• Please review our updated Terms and Rules here

Help needed trying to start a fake4032

dave_m said:
I hope you checked to see if that programmer will program the 2532 EPROM needed to replace the PET BASIC ROMs. If not, it may program the original Intel 2732 EPROM so a simple adapter can be built to to program the 2532.
Click to expand...
I'll check for him. I think there's a github for the list.
 
(THIS IS EDITED FOR ACCURACY) The T48 doesn't support the 2532, I think it does support the 2732. Here's a screenshot of that.
 

Attachments

  • 1772599014178.png
    1772599014178.png
    16.7 KB · Views: 5
Last edited:
Thanks so much to you all.
Your very kind help is very much appreciated.

So my guess is that in any case the PET cannot use 2732 ?
 
Hi there,

I did some homework ! ....................(trying to be a good student ! :) )

Seems I'm now able to read the TMS 2532 given with the PET.
Here is attached a reading of my EDEX 3.3 as well as 2 another chips labelled "?uM 04 9000" & "A9000" and one labelled "UTIL $A000 u+" (u=mu), so feel free to confirm if it worked well or not.
These are the 4 roms available by selecting switches on the PET Dashboard !
:)

My binary files

My guess is that I could now try to erase them to re-use them if needed in diagnosing/reparing this PET ?

Many thanks again.

Stéphanefirst_eprom_readings.jpg516881067_10162802436254361_3225795663117896727_n.jpg
 
Last edited:
Stéphane,
It's probably correct, but to know for sure that it read the data correctly, one must compare the checksum reported by the programmer to the known checksum of the binary file. If you have the original binary file, one could use a Windows program like HxD to calculate the 16 bit checksum by highlighting the entire 4096 bytes of code and using the analysis function.
-dave_m
 
Just a point have noted many times on the forum involving the PET.

While it is quite possible to use the 2732 Uveproms in the PET with a correct adapter, and also use them in a programmer too, either as they are, or the programmer set for a TMS2532, if you have an adapter, it does create a set of permutations that can add errors and problems. Also in light of the fact that a number of adapters can physically affect IC's sockets in unfavorable ways, that make fitting an actual IC to to them later, less than ideal.

I always advocate, especially in fault finding scenarios, minimising variables and using correct parts that don't need adapters. Your are already dealing with a fault, the smart move is to minimise every variable possible in the hardware, at least that is what Yogi Bear once told me and he was smarter than the average Bear, though I'm not sure about Booboo.

So I always suggest for PET repairs, ignore the 2732 parts completely, use the TMS2532 parts (not the TMS2532A parts) and your headaches will decrease by an order of magnitude.

Later of course, when all is working with the computer, play around with 2732's on adapters to your heart's content, but I would advise, only after the computer is actually working.
 
Hugo Holden said:
Just a point have noted many times on the forum involving the PET.

While it is quite possible to use the 2732 Uveproms in the PET with a correct adapter, and also use them in a programmer too, either as they are, or the programmer set for a TMS2532, if you have an adapter, it does create a set of permutations that can add errors and problems. Also in light of the fact that a number of adapters can physically affect IC's sockets in unfavorable ways, that make fitting an actual IC to to them later, less than ideal.

I always advocate, especially in fault finding scenarios, minimising variables and using correct parts that don't need adapters. Your are already dealing with a fault, the smart move is to minimise every variable possible in the hardware, at least that is what Yogi Bear once told me and he was smarter than the average Bear, though I'm not sure about Booboo.

So I always suggest for PET repairs, ignore the 2732 parts completely, use the TMS2532 parts (not the TMS2532A parts) and your headaches will decrease by an order of magnitude.

Later of course, when all is working with the computer, play around with 2732's on adapters to your heart's content, but I would advise, only after the computer is actually working.
Click to expand...

Thanks a lot for your advise, much appreciated.

