what is the best setup for BO??
#1
hello.


Just a quick question that might require a longer answer, but what is the "sweet spot" for Blitzortung?
Guess the stats are helping to tune for "ego" , but does the stats serve Blitzortung also?
@ what range from antenna do you get the best accurancy?
thinking more about what serves  the Blitzortung better, than my ego for getting most hits, or stats? -or are these "following the same path"?
Guess the answer will be different if a antenna is in a area with low coverage, and if there is high density of participants around, so i understand that the answer will be "depending on" Smile
For a triangulation perspective would not longer range give higher % accurancy than close range?
I know my signals are rubbish atm, but im working on that issue.. still what should i aim for to get my system RED to be best for BO?
Stations: 1145
Reply
#2
My personal opinion :

strokes/signals should go to maximum. If every signal is a stroke, then station has the best efficiency. But do not tune station to max. strokes, this will result in heavy signals and a lot of traffic.

Thomas
First station in Namibia (Southern Africa), look at #1305 !
Stations: 1006, 1305
Reply
#3
(2015-06-22, 19:40)frode dahl Wrote: hello.


Just a quick question that might require a longer answer, but what is the "sweet spot" for Blitzortung?
Perhaps the first 'sweet spot' is proper construction, recommendations for installation, and location for antennas. 

Guess the stats are helping to tune for "ego" , but does the stats serve Blitzortung also?

As we optimize, the 'stats' (Effectivity =  Having an intended or expected effect.) (Efficiency = accomplish something with the least waste of time and effort)
And yes, those give an indication of a station's "overall" "over time" "Quality" within some established parameters such as distance, station type, antennas, etc, and the ability of the station to send 'good' signals, consistently,, with good time stamps, data, etc related to 'signals' actually sent, for example.

As a very rough example... If all the strokes are 1500KM distant, for e.g. then I'm receiving the 'skywaves'... and none of the 'ground wave'.  The skywaves are of various delays, and distortions due to reflections, some frequencies are not 'reflected'.. so the quality and timestamp will not be as good.  But I still send a 'signal', yet it is so far away in terms of phase and time, that it may be discarded, and count against me! If it is 'recognized' it doesn't count against me.  In a similar fashion, strikes 20km distant may be so strong that the data is overwhelmed, distorted, etc, that again I send a signal (or signals!) that is useless, but counts against me!

@ what range from antenna do you get the best accurancy?
Perhaps, in perfection, we would want the initial discharge pulse, as a ground wave, and this would be 30 - 200km distant, depending on a whole lot of variables.
In Practice, this isn't possible, or even necessary.  This is an excellent system, with fantastic server algorithms.  Perhaps think in terms of an "E-field probe/amp assembly as designed for optimum at <300km and H-field at optimum <800km.

thinking more about what serves  the Blitzortung better, than my ego for getting most hits, or stats? -or are these "following the same path"?
Big Grin Perhaps a bit of both... If you are in a good environment, good system, etc.  Distance ability is a great EGO thing, but the ability to detect "across region boundaries, for example, helping our folks in South Africa, Hawaii, Oceania, etc as their network density grows is a good thing for Blitzortung.  In Americas, as
the density has grown, it is much less important for me to acquire signals from 2000km,... there are stations much closer to that stroke that will provide better data.
I still like to see my station "Up There" in rankings... but it will never be as high as it once was, simply because there is more 'competition' and a much stronger network.
Dodgy  And the '"location accuracy" (deviation) of the network strokes has improved from >8km to often <1km! 

Guess the answer will be different if a antenna is in a area with low coverage, and if there is high density of participants around, so i understand that the answer will be "depending on" Smile
Exactly!

For a triangulation perspective would not longer range give higher % accurancy than close range?
Again, many variables!  How Close?  How Far?  Also, 'triangulation' implies 'distance' and direction, so remember this is a "timing" system, not "direction finding' and anything that would distort, delay, the time/phase relationship  signal / timestamp too much introduces error, or causes the signal to be discarded.

