• Please review our updated Terms and Rules here

Plus/4 - Incorrect Characters Printed

D

DrAM19

Experienced Member
Joined
Mar 5, 2021
Messages
76
Location
Near Washington, D.C., USA
Hello Everyone,

I recently came into a Plus/4 that I have nearly working. I have a few days where I cannot troubleshoot more, so I figured I would post to the board, and see if you all have a better idea than mine for what to test next.

The Plus/4 boots to BASIC v3.5 just fine. When I type in a print statement, such as PRINT"PLUS/4", the text I type in displays correctly, but when I run the print command, all of the characters that are printed are replaced by the PETSCII character mapped to that key (for example, the S is a heart). I then noticed that the number printed in the BYTES FREE statement has the same problem, but the rest of the characters in the COMMODORE BASIC splash are all correct.

Based on this, I think the BASIC ROM is good and the Character ROM too. I am thinking there is a RAM problem (less likely), or an issue with one of the 74LS257 MUXes (more likely). What do you think?
 
(ClassHasClass, thank you for responding with your input. I appreciate it)

Finally got around to a bit of testing.

Piggybacking the MUXes got me the correct characters printed in the BYTES FREE statement, so I felt good about desoldering and replacing them. Once done, I however found no change in the output characters. I did find that while tapping on the chips and hitting the reset button, it would output different characters to the numbers in the BYTES FREE statement. Clearly the act of messing with the board to piggyback the MUXes caused it to change behavior, not the MUXes themselves.

Piggybacking RAM changed nothing, still that same odd behavior and changes in characters when messing with the board while resetting the system. Further messing with reset showed me that while the system has a preferred state of characters out, it will sometimes correct a few or all of the characters in the BYTES FREE. I tried more percussive tests on the board to see if maybe there's a loose connection or bad trace, but there was nothing reliable I could do to make the it behave, so its probably totally coincidental.

I started looking for a dead test equivalent for this series of machines, and found the Diag264 by Rob Clarke. I burned the NTSC Kernal to an EPROM and replaced U24 to test. It ran, and failed on BANK LOGIC (and HIGH RAM, but I bet it cannot pass HIGH RAM if BANK LOGIC fails). It specifically told me there may be a TED issue, since it had a problem with $FF3E. Unfortunately I do not have a functioning TED chip on hand to test as a replacement. All of the screen color and sound tests at the end of the Diag264 test looked good, so this must be a very slightly misfunctioning TED, if indeed that is the cause.

Now, before I go and buy another TED chip to try as a replacement, is there anything else I should be looking for?
 
Well, ClassicHasClass, you are correct. I moved the TED from this Plus/4 into an untested C16 that is giving a black screen, and that C16 now works perfectly. Lesson learned, don't trust the tools too much.

I'd love to know which address line it used to test $FF3E, but that's probably a bridge too far.

I'll focus hard on the address lines for the next while, for my tests.

I'm going to do continuity tests on all of the address lines tonight, as well as a careful visual inspection with a magnifying glass. That should give me a good idea on if its a board issue.

I don't have an oscilloscope, but I do have two logic probes that can also act as a toner, so hopefully I can find the problem that way.
 
Well, you were right ClassHasClass. I checked all the address lines and didn't find a problem. So in my frustration, I just desoldered all of the RAM (ripping one trace in the process =( so I did a bodge wire on the backside) and installed sockets. I inserted some 4164P-15s I had on hand (the board had 4264-15s originally) and now the BYTES FREE statement on boot looks good, and survives through some resets.

Thanks so much for your advice!!
 
You must log in or register to reply here.
Back
Top