CORE troubleshooting

[kokoon]

so. i've lurked for quite some time and then i've decided to make a move. mbsid and mbseq are 2 things i CAN'T live without. for start i ordered two mbsid+cs kits from joerg.

since i've never soldered in my life i needed to buy all the tools and accesories. :D

shortly - i've put together a core module. since it was my first effort of this kind it looks really bad, all the joints look really ugly. among other things i've managed to break 1 pcb line by heating it too much but i hope i've solved that with 1 extra bridge at the end.

when i finished i wanted to check if anything works. i've connected C64 PSU's +5/-5V to J1 on the core (by the way - which pin should go where?), set the jumpers on J3 and started measuring on the control points.

PIC 1 -> PIC 12 : 3.8V

PIC 11 -> PIC 12 : 3.8V

PIC 32 -> PIC 31 : nothing BUT PIC 29 -> PIC 30 : 3.8V (or -3.8V ... i don't remember)

6N138 8 -> 6N138 5 : -3.8V

after that i accidentally made a short circuit on the psu and i can't get anything to work since then. i opened it and apparently the 1.6A fuse went down. but still there's no output from it whatsoever (no 5V nor 9V)...

can anyone point me to anywhere ... should i post some photos of my core? i could do that if anyone wants to look at it... what should i do with the psu?

i'm really determined on finishing this whole thing. i've gotten much better at soldering and i know i'll be able to finish it. can you guys help me?

[pwx]

Hi!

I had similar problems with my first core module....

I think the first important to build is the complete psu shematics(www.uCApps.de) and connect the core to this like shown...

After that, check all voltages again....

If there is still no 5V at the sockets (do not install the PIC before!!!!!!!).....

check your soldering again and again and again and......

Perhaps it would help to buy a new core pcb and stuff it again ( soldering is learning by doing ;)

When I bought my second core pcb and suffed it again, the most problems where solved......

hope this helps ;D 8) ???

best regards,

pwx

[Jidis]

It should run regular +5v and ground right guys?

Mine comes off a supply which is supposed to be 9vDC , but it appears to put off something in the 12-13 range (it's heavy with high amperage capacity). The regulator/filter part does a clean 5v for me (or 4.9x).

I think with no chips, your supply pins at the sockets should be reading very close to what's at the regulator output pin (5v).

                                          (I'm not really an electronics wiz, but there's a bunch in here)

George

[illogik]

hey,

you just have the core right now, right (no other modules, no lcd)?

then it's maybe good to first use a regular wall-wart/adapter (7V-10V 500mA) for troubleshooting and fixing the core ; the c64 psu can be a bit confusing

as for the soldering; if you used too much you get blobs of dull looking solder (dull compared to shiny, good ones) and you probably have bad connections there.

get some desoldering wick/braid or a pump; remove the solder, and do them agian.

put your soldering iron not too hot (if possible) and be sure to touch both the pin of the part your soldering and the copper on the pcb pad, then put the tin against the tip/area to solder and if you do it right the solder will "flow" nicly in place

good luck, marcel

[kokoon]

wow so many responses!!!

firstly - i do have another core pcb but i wanted to build two complete synths. i was thinking about sacrifying this one for the learning purpuses but i'd be really glad if i could make it work somehow. and if i'll go practicing soldering i'll try to ruin something else, not the only remaining core module :) maybe DIN or DOUT.

anyway - i think the first problem was that i was using 5V supply instead of 9V. i'll try to get a 9V psu somewhere.

as for the soldering - i'll put on photos of both sides. maybe i did some trivial mistake :) .

[moebius]

Hi kokoon.

What.. +/-5v from C64 PSU?!? No way! ;) Check your connections. The voltages you measure could indicate, that you have connected PSU Ground and +5VDC to J1.. it won't work: The rectifier takes voltage down by diode drops and the regulator needs some additional voltage to work properly. (Polarity of the voltages connected to J1 doesn't matter - rectifier takes care of that).

See C64 PSU schematic options for the answers.

Bye, Moebius

[kokoon]

these are the photos  8)

moebius: hmm probably you're right, i had +5 and ground. nice to know that the rectifier takes care of the polarity.

illogik: yes i only have the core module at the moment.

