Weather Station Discussion

Forums › Gadgets, robotics, home automation, electronics (including wearables) › Weather Station Discussion

1 | ... | 20 | 21 | 22 | 23 | 24 | 25 | 26 | 27 | 28 | 29 | 30 | ... | 36

107 posts

Master Geek

#2871924 21-Feb-2022 12:24

dwl:

I’m sorry that his thread is looking like a Tempest one, and I have just installed one, but I have a question about rainfall accuracy. I got the Tempest partly based on rave comments here but also because I was getting tired of the cheaper tipping ones jamming. Over the weekend we had around 50mm of rain in the Wellington area and I was looking forward to more data. Surprised to see only 22mm recorded on Tempest, 40mm on tipping one and 43mm in a glass tube reference gauge.

I know that you can upload other data and Weatherflow will then push out calibration corrections, which I will do, but how bad should it be out of the box? I understand over reading can be vibration or even bird feet. Looking at the other stations in the wider area I see massive variations. Has anyone else out there focused on rainfall accuracy and been through calibration phases.

I did have the same issue and the readings were pretty close to what you have posted.

I did send the 2 other readings from both another weather station and a manual gauge (both within 5 metres of the Tempest) along with the Tempest readings for 3 separate occasions and they got back to me around 5 days later confirming that they had done a recalibration. We have only had 2 rainfalls since then but the readings now are virtually spot on.

Hope this helps.

Lenovo

Shop now for Lenovo laptops and other devices (affiliate link).

Technofreak

6525 posts

Uber Geek

Trusted

#2871974 21-Feb-2022 12:54

Niscoupe:

dwl:

I’m sorry that his thread is looking like a Tempest one, and I have just installed one, but I have a question about rainfall accuracy. I got the Tempest partly based on rave comments here but also because I was getting tired of the cheaper tipping ones jamming. Over the weekend we had around 50mm of rain in the Wellington area and I was looking forward to more data. Surprised to see only 22mm recorded on Tempest, 40mm on tipping one and 43mm in a glass tube reference gauge.

I know that you can upload other data and Weatherflow will then push out calibration corrections, which I will do, but how bad should it be out of the box? I understand over reading can be vibration or even bird feet. Looking at the other stations in the wider area I see massive variations. Has anyone else out there focused on rainfall accuracy and been through calibration phases.

I did have the same issue and the readings were pretty close to what you have posted.

I did send the 2 other readings from both another weather station and a manual gauge (both within 5 metres of the Tempest) along with the Tempest readings for 3 separate occasions and they got back to me around 5 days later confirming that they had done a recalibration. We have only had 2 rainfalls since then but the readings now are virtually spot on.

Hope this helps.

What I am curious about is did they recalibrate your unit or did they massage the data you see from their server that you see for your unit?

I'm pretty keen on one of these but I want to use the data directly from the unit rather than just the "corrected" data they send from their server. I will no doubt use the "corrected " data at times but I want to know that the raw data is a true reflection of what is really happening when I'm making use of it.

Sony Xperia XA2 running Sailfish OS. https://sailfishos.org The true independent open source mobile OS
Samsung Galaxy Tab S6
Dell Inspiron 14z i5

decibel

311 posts

Ultimate Geek

#2871984 21-Feb-2022 13:06

insane:

Can someone please explain briefly (don't spend too much time) why you might want to have one of these?

You can look outside and see it's raining, sunny, windy, snowing etc, and weather reports make some attempt of predicting future conditions. I just don't see why you'd want to spend money so that you can pull out your phone for it to say it's hot when you can already tell.

They do it for the same reason that I have a Stratum 1 NTP server at home as part of the public pool.

It is ... geeky.

neb

11294 posts

Uber Geek

Trusted

Lifetime subscriber

#2872060 21-Feb-2022 14:38

insane:
Can someone please explain briefly (don't spend too much time) why you might want to have one of these?

To see whether the rainfall rate is such that the street drains are going to overflow again and flush water down the driveway to the house.

To see whether the outdoors humidity has fallen to the point where we can open the basement windows to let air through.

To see whether the rain amount and rate is sufficient to avoid having to water critical garden plants for another week.

To see whether the wind direction and force is enough to push rainwater through the leaky window frame.

To notify that it really is that hot, or that cold, we're not imagining it, and we need to fire up the heat pump.

To ....

All of this has to be very local, even 500m away the conditions are quite different (we live on an inlet on the upper harbour surrounded by hills).

Bewildered

104 posts

Master Geek

#2872074 21-Feb-2022 14:54

Why you might want to have one of these?

In my case for the following automations:

When the washing line is out (there's a simple contact sensor for that) alert me that it is raining (our house is so well insulated we don't hear it at all)
When the light fades to the point of getting 'darkish' automatically turn on the evening lighting (so not dependent on time of day - cloudy days it triggers sooner)
When the temperature gets close to the dew point and the washing line is out issue an alert to get the washing in
When the light fades to the point of getting 'darkish' and the washing line is still out issue an alert to get the washing in
When the light fades to the point of being 'dark' activate automated motion-triggered lights around the house so stairways and hallways are illuminated automatically
Measure the temperature and rainfall for the day to set the irrigation schedule
If wind gusts exceed a certain level issue an alert to secure the sun umbrella
If the average daily temperature exceeds a certain value automatically disable the schedules for heat pumps and underfloor heating (great for spring/autumn where some days it's cold and other day's it's warm).

