• Please review our updated Terms and Rules here

Modified Model 1 shows weird characters

That complete Row of keys is fed by A5 going "HIGH" making Z1 Pin 3 "HIGH" and Pin 4 "LOW". Notice PULLUP Resistors R1 thru R8 Pulling the D0 thru D7 Data Lines "HIGH".
When you Press a Key such as the 8 you make contact from A5 to D0 which pulls Z3 Pin 2 "LOW" making a "1" on D0.

8 = D0
9 = D1
: = D2
; = D3
, = D4
- = D5
. = D6
/ = D7

Keyboard

The TRS-8O Keyboard consists of 53 single-pole.
single-throw normally open keys molded in a
plastic base. The base is mounted, together with
four ICs and associated resistors, to the key-
board PCB. As you can see from the Schematic,
this Keyboard does not output ASCII. It is scan-
ned, like calculator-type keyboards. Each key
represents a switch across a matrix node. When
closed, the switch will short out a horizontal
line to a vertical line. ROM software will detect
the node short and generate ASCII equivalents
for that particular key.

The Keyboard is accessed by decoder signal
KYBD*. When this signal goes low, it enables
tristate buffss Z3 and Z4. The inputs to these
buffers are normally held high by the pull-up
resistors (at the top of the keyboard schematic)
R1 through R8. All of the horizontal address
lines are made to go high at the same time as
KYBD* goes low. If the CPU detects a logical
"1" on one of the data lines, there is a key
pressed on the Keyboard. The CPU ROM will
then scan the address lines one-by-one until it
finds the "1" output on the data bus again.
After finding it, the ROM can instruct the CPU
how to generate the ASCII code for that parti-
cular key. At this time, the CPU also checks the
status of the two shift keys. If neither of these
keys is pressed, the ASCII code is not modified,
If a shift key is pressed, the ASCII is modified
accordingly.

Only one point should be brought up about the
Keyboard. The inverters on the address lines are
open-collector types. You may not be able to
see the address signal on Z1 or Z2's output
unless one of the keys associated with that out-
put is pressed. With no key pressed, there is no
voltage applied to the KR (Keyboard Row)
lines. When a key is pressed, the associated pull-
up resistor supplies voltage. Then you will be
able to see activity on a KR line.
Click to expand...

It should be easy enough to remove the Z80 CPU, Pull A5 High through a 4.7K Resistor and troubleshoot Z1 Pins 3 &4 along with Z3 Pins 2 & 3.
KYBD* will also have to be "LOW". The remaining Keys are just a continuation of Z3 & Z4 Pins for D1 thru D7. The remaining keys just
progress thru D1 to D7.

It should be easy to troubleshoot to the specific node of the IC. It could also be a bad Solder Joint, or a defective ribbon in that flimsy
keyboard cable. But, either way, it should be easy to locate.

If when you depress the 8 key you are getting the following keys @XHP instead of the 8, there is still a problem with the A0-A7 Address lines when it's
trying to access the keyboard.


Larry

Keyboard.jpg
 
Last edited:
ldkraemer: You were right.

The connection of A5 was broken causing the issue. It's resoldered and the Tandy is fully back to a working state!
Code: 
10 I% = I% + 1 
20 goto 10

This little program counts from 0 to 11392 in one minute on normal speed. On high speed it counts up to 22137 in one minute, effectively 2x speedup :)

Next thursday I'll have my machine back and I can start playing with the tons of cassette software I have for the Model 1 :D

Thanks all for advising and helping!
 


Yes, that's my Tandy's mainboard on the right side and the picture shown is the picture of that mainboard.. It's aliiiiive!

But, while the left guy was checking the Expansion Interface's powersupply for reasons why the Tandy broke down in the first time (because all the IC's broken, came directly from the bus). After some extensive measuring from the left guy, he declared the power supply not the issue and everything looked fine from there. So, we connected the keyboard to the Expansion Interface and after a brief 'hey, where's my diskette', the Tandy went into lockup showing a screen full of @. Disconnecting the EI and everything was fine. So, there's still a problem in there.

