WXforum.net

Weather Station Hardware => Blitzortung => Topic started by: dfroula on June 22, 2014, 11:38:21 AM

Title: Higher interference mode thresholds available!
Post by: dfroula on June 22, 2014, 11:38:21 AM
Tobi just posted this:

The new thresholds were available for my station this morning.

Don
WD9DMP

========================================================

"Hi,

we've recognized, that several stations could receive more good stroke signals, but the interference mode is limiting the rates. We want to avoid raising the original (hard coded) interference mode limits, but there's another solution now: Stations which are "not too bad", will get a higher limit by the remote control server. There's no exact definition of "not too bad", just don't send too much noise.

The new limits are 30 and 60 signals/s, the normal ones are 15 and 30 signals/s. You can check them at the bottom of the controllers status page. If your station got the higher limits, you can try to increase the gain values one or two steps.

This new behavior might increase stroke rates especially in America and Australia, as the lightning density in most areas there is much higher compared to boring Europe here Dodgy ."
Title: Re: Higher interference mode thresholds available!
Post by: Jumpin Joe on June 22, 2014, 11:44:08 AM
Same for me.

Joe
Title: Re: Higher interference mode thresholds available!
Post by: JonathanW on June 22, 2014, 12:03:13 PM
I'm actually seeing 60 sigs/2 sec, and 40 sigs in 60 secs.
Title: Re: Higher interference mode thresholds available!
Post by: Cutty Sark Sailor on June 22, 2014, 12:06:16 PM
 \:D/ YAY   =D>
Can't wait to give them dang axle factory welders a lickin'...
Title: Re: Higher interference mode thresholds available!
Post by: Cutty Sark Sailor on June 22, 2014, 01:20:51 PM
And the stroke counts went ballistic!!!! Wow...
Title: Re: Higher interference mode thresholds available!
Post by: Cutty Sark Sailor on June 22, 2014, 01:36:09 PM
OOPS
 If the system is reset, the new parameters are lost... ????  :sad:
Title: Re: Higher interference mode thresholds available!
Post by: JonathanW on June 22, 2014, 01:43:36 PM
Unless I'm reading it wrong, the interference thresholds continue to change.  Right now, mine are at:

BurstWhen > 60sig/s in 2s average. Disable when 50% below threshold
NormalWhen > 50sig/s in 60s average. Disable when 30% below threshold

Perhaps Tobi is continuing to modify them?
Title: Re: Higher interference mode thresholds available!
Post by: Cutty Sark Sailor on June 22, 2014, 01:55:00 PM
Unless I'm reading it wrong, the interference thresholds continue to change.  Right now, mine are at:

BurstWhen > 60sig/s in 2s average. Disable when 50% below threshold
NormalWhen > 50sig/s in 60s average. Disable when 30% below threshold

Perhaps Tobi is continuing to modify them?
Oh, I suspect they'll be dynamic and situational... mine returned to the 'new normal' after several minutes... don't know if it was 'manual' or through algorithm... suspect the latter, after server communication.
Title: Re: Higher interference mode thresholds available!
Post by: JonathanW on June 22, 2014, 02:03:41 PM
With the changes in interference modes, I'm leaving my station on fully auto for a bit.
Title: Re: Higher interference mode thresholds available!
Post by: Dr Obbins on June 22, 2014, 03:59:13 PM
I don't remember seeing such high strike counts before.
Great change!
Title: Re: Higher interference mode thresholds available!
Post by: Cutty Sark Sailor on June 22, 2014, 05:29:34 PM
689 hit 198 strokes (NOT signals!) per minute, moments ago!   WOW!
Title: Re: Higher interference mode thresholds available!
Post by: JonathanW on June 22, 2014, 05:54:38 PM
Look at the detections in the western U.S. and Canada!!  between the greater interference thresholds and the new stations that recently came online, the west and north have awoken. =D>
Title: Re: Higher interference mode thresholds available!
Post by: DaleReid on June 22, 2014, 06:01:06 PM
I see it must be thunderstorm season up north of the border, up Canada way, with all the dots appearing and lines darting up there.

Oh, What's that?  They've always had storms, we just weren't seeing them before, and now it's just the new station bringing enough reports to validate the other stations?

Well, I'll be darned.  Good to have these constant improvements and more stations coming on line.  Now when AZ and CO get lit up....

Title: Re: Higher interference mode thresholds available!
Post by: JonathanW on June 22, 2014, 06:42:46 PM
I really should have guessed that this was another reason why detections were lower in the U.S. than in Europe.  In addition to the "too weak to be picked up by 6 stations" theory, Tobi et. al correctly surmised that the existing N. American stations were simply being swamped by signals, configured as most of them are for long distance detection.  With the old interference mode settings, we either had to dial back the gain/threshold settings (making the "too weak to be picked up" effect happen), or go into interference mode.

