WXforum.net
Weather Station Hardware => Blitzortung => Topic started by: Cutty Sark Sailor on May 09, 2014, 05:10:09 PM
-
BLITZORTUNG Network
NEW Beta Test Real Time Lightning Mapping
http://www.lightningmaps.org/realtime?lang=en
This will really show how the Net Clicks! Especially this afternoon!
From Tobi: All strokes in last 60 minutes in all world network regions. New ones added almost in real time — delay is up to 5 seconds! The best results can be seen in Central Europe due to the high station density.
Please note that this is a very first beta version! There's still some work to do. We can not guarantee for anything!
-
It's really weird how once in awhile it will flag a flash, and then 'pull' the marker as location becomes more defined as more stations log it... Wild.
Example Video recorded 5/9/2014 Click pic, or use http://www.ourspecialnet.com/Weather/TwinHollies/i-gzjCJMN
(http://www.ourspecialnet.com/Weather/TwinHollies/i-gzjCJMN/0/L/LightningMaps.org%20-%5B2%5D-L.jpg) (http://www.ourspecialnet.com/Weather/TwinHollies/i-gzjCJMN/A)
-
I like it!!!!! \:D/ \:D/ \:D/ =D> =D> =D>
Thank you, Tobi!!!!
Most folks don't realize that to get lightning data like this in near real time from Viasala/USLPN would cost a fortune, or for that matter from Strikestar when you own a Boltek.
-
I like it!!!!! \:D/ \:D/ \:D/ =D> =D> =D>
Thank you, Tobi!!!!
I agree. This is phenomenal!
Interestingly, while the Texas Gulf Coast is getting hammered right now, the Texas stations' participation is minimal in the strikes my station is picking up.
-
I like it!!!!! \:D/ \:D/ \:D/ =D> =D> =D>
Thank you, Tobi!!!!
Most folks don't realize that to get lightning data like this in near real time from Viasala/USLPN would cost a fortune, or for that matter from Strikestar when you own a Boltek.
Yep - last I checked, it was >$90/mo subscription PLUS some sort of startup fee.
As an engineer, I'm very excited about where the Blitzortung system is going. I have a feeling this is just the start--there are threads at the internal Blitzortung forum discussing VHF interferometry. That's the sort of thing that allows you to capture branching cloud-to-cloud lightning, and the Vaisala systems that do it generally run $20,000 or more. Captive market :)
-
On a related note, it would be really nice to serve up the raw lightning data via some sort of push or streaming.
My immediate goal is to create rapid-update lightning placefiles for GR2Analyst.
-
Try the "Stations On" mode, it shows lines to which stations participated in the strike detection. This is too cool! 8)
Greg
-
Try the "Stations On" mode, it shows lines to which stations participated in the strike detection. This is too cool! 8)
Greg
I agree! It's amazing to just sit and watch the strikes roll in and see which stations participate, for instance seeing a strike down in the gulf off of LA and watching stations in WI and IL light up, or this one where stations out west of the Rockies were involved:
-
Try the "Stations On" mode, it shows lines to which stations participated in the strike detection. This is too cool! 8)
Greg
Oh, wow!
That really brings home how good the network has to be to be able to quickly locate multiple simultaneous flashes.
-
That is way cool!!!!!! \:D/ \:D/ \:D/ Awesome job, Tobi!!!!
-
Awesome job Tobi!!!
Its a work of art watching the higher station density midwest work the storms in MO right now.
~Lance
#724 - Owings, MD
-
This also points out the ongoing need for more station density. Big strikes are pretty clearly picked up by stations across the country, but I'd bet there are lots of little ones that simply don't get picked up where there's more sparse coverage.
Bring your friends, bring your family....
-
What's cool is when a stroke is also detected by our west coast stations... getting a stroke across those mountain ranges, at that distance...
then the maps 'lines up' across the whole continent...
Can you imagine what it'll be like during real widespread intense storms? Sheesh!
-
If you zoom in all the way, you can also see circles that, I assume, represent some standard deviation(s) of error in strike placement.
-
This also points out the ongoing need for more station density. Big strikes are pretty clearly picked up by stations across the country, but I'd bet there are lots of little ones that simply don't get picked up where there's more sparse coverage.
Bring your friends, bring your family....
Yeah... nobody can detect 'em all during intense situations... E-field may help with some of that, still we can only individually process so many at a time. One big reason a network of stations 50-200 miles apart is a goal... closer together is super cool. Can't see a signal if we're swamped with other sferics and local noise! Then there's Cutty's "Thor's Hamma Slamma Jamma" theory... :twisted:
-
This also points out the ongoing need for more station density. Big strikes are pretty clearly picked up by stations across the country, but I'd bet there are lots of little ones that simply don't get picked up where there's more sparse coverage.
Bring your friends, bring your family....
Yeah... nobody can detect 'em all during intense situations... E-field may help with some of that, still we can only individually process so many at a time. One big reason a network of stations 50-200 miles apart is a goal... closer together is super cool. Can't see a signal if we're swamped with other sferics and local noise! Then there's Cutty's "Thor's Hamma Slamma Jamma" theory... :twisted:
It looks like that theory might apply to northern Missouri this evening.
Out of curiosity, does the network run a server stateside, or are they all in Europe?
-
Amazingly cool! Fun watching the line west of Chicago....
-
Didn't notice the "stations" feature before - really cool.
Now all that is needed is a webcam on the ISS to watch the flashes from space at night.
(http://scijinks.jpl.nasa.gov/_media/en/site/aviation-game/earth_lightning-lrg.jpg)
-
As far as I can tell, even if you are in Region 3, the server that is pointed to is in Germany.
