loeliger@uxc.cso.uiuc.edu (Jon Loeliger) (07/29/90)
WAAAH! Sure enough. You go off on vacation, and when you come back nothing works worth a tinkers toenail. Well, it isn't *that* bad, but sure enough. My board died here mysteriously. So, after a saturday of wearing my feeble Hardware Jock hat, here's what I've learned: I am real certain that most all of the board is functioning just perfectly. Remember, this used to be a perfectly functional board that *has* done 10K digits of pi... Now, I get the SCSI leds. I get no console output. It appears that /DUART0 is never asserted at pin 39 of U41. Backing up a step, it appears that pin 15 out of U33 is always high. But we do see occasional pulses on pin 13 and 10, which should never be there. U33 is a 3/8 decoder that should pass /DUART through to each of the /DUART[0-3] according to BA04, BA05 and A06. The best I can tell is the address lines are working (many interesting transitions, nice easily discerned high and low voltages), the /DUART signal into U33 hangs high and blips twice by the resolution of my scope, but we never get a Y0 output. So, replace the chip. I did. Still nothing new. So I replaced it again. Still nothing new. I believe that the high volts coming into the 74F138 are in the 4.0 range. Anyone got a clue? Can anyone give me a clue? Do I need to drag the logic analyzer out of the closet and hook it up to determine exactly what the address lines are when the /DUART line goes low? Seems they should be 000, but are occasionally 110 and 011. I suspect the one of them is for /PARCLU signal as initialized by the EPROM, but could easily be, as they say, speaking out my ass.. Thanks, jdl k<bob> Oh, BTW, My minix boot disk -- Send it to Bruce or Dave or either?
news@daver.bungi.com (08/01/90)
John: I had peculiar problems like this with my Designer's Kit board before I re- alized the 532 wasn't completely pressed into its socket. So I pressed it in more firmly. We're talking Herculean insertion force here!