Rain calibration corrections

Thanks. This is an excellent demonstration of how much the specific installation/mounting can affect an individual SKY’s calibration. And it’s also a good demonstration of how well the SKY tracks the “shape” of the rain intensity - note how well the curves from both devices track each other. However, without in-situ calibration, the SKY will over- or under-report. That (in-situ calibration based on your SKY’s individual data) is what we’re working feverishly on - stay tuned!


I can do one better.

The “true” rainfall for the 10pm to 2am period as measured by 3 standard measuring cylinder gauges co-located was 8.0 to 8.2mm. (The logger that records the tipping bucket gauges was not working – forgot to press the record button).

I think Sky1 and 2 are in a beta group. They track each other very very closely over time and in the total recorded. Both 4.4 mm for the period
Sky1%20rain%2025_10_18 Sky2%20rain%2025_10_18

Sky 3 was the last to be added and has been in place a couple of months. It is way out at 31.2mm for the period.

This may have been asked already, but given that in-situ calibration is so important are you working on a way for those of us with alternative rain gauges at the same site to feed that data into the CL model?


you must really love rain. 3 gauges, a tipping bucket and three sky units. Me, on the other hand I hate rain :-), so I have only one sky unit (which stopped displaying accurate rain rates two months ago)


Yes, if your tipping bucket data is available online somewhere, we will be able to use it to help calibrate your SKY units.

I also have issues with under recording actual rainfall amount. I also have cylinder style rain gauge. The Sky measurement is close to 1/2 the cylinder number

It would be nice if we could calibrate Sky ourselves. In many weather programs I’ve used, we have an option to enter the “actual” recorded rain based off of the rain cylinder within the program.

Thanks for your report and for participating in the forum! We’re still working to fully understand all of the issues that affect rain calibration but we are making steady progress. Unlike tipping buckets, the individual conditions of your SKY’s installation can affect it’s calibration. There’s some more discussion on the forum here, like this thread:

We will be rolling out additional updates to the rain calibration process over the next few weeks.


Just my 2 cents here… if you are trying for ‘no vibration’ mount you do not want steel. Shock mitigation requires isolation not transmission. Plastics, rubber, etc. isolate. PVC is one way. Not stiff enough? Drop a broom handle inside or a length of wood same as the PVC

1 Like

As I previously reported there was a bug in some of the SKY units and an update to the firmware has taken care of the extreme over reporting. Great support from WeatherFlow. Steel mast is working just fine.


Station 5203


Is there something I need to do in order to update my firmware, or does that happen automatically?

This morning we received our second bit of rainfall since last winter. (Yes, it’s been a crazily dry year.) This time I have the exact opposite problem as my original post in this thread, where instead of my WF station reporting half as much rainfall as surrounding stations, today it’s reporting twice the rainfall that surrounding stations are reporting. :confused:


Just up the hill from you and I’m having the same issue. For the previous rain event, mine tracked really well with both my other station and surrounding stations. Now it’s about 3x surrounding stations and more like 5x my other station.

We had some rain today and nothing recorded on sky…

1 Like

First rain event for me since installing this past summer. Weatherflow read about half of what the Davis shows. My Davis generally reads very close to my real (non electronic) readings. The siting of my stations are very close and all in the open. Do not think that the sitting would account for more than a 50% difference.

I see lots of inaccuracy reports in this topic with both high and low reports. Is there hope this is a fixable problem?

07%20AM 47%20AM

Yes. Me too.
Mine is horribly inaccurate vs. my VP2 and manual gauges (which always agree).

It’s been several months now since most of us got ours. Hard to keep the faith. I know I certainly can’t recommend WF at this time to anybody, no matter how great their people and support have been (and they’ve been indeed pretty great).

1 Like

My rainfall measures appear to have gone crazy recently, as I am now reading double local stations and one other Smartweather unit. Today I have a reading of 4.4mm whereas the local comparison has 2.7mm. The other day mine read 18mm, whereas the other local unit read just over 10mm. At one stage it seemed stable, now it’s well out of wack…

We are inching closer to applying individual calibrations but currently all SKY units have the same “one-size-fits-all” settings. Thanks for your patience!


Just wanted to report that my Sky’s rainfall amounts appear to correspond with what surrounding stations are reporting. Not sure if I got an individual calibration from the good folks at the WeatherFlow mothership or the Sky recalibrated itself, but at the moment, rainfall appears to be accurate. Yay! And thanks!



It seems my Sky reports now about 1.2 too much rain it was factor 2.7 too much seems to be improved. Are callibrations done??

Hmm not yet i read elsewhere, strange🤔

When light winds and rain (windspeed 16kph or less) I get very good correlation between my Davis Vantage Vue and Weatherflow. Higher winds and the discrepancy seems to get worse (up to 3 times the Davis reading).

Does the haptic rain sensor take into account the wind speed when measuring rainfall? Light rain has a very low terminal velocity hence less effect on the haptic sensor. Wind will increase the apparent effect of the rainfall won’t it?

1 Like

Thanks for the report!

Not yet, but it most likely plays a significant role. We’re still evaluating how to best account for that variable (and others).

Overall, yes, although the effect is complicated. Wind changes the angle of attack, which tends to spread out the drop impact over the cap. This reduces the amplitude of the impulse but the total energy transferred is the same. Wind also causes more drops to impact the sides of the SKY which tends to add to the signal as well. We’re still working out the relationship between wind and rain, but we’re making progress!


My Sky station seems to be off by -0.3 compared to stations around me (non-weather flow stations). Just wondered if this was an issue or if my unit was bad. I have a clear view of the sky with no obstructions around the unit. Any thoughts? 58%20AM