GVP A530 free repair!

  • Thread starter Thread starter rimmer
  • Start date Start date
  • Replies Replies 29
  • Views Views 2490

rimmer

Active member
Banned
Joined
Oct 10, 2019
Posts
608
Country
Denmark
Region
DK
HI ALL

I'm in the process of reading back as many of the infamous PALCE-chips from the GVP A530 as I can, in an effort to compile an archive of fuse map code files for everyones benefit.

So far, I have been through the 16 chips in my two units, and managed to make a copy of the U53 in one of them.

Now, there are eight (8) PALCE-chips in an A530, so I'm nowhere near the target.

However, I cannot find, nor finance, enough units to track down a working set. But my findings confirm the thesis set forth elsewhere on the internet; there is a posibillity to compile a working set of code files, as not all chips have had their seciruty fuse activated properly, making a read-back possible.

These nice Italian folks did just this for the A2000 G-Force Combo 50Mhz card: https://www.nerdone.it/2020/11/22/l...p-g-force-combo-030-50-mhz-full-gal-al-mondo/

I call on anyone with a GVP A530, working or not, to either try and read back these devices themselves, alternatively send the chips to me.

There are several other members here on AmiBay with one or more devices, some working, some not - and all are looking for new chips, either to get a card going again or as backup. But nothing comes from nothing!

Let's keep this thread alive; let me and others know what you think and if you are able to help.

THANKS 🙏

// Rimmer
 

Attachments

  • IMG_2029.jpeg
    IMG_2029.jpeg
    530.8 KB · Views: 75
  • IMG_2030.jpeg
    IMG_2030.jpeg
    272.4 KB · Views: 62
  • IMG_2031.jpeg
    IMG_2031.jpeg
    438.6 KB · Views: 66
  • IMG_2032.jpeg
    IMG_2032.jpeg
    531.6 KB · Views: 72
  • IMG_2033.jpeg
    IMG_2033.jpeg
    487.3 KB · Views: 65
  • IMG_2049.jpeg
    IMG_2049.jpeg
    438.1 KB · Views: 52
Last edited:
So, I've just submitted an offer to the seller in this eBay listing, requiring only the A530, with the explicit purpose of trying to read back as many of the eight (8) PALCE's as possible.


If anyone else here is successfull in grabbing this set, I'd appreciate if an attempt was made to make the PALCE code available.

Yes, I've talked about this project for some time. Now it's time to pick up the pace :)

Thanks!

// Rimmer
 
I should receive an A530 Turbo in Australia in the coming months, I might be able to help. I like the idea.
 
I just purchased the three GAL fuse map files from ozkano, and I have confirmed they are from the Italian dump of the GVP Combo 50Mhz card, available on the webpage I referenced in my first post.

Next step is to try and program my GAL's with this code and try them in my working A530 - interesting if it works, as the Italian guys were pretty adamant the code *only* works for the 50Mhz card and the A530 runs at 40Mhz.

Stay tuned
 
OK, so it seems I'm not the only one where the U53 chip is the one causing the most issues.
In the GVP Combo 030 50Mhz, U53 is a GAL20 - in the A530 it's a GAL16...
So, I've begun to reverse engineer the code for the GAL20 version, aiming to re-code it for the GAL16 needed in the A530.
Unfortunately, it won't be until later this week, or possibly the weekend, before I get a chance to program a test-GAL and see if my second A530 boots with it.
 
I just purchased the three GAL fuse map files from ozkano, and I have confirmed they are from the Italian dump of the GVP Combo 50Mhz card, available on the webpage I referenced in my first post.

Next step is to try and program my GAL's with this code and try them in my working A530 - interesting if it works, as the Italian guys were pretty adamant the code *only* works for the 50Mhz card and the A530 runs at 40Mhz.

Stay tuned
I have used the files from Ozkano in an A530. It appeared to work for me.
 
How can I tell if the chips will read correctly or not? Do I have to read them then make a chip to test? Or can you tell from what it reads?
 
When you read your chips to a fuse map JED file, you can open that file in just about any text editor.

The file starts with a text description of the file and proceeds with contents.

If the contents if primarily "1"'s, it's read protected.