As it is, of course, this is partially a stopgap.  While it's true we have greater lightning activity here than in Europe, we definitely need more stations in Region 3, too!
Title: Re: Higher interference mode thresholds available!
Post by: dfroula on June 22, 2014, 07:53:58 PM
Pretty amazing, huh?!

Over 10,000 STRIKES (not signals) detected per hour for both your and my stations.

Tobi effectively increased the dynamic range of the system 5x. Amazing results!

Best,

Don

689 hit 198 strokes (NOT signals!) per minute, moments ago!   WOW!
Title: Re: Higher interference mode thresholds available!
Post by: Cutty Sark Sailor on June 22, 2014, 08:41:52 PM
Pretty amazing, huh?!

Over 10,000 STRIKES (not signals) detected per hour for both your and my stations.

Tobi effectively increased the dynamic range of the system 5x. Amazing results!

Best,

Don

689 hit 198 strokes (NOT signals!) per minute, moments ago!   WOW!
:lol:
I'm not going to look it up, but long ago in a galaxy far, far away, you replied to a PM with,
"Crank it up, Mike! You oughtta be getting 30 signals a minute!"

Or something similar!
 =D>

Title: Re: Higher interference mode thresholds available!
Post by: Dr Obbins on June 22, 2014, 08:58:19 PM
He went about it a different way than I had suggested, but I knew there was a software solution to register more strikes. What I didn't know was that Canada has lightning.  ;) :grin:
Title: Re: Higher interference mode thresholds available!
Post by: Cutty Sark Sailor on June 22, 2014, 09:19:08 PM
Region 3 peaked out today at 300 Strokes per minute. Unreal.  Prior to that we constantly struggled to get 80-100. Relaxing that threshold, and having a handful of newer stations allow us to register many, many more 'weaker' strikes previously un-validated, for one thing.

More surprises coming.  :roll:
Title: Re: Higher interference mode thresholds available!
Post by: JonathanW on June 22, 2014, 09:24:11 PM
My station had the highest stroke count per hour I've had so far - nearly 8000.  I think the repetitive noise/noise floor is still limiting my station somewhat.

Unfortunately, I'm off on a business trip again tomorrow, to return July 2nd.  So permanent siting will have to wait until I get back.
Title: Re: Higher interference mode thresholds available!
Post by: Dr Obbins on June 22, 2014, 09:41:58 PM
Modern Art:
(http://www.cavecountryweather.com/Pictures/modern%20art.jpg)
Title: Re: Higher interference mode thresholds available!
Post by: DaleReid on June 22, 2014, 09:51:01 PM
With more stations and better 'geometry' has the size of the error circle decreased too, or is that just my fertile imagination?
Title: Re: Higher interference mode thresholds available!
Post by: W3DRM on June 22, 2014, 10:35:29 PM
Well, it seems that I chose to move my antenna farm outside at the same time Tobi made his server changes!  ](*,)

So, I don't know if I am getting better results now or if it's Tobi's changes have made the difference. I'm see an incredible number of strikes from my system but not sure why. Anyone have any guesses?

My gains are higher than I've ever seen too. I've attached a screen shot of what I am seeing at the moment.
Title: Re: Higher interference mode thresholds available!
Post by: JonathanW on June 23, 2014, 12:43:08 AM
The only drawback I can see thus far is a greater displayed delay on the lightningmaps real-time page (up to 18 seconds now).   I'm not sure if that's due to greater demand on the server,  greater demand on my local machine for displaying the vastly increased lightning strikes,  or something else.
Title: Re: Higher interference mode thresholds available!
Post by: W3DRM on June 23, 2014, 01:03:09 AM
I noticed that too this evening. However, I observed delays of over 24.x seconds. Am wondering if it has anything to do with the extremely heavy lightning activity in north Texas and Oklahoma? There is an almost continuous bombardment of strikes in that area. My controller is in constant 'click' mode...
Title: Re: Higher interference mode thresholds available!
Post by: dfroula on June 23, 2014, 01:38:33 AM
Don, you're running in full automatic mode, right?

I find with the expanded range that I can run aggressive gains in full manual mode without auto-noise adjust enabled. I no longer seem to need frequent gain readjustments through the day as the lightning rate increases. That assumes you have local noise under control and the added gain is not just sending more garbage.

My real-time delays don't go much above 6 seconds. 4 seconds is typical with the same Oklahoma storms you are seeing. It may be a function of browser/computer speed. Try using the latest Chrome, if you are not. It processes JavaScript and PHP scripts much faster.

Regards,

Don
Title: Re: Higher interference mode thresholds available!
Post by: DaleReid on June 23, 2014, 06:24:31 AM
Don,
I have very low (3.9 to at most nearly 5) second delays, so I have seen zippo degredation in display of near real time strikes.

