On 16/12/2011 01:08, Nathan Hurst wrote: > On Thu, Dec 15, 2011 at 09:21:32PM +0000, Colin Law wrote: >> On 15 December 2011 15:44, Nick<[email protected]> wrote: >>> ... >>> After fiddling a little and getting nowhere, I brought in my laptop, LA and >>> some other kit to the office to explore more. Curiously, with my Dell laptop >>> (and same LinkUSB master), the network was now 100% reliable. Plugged into a >>> different Dell desktop in the office, again a disaster. >> >> This has got to be a big clue. One possibility that occurs to me is >> that the 0v on the desktop may be connected to the PC chassis and >> therefore earthed, whereas on the laptop it may not be earthed. Is it >> possible that that there is another earth connection somewhere on the >> bus so that you are getting an earth loop. this is possibly born out >> by the other points you note below. It is absolutely imperative that >> if you have an actual earth connection then that is at one and only >> one point. > > I'm not sold on the different pairs having different predictable > properties, they are all made the same way. I think this is just a > function of the specific cable used.
Thanks for the replies. Why it made a difference is probably because of using the earth that's paired with the data, rather than the earth paired with power. I don't think I tried using the orange/white earth from the LinkUSB connected to blue/white instead, and don't see why that would make any difference as I expect the earths in the LinkUSB to be commoned. I seem to recall guidance being not to connect other pairs to earth, but it made the difference here. My take is that you need to be prepared to experiment if things don't work by default. Each pair has different numbers of twists per unit length, but I've not found differences with simply choosing different pairs. The reliable home network uses the orange pair for power and green pair for data. I used Clipsal pink cat 5 there because the insulation is supposed to be approved for running into a space where there is mains and I was playing safe, though I suspect that other cat 5 is fine too, just never gone through approval. I have some vague recollection of Clipsal saying that their cat 5 was higher quality than some other cables, but maybe not. I have some Clipsal cable left so might try that, but I expect that there would be the same problems with the original wiring choice; would be an interesting result if the cable choice did make a difference though. So anyway, what fixed most of the errors in the office was switching the wire used for earth, wiring all the other earths (but only using blue/white for the slaves) and the pullup. I'm interested to see how behaviour changes with a few spurs too, i.e. does it get more reliable and would the original wiring choice then have been fine. Nick ------------------------------------------------------------------------------ Learn Windows Azure Live! Tuesday, Dec 13, 2011 Microsoft is holding a special Learn Windows Azure training event for developers. It will provide a great way to learn Windows Azure and what it provides. You can attend the event by watching it streamed LIVE online. Learn more at http://p.sf.net/sfu/ms-windowsazure _______________________________________________ Owfs-developers mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/owfs-developers
