I just want to leave some feedback. I've recently purchased one of the eBay units and it has been -mostly- sucessfully tested, great piece of hardware. Now I'll explain the -mostly-...

The expansion has been assembled in a EP64, easy to work as I don't need to desolder the EP128 64KB RAM board. It has been tested fine: EXOS 2.4 tests and even a IS-BASIC program (only the first three or four segments, I was pokeing all the addresses 0-16383!); but...it's extremely unreliable when I plug my EXDOS controller.
Sometimes I can create a RAMDISK; but it shows corrupt info after a couple of DIR commands. I've been able to format a disk; but it corrupts the same way as the RAMDISK after a DIR.

Facts:

- As the original PSU is a bit weak, I've replaced the original CPU by a powerful one with no changes.
- I've tested the EXDOS controller without flaws in the EP128.
- The controller has been upgraded to EXOS 1.3 thus, the 27128 original EPROM has been replaced by a 27256.

I'm starting to think that I'll need to swap the expansions and migrate the 512KB one to the EP128...

Anyone has a clue about what's happening?

Have you tried the EXOS 2.4 Advanced test when EXDOS attached?
Anyway, a good idea: clean the expansion connector.

Yep, I'd definitely try the advanced memory test first up to make sure it's all connected correctly.

I've not been able to test with EXDOS yet, as my EXDOS board is broken. I really need to find the time to take it apart, hook it up and do some tests to find out what is wrong!

I cant think of any reasons why EXDOS itself would cause issues, other than if it caused spikes on the address bus which would be missed by slower RAM. I know there were some issues like this on the Atari ST. The system wasn't very well designed, and the address bus is very noisy. Using faster RAM actually causes problems as it picks up the noise and causes writes to wrong addresses. The slower RAM misses these spikes.

jltursan! Before the upgrade you used the EXDOS card with your EP64? It is worked?

One idea: with the EXOS 2.4 you can start the machine in 64K mode. In this mode try to run 64K games. Also good idea copy games from disk to another, and check the copied programs also working? In this situation the memory expansion not used. If any problem detected then possible some problems with the EXDOS connection.

It feels like I have been waiting forever for this batch of PCB's...

Least the GALs arrived?

The EP64 is a newcomer and it has suffered a lot, maybe it has still some strange flaws I haven't noticed until now.

Then I suggest thorough cleaning of the expansion connector. The "not used for many-many-many years" connectors can make stability problems, especially on arabic machines which are stored in desert for a long time
Another possible problems with arabic machine the dried out electric capacitors, especially if your board built with PHILIPS capacitors.

Least the GALs arrived?

No, still waiting for those as well! Post seems to be very very slow....

I'm glad to hear the memory expansion is not the cause of the EP64 problems, though! I had some odd behaviour with my EXDOS expansion before it stopped working completely, so it sounds like yours needs some attention too, jltursan. I will be interested to hear what your problems are.

I had some odd behaviour with my EXDOS expansion before it stopped working completely

What happened with your card? EP don't start with the card, or starting but the EXDOS not detected?

A quick update...

I've replaced the 100uF electrolityc capacitor if the bridge (and the +5V regulator as a bonus), it was notoriously "inflated". The controller plugged to the EP128 seems to boot more stable now, I'll keep testing to be sure about this.

Sadly, the EP64 keeps failing, no black screens tho, only endless booting screen after pressing keys.
I suppose that there's some fault in the +9V voltage line of the bus. The controller seems to receive no power at all.