I know my signals are rubbish atm, but im working on that issue.. still what should i aim for to get my system RED to be best for BO?
Perhaps not as much 'rubbish' as you might think.  You're running, you're sending, and you're taking steps to optimize.  You'll do fine!
Some published articles in the past as TOA/TOGA has evolved have implied an Effeciency / Effective thing of 20-40%... yes, and no... with the standards Egon and Tobi, Richo et.al. are establishing, and our "current" reporting regions as large as they are, I might play with a rule of thumb of, say at any given hourly sampling on the "participant's " page of say - 10-50%, depending on all the variables.... and some type of 'over time' percentage of 20-40????

Cheers!
Mike


Stations: 689, 791, 1439, 3020
Reply
#4
Thank you for the reply Mike.
i know there are too many variables to give an exact answer to the questions, still i have to "porbe around".

About "sweetspot" to use that word, let me use a example:

In Norway we dont have that much strikes, so i can adjust to pick up more signals from places with higher desity as Russia, Turkey, Italiy, Spain (around 2500-3500km range)
in the range 800-2000 (central Europe) the density of stations are High, so my signal usually is not used or assigned. (system ticks, and register, but server dont use/assign)
So in my case i can run higher Gains, to get "better score" but i would assume in cost of accuracy at distance, and it might distort local strong signals?
Or i should set lower gains, to pick the fewer "local signals", but i would drop on the scoreboard.
As now, i get around 20ish% of local strokes out to around 500km, then stations in central europe takes over, and then a small increase in ragne 2000-3000 range, and a strange spike @ 3400km (40%)
Why the locals are just 20ish% might be part of my question, since i guess i loose some to interference/distorted signals, so should i drop the Gain and get more local strikes, and get lower stats?
Stations: 1145
Reply
#5
There is something wrong with your hardware !

I checked the signals, they are very noisy, a lot of "grass" and some channels didn't appear or are too low.

So, please connect h-field on amp1 and e-field on amp2. I see yellow lines, so what's your setup ? Do you use 2 e-fields ?

Thomas
First station in Namibia (Southern Africa), look at #1305 !
Stations: 1006, 1305
Reply
#6
As K suggests, get in sync.. H on amp 1 and E on ampl 2...  that will be very important. /.// also on the 'services page', as you login to BT, make sure you've entered the antenna types correctly! http://www.wxforum.net/index.php?topic=22710.0  toward the end of first posting. Also check that same thread for 'running both H and E field"... good foundation there, especially about "alternate channel mapping".. .

I'd suggest, if your station is running reasonably clean, as it appears to be, that 'distance' from cells, is the only thing lowering your numbers.
As we increase gain, lower thresholds, etc we are absolutely going to send 'un-useable' signals!  There are 'atmospheric' events that manifest as 'lightning', but really aren't... perhaps even meteor trails!  This is not a malfunction, or poor operation... it's just the nature of VLF / ELF reception!  For example, with the CMEs from Sol the last few days I've had some strange operation, probably due to GPS noise and interference... and anything interfering with time or phase can result in 'garbage' signals.

So, I grit my teeth, smile, and rely on the "network", not just my 'isolated' station,... we don't stand alone!... A little known paradigm, which may vary from region to region, I'm not sure,.... and may change without warning:

In terms of "time", and this is "for example"... First (8) stations are primary detectors.  Next (12) are secondary, for quality control, Others receiving are "detectors".  So while you're not 'top eight'. Your data is extremely important!!  And you'll want "quality" rather than the ability to detect signals in Canada, though your system may be capable of that,.

Play around with it.. enjoy the chase, find the 'sweet spot' where your station will 'normally' run, and not be in 'interference mode' http://www.wxforum.net/index.php?topic=20439.0
and read and learn all you can, and 'store up' your experience to pass on to others!

Cheers!
M


Stations: 689, 791, 1439, 3020
Reply
#7
thank you for replies Thomas And Mike..

I know my signal on H-Field is rather dead.. mangaged to break one screw on the terminal @ Amplifier, so i have some soldering to do...
And i just bought another CAT 6 FTP / LSZH cable to try check if that might help on the noise.
(guess my cheap loong cable works as antenna instead of signal carrier Smile

So im working a few hours each day to figure out... pretty much everything about this Smile
Stations: 1145
Reply
#8
(2015-06-24, 10:52)frode dahl Wrote: thank you for replies Thomas And Mike..

