Rain seem off (auto-calibration system not yet deployed)

[WFstaff]

Is this ‘array of other high-res precip data’ a live adjustment requiring connection to the WF servers OR is it only to aid the long term calibration?

Hi @iladyman . Two parts here. Calibration and ongoing QC.

Calibration: Well, if your station is not connected to WF servers, we won’t have your raw data to compare making it impossible to calibrate. You could stay connected until we apply initial calibration, then drop off, but you will lose the ongoing Continuous Learning benefits.

QC: For those who want the cool network stuff, the additional precip data will be used to QC the accum data and tune/correct if needed on an hourly (still to be confirmed) basis.

I wish to know if I have the latest firmware and calibration if I am not connected to the WF servers could my instantaneous rain amounts record a different amount compared to if I was connected to the WF server?

If you have the latest firmware and have received an initial rain calibration your raw haptic sensor should be dramatically improved. HOWEVER, it’s called a Continuous Learning system because it’s always learning, comparing, ensuring – but only when connected. If not connected to WF servers and only using raw data via UDP, it is quite likely you will see different accum amounts as the raw does not have the benefit of corroboration will all the other reference sources. You should, however, see the similar instant rain intensity values. Any disparities will decrease over time as capabilities improve.

[jblt]

My WF Station “Oak View Bluff” is still reporting less rainfall than my other rain guages. I have a Honeywell tip guage that reports in 0.03" increments, and a good old fashioned open ended rain guage. During the last series of storms the WF station reported a total of 0.86", the Honeywell reported 1.61" and the open ended guage had 1.74".
The WF station is definetly sensitive enough to detect the slight rain when it starts, but seems to no be able to handle the .75"/hr and greater rates we get here in Oak View, CA. The storm today, 1/31/19 has been very intense at times with WF total at 0.92", the Honetwell at 1.81", and 1.78 in the open ended guage. Rainfall rates have exceeded 0.75"/hr several times today, but the WF max rate is 0.65"/hr?
I have been patiently waitng for the CL process to correct the readings based on surrounding data, but still no change in the totall rainfall and rainfall rate.
Any suggestions are appreciatted.
Jeff Hall
Oak View Bluff

[eric]

Hello Jeff

I moved you post in this thread that will give the answers, I know, loooonnnggg read but even the title has the answer

[iladyman]

This morning between 2:45am and 6:30am my two plastic funnel gauges recorded average of 1.25mm rain. My Sky recorded zero. The wind readings in this display are correct. (We had a Southerly change cold front come through):

And other small rain events during January:
28th Jan 2019 Funnels=3.7mm, Sky 2.1mm
21st Jan 2019 Funnels=4mm, Sky 0.2mm

cheers Ian

[iladyman]

Hi @dsj,
For several hours light rain has been falling but nothing has registered on the Sky sensor. Here is what it looked like at the beginning:

The bottom of my plastic funnel gauge is dripping and some is collecting and the upwind side of the fence is wet

Water is dripping off the car:

It has continued since those photos for several hours with still nothing recorded.
The air is white fog due to cloud base being below me.

But on 21st Nov 18 during windy conditions it reported well mentioned here: Rain calibration since firmware update v94

I am currently on station firmware version 103, Sky version 43.

edit additional data:
The light rain continued and by 10:30am 2/2/2019 funnels avg 9.5mm, Sky 0.3mm.
Heavy rain commenced at 10:30 for 30min: funnels avg 4.5mm, Sky 2mm.
cheers Ian

[brad1]

Dear WFStaff,
My WF Station is about a week old, and went thru my rain yesterday, here are my findings. I have a Davis VP2 that I get off but want to make sure the sensors are working. For the Rain, the WF Station matched VP2 up to the first .10in after that, the WF was giving me errors. The final total rain was .95in and the WF showed only .52in. With this said, does the sensor “learn”. The sensor is mounted about 2 - 4 feet away from the Davis VP Tipping Bucket, and the sensor does NOT move/sway during high winds. The sensor is firmly mounted. Station ID 6238. Any and all information will be greatly appreciated.

Edited to add this. *** I will be a beta tester for the rain CL, going to have rain all weekend** in San Diego, California ***

Ciao~
BH

[lovemyram4x41]

I don’t if anything happened with calibration or CL since the last time it rained here but it rained all afternoon yesterday and things went in the other direction.

WF data and UDP were still off by 30% but this time WF data was slightly less than other local stations, especially the ones in direct line with path of storm (around .75"). On the other hand the UDP data was extremely high (around 2.25").

