Jump to content

Trivial desperate minimal MBSID first-timer questions


Flemming
 Share

Recommended Posts

I built a new core, looks good, works fine... i took the PIC from my former core and stuffed in the new core, and it adapted the precise booboos as the first core - my guess is that i blew my first PIC :( My theory goes like this; i powered both core and SID modules from the same PSU resulting in 60Hz hum from the SID - i cranked the voltage up a step and got rid of the hum... but also got rid of my PIC :( the cores vreg did get awfully hot.. too hot to finger it.

Now i'm pretty keen on not busting my last PIC so i'm thinking of a new PSU concept :) using a C64 PSU - could i power both core and SID module (8580) like this:

5V from PSU to core module

9V from PSU to SID module

... and what about the 5V to SID module? should i take it from the core or from the PSU????

I know 'optimized PSU' would be the best solution but i need to get hold of the parts first.. and decipher the schematic, which i'm very inexperienced with :) until then it would be so great to feed a single core/single SID setup!

Also, my first core module got so much juice that it busted my PIC - if i measure the right voltages on this core module would it then be safe to stuff it with another PIC, or is it possible that other parts of the core has had damage and may end up frying another PIC ???

Link to comment
Share on other sites

  • Replies 58
  • Created
  • Last Reply

Top Posters In This Topic

I would start with the assumption that you are going to use the Optmized PSU if you plan on having multiple SIDs. It's basically the same thing whether you have 1 or 8 SIDs anyway (except for having to run the power wires). The circuit itself is easy to build, but i you'd rather have a printed board to do it on, you can make your own via Eagle. I made one myself you can look at for reference.

You could also build your own PSU that outputs the correct voltages (+5, +9 and/or +12V), but it sounds like you might want to start simple and get a working CORE and SID. So, worst case, get two AC wallwarts and power the SID and CORE separately. I did that for a while and it works until you feel comfortable moving onto something better.

Link to comment
Share on other sites

@ Cimo: I will have frikadeller today actually :) Brinck is so expensive.. the wallwart i bought there cost me 100DKK - today i went there to get a 7pin female DIN and he charged me 25DKK.... it's like 5US$

@ m00: Thanks for helping me out and not judge me :) not pointing me to pointless quotes, or assume that i bought some cheap pin headers and threw the gold-plated ones from Smash away all out of the blue :) i read somewhere that this forum was very friendly and helpfull - i guess it must be you that keeps up the standard :) thanks for quick and helpfull answers!!!!

My plan can be changed :) if i power the core from my wallwart (5V) and power the SID from my C64 PSU (9V) it would be okay?? It would be the same solution as you using two powersources (well, it IS two powersources :))

But is two PSUs really neccesary?? Can't the C64 PSU power both the core (5V) and the SID (9V) without problems?? (i mean, it can power 4 cores and 8 SIDs, right?)

And, my biggest concern; Should the SID module then get it's 5V from the core or the C64 PSU?

I must say even though this is my first experiences with midiboxing i'm learning every day :) not just about MB but electronics in general - the only electronics experience i had before was from C64 modding

Link to comment
Share on other sites

@ m00: Thanks for helping me out and not judge me :) not pointing me to pointless quotes, or assume that i bought some cheap pin headers and threw the gold-plated ones from Smash away all out of the blue :) i read somewhere that this forum was very friendly and helpfull - i guess it must be you that keeps up the standard :) thanks for quick and helpfull answers!!!!

I think you misunderstood what Stryd was getting at. He's also been a very helpful resource on this forum.

Link to comment
Share on other sites

I think you misunderstood what Stryd was getting at. He's also been a very helpful resource on this forum.

I'm sure you are right :) .... His advice about 'stripping to the core' did reveal my problem for me, plus it cured a serious headache!! Furthermore i learned a valuable lesson about electricity ;) it had to cost me a lovely innocent PIC, but that's life, you know

Now, for the problem i'm facing :)

I'm having 1 core module and 1 SID module i want to be friends - running them from 1 PSU have only gotten me a fried PIC :( now i will try to use the C64 PSU instead (as it's really 2 PSUs) but i'm unsure how... I would like to avoid the optimized PSU solution by now because it would be too big an adventure for me at the stage i'm in.. i would just end up here with lame questions :) so to prevent that i came up with another idea;