But this brings up a concern I had last year and didn't get any comments.
This is a hobby for us, and of course the developers. I can't imagine the amount of time in design and production and setting up servers, say nothing about educating themselves about lightning and how to detect it.
On the other hand, there are a lot of contributors out there already, and if, don't jinx it now, if something strange would happen (accident, heart attack, etc), the system would be dead in the water.
I'm wondering if we could find a suitable situation where a server could be set up here, in the US? Obviously Egon would have the control, but unless he's tweaking code frequently, to upload stuff to him in Germany for EVERY noise and strike seems a bit of a waste, even in this era of ridiculously cheap internet across oceans. And those data could be sent packaged to him from the US. And if someone tripped over the cord in his office, this part of the system would stay up. If he and the rest got bored before we do, as long as the servers kept running, we'd still have this to play with. No further development nor bug fixes, but when he gets caught up on all the new stuff and fun things they have figured out, I wonder if they should be approached about having a 'remote' server here in the US.
By the way, I'm very pleased to see the enthusiasm for what the latest real-time mapping has brought to our computers. Indeed there was a time I was toying with robbing a bank and buying a month's worth of lightning subscription from Weather Tap just to have the high resolution. And a friend who worked for a major power distributorship had contacts within the company with such a subscription and they could determine, to some degree, which pole was hit or within a stretch of poles.
To have this system is an incredible toy and to me, who has dreamed of this network for over a decade but having neither the smarts nor organization to create it, am very pleased someone else does and I can be a small part.
-
Dale,
I know what you mean about waiting for a network like this. For a long time, the three things I really wanted in terms of weather data were near real-time, high resolution nexrad data, lightning data, and model data.
A NOAAPORT link would help with some of that, but mostly I just had to wait for bandwidth and availability. Nice that both are here now.
-
I sold a Boltek because I was wanting this type of network. I wish they could have put one together. I suppose it would be too much to ask if the Bolteks with GPS could send data to this network. Probably different data formats or something.
-
Gawd, this thing is now worse than a lava lamp.
I was out working all day blocking up trees and moving dirt, and after a relaxing shower and a little Cutty Sark, find it almost impossible to turn that display off. Anyone know if it is a drag on resources to leave it on and watch?
Damn you Tobi.
-
Dale I felt the same way earlier. I had to tear myself away from the computer so I could get ready for movie night with my girlfriend. Guess I'll be staying up late when I get home...
-
Kevin,
well, the storms began to quiet down a bit. Hate to wish for Tstorms that might hail or worse on someone.
With our weather luck, having been given this gift, we'll have few storms this season. But here's hoping.
And I'm wondering what the E-field antennas coming on line in the next few weeks will bring.
Enjoy. Hope you saw a good movie.
Dale
-
Anybody note a station in Europe :shock: hitting infrequently on those Missouri cells last night? :twisted:
Click Pic to play...
Hint: there's at least one in this recording from last evening...
(http://www.ourspecialnet.com/Weather/TwinHollies/i-BctM4SL/0/L/LightningMaps.org%20Real-Time%20Lightning%20Map%20--L.jpg) (http://www.ourspecialnet.com/Weather/TwinHollies/i-BctM4SL/A)
I'm pretty sure Tobias knows about it... :grin:
Mike
-
Saw that too last night, Mike. Station 913 is testing in Switzerland before going to Brazil. Incredible.
-
With all of the lightning near here this morning, this display is phenomenal. A new switch to me (I didn't see it yesterday, anyway) is 'Stations'. Turn it on, and you can see the stations that contributed to the strike in near real-time as the green lines radiate out from the strike to the contributing stations! But you guys probably already knew that... :-)
-
The data were already available from the strikes but us ordinary folk had no way of taking it and displaying it as we now are experiencing. It's been already said how 'cool' it is to have this visualization. I don't know about other users, but I had an ongoing question as to whether or not the strikes I send to the server were actually being used, and how did I compare (yeah, a man thing) to the other users sending data, beyond the efficiency statistics.
Now, in real time or nearly so, I can see my station participate. Sort of like watching your kid or someone you know play ball, puts a bit more interest in it.
My son (who does NOT play ball) wandered by and wondered what I was doing. I showed him and he indeed thought it was cool, so I guess the intergenerational gap has been spanned by Tobi.
But he almost immediately pointed out that it looks a lot like the Star Trek movies, (with their bigger production CGI budgets and high resolution) with battles that displayed the star ships moving and maneuvering very close to one another, with lasers or whatever those things were they fired at one another, sort of tracking as they moved past one another during a battle. And I have to agree it sort of does remind me of that, especially when there are two or more cells some distance or states apart and the participation-indicator lines 'paint' the strike.
Does it make the data better? I don't think so, but I'm still fascinated by the circles (I assume sort of like the DOP on GPS signals) and the visualization of what is going on.
What fun. And I really don't think it will end up being something we'll ignore in the future, I doubt it will grow old for me.
-
Klingons?
What's changed this morning? A few minutes ago there were only green participation lines.
Just now, before trying to go work, I looked again, just one more peek. Now from some of the areas south of the big Nebraska storms there are BLUE lines coming in? Non-Federation ships? Anyone know what the implication of the color difference is?
Dale
-
One more benefit of this display that just became visible is being able to see how close a station can be and still participate without being knocked out by signals from close storms.
I was watching the Omaha storm, and good ol' 798 began to participate in the storm a hundred miles or so to the east of him, despite a large storm just passing and still very close.
A side benefit to having this new capability.
Are there more things you've discovered and may not be obvious to the rest of us yet?
Dale
-
Are there more things you've discovered and may not be obvious to the rest of us yet?