I run FireFox.  I read one of Tobi's comments some time ago saying it was optimized for Chrome.  I have that and use it for some sites that it seems a little better, so I tried it.  I get the map display, but absolutely no participation lines, or strike animation!  I don't know how Chrome updates, or if you can even find out what version you are running, unlike the About or Help in Mozilla or IE, and maybe it takes care of all the updates itself without even asking or offering the option.  But on my system, Chrome isn't an option if I want to see the 'fun stuff' that makes the map display work.  It just lays there.  Grrr.
Dale
Title: Re: Higher interference mode thresholds available!
Post by: DaleReid on June 23, 2014, 06:39:13 AM
To update,
I went back to loading the beta test page into Chrome again.
I saw strikes animation, and then turned the Stations ON, and my version of Chrome seems to be finally up to speed.

Still a bit of a "Apple" type application (i.e., not enough control on my part, sort of a dumbed down browser as far as I can tell) but then again I like futzing, but not to the point of running Linux.

So all seems well on the latest version of Chrome that is on my machine now, and there doesn't seem to be any time delay advantage, so all seems happy no matter which browser I'm using.

As an aside, I do note that on rare occasion I need to leave Mozilla and go to a web page in IE since some boxes or other fancily programmed drop downs don't seem to appear in FireFox.  I would think that all programmers would follow the HTML rules and that all browsers would handle the standard, as well.  But in some instances, especially with on line ordering and filling in payment and other issues, things just don't work right.

Alas, the challenges of the new world, and choices.

Dale
Title: Re: Higher interference mode thresholds available!
Post by: dfroula on June 23, 2014, 10:44:57 AM
There have been many improvements to the html standard lately. Chrome seems best at implementing the new features first, IE the worst.

There are some graphing programs I wrote for statistical analysis at work that simply will not run on the latest IE, but work just fine on Chrome and Firefox. It all depends if the programmer is utilizing the newer features. I suspect whoever wrote the real-time display is using many of the latest html capabilities.

Don
WD9DMP
Title: Re: Higher interference mode thresholds available!
Post by: W3DRM on June 23, 2014, 12:09:07 PM
Don, you're running in full automatic mode, right?

I find with the expanded range that I can run aggressive gains in full manual mode without auto-noise adjust enabled. I no longer seem to need frequent gain readjustments through the day as the lightning rate increases. That assumes you have local noise under control and the added gain is not just sending more garbage.

My real-time delays don't go much above 6 seconds. 4 seconds is typical with the same Oklahoma storms you are seeing. It may be a function of browser/computer speed. Try using the latest Chrome, if you are not. It processes JavaScript and PHP scripts much faster.

Regards,

Don

The 24.x delays I saw last night only lasted for about 10-15 minutes and then dropped back to the normal 2-4 seconds. I am running the latest version of Chrome (Version 35.0.1916.153 m) on my HP i7 12GB desktop.

I am running in Manual mode. I did have 'Auto adapt to noise" on but have just turned that off this morning. Also, I have 'Alternate channel mapping' selected. I'll have to go through a full setup process again now that I moved my antennas. That will happen later today or tomorrow when I get some time.

Now that I've moved my antennas outside, I'm seeing a lot more strikes and higher ratios. Unfortunately, I moved my equipment at about the same time Tobi enabled the "Interference Mode Optimization" feature so I really can't tell exactly what improvements are due to my moving stuff. I have a feeling that the move did greatly impact my detections as my gains are higher than I have ever seen them before the move. Time will tell...
Title: Re: Higher interference mode thresholds available!
Post by: 92merc on June 23, 2014, 01:03:35 PM
I just looked at my station and noticed the higher interference settings.  I was running my station at 10*10, which is the highest I could get before the change before going interference.  I now pushed it to 16*8 and it's still going.

So good deal.  I should really be able to reach out and grab some strikes.  Now I just need to find time to finish building my housing to mount the E field outside...
Title: Re: Higher interference mode thresholds available!
Post by: Cutty Sark Sailor on June 23, 2014, 08:22:01 PM
Any body else turn off their 'beep on signal' buzzer with the higher interference limits?   :-({|=  I kindia miss mine, but it really became annoying.,.. :cry:
However, I'm definitely not for a return to the 'good old days!'
Title: Re: Higher interference mode thresholds available!
Post by: DaleReid on June 23, 2014, 08:43:59 PM
Oh, yes, that went a long time ago.

It used to be a reminder when a strike was plotted.

Now that happens so much and so frequently that it is indeed annoying, but not nearly as annoying as my wife complaining about the noise.

This fall and winter it no doubt will be turned back on again when the electrons are happy at a certain level above the ground and don't change positions so often.
Title: Re: Higher interference mode thresholds available!
Post by: dfroula on June 23, 2014, 08:48:10 PM
I am not 100% certain, but there seems to be a "catch 22" with the new higher interference mode thresholds.

A reboot will cause the controller to return to the old 15/30 limits until the new higher limits are downloaded from the server.

However, if the gains are set assuming the 70/80 limits, the system goes into immediate interference after a reboot, which seems to prevent the new, higher limits from taking effect, keeping the system in permanent interference mode. Or, perhaps the higher limits will always download after a time, and is not dependent upon whether the system is in interference - and I did not wait long enough?

