• Please review our updated Terms and Rules here

Commodore CBM 8032 video display problem

So glad I waited before RAMdomly removing chips possibly the ROM way! :)
Click to expand...

Nothing like a bad pun to brighten your day! :D

EDIT: I'm guessing c000 and a000 are suspect?
Click to expand...

Most of this conversation has been over my head, but I did some checking and found that a000 should be UD11. Is that correct guys? If so, then that can be ignored, because UD11 is an option ROM, which is empty. There's definitely supposed to be something at c000. I checked on my 8032 and there's code there. But being a newbie, I haven't been able to find which ROM corresponds to C000 yet, and I have to go to work!
 
Last edited:
Juleshsmith said:
EDIT: I'm guessing c000 and a000 are suspect?
Click to expand...

Jules,
The A000 socket (UD11) is unused so those readings are OK. So right now the main suspect is the "C" ROM UD9. Send me your mailing address using the private message system in this forum and I will ship you a replacement 2532 EPROM.


Also let's check some RAM by displaying data at 1000:
.m 1000 100f


Then using the cursor controls move the cursor over the first byte displayed and change it to 55. Change the second byte to aa. Then display them again and see if the data reads correctly.
 
Also let's check some RAM by displaying data at 1000:
.m 1000 100f

Then using the cursor controls move the cursor over the first byte displayed and change it to 55. Change the second byte to aa. Then display them again and see if the data reads correctly.
Click to expand...

Interesting. I assume that using 55 (0101 0101) and aa (1010 1010) are to check that each bit "stays put"?
 
KevinO said:
Interesting. I assume that using 55 (0101 0101) and aa (1010 1010) are to check that each bit "stays put"?
Click to expand...

Hi Kevin,
Yes, this is just a quick test to verify that each bit works in both the 1 and 0 states. Note that if somehow D0 and D2 were shorted together, this test would still pass. A comprehensive RAM test would incorporate many patterns such as a 'walking ones' test, etc to catch all types of addressing and data problems.
-Dave
 
jac_goudsmit said:
PS: I just thought of something: I (and others) wrote the commands in upper case (force of habit). But the 8032 initializes in lower case character set mode. I'm not sure if the MLM is case sensitive but you should enter the commands without using the shift keys, regardless of whether you see upper or lower case on the screen. I wonder if that's why the M command didn't work :)
Click to expand...

Jac,
Good catch by you and Eudimorphodon on the problem of upper case commands on the 8032. The manual for the Machine Language Monitor calls out capitol letters, but that is for the normal graphics keyboard. The business keyboard of the 8032 must use lower case. I was using upper case in my examples even though I have an 8032! Jules is well on his way to fixing his beloved PET.
-Dave
 
Juleshsmith said:
As requested Eudimorphodon!

So glad I waited before RAMdomly removing chips possibly the ROM way! :)

View attachment 8819View attachment 8820

EDIT: I'm guessing c000 and a000 are suspect?
Click to expand...

Defintely the C000 rom is suspect (there's still a possibility of a bad contact or borked address decoder but swapping the rom for a known working one should tell you).

The A000 data is irrelevant: That's usually where option ROMs and user ROMs go such as Simon's Basic (they usually are started with "SYS40960"). You probably don't have anything installed there, or if there IS something, it's not working but it won't harm normal operation.

===Jac
 
The dump of A000 was just a "sanity check". (I figured chances were that socket was empty, thus a dump of that region would show what an "empty" socket would look like on the 8032's physical hardware, something I can't verify right now. A dump from VICE might be misleading.) Comparing your dumps of d000 and b000 to VICE's results confirms that at least the first 48 bytes of those two ROMs are fine, so chances are pretty good that your problem is indeed limited to just the C000 ROM/socket.

Assuming your motherboard is the "Universal" you are indeed looking at ROM UD9 as your culprit. I'm guessing said chip is *not* socketed? If you have an oscilloscope (or a logic probe, perhaps?) you could try probing... pin 20? on the IC while running the command:

"m c000 cfff"

in the monitor. Doing so should cause a lot of action on the CS1 chip select line on the ROM. If you can verify you see activity there it would *almost* rule out anything besides the ROM chip itself. (If you don't have an oscilloscope or logic probe you could also, just for the heck of it, check for continuity between UD9 pin 20 and... UE12 pin 14. That's where the select signal comes from. A break in that line could also cause your problems.)

I wouldn't try piggybacking that chip. ROMs of the era these machines were built are notoriously fragile. Unless you're able to do the chip select test above you're probably about to the point where you should prepare to desolder UD9 and replace it with a socket.
 