Once you have one you're always cooking up new ideas, particularly when someone complains about something....you're like...."Oooo.....there's an automation for that!"

Deanonzl

51 posts

Master Geek

#2872102 21-Feb-2022 15:44

My Tempest is to supplement my Davis VP2 non Solar-UV.

These sensors are about an extra $400 each & only last 2+ years before our harsh UV kills them.

Rest of the Davis is ultra accurate rainfall & wind-speed measurement.

Tempest also provides lightning awareness as well.

So far, the Tempest doesn't measure the soft rain/ drizzle very well, but might get better over time as the software advances.

Even though I have linked the stations to the rainfall adjustment service, I've yet to see an automated adjustment.

Last big rain we had here, 50mm on the Davis, 32mm on the Tempest.

Also, it can be integrated into home automation when I get a round to it.

DjShadow

4055 posts

Uber Geek

ID Verified

Trusted

#2872166 21-Feb-2022 16:36

dfnt:
DjShadow:

I've e-mailed a fault through to Weatherflow, I don't think the "Feels like" temp is working correctly, the temp it shows always matches the actual temp instead of taking into account humidity.

e.g my station homepage says its currently 21 degrees with 88% humidity, the feels like temp is also 21 degrees. The station near me which feeds into MetService is saying 23 degrees, 88% humidity and feels like 27 degrees.

Get a response from them?

I noticed mine and several others around me all have the same actual and feels like temp

This is what they came back with:

The heat index will only be calculated if temperatures are equal to or above 80F (~26.66C) and RH is equal to or above 40%. If you're curious you can view the derived metrics here: https://weatherflow.github.io/Tempest/api/derived-metric-formulas.html

dwl

371 posts

Ultimate Geek

#2872879 22-Feb-2022 19:47

Deanonzl:

My Tempest is to supplement my Davis VP2 non Solar-UV.

These sensors are about an extra $400 each & only last 2+ years before our harsh UV kills them.

Rest of the Davis is ultra accurate rainfall & wind-speed measurement.

Tempest also provides lightning awareness as well.

So far, the Tempest doesn't measure the soft rain/ drizzle very well, but might get better over time as the software advances.

Even though I have linked the stations to the rainfall adjustment service, I've yet to see an automated adjustment.

Last big rain we had here, 50mm on the Davis, 32mm on the Tempest.

Also, it can be integrated into home automation when I get a round to it.

Good insights thanks. I have a second requirement for primarily wind at the top of the farm which gets the strongest in Wellington. Any comments please on the speed range and accuracy of the wind measurement of the Tempest compared to others.

dwl

371 posts

Ultimate Geek

#2873589 24-Feb-2022 07:21

dwl:

Good insights thanks. I have a second requirement for primarily wind at the top of the farm which gets the strongest in Wellington. Any comments please on the speed range and accuracy of the wind measurement of the Tempest compared to others.

Found some answers to my own question. https://community.weatherflow.com/t/feedback-on-accuracy-of-wind-sensors/11949 The reports of rain affecting the ultrasonic sensors seem to be a real problem and while Tempest is great innovation it is looking like very much indicative only. May look for spinning cup type for wind.

elpenguino

3359 posts

Uber Geek

#2873745 24-Feb-2022 11:08

dwl:

dwl:

Good insights thanks. I have a second requirement for primarily wind at the top of the farm which gets the strongest in Wellington. Any comments please on the speed range and accuracy of the wind measurement of the Tempest compared to others.

Found some answers to my own question. https://community.weatherflow.com/t/feedback-on-accuracy-of-wind-sensors/11949 The reports of rain affecting the ultrasonic sensors seem to be a real problem and while Tempest is great innovation it is looking like very much indicative only. May look for spinning cup type for wind.

You'll need a decent structure whichever type you use.

Cup type instruments are a maintenance overhead so you might decide it is better to put up with poor performance when it is both windy and rainy in return for less sensor work.

Or you could spring for an ultrasonic unit with higher transducer output levels e.g something in the professional range. https://r-p-r.co.uk/windsystems/windsonic.php

Of course, measurements on a ridge will be higher due to orographic effects and are therefore not comparable to measurements taken at lower elevations. Still fun to see how high they get though.

Most of the posters in this thread are just like chimpanzees on MDMA, full of feelings of bonhomie, joy, and optimism. Fred99 8/4/21

DjShadow

4055 posts

Uber Geek

ID Verified

Trusted

#2890322 22-Mar-2022 20:38

That was a good little storm that just crossed Wellington, I would of thought given how much Lightning there was it would of picked up more. Only seemed to detect the strong/longer flashes.

Bewildered

104 posts

Master Geek

#2890327 22-Mar-2022 20:57

Same here....it only counted 35 strikes which is very low given the amount of lightning activity....but I also noticed that it couldn't detect any strikes when the front got really close to us....seems ~10km is about the limit.