Dale
"We're not gonna tell you, we're not gonna tell you, we're not gonna tell you..." :-P :twisted:
-
Klingons?
What's changed this morning? A few minutes ago there were only green participation lines.
Just now, before trying to go work, I looked again, just one more peek. Now from some of the areas south of the big Nebraska storms there are BLUE lines coming in? Non-Federation ships? Anyone know what the implication of the color difference is?
Dale
Polarization or E- vs. M- field detection?
-
The color links seem regional, that is green above and blue below. E field would be shorter distance, and these are indicating participation stations from some distance away.
-
One more benefit of this display that just became visible is being able to see how close a station can be and still participate without being knocked out by signals from close storms.
I was watching the Omaha storm, and good ol' 798 began to participate in the storm a hundred miles or so to the east of him, despite a large storm just passing and still very close.
A side benefit to having this new capability.
Are there more things you've discovered and may not be obvious to the rest of us yet?
Dale
798, that's me. I've been running in 'Automatic' and 'Filtered' Modes for the past week or so with excellent results that far exceed what I was able to achieve in Manual mode. Currently my 'B' channel is stuck at 2*1*40 amplification, but overall performance is so good that I'm just going to leave it for now.
-
The color links seem regional, that is green above and blue below. E field would be shorter distance, and these are indicating participation stations from some distance away.
I believe the color has to do with the distance from the station to the strike. :-k
-
Watching mine, I'm thinking one antenna vs the other.
- Jim
-
Captains Log:
The Blues seem to be taking over. I, and a few remaining northern stations in the alliance, seem to be the only green stations participating.
Initially, we were all green. Then, without warning, the conversion to Blue began. Now, it seems other than a few strongholds left, almost all others are blue.
I wondered if it was indicating a degree of efficiency, but Jim at Mauston with his high levels remained strongly green, no hint of blue.
Then I wondered if it was ferrite vs. loop or ball antennas? So far I've not been able to research it enough to prove this, but with a large number of ferrites out there, I can't imagine they number so few on the display.
Since I'm still green and running a RED board, and the Rochester MN site is a green board and also showing green, this theory seems false.
My only hope is that we can hold out long enough to find out what has crept into the system to explain this blue conversion.
I hope I can survive being green before blue overcomes me.
-
Well, I've seen Jim in Mauston paint a Green and a Blue, so it must be distance from the station.
With the stations south of the Omaha and Evansville storms being distant they would be blue. Those of us a wee bit closer to the action would be green.
Is the mystery solved or is this a false theory that seems to fit the data?
-
Green=<500km Blue=>500km
-
Hey, you weren't supposed to tell us.
Thanks.
Anything else to divulge while we have you under oath and on the stand?
-
Hey, you weren't supposed to tell us.
Thanks.
Anything else to divulge while we have you under oath and on the stand?
Naw.. pay more attention to what Dr. Obbins suggests... he told you several posts above. ;)
-
The nodes and interconnects reminds me of old telephone switch routing diagrams, back in the days of hierarchical circuit switching, before everything became packetized!
Don
WD9DMP
-
I see the lighting dector only comes in a kit forum....to bad I am lucky I can program my cell phone let alone solder something
I do see there are some in Wisconsin that have one now....might be a over kill if I would have one anyway.
-
Hey, you weren't supposed to tell us.
Thanks.
Anything else to divulge while we have you under oath and on the stand?
Naw.. pay more attention to what Dr. Obbins suggests... he told you several posts above. ;)
It was either distance or Klingons vs Romulans - just a lucky guess.
-
Doc,
Well, there was speculation as to board color, antenna type, etc.
I thought the most logical explanation was Klingons....
This is all fun, hope no one takes things too seriously
I do note now that hovering over the stations selector says something about shade of blue.
My browser kept the stations selected as ON, so I didn't have to go near the button to hover.
Still neat. That is one heck of a storm in southern Iowa now.
-
Doc,
I thought the most logical explanation was Klingons....
This is all fun, hope no one takes things too seriously
I was enjoying you Star Trek commentary. My kids are all grown up.
-
If you zoom in all the way, you can also see circles that, I assume, represent some standard deviation(s) of error in strike placement.
I would say you are correct. After watching the system for some time last night and today it appears that the more stations report a strike from various angles the tighter the 'error ring' around a strike will be. I also noticed that the system seems to update those in batches, probably at an interval that is dependent on the amount of activity in the network i.e. more strikes = longer interval between error updates but I could be wrong. I just know that I was watching and trying to time how long it took to get the error determination for a few strikes when the system updated and the rings appeared around multiple strikes simultaneously.
-
Dear Mr. Wind,
Or Fox, if that isn't too informal.
I understand that in Europe, especially Germany where this program got started, there are more stations than beer steins. I am no expert, just look at my questions, but it is my understanding that the more stations, the better. If someone is off line with internet problems, with being knocked out by a storm, or whatever, your station might be feeding info the same as another guy a few blocks away until he tripped over the cord. And while distant strikes at a poor look angle don't get much improvement from a duplicate report just a little ways from each other, at certain angles there is a good contribution to the solution of where the strike is.
Everyone can speak for themselves, but my opinion is that: More stations the better. There is an expense but if you fish this pales in comparison to tackle bait, boats, motors, storage and beer. The kit really was easier to assemble than I thought, but if you have little or no experience, it would be tough to find the place that a ham-handed assembly was botched and then correct without ruining a component or the board trace. So perhaps in the case of someone anxious about their ability, look around for a local ham and be frank, saying you need some assembly work done, and just because he thinks he's a ham he can do it, I'd ask how long has he been a ham and has he built kits from Heathkit in the past, assembled his own projects, etc. You want quality help and if you ask pointedly but politely, you are entitled to know how experienced the guy is. Sort of like asking your surgeon how many of a particular procedure s/he's done. Your body, you have a right to know, and they expect to be asked nowadays.