Last edited:
MikeS said:
Looks like we have a new guru in the bunch? Hi Jac!
Click to expand...

Heh Thanks MikeS!

In the last few years when I became interested in retro computing, I've been studying a lot of schematics and listings and noticing how simple those old computers really were. The old Commodores are my favorites because my first computing experiences were with the PET-2001, CBM-4032 and 8032, and they don't have any custom chips like the VIC-20 or C64. I'm working on a project called Propeddle that started out as a replica for those early Commodores but is now a Software Defined 6502 Computer project in kit form.

I wouldn't call myself an expert on old computer repair: I just noticed how obvious this particular problem is. If the problem would be in the monitor, I'd be completely lost because I'm too scared of High Voltage. But when problems are this easy, I'll do my best to help!

===Jac
 
Last edited:
Eudimorphodon said:
The dump of A000 was just a "sanity check". (I figured chances were that socket was empty
Click to expand...

I have seen that before; a read from an empty socket seems to include the upper address in the data. It may be noise coupling from the address bus onto the tri-stated data bus.

I wouldn't try piggybacking that chip.
Click to expand...

Agreed; trying to align a 24 pin chip with your fingers might lead to shorting of adjacent pins.
 
So great to wake up to these awesome (and exciting) posts guys! It's Saturday here, so have the whole day to work on my beloved PET (and I have a feeling we are really, really close!)

Dave, thx so much for the offer of sending me 2532 EPROM. Fortunately I don't think I'll need it as I'll take the UD9 chip out of my other DOA PET (alas my good PET isn't socketed at UD9 but my other one is!).

Out of interested. UD9 has a 901465-20 chip, is that the same as "2532 EPROM" (I really am a newbie!). And... I presume I'd have to use specifically the chip from UD9 from my DOA PET as apposed to any of the other 901465-20 chips?

Guys, thx once again for joining forces and helping me! Some really awesome advice there (esp some great calls from relative newbies like Jac to gurus like Eudimorphodon!). I think this could be the day...

Gotta go out get some milk, then I'll start de-soldering, socketing and posting updates on 'our' patient's progress :)
 
Juleshsmith said:
UD9 has a 901465-20 chip, is that the same as "2532 EPROM" (I really am a newbie!). And... I presume I'd have to use specifically the chip from UD9 from my DOA PET as apposed to any of the other 901465-20 chips?
Click to expand...

The C ROM (UD9) has the specific part number of 901465-20. I was going to send you a copy programmed into a 2532 EPROM which is a pin for pin compatible part. The UD9 part from your DOA PET should do fine assuming it is a functional part. You cannot use any other ROM ( UD6,7,8 ) in the UD9 socket. Good luck on the unsoldering; you are doing a great job so far.
-Dave
 
GENTLEMEN...

Fixed1.jpgFixed2.jpgFixed3.jpg

This is indeed a fine day!

I can't thank you all enough for your invaluable contribution. You have guided me through this superbly with absolute spot on precision. Considering I knew nothing on how to fix an 8032 less than a week ago, to fixing not one, but two major issues feels extremely rewarding. So glad I bought the faulty 8032, I feel like I really have an understanding on at least a small percentage of what's 'under the hood'!

I hope the forum we've created will help others in the future as other forums have helped me.

Couple of question before I close 'official proceedings'.

1) Going to use the CN2 tape player to load software. I've heard you can download? software for the 8032, is this correct? If so, which sites would you recommend and, how would I loadl them via the tape player? I remember in the old days, I used to use a tape to tape player to copy software tapes. Are there 'sound bites' of programs I can record onto a tape then load?

UPDATE: Found this forum on software!

http://www.vintage-computer.com/vcforum/archive/index.php/t-22252.html

2) I have a 4040 which I got with my DOA 8032. Started cleaning it up (really really dusty, dirty) and plugged it in to the mains (not the computer yet). Churned away for a few minutes, then a puff of smoke came out from behind the transformer. I'm thinking it's that voltage spike regulator component/ resistor thing that's notorious for shorting with age. Apparently it's OK just to remove it and rewire. My only concern is plugging it into to my beloved now working 8032. Do I run any risk of harming the 8032 if the 4040 is faulty, is it mainly just data (as apposed to damaging voltages) that's shared between the two?
3) And finally, if I need to problem solve the 4040, I presume I would start a new forum thread?

Thx again guys!!! Jules :lol:
 
Last edited:
Congratulations! That's fantastic to see those screens. Collectively we saved two PETS in only a week or so. My 4016 was last week, and yours this week. I too have learned a bunch, and hope to learn more.

