It seems my feature request was beyond what’s available for consumer level use. Tempest has what I was looking for but at much higher cost & for professionals & business. Appears to be an optional addon for TempestOne systems. Lightning Data Solution | Tempest
The above link was already posted by @vreihen but it was interesting so I reposted it. I will stop pursueing this is & just do what I do… After a detection, simply open a radar app such as MyRadar to get ALL the consumer level, reduced accuracy position & direction information I need. Not a problem. Thank you for the reply.
I’m a bit late to the party, but it’s worth pointing out that direction detection simply isn’t part of the lightning sensor that is in the device which is on the unit itself. The unit can’t tell where a strike came from. In order to do this, Tempest would need to be able to preform a triangulation calculation from other Tempest sensors near by on the back end. This leads to another limitation of the lightning chip on the Tempest…the chip does not report the actual distance of a strike, it reports a calculated distance to the front edge of the storm. (I’m making these numbers of for the sake of argument) So if the sensor detects strikes within 60 seconds of 5, 1, 10, 15, 3 miles away…it’ll report all 5 of those detections back to Tempest as 1 mile, since that is the front edge of the storm. This is built into the firmware from the manufacturer, which isn’t Tempest so it is out of their control. So in this regard, all of the Tempest devices that could be triangulating data, would be using invalid data.
Simply put, the devices physical limitations make it impossible to do.
Actually … not the front edge of the storm … but, the closest lightning strike. Lightning can and does come from many parts of a thunderstorm, and, you can not infer the “front edge of the storm” from the closest lightning strike.
Appreciate the reply & I’m trying to learn. The AS3935 Franklin Lightning IC is cool tech. Here’s what I’m still not quite understanding… Tempest sends me notifications & well as records strikes on the horizontal time line & plots distance on the vertical axis. As strikes are detected, the distance of the newest vary quite a lot depending on where in the storm it occurred. So, based on seeing that 3 times now (new owner here), what did you mean about Tempest reports back all as one mile away? Reading about the chip on manufacturer pages, I don’t understand how it calculates distance. I’m a mechanical engineer rather than electronics . I’ve also noted that Tempest culls & removes strikes they determine to be false strikes on the backend? I’ll see that 11 strikes happened & 20 seconds later, only 7 or 8 are shown. It’s impressive how quickly their system responds. Am I saying/asking that properly?
yes - what you see on the app is a magic math manufactured value based on some undisclosed multiple sources of data that perhaps include your station. There are a lot of previous posts about this. It’s somewhat like the WF-altered rain values if you have that feature on.
Yes, I’ve noticed the rain ‘adjustments’. On first rain, Tempest was reporting .33 inches. I checked my old rain gauge/tube for a confirmation & it showed ~ .24 inches. Just moments later, Tempest downgraded it’s measured amount to .25 inches. For a split second, I actually wondered if Tempest had a built-in ESP chip (I read too much syfy) . What is unknown to me is whether this happens only on new stations during a calibration phase or does this happen as a regular tweak.
I wouldn’t be surprised at all if for its processing power it would use an ESP2866 or ESP32
Take a look inside a Purple Air and you’ll see an arduino board…
clearly an overpriced device. The replacement set of sensors is only about $50; asking 300 for adding an arduino…
Okay, you guys are messin’ with me head. Magic… Extra Sensory Perception ICs. I need a drink . There’s a front heading my way from the west. Tempest doesn’t see it yet. I’ll be on the lookout for anything supernatural.
if you drink more, those esp events are easier to spot. Cheers.
Don’t forget to let the dog out before the storm hits.
disagree - they have firmware, automation of uploads, a set of servers etc. to connect to, customer service, people behind all that stuff, manufacturing contracts, hardware/software designers, and stockholders. That stuff costs money to do.
of course you can cobble together something perhaps equivalent with the value of your time and blood pressure, but sometimes just paying for a shrinkwrapped solution that just-plain-works makes sense.