While I prefer both to be close, this is actually better since things like my Rainmachine use the WF data and when I get my new HA controller set up I’ll probably use websocket for my weather data.

[iladyman]

Hi @dsj,
I posted several posts complementing rainfall accuracy when I had Hub firmware revision 102 (between 17/11/2018 until 14/12/2018). It was fantastic, picking up light rain, not recording false rain, and being pretty accurate in ‘normal’ rain.
Since I have had Hub firmware revision 103 (14/12/2018) it has been fairly inaccurate. I dont know how an auto calibration will fix it because I can get 10mm soft rain without Sky even realising that it is raining. I was so happy with how it was during version102. But you didnt mention rainfall detection changes for version 103?

New Hub Firmware v103

The latest Hub firmware, v103, is rolling out to all stations starting today, 13 December. There are two significant changes since v98: Improved wind speed QC algorithms (addressing the issue discussed in this thread) Improved RH overflow correction for older AIR units (which should resolve the dreaded “overflow” issue for anyone still experiencing it) In summary: most users will not notice anything and some users may notice an improvement.

Is the rainfall detection changed without changing the Hub firmware revision?
I would like to be able to lookup when any calibration changes.
cheers Ian

[dsj]

Hi Ian. Thanks for the continued reporting! The answer to your question about firmware is that there was no change in the rain calibration function between v102 & v103 so any differences are coincidental.

Thanks to all for your continued reporting. We’re inching toward a major Rain CL update and your observations are helping drive that!

[brad1]

Dear DSJ,
My observations about the rain, for the past two storms we had, the rain was very accurate up to .10 in of rain, after that the readings are a bit low, for the day, the total rain was 1.10 in and the WF Reported .70 in. We have another storm today and tomorrow and will keep you posted, but the first tenth of a inch if perfect in two storms. I hope this helps you.

Ciao~
Brad

[bocestar]

@WFstaff
This morning I have about an hour of light rain.
Manual measurement recorded 0.5mm and WF recorded 1.0mm.
Some early measurements of rain recorded more rainfall with WF than manual measurements.

[bocestar]

Last night with drizzle. WF recorded only 0.02mm and manual measurement recorded 0.2mm.

[vidguy7]

@dsj-Add me to the list of recent undereporters. After first installing my Sky on 1/19/19, the first rain event was a bit over an inch, with my VP2 recording 1.14” and my Sky 1.13”, and impressive delta of only .01”. I was impressed. But subsequent to that it’s been going downhill.

With each subsequent rain event, my Sky has been undereporting. Thus far today my Davis has reported .32” and my Sky only .11”.

Yesterday my Davis reported .19” and my Sky only .12”. The Davis readings are consistent with trusted sources in my area, so the Sky is definitely underreporting. Nothing has changed in terms of location or mounting.

Station #6187

[khulguy]

Same results here as well. My VP2 recorded about triple the rain of my Sky however, results earlier in the rain event seemed more accurate. Things fell apart as time went on.

[tom06784]

I had same issue. Early in rain my WF and VP2 tracked closely with WF slightly ahead in rain total but by the end of the rain VP2 had more. Seems that WF can’t account as the rainfall rates drop.
Yesterday’s totals: WF .22 inch till midnight and nothing after midnight.
VP2 had .27 yesterday and .03 today.
Tom

[khulguy]

Thats interesting Tom, that your theory is the WF can’t keep up when the rain rate lowers. My thought was that maybe as the rain event continues and the top of the unit “puddles” perhaps the additional drops aren’t registering? Its interesting that the event starts out fine, but then fizzles. In addition, my wind direction was off this morning (after the rain was over) and I was getting false gusts. After drying off the unit both inside and out with a hand towel, the winds corrected.

[vidguy7]

I had the same spike in winds (recorded as 37mph when in reality they were about 10) as water collected in the area of the sensors this morning. However after drying my Air off, the winds returned to normal.

[tom06784]

I have had the problem with the wind direction once but it clears up after the unit dries.
I have had the same thought that existing water drops were keeping the new drops from being sensed but since it seems to happen at the end of the rain event no matter how long the duration I think it just has trouble sensing lite rain especially when it is already wet.

[khulguy]

I am looking forward to seeing how it handles a heavy rain event and if the results are in line with established weather stations. To this point, I have just seen it during a steady rain event

[dsj]

Thanks for the feedback, folks. Keep it coming. We are reading every word and looking at every data point! We’re hard at work on both issues (rain calibration and wind discrepancies due to rain on the plate).