So, the working Tandy went home again with the guy on the right (who did all the repairs so far), together with the EI and he's going to fix the EI this weekend. One of the IC's near the EI connector has probably a broken port (keeping the READ and WRITE signals both to low and creating a very out of band impedance on one of the pins of the EI-connector).
 
WOW! A Model 1 Monitor with a 5 1/4" Floppy drive inserted on the right side. Never in my life have I seen one like that before!

Good Idea.

You know you will never be able to repay these GURU's for all their work & time spent.

Larry
 
ldkraemer said:
WOW! A Model 1 Monitor with a 5 1/4" Floppy drive inserted on the right side. Never in my life have I seen one like that before!

Good Idea.
Click to expand...

Not sure if it'll work just yet. As far as I could see it gets the power from the monitor itself.

You know you will never be able to repay these GURU's for all their work & time spent.

Larry
Click to expand...

That's also one of my concerns, tough they do like the challenge of getting this to work. They are describing everything to me what they do and how they do it. I also want to learn to be sure to be able to support these machines for at least the next 40 years. Plus this club currently has 5 paying members which they need to bring up to at least 25, so I took that task on me to take the club to the next level. I'm more of the social media person who attends all kinds of retro fairs, so I have the contacts to get the club bigger.

The club can be found on http://www.trs80.nl and has some very interesting articles about the TRS80 in both English as Dutch (but Google Translate will help you there).

So, in that way I will pay them back. Taking their knowledge to the next generation and preserving these machines.

Funny enough, the guy on the right of the photo (who created a selector and high speed mod back in the late 70s) gave me a complete (boxed) Tandy TRS80 Pocket Computer 1, including tape drive, docking and a docking with printer. :D
 
It took a while, but the machine is back into it's fully working state:



As you may (or not) notice, the floppy drive has been removed from the monitor. This never could've worked (or worked stable) in the first place. The monitor (which is just an ordinary TV tube with the channel selection removed, in this case replace by a floppy drive) has no insulation inside whatsoever. The drive's head (and motor, it's a direct driven drive) was pretty much next to the main voltage regulator and the tube (a space around 5 cm in between). It didn't do anything (power cord was cut off and never could've come from the monitor since that's all 220v stuff and not 12v/5v). Even with an external power supply, the FDD didn't show life.

Yesterday we removed the drive from the monitor and attached an Amiga 2000 power supply to it and immediately worked. This is being confirmed when I put the power supply next to the drive, the drive refuses to read (motor doesn't spin even), but when I put it like on the picture above, it works. Gotta love old technique right?

Anyway, the machine is working without any problems. For those who are interested, The full description of the repair is below. This is the description of Rence Rozendaal who actually created quite some mods for the Tandy (and they are in this Tandy as well).

I found evidence on the floppies that this machine is used up until 2002 (!), at least backups were made in 2002.

I'm one happy guy to have the honor to have this late 1977 machine and show it to everyone over here.


Tandy TRS-80 Model-I and Expansion Interface Repair
Spring 2016

Upgraded Keyboard:
- Level II Basic (3-ROM set + 74LS42) - Cassette Read Improvement
- Lower case
- 64 kb RAM in Keyboard
- High Resolution Graphics
- Two times Speed-Up using 5.3 Mhz Z80-B and wo Wait states (PCB in Z80 Socket) - Selector (stacked on Speed-Up PCB in Z80 socket)

Upgraded Expansion Interface:
- (Original) RS232 interface
- Double density with Single/Double density boot switch - 512 kb RAM
The Selector is used to address the total amount of 576kb RAM and to support CP/M. Under Newdos the RAM can be used as Ramdisk, Printer spooler and text buffer in LeScript
The power supply had been replaced using a single transformer with at the secondary side two times 10V, i.e., center tapped 20V. A voltage doubler circuit is used to obtain +12V.