Since the SID module is housing a 8580 SID it needs just 9V (which the C64 PSU can deliver)

The core module needs the usual 5V (which the C64 PSU can deliver)

...now the SID module still needs 5V, and here's my concern :) should i take the 5V from the C64 PSU to take off strain from the core? or is it important to take the 5V from the core because of the current, ac/dc or what it's called? (i'm very sorry :) besides not having english as my main language i very much lack a tecnical vocabulary too, so if any of this sounds far out then please tell me and i will try to explain in other terms)

Link to comment
Share on other sites

i almost didn't post this, because of your earlier comments about help, but in the interest of saving a few sids:

the c64 psu puts out 9vac not dc, you will zap your sids if you connect that to them.

it would be good if you assume there's a good reason for the recommended circuits, until you know enough to question why...

Link to comment
Share on other sites

i almost didn't post this, because of your earlier comments about help, but in the interest of saving a few sids:

Hmm, you're angry too... why? I'm sorry if i misinterpret some of the answers in this thread as patronizing, and i'm very happy to get your (very relevant) input.. but is it wrong to appreciate m00 for helping me?

the c64 psu puts out 9vac not dc, you will zap your sids if you connect that to them.

Thanks for great advice :) i thought the bridge rectifier on the SID module dealt with that matter actually?. I'm glad i asked

it would be good if you assume there's a good reason for the recommended circuits, until you know enough to question why...

Hmm, i'm not questioning recommended circuits? all i'm doing is trying to get myself up and running :) until then i can only read, experiment and ask questions to learn, right? It's not like i'm saying the optimized PSU is wrong! I'm saying it's too big a task for me to build yet, so all i'm doing is looking for an alternative :)

Link to comment
Share on other sites

i'm not mad, no worries, it's just unlikely for folks to help if you take a shot at someone...

yes as long as you stuff the power supply components on the sid module you will be fine with 9vac

remember not to do so on the core, though, since the c64 psu outputs 5vdc regulated already.  check this with a multimeter, though, these guys are prone to failure.  you should see 5vdc (or very close).  also check it under load

Link to comment
Share on other sites

i'm not mad, no worries, it's just unlikely for folks to help if you take a shot at someone...

Man, i never meant to be personal in any way :) i just so badly wished for an answer not involving what pin headers i've used.. I deeply apologize if anyone fealt it was to attack anyone. I even wrote how much pain Stryd's advice saved me from :)

remember not to do so on the core, though, since the c64 psu outputs 5vdc regulated already.  check this with a multimeter, though, these guys are prone to failure.  you should see 5vdc (or very close).  also check it under load

My C64 PSU gives good 5/9 luckily, but i tried to feed the core with the 5V and it only came out as 1.5V in the end, so i guess that the 9V to SID will also end up a lot lower. The 9VAC can easily run the core though, so i can use my wallwart to run the SID :) If this is a success i will be running fine.. i'm testing atm

That thing about bypassing the power supply components does intrigue me though ;) To do such a thing i would only need to connect 5VDC on the core where the voltage regulator usually puts out it's 5V, right???

Link to comment
Share on other sites

Hmm, this is really puzzling me... now i have the core and the SID running off different PSUs, still the vreg of the core gets pretty hot...

I started up the core and it gave me only one sysex request indicating that MIOS is allready installed, so i uploaded the 8580 hex file, and now i have the exact same symptom on this PIC also... as soon as i upoladed the 8580 hex it begins to spit out endless sysex strings :( i think it's rebooting every time because i hear a short 'beep' in the speakers...

Am i uploading the 8580 hex in a wrong way? is it my MIDI interface or what??? This is the second core with accurately the same kind of symptom after uploading the 8580 hex... only that this time i used separate PSUs to not overload anything...

I use MIOSStudio to upload both MIOS and the 8580 hex file

Link to comment
Share on other sites

So i did a serious amount of testing...

Both my cores act completely alike; When i upload MIOS i get a single sysex upload request upon startup, but if i upload the SIDv2 app my core begins to restart in an endless loop (if i connect a SID module it's giving me a single 'beep' every time and sending a sysex request)

If i upload MIOS again i return to the start, having normal behavior and only one sysex string at startup..

I tried to upload SID app with 1000ms delays to make it behave, but still no cigar :( can anyone advise me how to get the SID app uploaded without problems??