[illogik]

so get a simple wall wart/adapter, because you probably want that for the mbseq anyway, put it on 9 volts (7-10V is good; 5V isn't like moebius said); connect to core and start troubleshooting (measuring)again; the soldering doesn't look really good but i've seen a lot worse.. ;D

cheers, marcel

[moebius]

Another possibility is to use mbhp_4xsid_c64_psu_optimized.pdf schematic and connect +5VDC directly to J2 (remember to remove the regulator and add wire between pins 1<->3).. check the schematics and be careful! (additional bypass-caps are not needed for just testing purposes with one core)

Well.. it's not the most beautiful soldering I've seen - but rather good for a beginner!! There might be some cold solder joints, but it might work anyway.

Bye, Moebius

[kokoon]

thanks guys i think i'll improve my soldering quite some more by the time i'm finished with a complete mbsid+cs ;D

i didn't know i should head the pcb's copper in the joint area. the first 10 joints were made totally wrong - i put the solder on the tip and then spent like 3 minutes per solder to get it on the joint :D :D :D the rest was made by heating the pin and then putting the solder on. it may be my imagination but i think that method is less and less efficient. the solder keeps climbing up the pin and gathers on the tip. i bought what i think now a very inappropriate soldering iron and i think the tip has gone worse quite a bit. i'll put up some pics of my soldering iron's tip.

[moebius]

Hi,

Non-temperature soldering iron should be around 25W one.. I'd say. That will keep the soldering temperature ok. Try heating the pin AND the PCB pad together for few seconds and then apply solder  to the joint. Once you have enough solder in the joint (less is more here - you can add more, if needed) - remove the solder and heat the joint for another 0,5sec ->1,5sec, remove the soldering iron tip and hope to see a beautiful shining solderjoint to form.

Bye, Moebius

[kokoon]

my soldering iron is a 25W cheap Weller. i got two tips with it. one is sharpened diagonally (the one on photos) and the other is like a small sharped pencil. the second one is far less precise... i think :)

current state of my tip:

[moebius]

You'll learn to solder with that one, it's ok. ;D Next time use the tip shown in "mbcore054lt.jpg" - keep the tip "tinned" (always have some solder on the surface) and clean it up occasionally with the sponge.

Good luck, Moebius

[kokoon]

hmm i use the sponge all the time... maybe that's what's eaten the whole tip...

also - my tip is removed/replaced by unscrewing it. it doesn not have a locking screw on side.

regarding the sponge - how wet should it be? what does you "occasionally" mean exactly?

what should i use to remove the solder i don't need on it? i'm using the wet sponge now...

where is that pdf?

[moebius]

Hi,

Wet sponge is ok. Just keep the tip "tinned", even if that means adding more solder and then rubbing it to the sponge. Don't leave excess of solder to the tip, anyway. Use the sponge to get rid of the solder..

also - my tip is removed/replaced by unscrewing it. it doesn not have a locking screw on side.

Well, that's normal for that Iron, then.

Bye, Moebius

[Jidis]

Make sure it's the solder sponge too. Regular ones supposedly can't handle the heat or corrode the tips. I've also heard that grinding the tips destroys part of their necessary coating, but I do it a lot myself in a drill, and a tech I really respect here does it too.

Moebius, thanks for the >5v tip too! I either didn't know, or had forgotten that :-\ I'm usually running a beefy supply before the regulator/rectifier stuff, but I always thought I was OK coming from a salvaged computer supply's 5v line

-George

[Davo]

Whether or not to grind the tip depends on the type.  Copper (with or without nickel plating) are supposed to be ground.  Ironclad ones should never be ground.

I grind my tips by hand with a single-cut file.  What do you press your tips against when grinding them in a drill?

[Jidis]

Hey Davo,

I've done the files too in a drill press, but they clogged really quick from all the friction being in one spot. Lately, I've been using really sharp pencil point tips, and every once in a while, I'll chuck them in a power drill at the studio. I wear a leather glove and "pinch" the tip sort of hard with progressively finer grits of folded up sandpaper, while the drill is spinning pretty fast. The last one ends up on that black emery cloth sandpaper, probably >400grit (nice smooth tips). I'm using the screw on tips like were mentioned a few posts up, in a Weller iron. Filing the spinning tips was sort of difficult to keep from putting too much lateral pressure against the point and bending them. The sandpaper works better, but things can get hot. (probably not the safest procedure in the world)

-George   

[kokoon]

hello it's me again.

i've bought a wall-wart and the measurements were okay. hooray.

then i proceeded to testing the midi. it doesn't work. the problem is that i don't get any sysex on power-up.

i had the PICs programmed by some guy (i asked on some forum if anyone would do me a favor) - so i know they're programmed, he did it in front of me.

i followed the midi troubleshooting:

    *  If you are not sure that the bootstrap loader has been burned successfully, use the verify function of IC-Prog - don't have a programmer

    * ensure that you've stuffed a 10 MHz crystal (parallel cut) to your PIC18F core module - OK. is the direction important?