I have had so much fun since staring on this project. Not only relearning the soldering and setup and all, but annoying the hell out of the others here with my incessant questions. Not only do I know more now, but there is a response history that others might reference. And almost all the chaps here are wonderful, except for a few and you know who you are, don't you? JUST KIDDING!
To have access to much of the better detail and zoomability and all is dependent (for now unless the system originators decide differently) upon being an active member.
So don't stand on the outside of the window looking in. Open the door and come on it.
-
Fox of the Wind:
I just looked up where DeSoto is, and I'm thinking you would be a great addition to the midwest. The nearest would be Jim up in Mauston, me in Eau Claire, a station in Appleton who's owner's name isn't on the tip of my tongue, and a Rochester MN station and one up on the other side of the Twin Cities. Then a few in northern Illinois.
I'm thinking from the European situation that we're actually very sparse here, and the system architects have been hoping for a very high density of stations, so give it some good hard thought. Dale
-
If you zoom in all the way, you can also see circles that, I assume, represent some standard deviation(s) of error in strike placement.
I would say you are correct. After watching the system for some time last night and today it appears that the more stations report a strike from various angles the tighter the 'error ring' around a strike will be. I also noticed that the system seems to update those in batches, probably at an interval that is dependent on the amount of activity in the network i.e. more strikes = longer interval between error updates but I could be wrong. I just know that I was watching and trying to time how long it took to get the error determination for a few strikes when the system updated and the rings appeared around multiple strikes simultaneously.
Actually, the system doesn't work on "angles" per se. Not being a world-class mathematician and program developer as Egon and Tobias specifically, I might explain it this way:
Being Time of Arrival, rather than DF, pulses created by the stroke are plotted where a series of hyperbolic curves would intersect, based upon the times of arrival at any series of stations, each station receiving a different GPS time stamp.
At the server, High Class Mathematics generate a different curve for each station, best 8 or so win the race (in the U.S.) and Mr. and Mrs. Algorithms twink their magic twangers and plot the location. The actual position N/S/E/W of the station to the stroke is irrelevant, which is why they say the actual orientation of the H field antennas does not matter. (And the single E field probe will, of course, be omnidirectional and mounted vertically).
Assuming you are referring to the white dot/red circle of the 'real-time' map, the 'relocation' of any dots is actually an artifact of your computer processing the data from the German servers. The location is already 'decided' by the time the server generates the image data for this plot... So, for your purposes, you could say that the location is being refined, but it would only pertain to your particular computer screen. ;)
Hoping I haven't goofed this after a nice busy Mother's Day with the spouse of 43yrs... :roll:
Mike
-
Congrats on 43 years! And never an argument, right?
You must know when it is important to win and when it is OK to let her think she won...
I agree in most part that it really isn't angle as we would usually think of it, but it is really close to what LORAN used to do. The reason they scattered stations where they could was to provide master and slaves that would give good solutions over a large area.
If one has say, six stations, almost all on top of one another, then the solution for the spherical hyperbolas (yeah, damned round earth) will be poor, especially at certain directions away from the stations and apparent base lines, whereas if you have the same stroke heard by a different six all in various areas then the solution is a little tighter. And as the developers point out in one of their discussions, it has to do with what wave (sky, ground) is seen and even some changes in propagation between here and there. Not many people listen to AM frequencies at night any more, but the 'shortwave' is still fun to listen to and hear all sorts of things that FM doesn't have with shifts and all. I doubt that it is enough to make much of a difference, but with the speed of light involved, even a microsecond one way or the other puts the error circle even bigger.
I recollect that the Vaisala network had close to eighty stations across the US, and their engineers tried to pick the locations as best they could for good math solutions overall, rather than just whereever a university or cooperating station would like to pick their setup location.
Great discussion, I am eager to learn more if the developers ever get caught up making fun stuff for us and have time to give a symposium explaining in more but not exhausting detail how it all works.
-
Mike, as always thanks for the explanation. I knew the system calculations were time dependent but knowing more of the method of how the system calculates the positions is always good information.
Assuming you are referring to the white dot/red circle of the 'real-time' map, the 'relocation' of any dots is actually an artifact of your computer processing the data from the German servers.
Not the 'white dot/red circle' as such. If you zoom all the way in on the real time map you will see most of the strike positions surrounded by a larger circle centered on their position but with varying diameters (see attached image 'no_MOE_circles'). These rings seem to denote a margin of error as it were for the strike position and based upon the explanation you gave it is probably related to the number of stations contributing data to the network for that particular strike and how confident the system is in the calculated position. If you notice in the 'no_MOE_circles' image there are four strikes with blue boxes around them (added in photoshop for clarity); at the time this image was captured off of my display these strikes had no circle denoting margin of error, but within a couple of minutes all four strikes had these rings as noted in the second image - again the blue circles were added in photoshop to provide clarity. For what it's worth on my display the strikes are never white, they range from bright yellow (with the red ring) to darker orange/brown depending on the age of the strikes; the red ring denoting a new strike disappears after about 30 seconds.
Hopefully this helps explain what I was referring to.
-
Mike, as always thanks for the explanation. I knew the system calculations were time dependent but knowing more of the method of how the system calculates the positions is always good information.
Assuming you are referring to the white dot/red circle of the 'real-time' map, the 'relocation' of any dots is actually an artifact of your computer processing the data from the German servers.
