Viet.Ho@SAMBA.ACS.UNC.EDU (11/26/90)
Anyone have a Frances working STABLE on an A500/Lucas combo? My Lucas works without a hitch, but Frances will usually give me a guru 81000009 when I issue the afm -m36c4c4 to set my memory (2Mbs). Only once in a blue moon do I get it going but the next time I turn it on, same thing... 8-( -Viet Viet.Ho@samba.acs.unc.edu
vincelee@tornado.Berkeley.EDU (Vincent H. Lee) (12/20/90)
In article <9011251927.AA11302@samba.acs.unc.edu> Viet.Ho@SAMBA.ACS.UNC.EDU writes: >Anyone have a Frances working STABLE on an A500/Lucas combo? >My Lucas works without a hitch, but Frances will usually give me >a guru 81000009 when I issue the afm -m36c4c4 to set my memory (2Mbs). >Only once in a blue moon do I get it going but the next time I turn >it on, same thing... 8-( > My frances works in my "500" just fine. I did, however, have to add some grounded PCB board to the underside of the frances board to get it stable. Seems the EMI was just too bad without the extra shielding. -Vince >-Viet > Viet.Ho@samba.acs.unc.edu >
chanson@isis.cs.du.edu (Chris Hanson) (12/21/90)
>My frances works in my "500" just fine. I did, however, have to add some >grounded PCB board to the underside of the frances board to get it stable. >Seems the EMI was just too bad without the extra shielding. I had this same problem when I built a Palomax HD interface for my 500 motherboard. Since it is just a motherboard really, I built it with a right-angle connector so that the interface would like parallel to the motherboard. This did not work. It would work, but whenever the machine would get cranking, it'd get flaky. During heavy IO, I could tap a key on the keyboard (any key, even a shift key) and have a good 25% chance of that action locking the hdgeneric8i.device up. Wasn't pretty. This makes me wonder how the CMI boards, the Supra internals RAM boards, and the PP&S internal RAM boards handle this. Chris - Xenon -- #define chanson Christopher_Eric_Hanson || Lord_Xenon || Kelson_Haldane I work, but you don't know who I work for. And they don't know I'm here. "Angels define the edge, beyond which only fools go." -Me
swarren@convex.com (Steve Warren) (12/22/90)
In article <1990Dec20.182929.23757@isis.cs.du.edu> chanson@isis.UUCP (Chris Hanson) writes: [...description of EMI problem deleted...] > I had this same problem when I built a Palomax HD interface for my 500 >motherboard. Since it is just a motherboard really, I built it with a >right-angle connector so that the interface would like parallel to the >motherboard. This did not work. It would work, but whenever the machine >would get cranking, it'd get flaky. During heavy IO, I could tap a key on >the keyboard (any key, even a shift key) and have a good 25% chance of that >action locking the hdgeneric8i.device up. Wasn't pretty. > > This makes me wonder how the CMI boards, the Supra internals RAM boards, >and the PP&S internal RAM boards handle this. [...] It makes a big difference when all your wires are etched on a PC board with a ground plane under them. The transmission line effects of this physical arrangement can help your signal lines to act less like antennae. A wire wrap project is especially susceptable to EMI because in addition to having long loops of wire out in the air you also have wire coiled around the wire-wrap posts. This worked great back in the days when clock speeds were slower (1 usec bus cycles), but at faster speeds these circuit features turn into antennae. In really harsh EM environments boards are built like a sandwich with power planes seperating signal planes from each other and from the outside world. EM waves cannot penetrate a solid sheet of conductive material that is held at a constant potential, so this "shielding" strategy works well. When researchers want to study particularly elusive EM properties they do it in a "screen room," where the entire room is layered with copper screen which is tied to ground. As long as the holes in the screen are smaller than 1/4 of the wave-length of the EM waves they want to eliminate, the waves will not penetrate. -- _. --Steve ._||__ DISCLAIMER: All opinions are my own. Warren v\ *| ---------------------------------------------- V {uunet,sun}!convex!swarren; swarren@convex.COM