Don
WD9DMP
Title: Re: Higher interference mode thresholds available!
Post by: Cutty Sark Sailor on June 23, 2014, 09:00:30 PM
I am not 100% certain, but there seems to be a "catch 22" with the new higher interference mode thresholds.

A reboot will cause the controller to return to the old 15/30 limits until the new higher limits are downloaded from the server.

However, if the gains are set assuming the 70/80 limits, the system goes into immediate interference after a reboot, which seems to prevent the new, higher limits from taking effect, keeping the system in permanent interference mode. Or, perhaps the higher limits will always download after a time, and is not dependent upon whether the system is in interference - and I did not wait long enough?

Don
WD9DMP
Yes, give it time to restore the server updates... what?  Maybe up to several minutes?  I noticed this, also... just wait it out. 
Title: Re: Higher interference mode thresholds available!
Post by: 92merc on June 24, 2014, 05:05:07 PM
Blitz MySQL database is down.  I first noticed participants page was down.  Then the forum is down.  My station has reverted back to the original 30/15 settings, which pushed my station into interference.

I was at 10x10 all day.   Had to drop it to 10x4/10x2 to get it back up and running.

I'm noticing the whole country drop in lightning strikes on the Lightingmaps.org site.  So a lot of people are probably into interference now that weren't earlier.  Just in time for the US to be covered in strikes...

Maybe us Region 3 people gave the servers a hernia??
Title: Re: Higher interference mode thresholds available!
Post by: W3DRM on June 24, 2014, 07:05:40 PM
Don, you're running in full automatic mode, right?

I find with the expanded range that I can run aggressive gains in full manual mode without auto-noise adjust enabled. I no longer seem to need frequent gain readjustments through the day as the lightning rate increases. That assumes you have local noise under control and the added gain is not just sending more garbage.

My real-time delays don't go much above 6 seconds. 4 seconds is typical with the same Oklahoma storms you are seeing. It may be a function of browser/computer speed. Try using the latest Chrome, if you are not. It processes JavaScript and PHP scripts much faster.

Regards,

Don

I was in full auto mode but as of yesterday have changed back to Manual Mode, Auto Adapt to Noise OFF and set my gains as follows:

Amp 1:
Channel A Gain: 10  10
Channel B Gain: 10  10
Channel A Threshold:± 120 mV
Channel B Threshold:± 120 mV

Amp2:
Channel A Gain: 5  5
Channel B Gain: 5  5
Channel C Gain: 5  5
Channel A Threshold:± 100 mV
Channel B Threshold:± 100 mV
Channel C Threshold:± 100 mV

Since going back to Manual Mode, I have noticed that my stats have steadily been improving over the past 12 hours or so. I am currently at an Effectivity L of 36% which I have NEVER seen before. This tells me that my move of the equipment outside (much quieter electrically) and the new settings have helped tremendously.

I had to turn my clicker off as my wife was getting pretty anxious about the constant "noise". I considered it "music to my ears" but she does see things a bit differently plus, my ham radio and computer stuff are in her "Sewing Room"...  ](*,)
Title: Re: Higher interference mode thresholds available!
Post by: Jumpin Joe on June 24, 2014, 07:33:31 PM
Don, you're running in full automatic mode, right?

I find with the expanded range that I can run aggressive gains in full manual mode without auto-noise adjust enabled. I no longer seem to need frequent gain readjustments through the day as the lightning rate increases. That assumes you have local noise under control and the added gain is not just sending more garbage.

My real-time delays don't go much above 6 seconds. 4 seconds is typical with the same Oklahoma storms you are seeing. It may be a function of browser/computer speed. Try using the latest Chrome, if you are not. It processes JavaScript and PHP scripts much faster.

Regards,

Don

I was in full auto mode but as of yesterday have changed back to Manual Mode, Auto Adapt to Noise OFF and set my gains as follows:

Amp 1:
Channel A Gain: 10  10
Channel B Gain: 10  10
Channel A Threshold:± 120 mV
Channel B Threshold:± 120 mV

Amp2:
Channel A Gain: 5  5
Channel B Gain: 5  5
Channel C Gain: 5  5
Channel A Threshold:± 100 mV
Channel B Threshold:± 100 mV
Channel C Threshold:± 100 mV

Since going back to Manual Mode, I have noticed that my stats have steadily been improving over the past 12 hours or so. I am currently at an Effectivity L of 36% which I have NEVER seen before. This tells me that my move of the equipment outside (much quieter electrically) and the new settings have helped tremendously.