1. I am going to let someone else answer about the C2N. I've been using C64s for years, and I've had several cassette players, and I have NEVER ONCE loaded a file from one, or even plugged it in. Always been a disk drive guy!

2. I too have a 4040 that actually works. My system is to download D64 images to an SD card and put them in my uIEC drive on my 128D and then use CBM-Commander to make them into real disks on the internal 1571. I found an archive somewhere (can't find it at the moment) with a bunch of PET games, and I've made one disk already. I was playing Lunar Lander just an hour ago! Obviously my method requires quite a bit more hardware that you probably don't have. There's also a user here named Gubbish who sells a cool disk product called PETdisk. I've ordered one, but I don't have it yet. It plugs right into the IEEE-488 in the back of your PET, and it stores programs on a Micro-SD card. There's a whole thread on it, if you care to look it up.

Although there might be more wrong with your 4040, I would agree. It sounds like an input filter cap or choke shorted. I'm not seeing that in my schematic. While I wouldn't bypass those permanently, you probably can for testing. Just remember that you're dealing with line voltage, (I guess you call it 'mains') and have to be careful. I think once you fix that, and if you can turn it on and the disks spin up for a few seconds and stop, it should be safe to plug it into your PET.

If you decide to work on the 4040, start a new thread. It helps people to find what they're looking for. I've got a 2040 that needs work, so when I finally get to it, I'd like to be able to look up your thread! A friend gave it to me two weeks ago with a DOA 8032! That's going to be a project for another time.

Again, congratulations on your excellent work.
 
Thank you on your excellent work Kevin!!

I've got a bunch of C64 tapes I've been trying with no luck earlier today (would load them on, but nothing came up with 'run' or 'list')... so went back to my trusted friend Google and found out why I had an issue!

http://www.portcommodore.com/dokuwi...ommodore:pet:pet_faq-cross-platform_emulation

Just about to test the first program I've loaded using the MLM mods. I'm sure this is probably common knowledge to most PET owners... but for me this journey is only the start of many other new journeys with my 'new' 8032!

Yeh, we call it 'mains' here (240v as apposed to your 110v). I'm presuming it has a filter cap/choke as that was the first thing to go on my DOA 8032. It's in the area between the line cable and the transformer (not on the PCB fortunately), so thinking it must be a cap/choke. Haven't opened that part of it up yet to check. Think I might play around with the 8032 for a few days before starting on the 4040.

Let me know how PETdisk works out for you, sounds like the way to go!

Til next time... :cool:
 
Here's some convoluted Basic to try on your working 8032:

Code: 
10c=39
20readi:ifi>=0then50
30c=c-i:ifc=91thenprintr$:end
40goto20
50iflen(r$)<=ithenr$=r$+" ":goto50
60a$=left$(r$,i)
70r$=a$+chr$(c)+right$(r$,len(r$)-i-1))
80goto20
100data3,-28,10,-2,5,8,13,-7,9,-1,12,-2,1,11,-3,4,-3,2,-2,7,-2,0,-2

(By the way this also marks the first time that I'm manually copying code FROM a Commodore screen to other media, instead of the other way around ;-))

===Jac
 
Here's some convoluted Basic to try on your working 8032:
Click to expand...

Hey Jac. First time I've entered basic into a computer in over 25 years! Was a real hoot, but I'm really rusty, and alas have got my first "syntax error" (oh, that brings back memories!)

Jac's prog.jpg

Silly question, should I have spaces between everything or type it out exactly how you've done?

BTW, been trying to record to tape converted (to WAV) PET .TAP files.

http://www.classic-computers.org.nz/blog/2011-02-27-pet-programs-for-download.htm

Haven't had success. I've learned that it could be the filtering on my sound card??

My next step is to put the WAV files on my other computer (I use for sound recording) which has got a high end audio interface... or alternatively, putting them on CD (nero burn) then recording the CD to tape with a better, metered tape recorder.

Anyone else had luck with this? Any suggestions?
 
dave_m said:
I do not know what the line is trying to do but there is an extra closed parenthesis at the end.
Click to expand...
Yup. Just paste it into VICE if you want to save yourself some typing...

Can you (Jules) load programs from tape that you've saved yourself?
 
MikeS said:
Yup. Just paste it into VICE if you want to save yourself some typing...
Click to expand...
Mike,
Nice trick with cut & paste into VICE.

The problem with code turns out to be a missing Open Prens just before 'len'.

The data statements don't look like valid ASCII characters but apparently are good PETSCII stuff.
 
You must log in or register to reply here.
Back
Top