Link to comment
Share on other sites

+1 buying an LCD. It helps tremendously and, plus, you can actually get some feedback from the MB-SID so you have an idea what patch you are on, etc. It's extremely helpful. In fact, I'm inclined to say that new Midiboxers should get a CORE + LCD working before they even tackle the SID stuff.

Link to comment
Share on other sites

The tests i did was with bare cores only, both cores act the same way.

Also i get the same results wether i connect the C64 PSU 9VAC or my wallwart (around 12VDC) to the cores.

I suspect my Steinberg Midex3 MIDI interface to be the problem (though i can send all 291 packs of SID app without errors) i tried sending with 1000ms delay, but still same problem; when SID app is uploaded it resets continuously :(

If i re-upload MIOS i get a single sysex upload request as expected.

I have an LCD but i'm afraid i fried it :( was too impatient to hook it up... only 3rd time the backlight worked, and now it shows nothing except distorted squares in glimpses. I'm not expecting to get it 'to work but i'm still trying :) if i got the LCD to work i guess it would show the copyright message and then 'ready' in a glimpse before resetting and showing the copyright message over again...

Link to comment
Share on other sites

More long-time testing... apparently my cores works fine! I can upload MIOS, I can upload the test-tone app and the software loopback test program - everything with normal behavior from the cores. I can on/off the cores multiple times and i only get the expected sysex upload request string every time. The cores passes 5V checks and can handle both banksticks and SID modules (bankstick and SID module does not change the outcome but to be sure i do all testing with bare cores)

Then i upload the SID program, and it only works normal about half the times i on/off the core. The rest of the time it start to reboot constantly (if i connect a SID i can hear the startup-beep each time) and i get the sysex string (usually i can get to play a midi note or two here) and then it resets. I also start to receive strings like this '[FF] System Reset' sometimes when i turn off the power.

Now if i upload another program to the PIC, voila :) back to normal behavior!

The cores are two identical regular Smash cores (not connected offcourse :)) the above result is the same with both cores, also it makes no difference if i use the 9VAC from my C64 PSU or my wallwart to run the cores. I used both MIOSStudio and MIDI-OX to monitor MIDI activity (all uploading done with MIOSStudio, works great)

I'm getting really suspicious about my USB MIDI interface (it's also the one factor where i can not test other alternatives) it's a 'Steinberg Midex3' running from an XP laptop. I searched the forum and it looks like other users are doing fine with Midex.

I'm just about lost for ideas :) i hope to get a cheap MIDI interface tomorrow to test further..

Link to comment
Share on other sites

Might be a silly question, but I didn't see that it was mentioned yet - you are using a PIC18F4685 right? This one is required for SID v2, though I'm not sure what happens when you try to install the SID app onto an older PIC. Anyways you might want to verify your PIC model just to make sure (it's a quick check anyway).

Link to comment
Share on other sites

Yeh... very unlikely / almost impossible that a dodgy midi interface would cause the core to reboot (if the mios+app uploads were successful in smart mode) I thought maybe wrong PIC, but then the test tone app would not make any sound... or skipped components you thought you didn't need (once someone skipped the diode)... or maybe LCD wired for 8bit (don't think it's that if you have no LCD heh)

I know you tried two PSUs and it is unlikely that both have failed, but maybe it's some other component (cap, regulator, rectifier) causing a BOR or something... Try testing the voltage while it runs...does it stay at 5V the whole time even when it reboots? Does the regulator get hot?

You asked about the 5V before... yes, the core and SID modules should both get the same 5V power.

I dunno... If the power is really all OK, It's mighty strange that, while running only with core and sid modules in place, the test tone app works, and makes sound and all,  every time... and yet the SID app dies. That almost certainly rules out a hardware error, because the sid app doesn't do much with the hardware that the testtone app does not. We know it's not a software issue because it's well proven (I assume you haven't modified the code in any way and are using a precompiled hex file?)... That points at a bad PIC.

Personally I wouldn't cough up for a new midi interface at this point. Unless you need a good excuse to give the wifey or something :D

Link to comment
Share on other sites

Hi m00 and Stryd :) thx for feedback!