Not the 'white dot/red circle' as such. If you zoom all the way in on the real time map you will see most of the strike positions surrounded by a larger circle centered on their position but with varying diameters (see attached image 'no_MOE_circles'). These rings seem to denote a margin of error as it were for the strike position and based upon the explanation you gave it is probably related to the number of stations contributing data to the network for that particular strike and how confident the system is in the calculated position. If you notice in the 'no_MOE_circles' image there are four strikes with blue boxes around them (added in photoshop for clarity); at the time this image was captured off of my display these strikes had no circle denoting margin of error, but within a couple of minutes all four strikes had these rings as noted in the second image - again the blue circles were added in photoshop to provide clarity. For what it's worth on my display the strikes are never white, they range from bright yellow (with the red ring) to darker orange/brown depending on the age of the strikes; the red ring denoting a new strike disappears after about 30 seconds.
Hopefully this helps explain what I was referring to.
Sorry, you are correct. With all the fun, I'd never zoomed in all the way on this particular map! In fact, I thought it might be restricted to logged in members only, and didn't concern myself with it.
Yes, the orange circles do indicate an 'error deviation'. The 'odds' are that the stroke actually fell within that circle. And as far as I know, the circles are calculated after the discharge is 'located'... it does depend upon number of stations detecting, but also what they detected, and where on the 'what' they triggered... if all triggered on a 'sky wave' rather than the initial discharge, for instance, the error might be greater... If they all triggered at various amplitudes on the same part of the pulse train, then the error might be larger. Of course there are exceptions, depending.
To oversimplify a bit--- If the stations are too far apart, and too far from the initial discharge, they'll trigger on reflected sky waves, rather than the initial ground wave (which has a distance limit) and each reflected sky wave will be distorted differently for each station, so the accuracy cannot be as well defined.
The lightning stroke has what some refer to as a "precharge" before the actual stroke, which produces a train of pulses... ideally stations would actually be running a low enough gain, have a low enough noise floor, in some combination, to detect this train prior to actual flash, thus producing better accuracy also. One reason for the ideal of "stations 50-150 miles apart"... and operating not for distance, but for quality. Also one reason they want the E-fields operating under a 300 mile range. Again, over-simplifying....
Please remember that I know just enough to be highly dangerous! Plenty of research and study by experts and experiments has been done on this... google various terms and phrases!
Mike
-
One thing that became apparent to me during yesterday's storm events here is that we need greater station density. There were many times when I saw lightning flashes and heard thunder, but the strokes didn't show up on the 'live' display. Is this a major problem? I doubt it because generally the strokes that do register provide enough information to identify/locate lightning/thunderstorm activity.
I just got mesmerized by the display again and I had to tear myself away in order to finish this...
-
One thing that became apparent to me during yesterday's storm events here is that we need greater station density. There were many times when I saw lightning flashes and heard thunder, but the strokes didn't show up on the 'live' display. Is this a major problem? I doubt it because generally the strokes that do register provide enough information to identify/locate lightning/thunderstorm activity.
I just got mesmerized by the display again and I had to tear myself away in order to finish this...
I'm guessing station density is one factor - as the Blitzortung folks note in their project document, the range at which a bolt is detected depends both on the sensitivity of the sensing equipment AND the current in the individual strike.
It may also be that VLF sensing equipment (such as the Blitzortung sensors) primarily picks up on cloud-to-ground lightning--VLF emissions from cloud-to-cloud discharges are much weaker.
A third possibility is there's still room for improvement in the system that correlates received signals from various stations.
I'm guessing the main culprits are #1 and #2, though.
-
I still suggest that some of this has to do somewhat with my dubious "Thor's Hammer Slammer Jammer" theory for sky wave signals, (http://www.wxforum.net/index.php?topic=19724.msg190881#msg190881) depending on other network stations' location.
Also, weaker discharges being overwhelmed by general noise. A signal of 120mv at a station flooded with 130mv signals and noise will not be detected. Observations for nearly a year have shown me that storms/strokes behind a very active storm between my location and their source are sometimes undetected, even though in range of my system... and they are basically the only sources around. Conversely, a very active storm W of me for example, with strong flashes, can obscure weaker flashes from storms E N S of me... happens all the time.
The energy from CG lightning generally is much stronger than IC or CC, true, building high charges from Anvil to Ground. And, according to some experiments, a magnetic pulse no matter what its source,is actually more easily sensed in a N/S direction as it more easily rides in the Earth's magnetic fields! Conversely, EW detection a bit more difficult. Weird, huh. etc.etc.
Remembering that our systems are specifically programmed with the current firmware, to go "Burst mode" interference at a 30 pulses/2 second average, and "Normal" interference at >15 sigs/min average. If you open your controller's status web page, and watch the "Tracker" section, I believe you'll see your mode etc change from Normal, to interference type off and on depending on storm activity... and if you're in burst mode for a second or two, you'll NOT detect signals... nor will you really know that's occurring unless you happen to observe status either on controller or web... then wonder why strikes are missed. As I was typing this paragraph, my controller 'burst' twice. I would not have known I was inactive for a total of 6 seconds if that page had not been open beside this one.
Keep always in mind, that this is NOT a single station system! It's about networking. If I'm in 'burst', you guys are supposed to "CMA" for me.. (Cover My Area!) and I'll try to CYA for you in return...
-
While mesmerized by the real-time display yesterday as a severe thunderstorm rolled through my area, I was startled by a lightning strike that hit within a few hundred feet of my QTH. I heard a loud fizzing sound. There was a momentary power failure, my USB hub lights started flickering wildly, and my UPSs took over for a few seconds. After I cleaned up the spilled coffee and changed my underwear :shock:, I noticed my R820T receiver dongle was only picking up a few Mode-S radar signals. "Darn!", I thought - the LNA and/or dongle is fried!
