Tempest voltage levels and power save modes, no rain detection etc

I used a 250W heat lamp placed far enough away to keep the Tempest temp below 100°F and has a couple of white Styrofoam pieces on each side forming a corner and was able to charge a very early Field Test unit.

I’m in Kelso so we have similar weather. So far all the Tempests since the early one has been staying charged so far. I wouldn’t worry about trying to charge the Tempest until it drops closer to 2.3V because it will mean you get even less data recorded during the artificial charging.

That said, what is your Station ID or link to it to see if maybe it has some charging issue.

I mistyped my voltage and said 3.4V but I meant to type 2.3V (since corrected above).

I do get a small amount of charging on some days, but not enough to turn on rain reporting, again.

If I knew the pin-outs on the 4-pin connector I might try my own charging (using a voltage and current limited power supply). This is a very low-voltage device but a few mA trickle should help. Seems like they could supply a charging adapter so we could buy a few days of operation at a time.

You have to leave the power switch on during charging and you need to leave the hub powered up so you can monitor the charging progress, but that means if you are indoors it’s sending bogus data to Weather Underground unless you deactivate the link. Very annoying.

What is your siting like? Is it getting more shade now than earlier in the year? If it is still getting a good view of the sky then maybe it has charging issue. That is why I asked what your Station ID was.

WF is working on an external power option which I think you’ll like. Not sure when it will be released but the design specs look really good.

For what its worth, 11 days later, my unit got just enough of a recharge to come back on in Mode 3… no interference on my behalf the entire time it was down. Got a notification on my apple watch while i was away from home and did a little public happy dance… now if only it could charge up enough to stay alive.

I agree with @general.delivery , an ability to charge via a wall-wart or a 12v battery and/or the much anticipated power-pack would be welcome as the sensor doesnt seem capable of currently surviving a dark PNW winter in my region surrounded by tall trees. Missing out on the most active weather period for the greater Seattle area is a bummer for me


Absolutely would be a nice feature to be able to do a charge for a day to buy a week or two in the usual dark, grey, wooded Seattle area. My Tempest has been offline now for a couple of months even while being in the sunniest part of our yard. But sunny is a very relative term when the sun has to cut through 100’ tall Douglas Fir trees because of the low sun angle.

And, yes, an external power charger should trigger an “offline” signal so that data is neither recorded locally nor transmitted to either WeatherUndeground or WeatherFlow’s AI prediction system.

Having same problem with low sun angles/cloudy days. First rain sensor shut down (happened before w/ low power) now, with 2.33 volts - offline

perhaps I should mention my suggestion here as well… add an option to set sampling rate not faster than once every 15 seconds (just as the sky). This should help the tempest unit running longer, before it drops to sampling once a minute or worse. (@dsj).

1 Like

I like your last two sentences. I’m patiently waiting for this item to climb aboard the “Tempest Train!”

I’m pretty concerned about this. I had a relatively sunny morning/early afternoon yesterday and it peaked at 190W/m^2. I’ve moved it to try to get as much direct sunlight as possible, but the forecast is grey for most of the next week. I love what this unit does, but if it shuts down for half the winter it’s pretty much a useless summer toy. I’m also not interested in charging every week or so. Dont suppose anyone has any ideas of how you might connect solar panels that are actually the size that should have been spec’d to charge this thing?

We have Tempests in northern Europe, Canada and Alaska that are staying charged as we head into the dead of winter. Vertically aligned solar panels are very efficient at capturing low sun angle. The panel size and numbers as well as the battery voltage requirements and even stretches of cloudy weather were well thought out. Siting limitations may be causing limited sunlight leading to low battery issues, if not, it could be a hardware issue instead. Please contact our customer support team so we can look into your case and assist! Thank you.

1 Like

Just had a look to your station, it is up since +/- 3 days ?? Where you are it should be able to charge ok once there is a sun ray, but direct sun on the panels, not just around and the station in the shadows as that won’t be efficient, especially this time fo the year. Can you try to have a look Wednesday or Thursday when there is a slight chance ou have some sun ?? Sometimes moving it a bit can do wonders. For now you still have enough to keep going for a week. A few hours direct sunlight will quickly charge it. I hope you can site it at best.

Thanks eric, I’ll keep an eye on it. I moved it today
to raise it and get a longer day, but I don’t have anywhere I can put it that will have clear sight of the sun all day at this time of the year. It is pretty low at the moment.
I like the ‘slight chance’ of some sun, my fingers are crossed, but it looks unlikely!
Thanks, Alan

do best you can, each minute of sun is one extra :slight_smile:
And for me no sun expected before next Monday with rain, rain, rain … wet south of France this winter :joy:


Is there a required minimum of W/m2 for the device to start charging?

Mine starts charging at about 100W/m²

1 Like

I would love to know that, but what is measured by the device as solar irradiance might not perfectly match as the solar panels are mounted vertically.

If you are interested you can mount a second tempest horizontally :wink:

1 Like

it is probably better to model the sun and atmosphere and just calculate the relation between the measured solar irradiation and charging (depending on location and time of the year).

I’ve sent a message to support. 1 week in and it’s dropped into the first power save level, that’s pretty pathetic for something that’s meant to record weather observations. I so like how everything has worked so far, but a battery life this poor makes all the other technology pretty pointless.

Hi @sunnysideal it’s possible there’s a hardware issue instead or a siting limitation. A perfectly functioning battery and solar power module will keep the Tempest charged, and out of any power saving modes unless bad stretches of weather occur.


You can use the table below to get a very rough estimate of the plane of array (POA) irradiance received to determine if it’s enough to induce a charge on the battery.

Determine the following information:

  • Installation latitude
  • Month
  • Peak solar irradiance on a previously clear day

Let’s assume a 45 degree latitude, in November, with a peak solar irradiance of ~250W/m^2.

We use the peak factor table to reference a “1.9” ratio based on 45 latitude and month of Nov.

Multiply 250W/m^2 x 1.9 = 475 W/m^2. This value is the plane of array irradiance on a South facing panel (which doesn’t actually exist!) but that gets us closer to a more representative value.
In this scenario, we might expect the 2 primary panels (SE, SW) to induce a charge on the battery around solar noon. Remember the charging threshold requirement is about 350W/m^2.

You can also use the ratio of the peak factor to quickly determine if the POA irradiance is going to be greater or less than the GHI value measured by the light sensor.

1 Like