Yes, the two PICs are both PIC18F4685 from Smash - bootstraploader v1.2b, MIOS v1.9f

Both have IDs with only zeros '0000 0000 0000 0000'

I have stuffed all components from the kits onto the PCBs, following the silkscreen and the wonderfull help on Smash's site :) I would not dare to do it myself as i'm very inexperienced with components.... the LCD i'm working on again (i wonder if it ever worked at all) and you say it has to be done 4bit, this means that i should leave D0, D1, D2 and D3 unconnected, right ??

Now i might appear quite ignorant again :) but may i ask, what is BOR ?? the voltage stays at 5V also during it's endless reboot cycle, and the vreg only gets hot when it needs to deliver the 5V to SID module also - that's why i wanted to try and feed the SID directly with 5V instead of pulling it from the core.

You are right about me not tinkering with the code :) i fully rely on the precompiled hex-files i download here. At times the SID app does work, but after an on/off i get that '[FF] System Reset' sysex message and then the resetting begins. If i on/off a couple of times i can be lucky to make it work again. The strange thing is that neither testtone or software loopback apps sends '[FF] System Reset', only the SID app ...... i'm not saying that the '[FF] System Reset' is causing the error, i rather think that it's a symptom of the error, but to be honest i'm pretty clueless atm :)

That's also why i thought that it might be my MIDI interface.. i imagine that the SID app is much more demanding regarding MIDI than for example testtone, and that my interface just didn't cut it... it sounds lame, i know :)

I also thought of bad PICs as the problem, but they have suffered no harm here with me :) also both PICs and cores show the same symptoms...

It's not like i would love to pay for a new MIDI interface, hehe... i'd rather save the dough and go for a MOTU 8x8 once and for all, but it would be kinda silly because i have reserved lovely gm5x5x5 boards courtesy of nILS to place my gm5 SMDs on eventually... this will in general cover my MIDI needs for quite a few years ;)

Link to comment
Share on other sites

Sounds lame? No, this is a confusing issue. Definitely a weird one. But I don't believe the sid app causes any extra midi traffic.

A BOR is a Brown Out Reset - meaning, the PIC automatically reboots itself if the voltage drops too low. Unless it's a very brief dropout, your meter would have picked that up, so I guess that's not it.

There's an off-board component required for the CAN connection to the SID, which is the diode I referred to earlier. Skipping that, does cause this problem, IIRC. Check out the schems on ucapps to be sure on that one.

Anyway you seem to be doing everything right... I'm a bit stumped!

Link to comment
Share on other sites

Stryd, i'm really puzzled too.. thank you so much for helping me out!

I tried looking into that CAN/diode thing but i did not find it. I will look harder :)

.....but now for some sunshine at last :) i was blind, but now i see!!

I'm not quite sure what's up with this one :(

dispbad.jpg

It has really been like this all of the time, hard to work with...

But i finally got hold of another one :)

dispgood.jpg

now for some serious debugging ;)

Initially i can mention that i tested the new LCD without MIDI connected to the core, and i did not experience any reboots :) i will look into this tonight!

Link to comment
Share on other sites

First of all i want to thank everyone who helped me out! It's been quite a learning curve for me allready but it seems i'm beginning to move in the right direction, thanks to you guys inhere :)

The status on my 1core/1SID/LCD/1bankstick project is: 100% :) no more odd behavior, it just rocks on and on...

My problem was the regulator on the core :( having an LCD connected made it very clear - when i powered up i got the error allmost immediately, and i saw the LCD flicker in psychedelic patterns. Now the error was not periodic anymore, it was consequent :) I got rid of the regulator and slapped a bridge in it's place instead, but then the LCD did not run (i gave the core 5V from the C64 PSU, but some was swallowed by the rectifier). The trick was to bypass the rectifier also, connecting the 5V directly in the regulators place... now the core runs fine driving both the LCD and the SID module :) the SID gets powered by the C64 PSU 9VAC (going through both rectifier and regulator, but no heat problem there).... so finally my core, SID, LCD and bankstick is runing, and even using just the C64 PSU (something i had expected to learn in a year or so :)) This may be far from OPSU but it works okay for now with no 60Hz hum, but OPSU will definately be my next step so i can hook up another SID and enjoy the sweet priviledges of a stereo sound image ;)

This is how it looks like for now.... i think it will end up living inside an empty C64 case

mbsid2.jpg

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
 Share


×
×
  • Create New...