Lots of False Positive Lightning

For the past 2-3 weeks now I have been getting lots of Lightning indications 1-2 times a day. usually 20km++
Never had this happen all summer unless there were storms in the area,

Just search for ‘false lightning’ and you will find tips on eliminating the issue.

I will twist it 90degrees and see if that helps.

1 Like

i wished they added an option to suppress those false lightning detections. My suggestion to report lightning only when there is another strike within 15 minutes, wasn’t appreciated as they prefer immediate response. To bad that this makes the lightning detector kind of useless for people with false lightning like you and me. It could be as simple as a setting to turn on ‘suppress false lightning’ (could give a warning when turned on for the first time ‘note turning this on might delay the detection of lightning with up to 15 minutes’)

Even 1 min would be enough, at least for my case. If you need 15 mins you must have some crazy amount of false strikes.

nah, the idea is that when real lightning is approaching you get a warning. But when is it real? if it is very remote, it might easily take 5 or 10 minutes between strikes. So if the first strike happens at say 11.00, nothing would be reported, if nothing else happens in the next 15 minutes, this would be considered a false detection, if however at 11.03 another strike is detected, both strikes would show up in the graph and a notification is send to the user (if enabled). That would cause a 3 minute delay, but only on the first strike, the rest are all detected in real time (until nothing happens in 15 minutes). if the second strike would happen at 11.01 there would be only 1 minute delay, but if the second strike happens at 11.14 you would get a 14 minute delay (I would consider both strikes real lightning if they happen within 15 minutes of eachother). This 14 minute delay is unfortunate, but for severe storms lightning is much faster and the delay is only minimal. Adding this as an option, so users without false strikes are not effected seems a no brainer to me, but apparently other people think differently (they prefer real time, but inaccurate weather in a super simple ui, without options)

so to answer to you remark. this 15 minute window is only useful if there aren’t many false strikes at all. A second false strike in 15 minutes would be considered real lightning.


A question to @WFmarketing: can you please add a simple option to switch between high (“indoors”) and low (“outdoors”) sensitivity?
I use this with my Franklin sensor and helps reducing detection of false strike events.

wouldn’t the filter I described above be good enough for both indoor and outdoor? reducing sensitivity probably also reduces the distance at which lightning can be detected. My false lightnings are mostly in the 10 to 20 km range and some in the 20 to 35 km range. reducing sensitivity to below the 10 km range doesn’t seem a good idea to me.

No it’s not the same. The Franklin sensor itself has this option integrated. One could build a detector with this chip and have it in house or somewhere outside. They don’t specify anything about detection range degradation. One of my sensors is placed outside and was set to lower sensitivity and detects lightnings more than 40 km away.

1 Like

Will investigate possibilities with the engineering team.

Thanks, I believe it’s not a big technical challenge but you’re probably busy with other more important issues.
Just a hint, look here:

…and go to Chapter 8.7 (AFE and Watchdog), Table 15.

1 Like

My ID btw is 2115

Yes and no. There are sensitivity settings on the Franklin chip but they are fixed in the device firmware (can’t be changed over the air). The engineers at AMS (the guys who make the sensor) custom configured the settings for our application, and seemed to work well in field testing. Also, according to AMS, changing the settings will upset the statistics used by the distance estimation algorithm and is “not meant to be a user setting”.

Note, the “indoor” setting is actually MORE sensitive than “outdoor” which means, in most cases, it will cause MORE false lightning (since there are usually more sources of such indoors).

We agree that it probably wouldn’t hurt to program the ability to change the sensitivity over the air into a future hardware device (future AIR). But, we would need to consult with the AMS engineers again and probably do a lot of testing before determining if the ability to change the setting in the app is a net good thing or net bad thing.


How long does it take for a strike to be recorded on the dashboard and graphs? We had a thunderstorm roll overhead a couple days ago and I was happy to note that the WF Air notified me that there were lightning strikes even before I got a storm warning from the NWS!

However, when the storm was overhead we had several close and powerful strikes that I never saw get recorded. I just wonder if they get recorded later or if there is a filtering mechanism. The graph shows a total of 7 strikes. I know the Air can report a lot of lightning. This summer a town about 20 miles from here had over 2000 strikes reported in about an hour.

Hi @stevencmorse . Ya, it’s pretty cool when the lightning alerts are instantly sent to your phone. We even had one user tell us that he saw the lightning flash, got the alert, then heard the thunder! (We gotta see that one to believe it!)

The Franklin lightning detector in the AIR is designed to alert users about incoming lightning activity. It is not designed or high powered enough to measure and report every single lightning strike like the more expensive dedicated lightning detection networks. There are two different lightning graphs in your Smart Weather App: one that shows a count of strikes, and the other shows the approx distance of activity. There is no filtering of the data shown on the graphs. There is a filter on the frequency of the alerts (to prevent 2000 alerts in one hour!) — the alerts are sent once every 30min unless a strike is detected closer to you than the previous one. There is also some directionality to the sensor antenna…so storms from one direction may under report vs. other directions. The important thing to note: the lightning detector is meant to be an early warning alert system to keep you safe.


I’m having the opposite problem. My AIR is not detecting lightning strikes that it should.

Halloween we had some storms late in the day. My Atlas picked up 25 strikes, but not a peep from the AIR.

I’ve noticed this for quite a while now. My local weather station KXAN puts out lightning alerts, and I get a text if they detect lightning within 5 miles of my location (typically my house). I hear from KXAN first, and once in a while I’ll get an alert from the AIR, but most of the time not.

Tonight, looking at the Status page, I noticed it has been saying “LIGHTNING_DISTURBER” all night. It’s still stuck that way.

Could this status be why I’m not registering lightning often? I’ve no idea what the disturber might be. It’s now 2:00AM and the neighborhood is quiet. My Atlas seems fine (it uses the AS3935, too). The Atlas is on the south side of my house, the AIR on the north side on an open-air porch to keep it out of the sun.

For now I’m going to move the AIR close to the Atlas and see what happens during the next storm. That’s going to screw up my temperature, though, as I don’t have it in a shield, but I’ll live.

Any other ideas as to what might be going on?


First thing is to rotate the AIR by 90 degrees. And see what happens.

1 Like

If you have an old little portable am radio, tune it to a free frequency and walk around the air, you’ll hear the buzzz from many little or bigger appliances that ‘disturb’ the air strike detection (meaning of the warning you gave = many false signals I suppress = low level detection)

Maybe a nearby electric fence can ‘hide’ real strokes … etc etc. Those little detectors are very sensitive and many electrical soues can disturb it, low energy lamp, plasma tv, cheap transformers, printer … name it …

Happy hunting :slight_smile:

1 Like

Right now there really isn’t anything running to check. The most likely thing to cause interference might be some LED plant lights I have (an Aerogarden). They blew out a couple weeks ago, though. Aerogarden sent me a replacement unit, but it’s not on at the moment. I just checked and I still have the “disturber” status.

Next most likely candidate would be the neighbor’s outside AC unit, but it’s now in the 50s, so the AC isn’t running.

I do have an SDR I can use to look at different frequencies. Any idea what frequencies I should consider?

Also, Can the AIR/SKY receive firmware updates?

My AIR is on firmware revision 20 and has an uptime of 330 days.