Woo hoo - Collection Completed!! TRS-80 inbound

[rubber_jonnie]

Looks like an address error. Check for solder blobs or the ribbon cable

The only ribbon cable is for the keyboard and I know that's ok as all the keys do actually work.

It certainly looks like an address error, seems like the video ram is being duplicated, I'll take a close look for any solder blobs tomorrow though, eyes are fading now.

It's currently in the state where there is 5v on pins 8 and 9 and does exactly the same thing as when I revert and put in the old RAM that was was working before, which makes me think something new has failed if the machine boots, can run programs and shows the correct amount of RAM when I do PRINT MEM.

Might be worth a quick count to see how many characters its shifted by.

On a side note, I have tested all my spare 41256s and they are all fine, so I have lots of known good RAM.

[Tomswork]

Time to download adrean blacks test rom



Tom

[rubber_jonnie]

Time to download adrean blacks test rom



Tom

Maybe...

[rubber_jonnie]

So I've been poring over the schematics for the Model 1 and it occurred to me that if I remove all the RAM it will let me see if the video RAM is working as expected, because the machine boots with garbage, which is subsequently cleared when the machine initialises.

No RAM means the Z80 has no workspace, and can't initialise, and the screen won't clear of garbage. So RAM out and power on, sure enough all of the screen is full with characters. So the video RAM looks to be fine.

Replacing the RAM and I get the strange doubling up of characters, and further to my earlier assertion that all the keys work, they don't, or at least not obviously.

Since I know the RAM ICs are good (I also need to get the board under the microscope to check for solder blobs and whiskers), a look at the schematics shows a pair of 74LS157s that deal with addressing, so I have ordered some replacements given that I have already had two other 74LS ICs fail on this machine.

I'll pull the originals and test in the TL866 Pro and fit sockets. Unless of course I find a solder bride first :)

[stephen_usher]

OK, the display is wrapping/duplicating every 8 characters. This suggests bit 4 is stuck when accessing the display sub-system. This suggests that Z65 (74LS157) may be the culprit as that deals with A0-A3, with Z49 buffering A4-A5 and some other signals and Z31 doing the rest plus A8 and A9.

The reasoning is that:

• Main RAM and ROM access is working correctly so the main address and data buses can't be broken.
• All display is buffered through those 74LS157s. The reason I mentioned a ribbon cable is that on another model of TRS-80 these signals are routed from one part of the board to another over a short ribbon cable as the symptoms I've seen in repair videos echo what you're seeing.
• I think your keyboard issue is a red herring as that's using a different sub-system hanging off the address bus separately.

[rubber_jonnie]

OK, the display is wrapping/duplicating every 8 characters. This suggests bit 4 is stuck when accessing the display sub-system. This suggests that Z65 (74LS157) may be the culprit as that deals with A0-A3, with Z49 buffering A4-A5 and some other signals and Z31 doing the rest plus A8 and A9.

The reasoning is that:

• Main RAM and ROM access is working correctly so the main address and data buses can't be broken.

I keep swinging both ways. I did some scoping before lunch and it does make sense. The machine boots, I can see the prompt (Albeit twice), I can run "PRINT MEM" and get the correct result for 16K, so I've changed my mind, and agree it must be video RAM now, despite the display looking correctly full of garbage when booting with no RAM.

[*] All display is buffered through those 74LS157s. The reason I mentioned a ribbon cable is that on another model of TRS-80 these signals are routed from one part of the board to another over a short ribbon cable as the symptoms I've seen in repair videos echo what you're seeing.

Yep, I've seen that cable on some of Adrian Black's TRS-80 repair videos, and it looks horrendous, though should be easy enough to replicate if you have one. I think it's the Model 3 that has it.

I have some video RAM on order and some

[*] I think your keyboard issue is a red herring as that's using a different sub-system hanging off the address bus separately.
[/list]

Yep, I agree 100% with that.

I think this afternoon I'll print out the 2102 datasheet and do a spot of scoping, see whats what. This machine has had the lower case mod done (By Tandy I think, it has an official label) to add an 8th bit to the video RAM in piggyback fasion. I'll check that is still wired as expected.

[rubber_jonnie]

Ok, I've had a bit of a probe around the video RAM and associated logic and it looks like A3 is not well, so it's a good job there are some 74LS157s on the way.

Bloody annoying that every time I need something 74LS wise, I don't have any in my parts drawer. I feel like I have almost every bloody 74LS made, yet never have the one I need!!

[PhilC]

Ok, I've had a bit of a probe around the video RAM and associated logic and it looks like A3 is not well, so it's a good job there are some 74LS157s on the way.

Bloody annoying that every time I need something 74LS wise, I don't have any in my parts drawer. I feel like I have almost every bloody 74LS made, yet never have the one I need!!

Know the feeling well, my collection of 74 series chips is starting to grown both in DIP and SMT format. Hopefully that'll be the end of the TRS-80 saga now :D

[rubber_jonnie]

Ok, I've had a bit of a probe around the video RAM and associated logic and it looks like A3 is not well, so it's a good job there are some 74LS157s on the way.

Bloody annoying that every time I need something 74LS wise, I don't have any in my parts drawer. I feel like I have almost every bloody 74LS made, yet never have the one I need!!

Know the feeling well, my collection of 74 series chips is starting to grown both in DIP and SMT format. Hopefully that'll be the end of the TRS-80 saga now :D

Fingers crossed, yes :)

Though, I am only half way through the 64k upgrade ATM, so there may be more shenanigans to come :)

[rubber_jonnie]

Ok, so the 74LS157 at location Z64 that deals with A3 has been removed and socketed, the problem still remains (Surprise surprise !) , but it's given me the opportunity to test the IC and see what's going on.

This was the result:

PXL_20220927_093100516.jpg

So it's dead Jim...

Unfortunately the post has been today and no LS157s arrived, but I'm at least ready for when they do. Video RAM is also on the way in case one of those has failed. They weren't in the post either...