I rebooted everything, unplugged and plugged the LNA back in and amazingly, all came back to life! The Polyphaser surge suppressor I installed on the rooftop ADS-B antenna seemed to really have done a great job protecting the LNA and receiver dongle! I also have the RG6 coaxial cable tie-wrapped along a 25 foot section of cold water copper pipe, so any serious lighting had a good chance of jumping the coax to the copper pipe.
Later, I checked my .kml files and the lightningmaps.org real-time display. The strike was located at the entrance to my subdivision, not far from the house. I took a drive and noted a fire truck about 50 feet from the indicated location, dousing a high-voltage transformer with foam. The uncertainty circle on lightningmaps.org was quite wide, about 3/4 mile in diameter. The location was dead-on however.
A close call, but a cool validation of the midwest location accuracy with all the new area stations.
Regards,
Don
WD9DMP
-
While mesmerized by the real-time display yesterday as a severe thunderstorm rolled through my area, I was startled by a lightning strike that hit within a few hundred feet of my QTH. I hear a loud fizzing sound. There was a momentary power failure, my USB hub lights started flickering wildly, and my UPSs took over for a few seconds. After I cleaned up the spilled coffee and changed my underwear :shock:, I noticed my R820T receiver dongle was only picking up a few Mode-S radar signals. "Darn!", I thought - the LNA and/or dongle is fried!
I rebooted everything, unplugged and plugged the LNA back in and amazingly, all came back to life! The Polyphaser surge supressor I installed on the rooftop ADS-B antenna seemed to really have done a great job protecting the LNA and receiver dongle! I also have the RG6 coaxial cable tie-wrapped along a 25 foot section of cold water copper pipe, so any serious lighting had a good chance of jumping the coax to the copper pipe.
Later, I checked my .kml files and the lightningmaps.org real-time display. The strike was located at the entrance to my subdivision, not far from the house. I took a drive and noted a fire truck about 50 feet from the indicated location, dousing a high-voltage transformer with foam. The uncertainty circle on lightningmaps.org was quite wide, about 3/4 mile in diameter. The location was dead-on however.
A close call, but a cool validation of the midwest location accuracy with all the new area stations.
Regards,
Don
WD9DMP
Many people consider Polyphaser the standard by which others are measured. I'll probably be remote-mounting the Blitzortung (controller and all) about 75 feet from the house, and an outdoor Polyphaser data/DC power surge protector will do guard duty on the incoming lines.
Glad to hear you came through the strike OK, by the way - and that the stike localization was accurate :) As you said, nice validation.
-
Remembering that our systems are specifically programmed with the current firmware, to go "Burst mode" interference at a 30 pulses/2 second average, and "Normal" interference at >15 sigs/min average. If you open your controller's status web page, and watch the "Tracker" section, I believe you'll see your mode etc change from Normal, to interference type off and on depending on storm activity... and if you're in burst mode for a second or two, you'll NOT detect signals... nor will you really know that's occurring unless you happen to observe status either on controller or web... then wonder why strikes are missed. As I was typing this paragraph, my controller 'burst' twice. I would not have known I was inactive for a total of 6 seconds if that page had not been open beside this one.
One possible solution to this (perhaps already thought of) would be to make "interference mode" a network-determined (or at least dependent in part on network feedback) condition -- pretty much by definition, interference should be a local affair, whereas rapid-fire strikes would likely hit large regions of the network.
Since the red systems enjoy feedback from the servers, perhaps going into interference mode could be stopped if the server indicates that a station isn't alone in getting lots of rapid-fire hits.
-
Dear Mr. Wind,
Or Fox, if that isn't too informal.
I understand that in Europe, especially Germany where this program got started, there are more stations than beer steins. I am no expert, just look at my questions, but it is my understanding that the more stations, the better. If someone is off line with internet problems, with being knocked out by a storm, or whatever, your station might be feeding info the same as another guy a few blocks away until he tripped over the cord. And while distant strikes at a poor look angle don't get much improvement from a duplicate report just a little ways from each other, at certain angles there is a good contribution to the solution of where the strike is.
Everyone can speak for themselves, but my opinion is that: More stations the better. There is an expense but if you fish this pales in comparison to tackle bait, boats, motors, storage and beer. The kit really was easier to assemble than I thought, but if you have little or no experience, it would be tough to find the place that a ham-handed assembly was botched and then correct without ruining a component or the board trace. So perhaps in the case of someone anxious about their ability, look around for a local ham and be frank, saying you need some assembly work done, and just because he thinks he's a ham he can do it, I'd ask how long has he been a ham and has he built kits from Heathkit in the past, assembled his own projects, etc. You want quality help and if you ask pointedly but politely, you are entitled to know how experienced the guy is. Sort of like asking your surgeon how many of a particular procedure s/he's done. Your body, you have a right to know, and they expect to be asked nowadays.
I have had so much fun since staring on this project. Not only relearning the soldering and setup and all, but annoying the hell out of the others here with my incessant questions. Not only do I know more now, but there is a response history that others might reference. And almost all the chaps here are wonderful, except for a few and you know who you are, don't you? JUST KIDDING!
To have access to much of the better detail and zoomability and all is dependent (for now unless the system originators decide differently) upon being an active member.
So don't stand on the outside of the window looking in. Open the door and come on it.
The more I read about it the more I want to join in.......but as for getting a ham radio person to build my station....hmmmm there has not been anyone around here for HAM radio since the 1950's I belive.....But I have not given up....I do have a cousin that is great at soldering skills....might send it to him...