I had to turn my clicker off as my wife was getting pretty anxious about the constant "noise". I considered it "music to my ears" but she does see things a bit differently plus, my ham radio and computer stuff are in her "Sewing Room"...  ](*,)


Tread lightly in the sewing room..... Been there done that.   :grin:

Joe
Title: Re: Higher interference mode thresholds available!
Post by: JonathanW on June 25, 2014, 04:25:25 PM
The new interference mode thresholds continue to amaze.  While I still need to solve the low-level noise issue at my station, today I've maxed out at more than 9000 strokes per hour.
Title: Re: Higher interference mode thresholds available!
Post by: W3DRM on June 25, 2014, 05:30:05 PM
It looks like they are still changing things:
Title: Re: Higher interference mode thresholds available!
Post by: W3DRM on June 25, 2014, 05:38:21 PM
Just noticed that my Effectivity L has really jumped quite a bit over the past couple of hours. I had watched my Effectivity values dropping so decided to experiment a bit by turning on the Auto Adapt to Noise. That had little to no effect at all. I then turned-on Automatic Mode and almost immediately I see my Effectivity levels rising steadily over the past two hours. I'm leaving it there for a while to see what happens this evening.

Have attached a screen capture of the current BO stats.
Title: Re: Higher interference mode thresholds available!
Post by: miraculon on June 25, 2014, 07:01:34 PM
I must be in the "bad boy club" since mine are still stuck at 30/15  :-(
From the real-time display it seems to be doing alright though...

Greg H.

Title: Re: Higher interference mode thresholds available!
Post by: W3DRM on June 25, 2014, 07:17:35 PM
I must be in the "bad boy club" since mine are still stuck at 30/15  :-(
From the real-time display it seems to be doing alright though...

Greg H.

I suspect those levels drop if you have strikes within a few hundred miles. I'm a very long way away from any of the strikes so I don't get degraded due to local thunderstorm activity. We are expecting some lightning activity in the SIERRA tonight so my interference level numbers may also drop.
Title: Re: Higher interference mode thresholds available!
Post by: Cutty Sark Sailor on June 25, 2014, 08:52:49 PM
I think they have to do with overall signal quality... we know we get scads of sigs that aren't tracked by other stations, either too weak, or blocked, whatever. The server knows that also. I think it compares avg controller settings over time, vs number of good signals actually sent, or similar... so letting her go into interference, without trying to 'ride gain', regardless, probably would be a good way to get the settings up...
Title: Re: Higher interference mode thresholds available!
Post by: W3DRM on June 27, 2014, 12:58:30 AM
Well, I'm eating my own words. This evening my station seems to have been in and out of interference mode for a long time. As I mentioned above, I switched over to Automatic Mode w/Auto Adapt to Noise ON.

For some reason, once the system went into interference mode it would not or could not auto adjust. I had to reset the controller several times but it would just go right back into interference mode. So, I have now gone back to the Manual Mode w/Auto Adapt to Noise OFF.

My Effectivity L numbers which had dropped down to 3% have already climbed back up to 30% in just a few minutes after the reset.
Title: Re: Higher interference mode thresholds available!
Post by: dfroula on June 27, 2014, 01:45:25 AM
The problem I have been having is that after going into interference mode, the interference thresholds revert back to 15/30 sigs/sec from a max of 70/80 sigs/sec.

If you set your gains higher to take advantage of the higher interference thresholds, you may never come out of interference mode once a noise burst pushes you over the edge.

MORAL: Set your gains assuming the old interference thresholds of 15/30 sigs/sec. That way, the system will stay out of interference longer and will recover once the transient noise or heavy local strike activity stops.

I believe the developers cleverly crafted the behavior to work this way!  :-)

Regards,

Don
WD9DMP
Title: Re: Higher interference mode thresholds available!
Post by: JonathanW on June 27, 2014, 07:28:12 AM
The problem I have been having is that after going into interference mode, the interference thresholds revert back to 15/30 sigs/sec from a max of 70/80 sigs/sec.

If you set your gains higher to take advantage of the higher interference thresholds, you may never come out of interference mode once a noise burst pushes you over the edge.

MORAL: Set your gains assuming the old interference thresholds of 15/30 sigs/sec. That way, the system will stay out of interference longer and will recover once the transient noise or heavy local strike activity stops.

I believe the developers cleverly crafted the behavior to work this way!  :-)

Regards,

Don
WD9DMP

Don - I've been having the same issue.  There seems to be a time threshold as well -- the reversion to old interference mode levels doesn't happen instantly upon going into interference mode with the new levels. 

The biggest problem with the reversion is taking advantage of the new interference mode levels becomes impossible.  If you set your gain/threshold such that you don't violate the old levels (so that you recover after an extended interference mode event), then you won't be capturing any more legitimate signals than you used to.  Or, if you set your station to handle a higher level of legitimate signals and you experience an interference-mode event, on an active lightning day, your station would stay in interference mode until all activity quiets down.

The new levels with old gain/threshold settings DO allow for a higher "dynamic range" of system counts, as it were.  But it's still a problem.

