AlfaRAM A1200 how to replace faulty PAL chips

[tbtorro]

OK, so I have a few remainder Alfaram A1200 cards. (http://amiga.resource.cx/exp/alfaram1200) They have been tested but all suffer from the same problem: when jumpered to enable the ZIP RAM, the A1200 won't boot (black screen). When the ZIP RAM is disabled, they boot fine and the 1M on-board RAM is visible and working fine. Strangely the cards also work when the ZIP RAM is enabled but J6 is set to NON-test mode, but only 1MB is visible still.

Now I'm pretty sure this has something to do with the PAL chips that I suspect drive the RAM logic, because I had one last card fully working and I took the PAL chips and moved to another card which started working. However when I tried to change the chips one-by-one to find out which one was faulty somehow the last set of working chips became faulty, so none of the cards work at the moment (above the 1MB that is)

I actually have new PAL replacement chips for this card specifically, they are marked U5, U8 and U9, but when I try to replace them the symptoms are the same (no visible ZIP RAM).

I wonder if someone could enlighten me:
- Could it be that mixing good and faulty PAL ICs somehow corrupts the good chips?
- Do these new PAL chips need flashing/programing somehow, i.e. is it possible that I have blank PAL chips, and I need to load the ROM on them before placing it into the RAM card? How could this be done?
- Can anyone guess which PAL chips do what? I have replacement chips for U5 U8 and U9 but not for U6 and U7.
- Perhaps there is some kind of mode where the PAL chips "learn" how much RAM is in the card?

Hi-res photos included, and here are jumper details according to the manual:

J1: enables/disables the RTC
J2: N/A
J3: enables/disables ZIP RAM
J4: enables/disables on-board 1MB
J5: switches between 4MB and 8MB ZIP RAMs installed
J6: enables/disables ZIP RAM test mode

I'm on completely unknown territory here so please forgive me if some of my questions are silly. Any advice would be much appreciated, as I would love to get these last 4 cards working. Any other advice apart from the questions above are also much welcomed.

 

[Zetr0]

@tbtorro

my friend, if I am not mistaken these cards can string up 9MB of RAM on the A1200.

To do this they need to ADDMEM to different portions of the Amiga A1200 Memory Map. This is the primary reason for all that PAL logic, to setup the first 1MB in the RANGER address space of memory.

I suspect that atlest U7 and U9 are resonsible for this part of the card. The other three U5, U6 and U8 will be responsible for the jumper configuration setup. These cards DONT have any real smarts they just report back as to what the jumper configuration is set, the assertion addresses (as the 9MB will be in atleast two blocks) will also be mapped by the PAL logic.

If you have tested a known working set of PAL logic then its another problem with the adapter card, looking at the card theres little else to go wrong - I would do the following

1. Check the Mechanical

Inspect the ZIP sockets and track traces around the Battery, buzz it out in places that you are unsure of, I would also check for oxidization of all the sockets (memory / PAL / FPU and A1200 connector) you may have a dry solder fault on the RW or IO lines

2. Replace the Capacitors

There are some wetware radials that could do with a replacement, although to be honest, these look like they are needed for the Clock and Oscillator, saying that there are a coupleat the bottom and right of the PCB that I would certainly consider chaging first.


3. Replace the Hard Logic

you will notice 5 SMD 74x logic IC's, U10, U11, U12 and U17, these would be the last port of call for the card. if nothing happens at this stage, then I would hazzard a statement that either printing fault of the PCB's (if they are NOS and never worked) or perhaps damage between the layers of the card have caused a short.


Good luck on your quest with these, I do wish I could help more, albiet the best I could suggest is the above. These adapters are quite unique, since they are the only one I know that was marketed allowing you upto 11MB of RAM (2MB CHIP / 9MB FAST) on an A1200 with RTC.

This adapter would take a stock A1200 of 1.1MIPS upto 2.8 MIPS - quite nice indeed!

 

[tbtorro]

OK so here is the thing, when I replaced all the PAL chips (from a working card) everything started working in the previously faulty card.

Then I placed back only two (U8 and U9) of the old (faulty) PAL chips and the card stopped working.

Then replaced again U8 and U9 from the working set and still the card remained faulty.

So now no combination of chips work on either card.

Does this make any sense? The working chips somehow developed a fault by mixing them with the faulty ones? Because in that case I have no hope of fixing this as out of 5 chips I'd never know which one is faulty when good chips go bad with only one trial.

And do I understand correctly that these PAL chips doesn't have any programmable logic? U6-U9 seem to be the exact same type of chips yet if I say interchange U8 and U9 the card won't boot even with just the 1MB RAM enabled.. So therefore I assume the chips have different logic to them.

In short, I'm pretty sure it's not mechanical/capacitor/layout related. But what is the best way to troubleshoot the PAL chip setup? As I said I have brand new replacement chips but how to check if they have the correct logic and they aren't faulty?

@tbtorro

my friend, if I am not mistaken these cards can string up 9MB of RAM on the A1200.

To do this they need to ADDMEM to different portions of the Amiga A1200 Memory Map. This is the primary reason for all that PAL logic, to setup the first 1MB in the RANGER address space of memory.

I suspect that atlest U7 and U9 are resonsible for this part of the card. The other three U5, U6 and U8 will be responsible for the jumper configuration setup. These cards DONT have any real smarts they just report back as to what the jumper configuration is set, the assertion addresses (as the 9MB will be in atleast two blocks) will also be mapped by the PAL logic.

If you have tested a known working set of PAL logic then its another problem with the adapter card, looking at the card theres little else to go wrong - I would do the following

1. Check the Mechanical

Inspect the ZIP sockets and track traces around the Battery, buzz it out in places that you are unsure of, I would also check for oxidization of all the sockets (memory / PAL / FPU and A1200 connector) you may have a dry solder fault on the RW or IO lines

2. Replace the Capacitors

There are some wetware radials that could do with a replacement, although to be honest, these look like they are needed for the Clock and Oscillator, saying that there are a coupleat the bottom and right of the PCB that I would certainly consider chaging first.


3. Replace the Hard Logic

you will notice 5 SMD 74x logic IC's, U10, U11, U12 and U17, these would be the last port of call for the card. if nothing happens at this stage, then I would hazzard a statement that either printing fault of the PCB's (if they are NOS and never worked) or perhaps damage between the layers of the card have caused a short.


Good luck on your quest with these, I do wish I could help more, albiet the best I could suggest is the above. These adapters are quite unique, since they are the only one I know that was marketed allowing you upto 11MB of RAM (2MB CHIP / 9MB FAST) on an A1200 with RTC.

This adapter would take a stock A1200 of 1.1MIPS upto 2.8 MIPS - quite nice indeed!

 

[Zetr0]

PAL (PALIC) is an abbreviation of Programable Array Logic Integrated Circuit.

These chips are specifically programmed Input to Output signals, thus you cannot mix or match them. Each chip is programmed for a specific task, thus you cannot take a chip that was programmed to deal with signals in U8 position and put it in U9 as it wouldn't deal with the signals properly.

PAL logic replaces large 74xxx logic circuits - reducing both complexity, popagation delays and power drains - for more info have a read here