Like I said I would LOVE to come in....looks fun.
-
One thing that became apparent to me during yesterday's storm events here is that we need greater station density. There were many times when I saw lightning flashes and heard thunder, but the strokes didn't show up on the 'live' display. Is this a major problem? I doubt it because generally the strokes that do register provide enough information to identify/locate lightning/thunderstorm activity.
I just got mesmerized by the display again and I had to tear myself away in order to finish this...
Following up on this - we just had multiple cloud-to-ground lightning strikes here in the Gaithersburg area, with nothing on the map. Hopefully, once I get my system red built and installed, that won't be a problem anymore :)
-
Following up on this - we just had multiple cloud-to-ground lightning strikes here in the Gaithersburg area, with nothing on the map. Hopefully, once I get my system red built and installed, that won't be a problem anymore :)
My understanding is that local strikes tend to overload the receivers so you probably won't contribute to data in your immediate area, but with enough coverage in the future you should start getting those areas filled in.
On a side note I noticed over the past few days that it looks like they've been making adjustments to the real-time display code; previously I had commented that it took the system a couple of minutes to populate the narrow-field (high zoom) views with the rings denoting the location deviation/margin of error for strikes. Over the past two days I've noticed that they are now appearing simultaneous to the strike displays. So another tip o' the hat to Tobi, Egon and the crew! \:D/
-
Has anyone noticed that the real-time display chews up huge amounts of CPU? There is a warning about this on the popup bubble where station display is selected.
I find that my dual-core Athlon 7550 system running at 2.5GHz is using near 100% of both cores when the display is running.
The only bad effect of this for me is that PlanePlotter (used for tracking aircraft ADS-B transponder signals) can't process data quickly enough from the R820T dongle and dump1090 decoder software. The connection halts and doesn't get re-established until 5 minutes later by PlanePlotter, unless it is re-started manually.
Worth it though!
Regards,
Don
WD9DMP
-
I haven't noticed it much on an i7 2.4 GHz machine, but it does seem to claim a lot of memory (an ongoing problem in Firefox, not just for this page, but there's also a lot of memory being consumed if I open the page in IE).
-
I did notice it bogged down the other night when I had the real-time page running, to the point that I had to close Firefox and re-open it and that's with a six-core processor and 8GB of ram. Today it doesn't seem too bad but I also didn't view it for very long.
-
I just ran a test using IE, FF and Chrome simultaneously on my HP i7 2.80GHz w/8GB of memory. All three are running the Real Time (Beta) display. I can detect no degradation of my system. It is showing around 14% CPU usage and RAM at 68%.
-
There is now a full-screen mode available on the real-time display at lightningmaps.org. Nice!
Don
WD9DMP
-
There is now a full-screen mode available on the real-time display at lightningmaps.org. Nice!
Don
WD9DMP
Also, click on a station dot and go to its info page!
-
Some observations during today's storm here in Nevada:
- First of all - I'm not convinced the line colors (blue & green) are set for >500km and <500km respectively. I'm seeing many green lines that appear to be much longer than 500km in length. That value seems to be closer to 1000km.
- It would be very nice to have an adjustable "persistence" settings for the strike lines. They disappear too fast to get a good look at them.
- I would also love to have the capability to play-back the data from time A to time B.
- I notice a station in Columbia (935) that is regularly detecting strikes throughout the US.
- We had several strikes within a few miles of my location this afternoon. I detected none of them and, they never showed-up on Blitzortung at all.
- About 2 seconds before the second bolt was observed, my station went into interference mode. It took another 10 to 15 seconds before it came out of interference mode. There was nothing observed before and after the first strike. The strikes were about 15-minutes apart. No flash was observed for either but the thunder shook the house.
-
Some observations during today's storm here in Nevada:
- First of all - I'm not convinced the line colors (blue & green) are set for >500km and <500km respectively. I'm seeing many green lines that appear to be much longer than 500km in length. That value seems to be closer to 1000km.
- It would be very nice to have an adjustable "persistence" settings for the strike lines. They disappear too fast to get a good look at them.
- I would also love to have the capability to play-back the data from time A to time B.
- I notice a station in Columbia (935) that is regularly detecting strikes throughout the US.
- We had several strikes within a few miles of my location this afternoon. I detected none of them and, they never showed-up on Blitzortung at all.
- About 2 seconds before the second bolt was observed, my station went into interference mode. It took another 10 to 15 seconds before it came out of interference mode. There was nothing observed before and after the first strike. The strikes were about 15-minutes apart. No flash was observed for either but the thunder shook the house.
1) You may be right,.. Tobi is still working on the script, but I believe those distance estimates are the general intent.
2) Tobi even turned off the 'persistence' in an effort to find solutions for some Apple products, and various browsers. It is browser intensive, and a few of us begged him to turn it and transparency back on. He's running a tough compromise right now, I think.
3)I'm not sure that could be possible, at current stage. Here's a quickie, though, if you want to manually do it: http://applian.com/replay-video-capture/download $40 if you purchase, or get a whole buncha stuff for $80. I'm sure there are other ways, possibly even free, but this is a darned handy product I've had for quite a while.
4) Yes, ain't it great? We've stations down in Brazil I'd love to see bounce up their gains enough also, and a couple in the islands... We also need to get some help for old Lonesome Stan Silversword out in Maui...
