On Saturday, 10 February 2018 22:58:32 UTC+10, mwall wrote: > > On Saturday, February 10, 2018 at 3:26:22 AM UTC-5, Cameron D wrote: >> >> Hi all, >> I am back after a long period of doing other things and have started >> merging Matthew's latest code into my version of the driver, which >> uses only PyUSB v1 >> I am looking to create a unified version that works with both V0 and V1 >> APIs. >> > > camera, > > i think we have the libusb/pyusb issues under control. it was basically > and issue of detecting timeouts and 'no data' "errors", since the wmr300 > driver needs to ignore those. since v1 is backward compatible with v0 api, > we should be ok. > > at some point it might be useful to make weewx aware of different pyusb > and libusb versions, but imho that should be done in a weewx.usb module, > not in each driver. (see the 'usb' branch for one way of doing it - that > generalization also makes it possible to support platforms like macos and > other hid implementations) > > i think the remaining issues for wmr300 driver are: > > 1) figure out how to clear the history properly. the rc5 driver does this > (based on the code you wrote!), but ruben's station seems to respond > differently than others, so we're still not quite there. > > 2) figure out how to reset the rain counter from software. this might not > be possible. > > 3) address any of the long-term issues you have discovered. did any of > those cause weewx to stop? > > thank you for continuing to collect data and for the detailed analyses. i > suspect we might have to make some of that instrumentation easy for any > wmr300 user so that we can figure out what is happening with hardware like > ruben's > > m >
Hi Matthew, 1. I might load RC5 onto my system and see how it behaves, Do you have more information on Ruben's problem than posted here? 2. Resetting the rain counter - I decided that it is so unlikely they have allowed for that possibility that I am not looking to test it. 3. It stopped once, but restarted, so not an issue. I put it down to possibly a hardware glitch. Other issues I might look at 4. Battery levels. - I have pored over captures when the console and Windows software changed the display indicating battery issues, but I just cannot work out which bits did it. 5. clock drift - they did not even consider the possibility of customers outside North America or Europe, so their windows software does not even worry when the clock is miles off the actual time. I am sure there is no way to remotely fix it so I am hoping for some way to issue an alert. Mine drifted by over a minute in about 10 months. Cameron. -- You received this message because you are subscribed to the Google Groups "weewx-user" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. For more options, visit https://groups.google.com/d/optout.