Problem
Several ic’s in the keyboard and in the Expansion Interface had been broken, presumably as a result of a high current through the cable connecting keyboard and expansion interface. This high current could have resulted from switching (off and) on the keyboard with power supplied to the Expansion Interface. Doing so, would add an additional 2200 uF capacitor of the 12V circuitry in the keyboard in parallel to two 2200 uF capacitors in the Expansion Interface (one in the Expansion Interface and one on the replaced power supply).
In this event, the current loading the 2200 uF capacitor in the keyboard may be very high.
The following ic’s had been broken (9 in keyboard and 4 in Expansion Interface):
- OUT* (keyboard Z22 and Z25, LS367 and LS32, respectively)
- CAS* (keyboard Z67 and Z72, both LS367)
- A10 (keyboard Z36, Z38, Z51 and Z52, LS32. LS367, LS157 and LS04, respectively) - Wait* (Speed-Up print, LS367)
- A12 (Expansion Interface, Z44, LS244)
- A6 (Expansion Interface, Z45, LS244 output side sticks to zero level)
- RD* and Write* (Expansion Interface, Z30, LS243)
- CAS* (Expansion Interface, Z38, LS243)

A total of 10 ic’s were replaced. Of the other 3 ic’s unused ports were used to restore function (Speed-Up Wait LS367 and Expansion Interface Z30 and Z38, two LS243's).
One could wish to replace the remaining four LS367's (Z39, Z55, Z75 and Z76, i.e., A4-A9 and Data in/out) that output to the expansion connector of the Keyboard.
Additional changes:

In general, RAM in the Expansion Interface, and especially in case of a Speed-U,p is error prone. Probably to reduce RAM errors, the RAS, MUX and CAS signals had been loaded with 1k pull-up to +5V and 220 ohm pull-down to ground.
In addition, a 220pF capacitor was added to MUX (i.e., between MUX and Ground).
Low resistance loads are well known of the SCSI-bus, in which case the load consists of 220 ohm pull-up and 330 ohm pull-down. A 220/330 ohm load results in a logic high level (3 V). So a 220/330 ohm load may be driven by LS-TTL.
However, an LS367 is not specified for driving a 1000/220 ohm load. Two options could be considered:
- change the 1000/220 ohm load to 220/330 ohm
- change the driver

In order not to interfere with the load of 1000/220 ohm, of which can be assumed that it has resulted in a reliable TRS-80 Model-I with 576 kb RAM, a 74HCT367 was used to replace the 74LS367. The HCT367 may supply up to 35 mA per output (compared to 2.6 mA for an LS367). To obtain a logic high level with a pull down load of 220 ohm would require at least 11 mA (220 ohm 11 mA is 2.4 V). A 20 mA current would give about 4.5 V logic high level.
The 220pF load in MUX was simply removed...
 
Al Hartman said:
I wish I had guys like that near me in the U.S. - My Model I is dead, and I'd like to get it fixed.
Click to expand...

Just come over here and take the Model 1 with you, it'll be repaired. You can visit my personal museum, the big computer museum and all the beautiful places to see in NL :)
 
@whizzi glad it is all working now. You have some very talented friends over there who really know the Model 1 in and out. Ok now I want a speed up board too!:)
 
Al Hartman said:
I wish I had guys like that near me in the U.S. - My Model I is dead, and I'd like to get it fixed.
Click to expand...

Interesting thought. Is there anyone in the U.S. with the kind of mad TRS-80 skills that we have seen in this thread? If so, let yourself be known!
 
@Whizzi do you have the lower case mod in your computer? I think you did mention it was installed. If you use LDOS 5.3.1 you will see the lower case letters. See my screen picture.
 

Attachments

  • 100_1800.jpg
    100_1800.jpg
    102.3 KB · Views: 1
Yeh, I have the Lowercase mod installed. It's one of the few things I haven't tested out yet. I don't have LDOS disks here is the main issue.

I recently got myself an old P4 machine with network-card where I can put some 5,25" drives in and hopefully I'll be able to write some disks, including LDOS.
 
The Pentium system has a Adaptec SCSI controller in it. I don't use the SCSI portion only the floppy controller part cause it can handle Single Density.
And I use an emulator to make the disks.

I think any 486 system should be a good choice. Just need to have ISA slots for the Adaptec. IMO!
 
You must log in or register to reply here.
Back
Top