• Please review our updated Terms and Rules here

TRS80 Model 1 cable or RAM

cavaughan

Member
Joined
Mar 30, 2021
Messages
11
Became a proud owner of a TRS-80 Model 1. YEAH! Of course there is an issue. Some of the keys do not print the correct letters. For example, E comes out as G. This is not just in the keyboard, but also when an Error message is produced it is spelt GRROR.
One possibility is that it's the keyboard flex cable. But that wouldn't explain the E <> G switch for messages, I think. So the other possibility is that it is the video RAM.
First question is whether anyone has any other thoughts on the problem. But whereas it seems to me that it is more than likely the video RAM, does anyone have any detailed description on how to go about diagnosing and repairing the RAM. Note, I am very much a neophyte as far as electronics are concerned. I am trying to learn though.... I am fine at soldering, I just can't really read a schematics, etc.

Thanks for any help in advance!
 
According to the manual, Z11 is the faulty ROM if characters appear wrong. It would seem that I might also want to check out Z8, 29, 36 connections.

Also...
I think I have a very early board maybe? For one the CPU LOGIC PCB in the manual is not the same as mine. In the manual it also says to look for a printed sign on the circuit board with TEC and other information. I don't see any such sign. Finally, it came with documentation, "Preliminary User's Manual," which is just xeroxed and stapled together and the first page of which states, "we decided not to delay your shipment any further ... and to send you this Preliminary Manual (along with some game tapes and other materials)."
 
The only manual I could find was the Japanese version, which is not in Japanese. But maybe that's the problem. Nothing is looking the same. Where can I get the right service manual?
 
Is data entered at the keyboard "correct", just displayed incorrectly? IE, if you type: `print "hello "` does it work, just displaying the wrong characters?

If it's just the video that's incorrect then translating "E" to "G" could be a stuck second bit. (IE, "BIT 1" on the schematic, they start at zero.) Here's a chart of ASCII values to their binary equivalent If this bit is stuck as a one then that chart can clue you in what the substitutions will be; add 2 to every letter that has a zero in the second least significant position. (A will be C, D will be F, E turns into G, etc.) Bit 1 is chip Z47.
 
For the service manual for the TRS-80 Model 1, goto http://cpmarchives.classiccmp.org/trs80/Library/Manuals/Hardware/ The manual is called TRS-80 Micro Computer Technical Reference Handbook 1st (1978)(Radio Shack).. There are two versions i.e. 1st and second. I think the schematic's are the same in both, but the 2nd edition includes some extra information on level ii basic..
You should also get the basic reference manual, level i or ii at the same location.. They are handy to have if you need the ascii codes and binary values..
 
so in response to Eudimorphodon.
When I type PRINT "HELLO", yes it does respond with HELLO and then READY. Of course given the transposition of letters I had to map out how each key is generated to figure it out.
Here is the mapping. I'm just providing letters for now. Numbers and symbols also have varying mappings. Now to crack the ASCII values and see if there is any rhyme or reason, but still don't know what exactly I should look at to fix this issue.

A - C
B - B
C - C
D - F
E - G
F - F
G - G
H - J
I - K
J - J
K - K
L - N
M - O
N - N
O - O
P - R
Q - S
R - R
S - S
T - V
U - W
V - V
W - W
X - Z
Y - [up arrow]
Z - Z
 
Ok, I see instantly that every other 2 letters are being substituted. Each of these have several things in common, but in particular the 2nd binary digit is 0 and their getting transposed to the letter for which that digit is a 1. I think is exactly what Eudimorphodon was saying. So Z47 is the guilty party?
 
I assume I have to replace Z47. What's the best source for this IC?

If my read of the the ridiculous six-page taped-together printout I made of the schematic in the original TRS-80 service manual are correct then, yeah, Z47 is the SRAM that holds the bit that would cause the symptoms you're seeing. On the bright side 2102s are not hard to find as ICs that ancient go. Even Jameco has them.

When you get the computer open it's possible you may find a weird situation like, for instance, maybe someone added lowercase to the machine and chose to piggyback the necessary additional RAM chip on top of Z47, and something has gone bad with their soldering that's fouling things up. My "new" Model I is full of homebrew hackery that makes tinkering around inside it exciting.
 
cavaughan
A = 0x41
B = 0x42
C = 0x43
D = 0x44
E = 0x45
F = 0x46

A = 0x41 = 0100 0001
----------C = 0100 0011

D = 0x44 = 0100 0100
----------F = 0100 0110

E = 0x45 = 0100 0101
---------G = 0100 0111

So, it's looking like Bit 1 (right to left 0..7) is the culprit.

Larry
 
Back
Top