I know my signal on H-Field is rather dead.. mangaged to break one screw on the terminal @ Amplifier, so i have some soldering to do...
And i just bought another CAT 6 FTP / LSZH cable to try check if that might help on the noise.
(guess my cheap loong cable works as antenna instead of signal carrier Smile

So im working a few hours each day to figure out... pretty much everything about this Smile

heh, heh,... Wink
I also did the same thing with the 'ground' block on my controller... drove me nuts for a few days... Replacing the block fixed it. 
...did something similar with the F connector on E field preamp... over tightened the nut, and broke the connector...  Dodgy  I suspect there have been examples of antenna blocks broken in a similar fashion...  silly things break between the block and the board... weren't designed for us super-humans  or electric screwdrivers... , Lightning

In your post on WxForum.. you mentioned an "alarm" system...  the 2.4 / 5mHz probably isn't an issue, unless it's the power supply as Greg (Miraculon) mentioned.  However, you might have a 'sensor' on that alarm that operated at 74Khz,or something similar, that could????


Stations: 689, 791, 1439, 3020
Reply
#9
did manage to solder my Amplifier yesterday, and make new pole for my H-Field, and change the cable between Unit and amp to 20m Cat6FTP..
Seems like it helped some.. and my signals are around 200ish last hours, so no spikes... (change was made around 22:00 yesterday)
Still i had something @ aound 02:00 that is causing a mess, wich can't be dishwasher, alarmcentral, wifi, TV or other...so i have to figure the soruce of that one....
Im in the middle of making a terrace, so i can later move my Antenna around on the railing to see if i can reduce noise, but atm im pretty stuck on location...
@Knickohr where do you look at my signals?
http://www.lightningmaps.org/blitzortung...n_id=14057 ?
I have one H-Field Coax loop antenna, and one 1.5mm E-field probe.

*edit* now its a mess again..... and im at work... picking alot of signals... Sad
Stations: 1145
Reply
#10
there is something strange here...
My channel 1B suddenly sparked with a noise floor of 400+ mVpp
When i set threshold to 555 it stopped and calmed down again.. so now its @ 61-23mVpp

I unplugged my Alarm, still i got the high Noise. so i took donw H-Field for "service"....

*edit* Now im running only E-Field.
(here is the trouble about reading around, because E-Field shall be AMP2, but I have also read that if you use one antenna, it should ALWAYS be amp1) -still for test/faultdebug it should not be a problem..
I have tried E-Field @ both amps, to see if i get the noise on Amp1, but none of the amps gives the noise when using E-Field. (still i might have been "lucky" to try while the noise was not present.
and about 10 hours later, there has been no spikes on my E-Field.
http://www.lightningmaps.org/blitzortung...o_sid=1145
I started with low threshold and my E-Field was ticking away in sync (with a few sec delay) with signals on the eastern side of USA..


If im making a crude test Antenna, Coax loop is the easiest one, but if im not using transformer, should i not add a resistor? what value? (or am i remembering wrong now? )
Stations: 1145
Reply
#11
well.. x fingers...
was tired of pulling my hair in frustration, and went over the AMP with the iron... and seems like that helped Smile

Ill see if it does the usual trick with working ok first, and then give me the usual Noise...
Stations: 1145
Reply
#12
Whowww !

Much better signals Wink But very high gains for amp1. Is this normal for an coax-loop ?

One point left. Disable channel C on amp 1 (yellow).

Thomas
First station in Namibia (Southern Africa), look at #1305 !
Stations: 1006, 1305
Reply
#13
Thank you for pointing me in the correct direction Thomas. I have to be honest enought to tell you my coax cable is old and bad. . Bought first 200m of high ql cable in a hurry . And grabbedvone with alu shielding Sad and had about 100m of similar.
So i found some really old coax that probably worked best last decade Smile so it is on the todo list. Had just my faulty soldering to figure out. . So now i can focus on a smaller house project before i make new antenna Smile
Stations: 1145
Reply


Possibly Related Threads…
Thread Author Replies Views Last Post
  Looking to setup a new station near Barcelone VLorz 5 29,120 2016-04-02, 09:57
Last Post: VLorz

Forum Jump:


Users browsing this thread: 1 Guest(s)