Hacker News Reader

Comment by bri3d | Hacker News Reader

I was recently introduced to fully integrated lightning sensor ICs: https://www.digikey.com/en/product-highlight/a/ams/as3935-fr... - supply it a repurposed RFID coil and it will tell you when it detects lightning and how far the storm front is. I have one in my weather station just for fun, but it's a highly practical device for golf courses, swimming pools, sporting facilities, etc. where knowing how far away lightning is is operationally important.

View on HN

Better yet, this network

https://www.lightningmaps.org/#m=ses;t=3;s=0;o=0;b=;ts=0;

triangulates lightning strikes detected by this method. Right now I see a front moving between the Triple Cities area and Scranton.

Here is a board ready to use:

https://www.sparkfun.com/sparkfun-lightning-detector-as3935....

I was heavily involved with the local soccer community, and as a coach/referee I carried my own detector that I paid less for a ready to go version than this DIY shield. I was not going to depend on someone else's gear functioning when these were so cheap.

Got a link to that detector? The cheapest self-contained lightning sensor in a portable form factor I know of nowadays is the Acurite 02020 and it's around ~$50 nowadays.

That's interesting. That's the same one I have, but I only paid $27.99 for it. I couldn't remember what the model was, but Amazon shows that I purchased that model and clicking the View Order link it shows $27.99. I guess I bought at the right time??? Maybe it's tariffed to death?

https://camelcamelcamel.com/product/B00EO1H3X8

Looks like it's steadily increased, labor maybe? Just a WAG on my part to be sure, I'm too far away to legitimately speculate.

likely the chip in it is out of production

lots of products like this are increasing because nobody is running the older cheaper fabs

I was considering one of these recently. By chance do you have any feedback on how well it works relative to the detection and the storm front tracking?

I hate these chips after a multi-year crusade to get one to work. I've bought 3 different variations of it on breakout boards (SparkFun, DFRobot, and another one off of Tindie) and spent 10s of hours fiddling with them trying to get them to work with no luck. That includes multiple attempts at tuning the antenna and calibrating the capacitance, tuning all the other detection parameters, multiple different microcontrollers, several different power supplies and physical arrangements etc in case of interference. Anything and everything I could think of and test.

Everything I did failed and I have never been able to get anything other than noise out of these sensors at best. YMMV, as I know these chips have a history of problems but supposedly do actually work. After I finally gave up on these things I ended up getting an Acurite 06045M, a cheapo SDR dongle, and a Pi Zero W running rtl_433 for like $60 and that has worked perfectly for years and is fairly straightforward to plug into Home Assistant over MQTT. It very reliably detects strikes and gives a pretty good estimate of distance (but NOT bearing), plus you get an outdoor temp/humidity sensor for free.

I think the 06045M might actually use the AS3935 internally so presumably you can get them to work and when they do work they work very well, but nothing I could do as a (pretty advanced and well-equipped) hobbyist could make them work, and I've heard others had similar issues with them. So maybe stick to buying something commercial with it inside?

Yes, that Acurite sensor and the one I have connected to my weather setup (Shenzhen Fine Offset / Ambient Weather) are all AS3935 based. I also have an Aliexpress "JMCU" AS3935 breakout/antenna board that I bought to play with after being pleased with the Ambient Weather sensor, and it seems to work OK to me as well - I had to tune the watchdog thresholds and gain to eliminate false positives, but after doing so it works great.

I did try tuning the watchdog thresholds/gain too but never had much luck. I read a few places that there might have been problems with some of the early AS3935 chips (but could never find too much info) and I know there were problems with some of the breakout boards on Aliexpress having the wrong capacitors on the board (which was another thing I tested/fixed). The one thing I never tried was all my boards are fairly old so they might have whatever the original hardware flaw is and maybe newer ones work as intended.

I've been mighty tempted to try again with a newer one as I really want to build a compact portable lightning sensor linked to Meshtastic but I dread going through the whole ordeal again.

Your trials around the AS3935 piqued my interest a little and I thought that maybe the Blitzortung/LightningMaps people may have tried working with them. It turns out they have! The earlier generation Red board documentation [1] has a section on them and there was apparently internal-to-operators discussion [2] which isn't directly public, but which points to wxforum.net where there were quite a few results with at least one [3] detailing struggles to get it working.

This probably isn't directly helpful but I found it a fun trail to follow for a bit.

[1] https://www.blitzortung.org/Documents/TOA_Blitzortung_RED.pd...

[2] https://forum.blitzortung.org/showthread.php?tid=1022

[3] https://www.wxforum.net/index.php?topic=22235.0

Thanks for that, many of the problems described there mirror my own experiences. Part of me really wants to take yet another stab at it with a newer board but I also am so burned out on it that it feels like borderline PTSD.

Hmmm so they're fiddly is what it comes down to. I was thinking of the one from Mikroe. The 06045M has the same 40km as that chip so I believe the other commentor that it's the same chip. Thanks

I get reliable detection out to about 25-30 km, but it depends on your placement and local RF environment. I hung my Acurite sensor on a nail up under a roof eave out of direct sunlight so as to get reasonably accurate air temperature readings and that seems to be pretty good for lightning as well. I'm in a bit of a valley which probably limits my range somewhat, I am fairly confident the 40 km range is possible with the right placement.

Good to know, thank you

I'm interested in this type of device in a boating situation. The network based systems wouldn't work for me.

Can somebody provide clarity on if any of the boards really work?

It seems that bri3d has faith in the ScioSense.

They do but without information from additional locations all you can get is distance and sometimes direction (dependent on specific chip / circuit). The ones with direction will probably good enough for most boating awareness situations.

If you want to know more specifically you need to have separate sensors that have with reasonably synchronized clocks so they can trilaterate or triangulate strike locations. Those sensors probably need to be further away from each other than most boats allow for. The further away the sensors are from each other, the tighter the time synchronization between them, and the more sensors you have will determine the overall accuracy your system will be able to attain.

> so they can trilaterate or triangulate strike locations

I don't think that's feasible. These sensors work by detecting radio waves which propagate near the speed of light. Even if you had two sensors separated by 1 km, they'd both receive the radio signal within ~3 µs of each other. I don't think the sensors guarantee detection time to that level of precision, nor that it's feasible to keep clocks sufficiently synchronized that you can timestamp events at a sub-microsecond level.

You're partially right for the case of trilateration, usually these would be placed about 10km apart with a GPS PPS clock (at least) and would be using sensors with a much faster response rate.

With triangulation you only need the clocks accurate enough to correlate the events. This will largely be determined by the rate you see events, and you only need to distinguish specific events if you want to differentiate locations of individual strikes as opposed to a storm front. For a boating case you probably only care about the location of a storm front.

Two sensors measuring angles from a couple dozen meters apart can get you a pretty precise location of storm front especially when aggregating results from multiple detections.

It's quite feasible. There are a few approaches. It's common to build lightning detectors with four antennas in a square, which gets you a rough vector to the s strike. That's 1950s ham radio technology.

Synchronizing multiple receivers is common. That's how passive ADS-B locators work.[1][2] This works so well there have been attempts to stamp it out.

[1] https://www.adsbexchange.com/what-is-the-exchange/

[2] https://adsbglobal.com/

Thanks all. I will look into the square configuration but I also have doppler RADAR direction might not be as crucial as knowing that static charge is building up.

Reply on news.ycombinator.com