Image Map Image Map
Results 1 to 6 of 6

Thread: Model 1 ROM 1.3 Upgrade with 2732 EPROMs

  1. #1

    Default Model 1 ROM 1.3 Upgrade with 2732 EPROMs

    Hi guys,

    looking to replace the 3 ROMs (A, B, C) on the daughter board with Version 1.3 (mainly for keyboard debounce).
    My motherboard is a revision "G", and I have the "Alternate Version" of the 3-ROM daugherboard (page 107 in the Technical Reference).
    It seems that I currently have version 1.1 installed in these 3 ROMs.

    Now, I found a ROM 1.3 BIN file here:

    https://drive.google.com/open?id=0B9...2JPb0cxMmF3RFk

    I have split this in 3 parts of 4096 bytes each, and burned three 2732 EPROMS: ROM A, ROM B, ROM C.

    Next, I tried building an adapter rewiring the sockets because I do not wish to modify the motherboard.

    Hence, my adapter connects the ROM socket pins to the 2732 EPROM pins as follows:

    21 -> disconnected, BUT see discussion below!
    A11 signal 18 -> 21 (A11 from ROM socket goes into pin 21 of the EPROM)
    CS signal 20 -> 18 (pin 18 of the EPROM is ~E, Enable - another choice might be pin 20, which would be ~GVpp - not sure which is better for "CS")

    However, it didn't work... and the reason is likely the 2 different select lines 21, 20, and me trying to map this to the 2732.

    What I am unclear about - some people claim that the 3 ROMs are drop in replacements for 2532 EPROMs. I can see that the main chip select line is pin 20 for the 3 ROMs, they come from the 74LS32 "one hot" encoder, hence, this must be "active high".

    However, what's really confusing is pin 21: it is low (GND) for ROM A and ROM B, but high (VCC) for ROM C! This can be seen in the schematics, and I actually confirmed this by measuring it. Now, why is this? Is this just because this is what Tandy had available, or is there some deeper logic behind it? I am assuming that A, B need GND + 21 high for enable, and C needs VCC + 21 high for enable. Hence, trying to do this with 2732 EPROMs with either PIN 18 or PIN 21, both are active low, would require at least an inverter, right? And I can probably just ignore pin 21....

    Does that reasoning make sense?

    Anybody else replaced the 3 2332 ROMs with 2732 EPROMs successfully before? If so, please share some insight, I really don't want to modify the motherboard. So, it seems it is not that straight forward - at least, an inverter is required for the three ROM A, B, C, pin 20 select lines. Or?

    Thanks in advance for any ideas and insights!

  2. #2
    Join Date
    Mar 2013
    Location
    Chaffee, MO
    Posts
    1,338

    Default

    I've changed my TRS-80 Model 1's ROM's from Ver 1.1 to Ver 1.3. I used the Texas
    Instrument EPROMS. It has worked Perfect. I had the 3 ROM Set Board in my Model 1
    and I did have to add one PULLUP Resistor to the ROM C to make the Logic work.

    It was all documented years ago, and attempts to get it published failed. But, Dusty has
    it in the TRS8BIT issues 0801. I can send you all the documents but will need you to PM
    me your email address.

    You might want to check for the Flying "a" in the Lower Case Characters. If you have that,
    I'd also put the correct Character set in an EPROM. I used the KGS Lowercase Conversion kit
    that I purchased years ago. There is an easy upgrade for Lowercase depending on the exact
    Version of Character ROM that is installed in your Computer. Post the exact numbers on your
    Character Generator.

    Larry

    flying-a.png
    Last edited by ldkraemer; January 26th, 2020 at 04:53 AM.

  3. #3

    Default

    Quote Originally Posted by ldkraemer View Post
    I've changed my TRS-80 Model 1's ROM's from Ver 1.1 to Ver 1.3. I used the Texas
    Instrument EPROMS. It has worked Perfect. I had the 3 ROM Set Board in my Model 1
    and I did have to add one PULLUP Resistor to the ROM C to make the Logic work.

    It was all documented years ago, and attempts to get it published failed. But, Dusty has
    it in the TRS8BIT issues 0801. I can send you all the documents but will need you to PM
    me your email address.
    flying-a.png
    Hi Larry,
    thanks for the reply and I read your article in TRS8BIT 0801.

    Two more questions though:
    - my programmer does not support 2532 EPROMs, so I need it to work with 2732s. It is becoming very difficult to program 2532 with "modern" EPROMers...
    - I still don't understand how the Chips Select / Enable can work... the 72LS42 gives a "high" for Chip Select for the 3 chips, and the 2532 needs a "low" for Chip Enable (~E) AFAIK. The same holds true for the 2732, right?

    There are adapters for 2532 -> 2732 conversion, this simply rewires (along the lines what I did):
    http://www.simonsdialogs.com/2015/08...e-every-50-ms/

    So you are saying that this would definitely work with the 2532 (or a 2732 -> 2532 adapter such as the one above), and the 3 ROM adapter daughter board, and that no further conversion logic for CS is required, right?

    Then I will simply try again. Thanks!

  4. #4

    Default

    Regarding the 2732 -> 2532 adapter, there are two variations possible:
    - pin 20 (CS from the TRS80) can go to either ~OE (Pin 20 of the 2732), or the ~CE (Pin 1
    - the other pin should go to GND
    Most adapter use ~OE for CS, and GND for ~E, e.g., this one:

    https://www.arcade-cabinets.com/boar.../2732-to-2532/

    I am wondering if the TRS80 has any very specific timing requirements that either demand ~E or ~OE for CS.
    Well, I will keep trying!

  5. #5

    Default

    Actually, I figured out what was wrong in my thinking!

    Looking at the 74LS42 datasheet

    http://www.datasheet39.com/PDF/37564...datasheet.html

    I am now realizing that the output of the encoder is "low" active, not "high active" as I had assumed:

    0 -> LHH
    1 -> HLH
    2 -> HHL

    That means, the CS line uses Low for enable. And, now I remember that the 2332 ROMs with their 2 Select Lines could be programmed to either enable for Low or High. So it seems that Tandy arranged for ROM A and ROM B to enable for LOW and LOW (Pin 20 and 21), and for ROM C they used (for whatever reason!) LOW and HIGH. Indeed, PIN 21 can simply be ignored, and Socket Rom Pin 20 either goes to ~E (Pin 1, and Eprom PIN 20 (~GVpp) to GND, or vice versa. Both should work actually.

    Clear now, thanks.

    Will try again and post an update later.

  6. #6

    Default

    Alright, THANKS MUCH TO LARRY for the PM, we sorted it out now!

    So, it was a combination of bad EPROM and (probably) bad ROM files... Pics attached. Not pretty, but works!

    DSC09521.JPGDSC09522.JPGDSC09527.JPG

    The only drawback is that this double stack of DIP sockets prevents the top shell from closing, so I need to remove the mounting styrofoam under the daughter board to make it fit. But not a big problem.

Tags for this Thread

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •