[rescue] Anyone Familiar with VMEbus?

[Dave McGuire]

On 9/17/21 10:30 AM, Christopher Purdy wrote:
> I picked up an old ('92) Unimation /Staubli Puma 560 industrial robot arm.
> The controller is VMEbus based and after messing with it for a couple hours
> not being able to get it to talk to my VT420.... I popped the case open and
> saw some error LEDs.
> 
> Of primary concern - the VME motherboard status lights show "sysfail" and
> both cards show "err".  I
> 
> 'm assuming this is a mainboard issue as I pulled the cards and it still
> says "sysfail", then adds "fault" I assume because now there is no CPU card
> attached.
> 
> I have reseated as many chips as I can access with no luck.
> 
> Anyone have experience with these - or better yet, have a Staubli UNIVAL
> PUMA controller box?

   I have a lot of experience with VMEbus, but this sounds less like a 
VMEbus issue and more an issue with your boards.

   SYSFAIL is a signal on the VMEbus that any board can assert.  It's an 
active-low, open-collector signal, which means it can be pulled down 
(asserted) by any board in the backplane, but has to be pulled up by 
something with a pullup resistor.  Sometimes that's done by the CPU 
board (or other bus master) but sometimes that pullup resistor is on the 
backplane.

   If there's no SYSFAIL pullup resistor on your backplane, then it 
being low (and lighting that LED) when there are no boards in the 
backplane is the expected result.

   "Fault" is not a VME signal that I recognize.

   I too have a PUMA robot here, but I don't have the original Staubli 
controller for it, at least I don't think I do.  I will check the next 
time I'm in that building.

              -Dave

-- 
Dave McGuire, AK4HZ
New Kensington, PA