Answering my own question for posterity:

> However, test 1, the RAM test, should show long flashes of the green LED if 
> RAM
> is bad. I was prepared to see all long flashes which might implicate the
> buffers or address decoder, but instead it won't blink the LED at all in that
> or any of the other tests. The red LED remains lit and appropriately
> extinguishes when the RS button is down.
> 
> Again, the board works correctly and fully certifies the other two KIMs.
> 
> What would cause it to hang (?) in the RAM test on the defective one?

Dwight suggested something wrong with the address lines. That made sense, since
the continuity problems I had on the board were on the address lines (turned
out to be lifted traces). However, after spending a couple hours more with the
tester, the actual problem was two address lines that had an intermittent
short. I cleaned that up and everything passes.

The original fault, a bad 2nd bit from $280 to $2bf, was indeed a single RAM
chip gone bad in a single row. Now that it's replaced and the board is fixed,
the fault is gone. Buffers and address decoding check out just fine.

-- 
------------------------------------ personal: http://www.cameronkaiser.com/ --
  Cameron Kaiser * Floodgap Systems * www.floodgap.com * [email protected]
-- Not sun-worshippers: Son-worshippers! -- Uhura, Star Trek "Bread & Circuses"

Reply via email to