Patrick's memory expansion boards can be great, but they are quite difficult to install perfectly, and even the best installers sometimes have trouble.

This can result in a player with an installed board which doesn't actually work reliably. When that happens, it is often best to just leave the board in-situ, and flip all four bank enable switches off, effectively disabling the board.

The alternative, desoldering the board and removing it, bears a high risk of destruction to the player mainboard.

I have now removed three memory boards. The first time was on one of my own, and I got lucky in that it removed just fine with no damage.

The second time I removed one (same board, actually) resulted in much destruction of test pads and attached traces, requiring a couple of hours of work tracing signals and solding in replacement jumpers. That player has been in good health ever since, though.

The third time was this evening. I began by solder sucking all of the connections, and then gently pried up the VCC, SCL, and GND connections (while heating them with a soldering iron), one at a time to release the bulk of the memory board. So far so good.

Now for the two big groups of test pads.. this is where all of the damage happens. I started heating/prying, but then got a clue and stopped.

Down to the workshop, bore a 1.3" hole in a scrap of plywood, place hole over the test pads area, masking off the rest of the player & memory board. Apply heat gun on "high", and wait a few seconds.

The memory board promptly fell off, with very minimal damage to the pads -- probably all inflicted during installation and/or the inital prying beforehand.

The player boots up and works perfectly, and the memory board looks like new.

So.. if you ever have to do this, Don't Do It !!!. Send it to me instead.

Cheers!

Mark

And here's a photo of the basic idea.
The white thing is a slim pry bar that I used to separate the two once the solder had melted.



Of course, a better approach would be to have a hot-air desoldering machine.
But not many home labs are likely to have one of those!

Cheers

Hey Mark,

Excellent suggestion in lieu of a proper hot air setup. I've run into one sitution where the memory board would not work at all, but luckily the switches did the trick and at least got the Empeg booting again. In the end I just left the board in place and did the much more reliable memory stack to 32 MB.

Have you had any thoughts to just doing continuity tests and jumpering the ones that aren't connected? I tried this, but it didn't work. Perhaps the jumper wires added too much capacitance to the lines. I'm not sure.

Stu

Are the boards still available or are they really not recommended?

Mine's available. I don't know if it's in operating condition or not. I've been considering asking Patrick if he'll look at it and repair if necessary.