Example from nerdone.it: (data part marked with yellow in the margin)

1694413339865.png


Thanks for helping out, Dale :)
 
I’ll dig into it. I have a good A530 and a non working one I believe to be a MACH chip issue that I can also try the PAlLs with.
 
I really want this project to be finished. There are 4 A530 turbos. New PAL's are waiting.
 
I've got one of these and a programmer - no time to attempt this right now unfortunately, but I will try to have a look next week!
 
TLDR; I have no idea what I'm doing, but I think I'm making progress.

--- The longer version ----
So, slight progress update.

I've been sent the original PAL fuse map JEDEC files from @digitalh (thanks!), who is the person responsible for the webpage and work done at nerdone.it - see link above.

The main concern with the files from the GVP Combo 50Mhz card is U53; this is a PALCE20V8L chip on the Combo, and a PALCE16V8L-chip on the A530.

I've successfully de-compiled the fuse maps to a binary source code files, and can see all input-pins are used on the U53 PALCE20V8L - which is more than the U53 PALCE16V8L on the A530 has.

Now, this is not necessarily an issue, as I think the binary code describes *all* inputs, also if the input is not used.

Since I am an optimist and generally thinks nothing is impossible, next step for me is to dig out a Combo card and see if I can isolate inactive pins of the PALCE20V8L - if there are any.

It would be strange if the Combo is substancially different from the A530, since they share functionality (Processor and SCSI), target the same basic computer architecture (A500/A2000), share basic structure and the other PLACE glue chips seems interchangable between the Combo and the A530. I might be in the wrong, but time will tell.

If I can isolate the inactive pins, it would be somewhat straight forward to re-write the binary source code and compile a test-GAL. Otherwise, I'll cook up a few tests based on gut feeling...

Please feel free to point out any obvious flaws in my reasoning; it should be noted I have *zero* idea what I'm doing and just making this shit up as I go along.

Thank you for your time. To be continued...

// Rimmer
 
Any progress? Can anyone share files? No problem if paypal is needed for that to happen ;)
 
Last edited:
I have also 2 A530 Rev. 5 Boards including chips with me. Unfortunately both are not starting up. On one U36 is now broken (this device was running in the past). I tried to replace the chip with a ATMEL GAL as described on nerdone.it, but I do not get it working (green or red screen). It maybe because my Amiga 500 is "tuned" a little bit (1MB chip RAM). I am currently looking for another not tempered working Amiga to see if this is the problem.
The other unit was already not working when I received it 20 years back. I checked the chips and all are protected, but 2 are broken. U36 again and U38.
Unfortunately comparing the checksums of my chips with the chips from nerdone.it, they are all different except for U36. So I am not sure if I can use the U38 file there. Furthermore, because the device was always broken I do not know if it will ever work or if there are more damages to it. One of the LEDs is missing too...

So in case there is any progress or I can somehow help, please add it here. Thank you!
 
I have also 2 A530 Rev. 5 Boards including chips with me. Unfortunately both are not starting up. On one U36 is now broken (this device was running in the past). I tried to replace the chip with a ATMEL GAL as described on nerdone.it, but I do not get it working (green or red screen). It maybe because my Amiga 500 is "tuned" a little bit (1MB chip RAM). I am currently looking for another not tempered working Amiga to see if this is the problem.
The other unit was already not working when I received it 20 years back. I checked the chips and all are protected, but 2 are broken. U36 again and U38.
Unfortunately comparing the checksums of my chips with the chips from nerdone.it, they are all different except for U36. So I am not sure if I can use the U38 file there. Furthermore, because the device was always broken I do not know if it will ever work or if there are more damages to it. One of the LEDs is missing too...

So in case there is any progress or I can somehow help, please add it here. Thank you!
From my knowledge reading protected PAL chips is nearly impossible. The only hope for A530 users for a full set of working Gals is that someday there might surface a unit that is not write protected. So for now... let's sit and wait another few years.

PS. u38 u36 are already available. So just buy them and check.
 
He ordered U35 - this one was also already read. I think he tried to read U35. I wonder if he knew it was protected or it was indeed dead. Or dead and protected.
 
Back
Top Bottom