cool I’m looking forward to it and will try and be your first order. I want to give them out as gifts plus being a now a real weather point on the TX Gulf Coast.
My wind reads low at higher wind values compared to a couple of Davis (one 5 meters from WF). Rain is very unreliable either with or without rain check.
Hope things improve with future updates but for now Davis is my go to. Not going to new hardware until I see some solid performance from what I have
The advantage of the old WF model is that it has separate parts for temperature and relative humidity of wind and precipitation. So I can place the Air unit in adequate Stevenson shelter at 2 m. With the new model this is not possible and for that reason I will not change the Air. Apart from this, I do not see that the problem with unreliable precipitation measurement has been solved.
I would be very interested in being a beta tester and could compare it to my existing Air and Sky system
Thanks
I think that having the temp sensor in the New Sky unit makes the reading equal or equivalent to all the others throughout the entire network of New Skys.
Right now its impossible to understand where it Air is actually mounted because each location is different and they effect temperature.
sky units are also placed in very different locations, different heights and on top of different surfaces. I’m assuming that they do use some AI to compensate for it. That AI could just as well be applied to the Air systems. (but I would prefer if it didn’t and keeps showing my local temperature)
Is there anything different about the new HUB vs the old one? I’d get the new HUB too if there were any substantive changes.
wouldn’t it be great if it came with humidity, temperature, co2, no, voc, o3, ozon sensor or something similar (a.k.a. build in breath ), but if not, what would you possibly like to have added to the hub. It seems to be doing its job just fine.
We’re playing with a few minor HUB tweaks, but nothing substantive. Purpose built for simplicity so that when network technology changes it’s both easy and inexpensive to upgrade (read:future proof). Backwards compatible with the forthcoming Tempest device.
Ponder this: when you know the solar radiation, the temp, the wind, and the thermo-dynamic properties of the physical device you can apply a crafty mathematical formula to correct for sun expsoure and effectively eliminate the need for a clunky Stevenson shield or waste power on aspiration. (Or you can opt to put the AIR as you have in your Stevenson at your choice.)
Plus, many precip performance tweaks we haven’t told you about yet. Have faith @bocestar.
If you need a BETA tester outside US in Finland, I’m interested
Would this unit be good for agricultural use or would herbicide sprays etc. damage the sensors? Or the only issue to worry about is false rain readings?
The Tempest is specifically designed for the common home environment with available AC power and WiFi. We would build to a different design spec for remote autonomous operation in commercial environments such as agriculture (ie. cellular comms, fully self-powered, +++)
Thanks. AC power and wifi will be available. Was just curious if the sprays would damage the sensors.
Will this be in ‘realtime’ or will it be a kludge like RainCheck a day later (making realtime data meaningless??
Interesting that it seems to combine the Air + Sky hardware into a single unit. I like that from the standpoint of consolidating everything together.
The Kickstarter preview again talks about what a good Smart Home player it is. Does that mean that this version will include the HomeKit support that was pulled from the original version? And while we’re talking about that, will the iOS software ever support Shortcuts? I could see some interesting workflows built from the device.
from the kickstarter page… “with Guaranteed Proven Accuracy”, sounds like a bit hollow phrase if you think about it. It guarantees that the accuracy is correct, but it still could be like 20 mm rain +/- 25%. The statement only tells us that the 25% is guaranteed. 25% still would be pretty bad. So what is the precision of the tempest data? (before raincheck a day later)
I would gladly test one next to my Air in a ventilated shield.
I would gladly do so as well.
Design notes: Did you know that WeatherFlow has an entire team of PhDs in Atmospheric Science working on progressive modeling techniques? A big component of the new Tempest system revolves around better forecasting . The general premise is to start with the absolute best global forecast models, then enhance performance using ground-truth observations to understand where/when/how the model actually gets it right/wrong/etc. It’s super technical weather nerdy stuff, but here’s a basic description of what we’re doing:
WeatherFlow starts with the most advanced global forecast model, the ECM/WF – the “European Model” – as well as various NOAA models. WeatherFlow then runs our own hi-resolution Numerical Weather Prediction (NWP) model which is further enhanced with observation data from thousands of proprietary weather stations. The output is a hi-res gridded forecast model that exceeds the accuracy of pretty much anything else out there.
Then, data from the WeatherFlow station in your backyard in used to inform an AI-powered post-processing technology that dramatically improves the accuracy at your exact location over time. Cool, huh? OK, now back to work for us.