On Sat, 7 Nov 2015 09:26:24 +0000 Colin Law clan...@gmail.com wrote: >> Jan Kandziora j...@gmx.de wrote: >> When you connect the ground line to data-, you have a different current >> on the common ground/data- line because the ground line also carries the >> supply current. So there's no way the receiver can successfully >> eliminate line noise. Instead, it picks up supply current noise. It just >> doesn't work. > > Does that argument hold in the case of parasitic power, where the data > and common lines do carry the same current?
And what about when you have an out-and-back path through the same CAT5, to maintain a single chain and avoid a "star" configuration? Or when you have other signals (relay coils!) in the same CAT5? I can't say I've tried the experiment of putting ground and data on different twisted pairs, but since I eliminated all the star branching from my (parasitic) bus (which always keeps data and ground on the same twisted pair) I can go for months of reading every ten seconds without a single error. Despite the total path now being much longer due to out-and-back. One other trick that seemed to help stop the last few rare errors on this long, perilous route was adding termination on the far end - a (reverse biased) 1N5711 Schottky diode and a 1.5KE20A-T TVS in parallel. I know, separate cables for 1-wire would be better, but it seems no matter how much cable I pull around this house, there is never enough! | Loren Amelang | lo...@pacific.net | ------------------------------------------------------------------------------ _______________________________________________ Owfs-developers mailing list Owfs-developers@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/owfs-developers