To summarize :
  • I just built a quick adapter in order to try to make my very first readings done on the 4 chips 2532 I received with the PET as optionnal ROMS selectable with custom switches first owner drilled on the dashboard :
    • interested in knowing if attached files of these four TMS2532 may be readable by others, means my very first reading is ok !
    • interested in knowing, except regarding the I assume very known EDEX 3.3, what are the 3 others ROMS, and if they were already dumped ?
  • If my dumps are ok, my assuming in the last post is that we could now maybe erase one of these 2532 to re-use it/them to diagnose and or repair this PET ?
    The question is : will my just acquired T48 accept to write the 2532 as it seems to accept to read them (the T48 = TL866v3 should able to write 25V eproms ? but I will only be able to set the T48 in 2732 mode as for reading)

  • Last point : I've just tried to read/ dump my original character ROM labelled 901447-10 (dated 4680) : given checksum is 0x0009 7367 (I'm adding this 5th file in the shared Google drive).
    • >>> ++++ I read the zimmer's matching file name and checksum is 0001 7B67 (zimmer's file)


Many thanks again to you all

Stéphane
 
Last edited:
I have never recommended a T48 programmer (not that there is anything wrong with it at all) for programming parts for the PET. I am just unfamiliar with it.

All that I can say is, with 100% confidence is that the GQ-4x works perfecty for the application with no hitches. It can also prgram 21 or 25V parts. It has easy install software on a real disk with a friendly GUI, works on a usb link.

For me that is important "it works" because if something is going to go awry with some "modern" product & its firmware, trying to deal with vintage parts, I am like a magnet to find it. While I am ok with hardaware, software and firmware issues/bugs are not my strong point.

So far, for me, the GQ-4x has worked perfectly for programming & reading a whole host of vintage computer memory parts. Only on rare occasions do I need to go to a vintage programmer like the BP-1400. In any case, if you look at the GQ-4x, you will see the very wide range of supported devices. It is well made in a nice housing.

I'm sure somebody with the T48 will respond to help you.
 
stansteph54 said:
  • Last point : I've just tried to read/ dump my original character ROM labelled 901447-10 (dated 4680) : given checksum is 0x0009 7367 (I'm adding this 5th file in the shared Google drive).
    • >>> ++++ I read the zimmer's matching file name and checksum is 0001 7B67 (zimmer's file)
Click to expand...
Stéphane,
I checked your reading of the character ROM in the google drive and the one in zimmers and I get a 16 bit checksum of 7367 for yours and 7B67 for zimmers. This may be a problem.
 
Thank you so much Dave for your feedback.
I'm not very familiar with eprom yet but I compared the checksum too following instruction received in this tread 🙏

What could be the next trial ?
1)Erase one of my "optional" 2532 and try to write a new character rom on it with T48 settled on 2732 + my adapter + 25v modified settings ?
2)burn a 2732 and try it on the PET with my same adapter ?
3)burn the PETESTER rom on 2532 or 2732 to first try to confirm the suspicious character rom ?

Thanks again to you all 🥰
 
Stan,
I'm not sure the schematic on message 48 will work to program and read a 2532. It has to do with pin 21 (Vpp) needing to be at +5V during a read operation. Let me look into it. For the time being do not erase your 2532 chips.

One thing you can do is to program a 2732 to work as a PETTEST4. Note that the 2732 is a 4K Byte EPROM while PETTEST is a 2K Byte file. But due to a happy quirk of pinouts, it will work in the EDIT socket but only if you program the 2K contents into the upper 2K of the 2732 address space i.e., from Hex address 0800 to 0FFF. This is an offset of the starting EPROM address of 0800 Hex.
If this does not make sense to you, read about starting 'Offset' in your T48 manual. I do not have a copy of the manual.
 
Stan,
It has to do with pin 21 (Vpp) of the 2532 needing to be at +5V during a Read operation while pin 20 of the 2732 will be held Low during a Read and Verify operation by the programmer.

The 2532 EPROM is not specified to work with Vpp at a Low level. See the spec sheet: https://www.qsl.net/n9zia/vision/TMS2532.pdf

Adding a diode and a resistor to the adapter should fix this issue. Let's have Daver2 take a look before you try it. I don't want you to blow up your T48 or the 2532.

With this circuit, the 2532 Vpp pin will be close to +25v during programming, but pulled up to +5V during a Read or Verify due to reversed bias on diode when pin 20 of programmer is at Low.

EDIT: Nuts that may not work either! The Ipp1 current specification for Read operation with Vpp at 5V is 12 mA Max. With a 1K Ohm resistor there will be to much voltage drop across the resistor. Maybe a diode rather than the resistor? It would have to be a diode with a low forward voltage like a Schottky diode. This adapter is not looking like a good solution to programming a 2532...
 

Attachments

  • 2532-2732 Adapter.jpg
    2532-2732 Adapter.jpg
    122.5 KB · Views: 10
Last edited:
You must log in or register to reply here.
Back
Top