I have a feeling the new system will still require some tweaking, and I have a feeling that's planned.  The developers haven't let me down yet :)

The best of all worlds: eliminate all sources of true interference, then set your gain/threshold levels to not hit the new interference mode levels on an active lightning day, and ride herd on your system if a thundercloud goes over...
Title: Re: Higher interference mode thresholds available!
Post by: dfroula on June 27, 2014, 07:59:59 AM
Yes, I find myself "locking" into interference mode a few times a day, which knocks my station off-line for an extended period of time, until I manually set the gains very low. wait to come out of interference mode, wait some more until the interference thresholds set back to the higher levels, then readjust the gains.

The added range helps in one direction - it keeps you out of interference mode longer,but once you go over, the recovery criteria revert to such low values that you stay in interference mode for a long time.

My point about setting interference values assuming the 15/30 values is that you will take advantage of the new thresholds and get better rates, but you need to assure that the rates will drop below the (old 15/30) reset percentage criteria when the interference/local storm passes. You don't want steady-state signal levels from normal aggregate distant lightning levels to EVER cross the new higher threshold, or you will stay in interference until the nationwide aggregate rate for your station drops, which could take hours. Hitting the threshold from a local storm or temporary interference should theoretically clear more quickly.

Don

Don
WD9DMP
Title: Re: Higher interference mode thresholds available!
Post by: Cutty Sark Sailor on June 27, 2014, 08:32:59 AM
If the system resets, then it reverts. It then will not return to 'increased' thresholds until next server communication, which may be up to, what, 7-10 minutes?
'Burst' mode doesn't reset the controller, and therefore, doesn't reset the thresholds. Also if you have any numbers in the setting box for "system daily reboot' you can count on reverting to default at least once daily!  That field should be 'blank" !

I suppose I'm lucky... my normal H field and E field settings have been pretty much established, with or without the increased settings, they run the same, normally..
What I do get, is more good signals, and a 'decrease' in 'distance from cell activity before interference' from about 200 miles to something much lower. Now the 'weaker strokes' not detectable by distant stations,  that once drove me into interference, can be added to, say, 5 stations at moderate distance, and I become the 6th station. Viola! More Strikes!

And guess what.. more of those 'weaker' strokes will be ground waves, previously not used, instead of reflections, and Egon's algorithms have better TOGA data with feelers, streamers, etc previously undetectable, and therefore unusable.  Now the option to use more than the current 1 channel of data comes into play dramatically...

As said above... turning my 'normal' settings up is a waste of time and strokes.  I've experimented with it. It's as much about geography as anything. West coast stations could potentially raise theirs for example, but not ol' Mike here in Central KY... too centralized...  If anything, I'm prone to back 'em down during nearby heavy activity...

Then, again, especially on H Channel A,... I never know when the damn axle factory south of me will fire up line three welders.
Title: Re: Higher interference mode thresholds available!
Post by: dfroula on June 27, 2014, 08:37:34 AM
....and after a reset, the higher settings will NEVER be set by the server if you go immediately into interference mode!
Title: Re: Higher interference mode thresholds available!
Post by: JonathanW on June 27, 2014, 08:41:46 AM
If the system resets, then it reverts. It then will not return to 'increased' thresholds until next server communication, which may be up to, what, 7-10 minutes?
'Burst' mode doesn't reset the controller, and therefore, doesn't reset the thresholds. Also if you have any numbers in the setting box for "system daily reboot' you can count on reverting to default at least once daily!  That field should be 'blank" !

Mike, I've seen it revert after an extended "interference mode" event, without system reboot.  Perhaps it was due to the non-burst IM level being triggered.

Also, you should see if Tobi or Egon can reprogram that axle factory to fire up only when there's no lightning.
Title: Re: Higher interference mode thresholds available!
Post by: Cutty Sark Sailor on June 27, 2014, 09:39:10 AM
....and after a reset, the higher settings will NEVER be set by the server if you go immediately into interference mode!
Exactly!!!
Title: Re: Higher interference mode thresholds available!
Post by: Cutty Sark Sailor on June 27, 2014, 10:09:51 AM
If the system resets, then it reverts. It then will not return to 'increased' thresholds until next server communication, which may be up to, what, 7-10 minutes?
'Burst' mode doesn't reset the controller, and therefore, doesn't reset the thresholds. Also if you have any numbers in the setting box for "system daily reboot' you can count on reverting to default at least once daily!  That field should be 'blank" !

Mike, I've seen it revert after an extended "interference mode" event, without system reboot.  Perhaps it was due to the non-burst IM level being triggered.

Also, you should see if Tobi or Egon can reprogram that axle factory to fire up only when there's no lightning.
Well, after extended interference mode, my system will generally reset, and revert... then I wait for a server communication, unless I go immediately into interference. Reset it, or reboot it, it reverts. 
Now, I suppose the logic is that if you've gone long interference at higher count threshold,  then we'll make sure you don't keep sending junk, at least until you get below 'minimum' threshold...  do something ... find the issue...

