Hello, I install Air and Sky yestetday afternoon. My first impressions were very good. I put Air device in Stevenson screen. I measure data for National hydrometeorological service of Serbia for almost ten years. I have thermometer with mercury. Night readings from WF was nearly same as classical thermometer. Today I have sunny weather and temperature rising very fast. Temperature data from WF rising slowly then from other thermometer. Now temperature going down and WF temperature data again delays. My notice is that temperature sensor has a problem with fast temperature changing.
2 Likes
Hello
What is happening is the fact the Air is massif compared to your other thermometers. The mass needs time to react (both ways). I’m pretty sure the sensor is sensitive enough but the plastic around has to warm or cool down … Not much we can change except take out the housing and put it naked in the screen. But it isn’t build to be taken apart.
5 Likes
Thank you eric for fast answer! I notice too that air device have much mass then other similar stations.
I installed the Sky device on the roof of my house 13. December. On Sunday, during the period from 4 am to 9 am, snow was moderately intense, and in the period from 9 am to 1 pm the snow of high intensity was falling. I noticed that wind speed measurement at that time was somewhat lower than the actual situation outside. It seems that heavy snow precipitates accurate wind speed measurement.
Hi @bocestar . Welcome to the family. Regarding rate of temperature change: it is less about the mass of the device and more likely due to ambient air flow (or lack of) inside your Stevenson screen. The relatively thin plastic casing of the AIR does not retain much heat/cold. To experiment, place your comparison thermometer inside the same Stevenson screen. Also, if you have a prevailing wind direction at your location, rotate the AIR such that the vent port on the side faces the incoming wind direction for optimal air flow exposure. Let us know the results.
(Design note: AIR was designed to be placed outside in full shade. If full shade is not available, place in a properly ventilated radiation shield, preferably actively fan-aspirated to ensure adequate airflow to conform to changing ambient conditions.)
2 Likes
Hi @WFmarketing
I came to conclude that the ventilation of the shelter was the reason. I have another shelter with better ventilation and there is a different situation in it. Thank you for your support!
What do you say about my observations regarding wind measurements during heavy snowfall?
1 Like
Hi @bocestar,
Glad to hear the more ventilated shelter is helping.
About the wind readings during heavy snow… sampling the wind becomes difficult in heavier precipitation. Any object in the path between transducers will affect the reading; rain drops, snow, ice even insects can have an effect on wind readings. See this topic: Can Rain Affect Wind Readings - Looks like Frost can as well (Updated)
2 Likes
Hi @WFstaff ,
Yes and to me it was quite logical that heavy precipitation with a large number of droplets or flakes prevents the passage of air to sonic sensors. An interesting idea is to reduce or prevent this problem with grids.
@WFmarketing After the measurement in the first few days, are the calibrated sensors for wind direction or just for speed of the wind?
We are currently experimenting with nano-particle coatings that claim to drastically reduce water droplets and snow sticking to SKY.
When first installed, SKY will go through a self-balancing routine to ensure the sonic transducers are correctly aligned — critical for timing of sonic signals and determination of wind speed. See SKY / Tempest: wind sensor initial self balancing.
Wind direction is determined by the user installation. Point the “N” marking on the side of the SKY device facing directly North. (Tip: use a compass app on your phone to determine North.)
2 Likes
Good news 
My WF smart station has been working for ten days and I am very pleased with the work of all sensors. I only think that UV and solar sensors still give higher values ​​than real ones. But as I understand you with the sensors, it takes about a dozen sunny days to calibrate these measurements. I just did not have much sun in these ten days and I probably will not have it in the remainder of December. The winter season is here in the east of Serbia, usually with few sunny days.
Is there any plan for your few thousand existing users who don’t have that (hopefully coming soon) magic on their installed units ?
@bocestar . Great! We’re thankful to have you as a customer. Regarding UV, the key is to have a number of days with no clouds around the time of solar noon so that we can benchmark the UV value at your exact location. See this: CL system for UV explained . Once your first calibration is set you will see a pretty drastic improvement that day, then just very minor tweaks (likely unnoticeable) on an ongoing basis.
@vinceskahan Options pending results of the application testing. Don’t worry so much, we’ll stand behind our products and take care of you. It’s all part of the process to develop something new and better. If that’s not for you and you’d rather send your WF station back for a full refund please let us know via DM.
3 Likes
After a few days with intense snowfall last week, in the last few days I had a rainy days. I own a classic rainwater calibrated by the Republic Hydrometeorological Service of Serbia for which I have perform climatological measurements for the city of Knjazevac, eastern Serbia.
Here are comparative data:
22.12.
Manual 0.4 mm
WF 0.3 mm
23.12.
Manual 0.1 mm
WF 0.1 mm
24.12.
Manual 1.2 mm
WF 1.0 mm
25.12
Manual 3.4 mm
WF 5.1 mm
My observations are that during light precipitation in a short period of time, WF is extremely precise, while in the case of rainfall that lasts more hours and precipitation with a higher intensity, it records more precipitation than the real.
@WFmarketing Unreliable precipitation measurements continue. Very light rain started last night and fell with the breaks. WF did not register anything. Rainfall was less than 0.1 mm. From the early morning there is a steady stream of rain. Manual rainwater measure 1.0 mm, and weather station WH 1080 measures 0.9 mm. At the same time WF shows only 0.1 mm and only occasionally detects precipitation.
Hi @bocestar . Thanks for the report. We have yet to launch our Continuous Learning system for Rain. This system is designed to evaluate the calibration of each and every station once mounted in situation. In addition, the system will us e the raw rain intensity data from SKY as one of many input variables to ultimately determine the accumulation values. Stay tuned and see this thread: Rain Gauge Reading Seem Off (auto-calibration system not yet deployed)
@WFmarketing @anon84912554
Here is a comparison of today’s precipitation. Very light rain, can be said drizzle, started early in the morning. After the first couple of hours of drizzle, after that the rain fell with light intensity.
My handy rain meter measured 0.9 mm.
The WH 1080 station also recorded 0.9 mm.
[quote=“bocestar, post:237, topic:930, full:true”]
@WFmarketing @anon84912554
Here is a comparison of today’s precipitation. Very light rain, can be said drizzle, started early in the morning. After the first couple of hours of drizzle, after that the rain fell with light intensity.
My handy rain meter measured 0.9 mm.
The WH 1080 station also recorded 0.9 mm.
Weather Flow records 0.7 mm.
There is a lack of precipitation from the beginning when a very light rainfall or drizzle fell.
@anon84912554
I install Sky unit 9. december and then reset data 13. december when I set Sky on my roof.
In a period from 9. to 13.12 I have one sunny day, after that in december I don’t have any more sunny days.
In January I have nearly 9 sunny days. No full sunshine but mainly sunny in more of this days.
My Sky and further give bigger values then real.[quote=“bocestar, post:196, topic:1341, full:true”]
Before calibration can happen it needs several good sunny days. Maybe the few days you had weren’t enough to have a stable and reliable sample.
It is winter for us, so even the noon sun isn’t high nor strong. Be patient. It will kick in soon when days are really becoming longer
1 Like
@bocestar . Our computers (CL System) report only one day recently, 17Jan, at your station location with clear skies at solar noon – not enough data yet to confidently set your UV cal. The system does realize the max UV at your location on this date should be 1.6 . We’ll have the humans look at it.
1 Like
@bocestar . Our data scientists applied an initial UV calibration to your station today. As the system collects more full sun (during solar noon) days, the sensors will be tuned on a continuous basis.
1 Like