5) Station density, pure and simple. Rest of us eastward were pretty much swamped with Midwest cells... at one point, running gains near minimum, I was polling over 50,000 signals in an hour... I'm not sure, but I think #of stations currently required to track a stroke in US is 6 so and you were probably too close to count as a station... which gets us to
6) There's a whole buncha stuff in the flash before the actual discharge... what I call 'precharge'... they're strong enough to trigger interference. If Egon had his way, we'd all be detecting and signaling some of those early events so his algorithms can get a super accurate "Time of Group Arrival". I notice with my E field, what appear to me more triggers back way early in the "group", but this may be just an illusion, as I've not done that much experimenting with systems....
I do know that those filters Richo and the team designed are really brilliant. All ya gotta do is look at the spectrums compared to just loops. This is gonna help folks with specific and consistent interferences, when the gains and thresholds are fine tuned for their stations.
Have you taken these questions to the BO Forum? Why don't you do that, be interesting to see how my surmises fit the truth.. :roll:
-
Don, Here's a full screen recording... click the image, when the page opens, lower right of the player, call full screen...
(http://www.ourspecialnet.com/Weather/TwinHollies/i-Rxcgkgt/0/L/LightningMaps.org%20Real-Time%20Lightning%20Map%20-%5B6%5D-L.jpg) (http://www.ourspecialnet.com/Weather/TwinHollies/i-Rxcgkgt/A)
-
4) Yes, ain't it great? We've stations down in Brazil I'd love to see bounce up their gains enough also, and a couple in the islands... We also need to get some help for old Lonesome Stan Silversword out in Maui...
Last night I saw the Brazil station detect a strike ~8,200km away in the north west USA.
-
Hi All,
Just noticed that just a little while ago that the Station On is is not working for me for some reason. There are strikes but no lines and no green dots showing up. I was viewing with FF 29.0.1.
Anyone observing this?
Regards,
--Stan Y.
KH6HHG
Maui, Hawaii
-
Hi All,
Just noticed that just a little while ago that the Station On is is not working for me for some reason. There are strikes but no lines and no green dots showing up. I was viewing with FF 29.0.1.
Anyone observing this?
Regards,
--Stan Y.
KH6HHG
Maui, Hawaii
I think Tobi's working on it. That feature has been off and on all afternoon at times.
-
Thanks for the info Mike.
I thought that I was doing something wrong again.
Regards,
--Stan Y.
KH6HHG
Maui, Hawaii
-
Thanks for the info Mike.
I thought that I was doing something wrong again.
Regards,
--Stan Y.
KH6HHG
Maui, Hawaii
Ain't it great? Part of the fun in this is learning a bunch of new things! Some of it the hard way... :oops:
I don't know what I'd done early on if it hadn't been for the original green station operators passing along clues, and hints. Don F and Greg H and I kinda worked through a lot of stuff early on with the reds, and I'd be in sorry shape without those two.
You "new folks" have a lot of help, don't ever hesitate to ask... we'll find some way to add to your confusion.
Don't forget, the Blitzortung forum up on the website. There are countless issues and resolutions and tips up there. Read it all.
Mike
-
Some observations during today's storm here in Nevada:
- First of all - I'm not convinced the line colors (blue & green) are set for >500km and <500km respectively. I'm seeing many green lines that appear to be much longer than 500km in length. That value seems to be closer to 1000km.
- It would be very nice to have an adjustable "persistence" settings for the strike lines. They disappear too fast to get a good look at them.
- I would also love to have the capability to play-back the data from time A to time B.
- I notice a station in Columbia (935) that is regularly detecting strikes throughout the US.
- We had several strikes within a few miles of my location this afternoon. I detected none of them and, they never showed-up on Blitzortung at all.
- About 2 seconds before the second bolt was observed, my station went into interference mode. It took another 10 to 15 seconds before it came out of interference mode. There was nothing observed before and after the first strike. The strikes were about 15-minutes apart. No flash was observed for either but the thunder shook the house.
1) You may be right,.. Tobi is still working on the script, but I believe those distance estimates are the general intent.
2) Tobi even turned off the 'persistence' in an effort to find solutions for some Apple products, and various browsers. It is browser intensive, and a few of us begged him to turn it and transparency back on. He's running a tough compromise right now, I think.
3)I'm not sure that could be possible, at current stage. Here's a quickie, though, if you want to manually do it: http://applian.com/replay-video-capture/download $40 if you purchase, or get a whole buncha stuff for $80. I'm sure there are other ways, possibly even free, but this is a darned handy product I've had for quite a while.
4) Yes, ain't it great? We've stations down in Brazil I'd love to see bounce up their gains enough also, and a couple in the islands... We also need to get some help for old Lonesome Stan Silversword out in Maui...
5) Station density, pure and simple. Rest of us eastward were pretty much swamped with Midwest cells... at one point, running gains near minimum, I was polling over 50,000 signals in an hour... I'm not sure, but I think #of stations currently required to track a stroke in US is 6 so and you were probably too close to count as a station... which gets us to
6) There's a whole buncha stuff in the flash before the actual discharge... what I call 'precharge'... they're strong enough to trigger interference. If Egon had his way, we'd all be detecting and signaling some of those early events so his algorithms can get a super accurate "Time of Group Arrival". I notice with my E field, what appear to me more triggers back way early in the "group", but this may be just an illusion, as I've not done that much experimenting with systems....
I do know that those filters Richo and the team designed are really brilliant. All ya gotta do is look at the spectrums compared to just loops. This is gonna help folks with specific and consistent interferences, when the gains and thresholds are fine tuned for their stations.
Have you taken these questions to the BO Forum? Why don't you do that, be interesting to see how my surmises fit the truth.. :roll:
Have posted the questions on the BO forum, as suggested...
- http://www.blitzortung.org/Webpages/index.php?lang=en&page=6&session=1 (http://www.blitzortung.org/Webpages/index.php?lang=en&page=6&session=1)