Axles?  Gotta have 'em to keep rollin'... What's cool about it, I took the time to locate the source last year, found it exactly due south, and know exactly what to change, and by how much. H field, Amp1 channel A, from 10x10 to 8x5 or 8x4 if they're running more than 1 welder line on the north side. All I need do is confirm its H Amp1 Channel A, and a couple of clicks in settings, then go on about life.  Once in awhile, check to see if it's still there, If not, bring it back up. Heck, can predict if it's going to run by driving by the factory and observing the parking lot at shift change... "probability of Weld Strikes tonight? 80% after 6pm with intermittent discharges until 8am"...

Now, a failing vapor street light is a different issue... about all I need do now, is ring up the utility co, they see the incoming phone number and immediately dispatch a new bulb and a crew... don't even have to talk to 'em anymore... :twisted:

Neighbors' plasma TV's and bad heat pumps and HVAC are a bit more difficult... however....

None of this stuff bothers E field, at least so far. 'Bout the only way to throw that bugger into interference is crank up the gain 'till signals overwhelm it.... my environments pretty E noise clean.
So for my location, I love the higher thresholds, and for the most part, an issue is predictable, or easily mediated. Now, it took time and patience to reach that stage... each of us has to do it to some degree, especially when sources mask each other, and there is no single reason for an issue that might appear single source.
Title: Re: Higher interference mode thresholds available!
Post by: Jumpin Joe on June 27, 2014, 10:18:37 AM

"Heck, can predict if it's going to run by driving by the factory and observing the parking lot at shift change... "probability of Weld Strikes tonight? 80% after 6pm with intermittent discharges until 8am"...



I think you need to put a "Loggie" on the lot!!!!    :lol: :lol: :lol:
Title: Re: Higher interference mode thresholds available!
Post by: W3DRM on June 27, 2014, 11:11:06 AM

"Heck, can predict if it's going to run by driving by the factory and observing the parking lot at shift change... "probability of Weld Strikes tonight? 80% after 6pm with intermittent discharges until 8am"...



I think you need to put a "Loggie" on the lot!!!!    :lol: :lol: :lol:

Wow, lots of good info in this thread!

Mike - As for your welding noise, perhaps they do have some kind of a weekly/monthly schedule for axle welding. Have you tried to log the date & time of these events? They might be periodic in a way that would allow you to know when you're going to get hit with the noise bursts. Just a thought...
Title: Re: Higher interference mode thresholds available!
Post by: Cutty Sark Sailor on June 27, 2014, 11:12:49 AM

"Heck, can predict if it's going to run by driving by the factory and observing the parking lot at shift change... "probability of Weld Strikes tonight? 80% after 6pm with intermittent discharges until 8am"...



I think you need to put a "Loggie" on the lot!!!!    :lol: :lol: :lol:
EXCELLENT IDEA... without doing any math, though... would need a whole bunch of extenders.
wait for it... wait for it..

What about an H FIELD AMP to detect the welders?  Hmmm.... .
Title: Re: Higher interference mode thresholds available!
Post by: W3DRM on June 27, 2014, 11:18:58 AM
Or, you could get friendly with some of the workers there and see if they can tell you anything about their work schedules and when welding is done...
Title: Re: Higher interference mode thresholds available!
Post by: Cutty Sark Sailor on June 27, 2014, 11:23:40 AM

"Heck, can predict if it's going to run by driving by the factory and observing the parking lot at shift change... "probability of Weld Strikes tonight? 80% after 6pm with intermittent discharges until 8am"...



I think you need to put a "Loggie" on the lot!!!!    :lol: :lol: :lol:

Wow, lots of good info in this thread!

Mike - As for your welding noise, perhaps they do have some kind of a weekly/monthly schedule for axle welding. Have you tried to log the date & time of these events? They might be periodic in a way that would allow you to know when you're going to get hit with the noise bursts. Just a thought...
Good thought, except it depends on their daily orders, ... besides, it's easy to live with now... which is part of my emphasis when I keep saying that operators, especially new ones, need to explore and understand their electromagnetic environment, from season to season, day to night, and when they do, they'll be more efficient, and that's a huge step toward sending quality data.
Furthermore, if I went away for a few days, knowing my environment, I'd just set the gains at 8x4 on H1A and 8X8 on H1B, and go enjoy myself.
Title: Re: Higher interference mode thresholds available!
Post by: Cutty Sark Sailor on June 29, 2014, 07:04:34 PM
What is a rudimentary ground???


Title: Re: Higher interference mode thresholds available!
Post by: Cutty Sark Sailor on June 29, 2014, 07:16:24 PM
What is a rudimentary ground???

Connected to household power earth ground.  Nothing fancy, just an earth reference to the controller ground lug (which up until now was floating).
https://www.isa.org/templates/news-detail.aspx?id=125975
Title: Re: Higher interference mode thresholds available!
Post by: Cutty Sark Sailor on June 29, 2014, 07:44:43 PM
I've found over 45 years of career and experience... the best true ground normally will be found at the shield of a cable TV at the entrance to your home. Period. Residential grounds,. utility ground... maybe... maybe not... CATV folks don't want their crap buggered.  Best Ground Going, normally.
Title: Re: Higher interference mode thresholds available!
Post by: DaleReid on June 29, 2014, 08:30:43 PM
Mike,
Its sort of like whether the cable installer initially took the time to ground properly or not.