    * Are all 5 jumpers connected to the programming port? - OK

    * Check it visually for bad solderings or missed junctions - OK (i guess)

    * Check Vdd of the power supply like shown here: mbhp_core_extract_measuring_vdd.gif - OK

    * Check the ground of the power supply like shown here: mbhp_core_extract_measuring_gnd.gif - OK

    * Check the polarity of your MIDI plugs: mbhp_core_extract_midi_plugs.gif - OK (i think - i soldered it on the connectors the way it's in the gif

    * Check the polarity of the protection diode D1: mbhp_core_extract_midi_d1.gif - OK (i guess - the black stripe is to the midi-out side

    * Check the resistor values at the MIDI Out Port: R8 and R7=220 Ohm (resistor code: red-red-brown) - OK

    * Connect a LED to your MIDI Out port and check if it flickers: mbhp_core_extract_out_led.gif - NO FLICKER!!!

    * Just to ensure: this diagram shows a crosslink between the core module and your PC: mbhp_core_midi_crosslink.gif - OK

    * If you notice a lot of request messages like shown in this snapshot, then there is a short circuit between the Rx and Tx pin of the PIC. Check the tracks which are routed from the MIDI-Link port J11 to the Rx/Tx pins for direct connections (see the layout and schematic), you could scratch with a screw driver between the tracks to ensure that they are not connected together. - NO SIGNAL!!!

    * Somebody noticed in the forum, that the MIDI Out of his core module didn't work because of an "incompatibility issue" with the bench supply he used and the switching PSU of his PC. The solution was to disconnect the middle pin of J12 (ground line of MIDI Out port) - tried, no change.

now - i know my midi on the computer is okay, i've used it many times. the interface is roland's MCB-10. it's an ancient one but it looks really solid and bullet-proof. it's worked with all the midi devices i've had so far. and if i connect a led to the MB midi-out it doesn't flicker so i guess it's okay to assume the problem is on the core module.

what should i try now? i'm building a MBSID and i'm really anxious to try out the first thing i've soldered in my life (this MB core)...

heeeeeeeeellppppppppp ;)

[kokoon]

anyone?

[illogik]

hi,

so this could be 2 things (i kan think of right now);

- no bootstraploader on the PIC>> but you bought allready burnt PIC from MBSTore? then it's probably not that

- faults in the midi part of the core (dead optocoupler, bad connections, wrong polarity of components)>> i know this isn't what you wanna hear, because you checked it allready, but hey.. check again, youll be surprised how many times i looked over something that was (too) obvious. Try to check the connections in a different order then you did before, to avoid systematic errors.

cheers, marcel

dontgivup ;)

[kokoon]

which of the core components must be oriented the right way (polarity)? the resistors not, right?

[illogik]

hi;

not the resistors, but the diode and the optocoupler; triple check midiport wireing

have you got a spare opto to check (first make sure right voltages so you don't fry another)? 

and maybe get another PIC or make a burner to burn the bootstrap again.

sorry i don't have the magik answer :P

cheers, marcel

BTW i dunno how optocouplers and PIC's react to stactic electricity bursts, so maybe this won't help you;>>but do you ground yourself when working with sensitive components?

static electricity (which builds up on you when you for instance walk on carpet) can kill certain components when your body releases it. Ever had a shock when you touched a metal closet or something? its the same thing

If you don't have equipment to ground yourself just sit next to a big metal object (big heaters/pipes/closets stuff like that), touch it now and then, and don't move your feet too much.

[kokoon]

assuming all the measuring tests showed okay results - is it still possible that the octocoupler is broken?

i'm not completely sure how the opto should be turned - i assumed that the pin under the + sign (roughly "under") is the "+" one. hmm. i'll check it again when i get back home.

[kokoon]

oops that wasn't optocoupler. anyhow, everything is right on place. i also tried to switch the optocoupler with a new one but it still doesn't work.

maybe the BC337 is turned 180deg but i can't find out what brand it is (two different datasheets of that transistor say the oposite direction) but could that affect anything? i mean - it's for lcd right?