was going to experience some crappy filter issues. But that wasn't my problem:

I pushed LINK, then:

My problem was that only SID#4 was playing. Not even the Master was playing notes.

After fiddling around for a few minutes and realizing that I had placed the pics in the wrong sockets and that IDx1 -- button SID#2 on my control surface -- was controlling #4, I turned off and switched to the correct PIC setup.

Now, predicting only that SID #2 would work (the pic is good and all the others have errors), or only SID #4 would work (the SID is good and all the other SIDs are dead), I turned on and found only SID#3 to be working.

So,

IDx1 in socket #4 makes SID#4 work, but IDx1 in socket #2 doesnt make SID#2 work

IDx3 in socket #3 doesn't make SID#3 work, and IDx3 in socket #4 doesn't make SID#4 work.

but,

IDx2 in socket #2 doesn't make SID#2 work, but IDx2 in socket #3 makes SID#3 work.

other clues:

I am getting the start-up "pop" from every sid, and a high pitch buzz from both the 6581's. So there's something going through there. Also when I move certain rotary encoder knobs I'll get a quiet 'click' noise from every sid, letting me know the whole thing is functioning in some way.

Anyone know what the problem could be?

[Wilba]

I don't know what the problem is, I can only suggest you prove each Core and SID module (and SID) works independently. You're trying to debug by just swapping PICs around, when it could be a faulty PIC, a faulty Core PCB, a faulty SID, a faulty SID PCB, a faulty Core->SID cable, faulty power wiring, faulty MIDI link cable, etc. So I suggest:

1. Plug the MIDI out into each Core and prove it's working.

2. Plug each SID module into the master Core and prove all the SID modules work.

Once you know each SID module works...

3. Connect ONE slave to master, test a known good SID module with it. If it doesn't work, the most likely cause is the cable between the slave Core and SID.

4. Repeat step 3 with each slave.

Make sure you fully disconnect everything else when testing one part, including power. It's scenarios like these that it pays to have used cables with header connectors instead of soldering cables.

BTW, double check if the MIOS 1.7 hardware mods are not the problem: http://www.ucapps.de/mios_v1_7_hardware_mods.html

[tarzan boy]

OK,

a few days later and I've learned an important lesson: Don't ever take your MBSID in mid-construction on 6 different airline flights and expect everything to be cool and hunky dory like you left it 3 months before.

Right now I have isolated my problems and gotten quite a bit of it fixed.

Working:

-All Core boards and Sid boards will function with confirmed-working PICs in a one-on-one relationship (no slaves running off it)

-Bankstick with 3 banks, A,B,C. I've uploaded some presets to A so far.

Remaining problems (not mentioned in http://www.midibox.org/forum/index.php?topic=6464.0) :

-definately PIC IDx3 is either dead or somehow got completely wiped. I'll try to re-burn it (NIGHTMARE) with my broccoli18 setup later this week when i have hours and hours of patience.

-although all SID modules are confirmed working now, i can now only get simultaneous sound from IDx0 and IDx2 sids (the two 6581's.) Still trying to trouble shoot this.

-Row 3 of my LED matrix is mysteriously dead.

  -All the connections are correct to the DOUT board, i can get each LED to light up by itself in that row by touching the contacts to my multimeter in Testing mode, but it remains unlit when my CS is powered on.

  -Am I naive enough to see if switching around my 74HC595s will work? Yep, coming soon.

Any suggestions on what could be the problem with the LED matrix? The other stuff, my current path of poking around I think will get me to an answer, but this problem is a new kind of monster, not really sure what I can do about it.

[stryd_one]

Aircraft (in particular, the things that happen to the luggage before it gets on the aircraft, and the extreme cold and damp of high altitudes) and electronics do NOT mix!!!!

If you want to do this in future, then arrive at the airport early. Most airlines offer a service where they will carry fragile/sensitive luggage separately, sometimes for a very small fee - but always worth it, for obvious reasons.

Won't hurt to pack your gear in airtight anti-static bags and bubble-wrap either. If you can't get those, then just use normal antistatic packaging and put it in a ziplock bag.

Hope this helps

stryd_one - ex-pilot :)

[tarzan boy]

I have bad luck with the special handling people on airlines; my guitar has been lost 2 times and the case ripped in two another time.

I had this packed up in its almost completed case (a C16 shell) inside an original C64 box, all chips packed in static bags taped to the styrofoam of the C64 box where the power supply would normally be packed, then this whole thing put into my suitcase surrounded on all sides by either by clothes or books and clothes.

But i suppose none of it could have prevented the moisture and cold.

didn't want to risk it getting scanned for carry-on and having to explain myself with these homemade electronics I'm bringing on board, when I already have a suspicious passport and get questioned all the time at customs (lots of travel stamps and a birthplace in one of the axis of evil countries.)

god bless the USA...

[stryd_one]

Axis of evil LMAO don't get me started on US politics  >:( .... Racist sons of.... Only reason they don't have civil wars like everywhere else is that they slaughtered and bred out their adversaries when they arrived. But no, there's no evil in New York or DC or Chicago or any other large US city, is there? Noooo!

OK I'm going to shut up and get back on track now. But yeh, Nil illegitimi carborundum ;)

Yeh well the bags that go into the cargo bays get scanned WAY worse than the hand luggage. Not just XRays but all kinds of nasties... And this was before 9/11, so I hate to think what they do now.

If you register it as static and EM sensitive baggage then they will check it by hand rather than putting it through a machine. The clothes and all may actually help with the damp and cold to a degree but those x-rays will mess with your kits like crazy and unless you have lead-lined suitcases you're screwed. If you tell them that it is computing equipment then they will have guidlines to follow which will ensure its safety. Often there is a small charge for this kind of thing, but it's well worth it.

Of course, the easiest way is to get some insurance on the bags - usually only costs 20 bucks and it beats spending a few hundred on new chips :)

[tarzan boy]

So there is a chance that my mysteriously non-responding PIC is the result of some airport tomfoolerly? because honestly I am dreading setting up my crappy unreliable broccoli18 pic burner again to re-burn the ID onto this thing.

would it be more likely that the PIC is completely fried or just wiped blank? The other 3 are still fine.

[stryd_one]

Man that's a tough one... It's far more likely to have been wiped than destroyed, but that all depends on what they did to it...

You might have to try re-burning to find out :(