Gary, after carefully watching data from an anemometer and thinking about this issue more, I have decided that a 10-minute moving gust-window is more reasonable than I had thought. So unless you are already underway with what I described below, let's forget about it.
Bob On Thursday, February 23, 2017 at 8:53:39 AM UTC-8, tempus wrote: > > With a 10-minute gust window, a high-velocity-gust 10-minutes ago still > determines the indicated gust level, even though gust velocities since that > time may have been much lower. The gust indication in that case is > 10-minute-old information. I am wanting gust information that is much > closer to real time. A min_interval <= loop-period with min_interval set > to 3-seconds may be too short to provide useful data, because of the > high-frequency filtering-effect of cup-rotational-inertia. If so, some > small multiple of that interval can be used instead. What I am wanting is > a gust-window that is much shorter than 10-minutes, but of course long > enough to differentiate between very recent gust velocities and the > periodic latest-point-in-time wlatest velocity samples. > > Bob > > On Thursday, February 23, 2017 at 12:31:21 AM UTC-8, gjr80 wrote: >> >> On Thursday, 23 February 2017 17:55:12 UTC+10, tempus wrote: >>> >>> I am wanting glatest to be the peak gust over the moving rtgd >>> 'min_interval'. I realize wgust already provides peak gust over a 10-minute >>> moving window, but in contrast, glatest would provide near-real-time gust >>> information if min_interval is made short. >>> >> >> I really am having a hard time coming to grips with this. I understand >> your definition of glatest; but *glatest would provide near-real-time >> gust information if min_interval is made short*? I don't see that a 10 >> min gust calculated now or a 30 second gust calculated now are any more or >> less 'near-real time'. One uses a smaller window than the other but they >> are both just as near real time as each other. If decreasing min_interval >> makes glatest 'more near real-time' does reducing it to 0 may it as real >> time as it gets? On one hand you seem be advocating pumping out wind data >> as fast as you can (which implies min_interval == 0) but on the other hand >> you want max(windSpeed) over min_interval. On a Vantage, and I suspect most >> all of the other stations, glatest=wlatest for min_interval <= loop period. >> >> Nothwithstanding, I will look at doing adding it.All of the data is being >> cached so it should be easy enough to do. >> >> I realize SteelSeries Weather Gauges enhancement is beyond the scope of >>> rtgd. However, standard installations would ignore glatest and remain >>> compatible and I can provide a modified gauges.js file with a >>> near-real-time gust pointer to weewx users who want it (*or ideally to >>> the SS Weather Gauges author if he is willing to incorporate it for weewx >>> users*). Unrelated to the gust issue, I have fixed gauges.js coding >>> mistakes that should be submitted to him anyway. >>> >> >> No problems, just wanted to make it clear I wasn't touching gauges.js. >> >> Gary >> > -- 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.