I think we are saying the same thing. The terminology the Franklin AS3935 chip that the Tempest uses, straight out of the spec sheet:
The AS3935 is a programmable fully integrated Lightning
Sensor IC that detects the presence and approach of potentially
hazardous lightning activity in the vicinity and provides an
estimation on the distance to the head of the storm
where the leading
edge of the storm is defined as the minimum distance from the
sensor to the closest edge of the storm.
In this regard, “Leading edge of the storm” and “closest strike” are synonymous. I guess if anything, the “Leading edge of the storm” can be confusing as it doesn’t really mean “leading” it means “closest edge” which is actually the tail end of the storm once it passes you.
https://www.mouser.com/datasheet/2/588/ams_AS3935_Datasheet_EN_v5-1214568.pdf
In any regard, the distance value returned from the strike is not necessarily the actual strike distance, therefore, it cannot be used in triangulation.
Well … from my knowledge of meteorology and experience, the occurrence of lightning has little (pretty close to none) of where any part of the storm is located, “head” (not really a meteorological term) or otherwise. Unless this sensor is also obtaining a visual of the storm (which it obviously isn’t) there is no way from a lightning strike to tell where any part of the storm is. Lightning can come out of the back and the sides of a thuderstorm, from under the base of a thunderstorm, from the anvil.There are things as “a bolt from the blue” (lightning striking 10+ miles from the thunderstorm) and “anvil crawlers” (lightning that traverses the underside of an anvil … usually staying cloud to cloud, but, some times it does produce a CG strike).
So, I don’t see how the sensor can give you the “head of the storm” or the “leading edge of the storm” or the “closest edge of the storm”.
The NWS advice is, if you can hear thunder, take appropriate action (for people, the general rule is go indoors, but, the dog mentioned previusly seems like a special case). They have the saying “when thunder roars go indoors” and also advise that, to be safe, you wait 30 minutes after the lightning seems to subside before going back out.
Sadly & strangely, every time I review the tech doc on that IC, my confidence in it’s ability to ‘calculate’ distance suffers a huge hit & my disappointment in their terminology & explanations increases. I’m no weather expert or electronics buff, but my bad smell meter keeps going off. I suppose for the money we all paid with no subscription fees, it’s the best we’re gonna get. In general, I’m satisfied with my purchase of the Tempest (even though I missed the sale price by just a few days).
The Franklin AS3935 was designed as a mass-market chip, to be embedded in all kinds of electronics so that soccer moms could get their kids off the pitch before the storm arrived. As I’ve posted here a gazillion times, trying to use it as a scientific instrument is fruitless bordering on downright dangerous. It works great as a layer of the onion when combined with other resources in a protection scheme, but not as the only source of information.
“For entertainment purposes only”
We’ve had training thunderstorms here most of the day. My Blitzortung alerts have been much more reliable than AS3935’s in my Tempest and two old WF Sky units.
Yes, I agree with all you said. I will continue to give Tempest kudos for their nicely packaged, compact & no-moving-parts weather station. But like you, will also tell folks not to put all their eggs in one basket. My secondary & backup GoTo is MyRadar, but there are many others that can & should be used to ‘confirm’ weather. The Tempest station provides what I was wanting without breaking the bank. This new customer thanks you for the reply!
PurpleAir offers a fine device & besides, what’s NOT overpriced these days? What I like best about PurpleAir is that a couple of them are located in my little town of 2600 people. Using the IQAir app, I see their data, get AQI, PM2.5 & detailed pollen levels. I see history, charts & forecasts for air quality & even get notifications if things are becoming ‘dirty’. Heck, one of them in town is operated by the US National Park Service. This is all 100% free & that works for me. Ain’t technology great!
Related to the IFTTT integration… I bought Tempest mainly for the Lightning detection. I have a few robot mowers that I want to park and power down when there is lightning in the immediate vicinity. When I configure the IFTTT trigger for Lightning detection, there is a configuration parameter for Units (Miles or Kilometers) but nothing about distance. RIght now, I’m getting triggers for lightning detected 25 miles away, but for my purposes that would be too far away to want to react to. I’d like to trigger an action only when lightning is within 10 miles. Can the IFTTT integration be enhanced to allow configuration of the distance?
Can we PLEASE get this? It would make the lightning trigger MUCH more effective and useful