0TT-C head scratcher

Hi all,

I've spent nearly all day with this problem and I'm stumped. The machine is a Mori DL20MC circa 1990 with a 0TT-C control. The problems started last week with intermittent 930 CPU interrupt alarms. I cleared them and moved on. This morning, the machine woke up with a 914 RAM parity (servo) error. This machine has six total axes- X,Z,C on main and sub. When I try running the program, I get a 411/421 X/Z excess error right away.
I swapped drives between the main and sub. This did nothing.
I swapped the 5/6 axis card with another machine. This did nothing.
I swapped power supplies just for fun. That did nothing.
I did further investigating of the issue, and what I found is that I can rapid the X and Z (of the sub) without any problem UNLESS the C axis (of the sub) is engaged. As soon as I activate the C-axis, I cannot rapid either X or Z any faster than 5% rapid. I looked at the following error diagnostic (800 and 801) for these axes and saw a change in the following error depending on whether the C axis is engaged. For example, at 5% rapid and with the c axis disengaged, the Z axis shows a value of 167. If I engage the C axis, that number becomes 870. If I move the axis any faster, it exceeds the following error and alarms out.
The main side is acting perfectly normal. I'm completely stumped as to why the C axis being engaged should have any kind of effect on the X and Z. (The C axis is controlled by a Mitsu drive, btw.) I'm hoping someone here will have an idea. Otherwise, I'll be swapping main boards with another machine which is not something I really want to be doing.

Thanks!
wmpy