Author Topic: Popular Electronics - Peel programmer  (Read 5194 times)

0 Members and 1 Guest are viewing this topic.

Offline fabiodlTopic starter

  • Frequent Contributor
  • **
  • Posts: 286
Popular Electronics - Peel programmer
« on: January 21, 2020, 01:19:52 am »
I was looking for options for programming a PEEL18CV8 and found this article
https://www.americanradiohistory.com/Archive-Poptronics/90s/90/PE-1990-01.pdf
(from page 33) however there's no listing of the source. How did they provide the code? was there an associated floppy with the magazine?

 

Offline westfw

  • Super Contributor
  • ***
  • Posts: 4671
  • Country: us
Re: Popular Electronics - Peel programmer
« Reply #1 on: January 25, 2020, 10:02:55 am »
Looks like you were supposed to buy the pre-programmed eprom from the beginning author...
 

Offline rstofer

  • Super Contributor
  • ***
  • Posts: 10093
  • Country: us
Re: Popular Electronics - Peel programmer
« Reply #2 on: January 25, 2020, 11:08:58 pm »
I was wondering if the device could be programmed with an Arduino or something similar.  I grabbed the datasheet and, lo and behold, they just state that you have to use their programmer and software.  Not a clue about the algorithm.

http://www.komponenten.es.aau.dk/fileadmin/komponenten/Data_Sheet/Memory/18CV8.pdf  Page 4

I guess you can use Atmel WinCUPL but there is still a shortage of information

http://ww1.microchip.com/downloads/en/DeviceDoc/doc0737.pdf

I have a feeling that this device has long ago moved past end-of-life.

 

Offline fabiodlTopic starter

  • Frequent Contributor
  • **
  • Posts: 286
Re: Popular Electronics - Peel programmer
« Reply #3 on: January 27, 2020, 12:41:28 am »
Yep, it is well beyond its end of life.
The only reason I am interested in is reverse engineering a mega-cd daughter board (details here
https://www.eevblog.com/forum/projects/help-with-reverse-engineering-unexpected-behavior-of-peel18cv8-223204/msg2840134/ )
I indeed think an arduino based programmer is totally feasible, similar to this https://github.com/ole00/afterburner
user up8051 of this forum provided a datasheet of the protocol
https://www.eevblog.com/forum/projects/protocol-for-programming-peel-chips/?action=dlattach;attach=913252
I am now working on it.





 

Offline builder68

  • Newbie
  • Posts: 1
  • Country: us
Re: Popular Electronics - Peel programmer
« Reply #4 on: January 04, 2025, 03:40:51 pm »
I'd like to revive this post!

Have you successfully implemented the programming protocol for the Peel 18V8 in the Afterburner?

I have a couple of 18V8 devices that I intend to use as replacements for Altera E320s. I have already converted the JEDEC files, but I'm having difficulty determining how to program them using the Afterburner.

Cheers!
 

Offline fabiodlTopic starter

  • Frequent Contributor
  • **
  • Posts: 286
Re: Popular Electronics - Peel programmer
« Reply #5 on: March 17, 2025, 02:02:48 pm »
Hello, sorry for the late reply. Are you still interested? I can dig up the software.
Yes, I made it working. But I built a custom pcb, I did not use afterburner. I believe it should be possible to port it to the afterburner hardware with some effort.

 

Offline olin

  • Contributor
  • Posts: 18
  • Country: 00
Re: Popular Electronics - Peel programmer
« Reply #6 on: October 11, 2025, 09:19:36 pm »
Hi fabiodl. I'd like to add support of the IC to Afterburner. I built an adapter for afterburner which seems to work (PVE, PVP, ALE/ERA voltages now work OK). I can erase and write and read fuses, but I'm bit confused about the fusemap addressing. Especially the TGA number, the IAD address number - how they match the fusemap JEDEC format.
1) the fusemap in Afterburner is read as a linear array of bytes. Then Table 6 shows TGA 0 as Inputs (0-35) which is 36 bytes then  Config (36) which is 1 byte - are the bytes read from the fusemap in that sequential order, or is the Config byte located in a different place in fusemap? It is not obvious to me where the Config byte is located in the fusemap (I guess it is starting from fuse 2664) and whether to set only 4 bits or 8 bits (as the bus width is 8 bits) to IAD pins.
2) The table 4 (Input line addressing) does not say whether the linear fusemap matches column 'InputLine' or column 'Input Function'. I understand that ultimately whichever line I select from the Table 4, the IAD pins must be set according the column IAD(7-0).

I feel I could spend weeks figuring the right fuse order (and possibly few damaged ICs as well).   If you can shed some light on that it would be appreciated. Or, If you can dig up your code I'll credit you in the PEEL18CV8 Afterburner's implementation. Thanks.
« Last Edit: October 11, 2025, 09:24:21 pm by olin »
 

Offline olin

  • Contributor
  • Posts: 18
  • Country: 00
Re: Popular Electronics - Peel programmer
« Reply #7 on: October 17, 2025, 03:41:22 pm »
I got it working now, support of the IC will be added soon to Afterburner.  It will probably be in Beta phase for some time before the hardware adapter is finalised. So, check the Wiki on Github if you'd like to help to troubleshoot or want to have an early access to the code.

BTW. To answer my questions (if somebody is interested):
1) When writing bytes of the Product terms (fuses 0 to 2303) each macrocell has 36*8 fuse bits. 36 bytes need to be written for each Macrocell. The 1st byte contains fuse bits 0, 36,72, 108,144,180,216,252. The Second written byte contains fuse bits 1,37,73,109,145,181,217,253. The 3rd byte contains bits 2,38,74... etc. OE terms are similar, just start from fuse 2304 and the first OE term byte is comprised of fuse bits 2304, 2340,2376,2412,2448,2484,2520,2556. The second OEM byte is comprised of fuse bits 2305, 2341, 2377 etc. The important thing is that bit order is reversed when such byte is placed on the 8 bit bus (ie. bit 0 goes to IAD7, bit 1 to IAD6 etc.) - that is both for Product terms and OE term..
2) The table 4 in the programming document is linear, first byte written (of certain macrocell) corresponds to Input Line Number 0, and the IAD0-7 value  is taken from the column of the same name -  value 1. 2nd byte written uses IAD0-7 value 3 etc.

No IC was harmed during the trials and errors :)
 
The following users thanked this post: oPossum, PCB.Wiz


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf

 

- - -
-->