The older homes, with ATT or MaBell doing installs, had wonderful grounds.  Usually the power company did a bang up job of grounding outside before powering an entrance panel.

Maybe so maybe not with the recent cable company installs I've seen. 

Mine cable cable's ground, which brings in internet, is jumpered over to the telephone ground.  Go figure. But I'm sure it saved someone three minutes and ten bucks, so they could have coffee on the way to their next job.



Title: Re: Higher interference mode thresholds available!
Post by: txweather.org on June 29, 2014, 10:52:35 PM
I've found over 45 years of career and experience... the best true ground normally will be found at the shield of a cable TV at the entrance to your home. Period. Residential grounds,. utility ground... maybe... maybe not... CATV folks don't want their crap buggered.  Best Ground Going, normally.

Lucky you.

4 years ago a lightning hit came in through the cable (Coax) line and fried everything connected to my network.
The same happened to all of my neighbors and their stuff.
Title: Re: Higher interference mode thresholds available!
Post by: dfroula on June 29, 2014, 11:39:34 PM
A US residential power installation, using 220VAC divided into two phases for 110VAC, NEEDS a proper ground to work properly. 15 years ago in my last house, the service panel lost its ground reference due to a loose connection in the meter housing. My voltages across the two phases swung around wildly, depending on the load on that phase. I lost a good deal of electronic equipment at the time because of the surges. Light bulbs got very bright on some circuits and dimmed to nothing on others. Very weird experience.

Whether the service panel ground is also a good RF ground is another matter. In my house, the service panel has braided copper strap in a conduit going to the copper cold water pipe on the city side of the water meter. That ground has proven to be a pretty good RF ground as well, at least at VLF.

Every few years I pull the main breaker and perform some maintenance on the ground clamp connecting the service panel to that water pipe. It can corrode over time and cause problems.

Don
WD9DMP
Title: Re: Higher interference mode thresholds available!
Post by: W3DRM on June 30, 2014, 12:57:20 PM
Nice -- over 12,000 detected strokes per hour peak this afternoon at one point.

That's mostly because I added a rudimentary ground.  But I was also experimenting earlier with a first crack at a linear regulated power supply -- the traces were really nice :)  But inadequate heat sinking was pushing the regulator temperature too close to the operating limit, and adding a fan seemed to add some E-field noise.

Now, I just need the remaining parts to arrive from Mouser to build my production power supply, and install in the attic.

Care to share your linear designs with the group? Might be something we'd all like to think about doing...
Title: Re: Higher interference mode thresholds available!
Post by: W3DRM on June 30, 2014, 01:05:28 PM
I should have read through the rest of the thread before posting my reply above.

Anyway, here are a few links regarding UFER grounding. That is what I have here in my house. It was built in 2005 and local codes require UFER grounding. I suspect it's mainly due to our desert environment here along with only a few inches of precipitation on an annual basis.
Title: Re: Higher interference mode thresholds available!
Post by: dfroula on June 30, 2014, 01:33:25 PM
You know, I thought about this, and what I stated is not quite what happened.

What was lost in the meter housing was the neutral connection back to the center-tapped power distribution transformer. The neutral connection is also connected to the service panel ground. When the neutral connection was lost, all return current was forced to pass through ground, creating a high-resistance path and the power balance issues that I mentioned.

Losing both the ground and neutral both probably would have completely interrupted power, as neither 110VAC phase would have had a path back to the center tap of the distribution transformer. Losing just the ground, power would have been unaffected, but it would create a significant safety hazard.


Don
WD9DMP

A US residential power installation, using 220VAC divided into two phases for 110VAC, NEEDS a proper ground to work properly. 15 years ago in my last house, the service panel lost its ground reference due to a loose connection in the meter housing. My voltages across the two phases swung around wildly, depending on the load on that phase. I lost a good deal of electronic equipment at the time because of the surges. Light bulbs got very bright on some circuits and dimmed to nothing on others. Very weird experience.

Whether the service panel ground is also a good RF ground is another matter. In my house, the service panel has braided copper strap in a conduit going to the copper cold water pipe on the city side of the water meter. That ground has proven to be a pretty good RF ground as well, at least at VLF.

Every few years I pull the main breaker and perform some maintenance on the ground clamp connecting the service panel to that water pipe. It can corrode over time and cause problems.

Don
WD9DMP
Title: Re: Higher interference mode thresholds available!
Post by: Cutty Sark Sailor on June 30, 2014, 01:40:06 PM
I thought you catch that, Don... !  :-P  That's why I like the Cable ground... the folks here run a ground at least every third pole... rarely do we have any local outages, but sometimes it does take the main head out!

As you said, neutral and ground are not the same thing at all... and the instructions specifically, and one time, I suppose they still do, stated specifically NOT to connect to mains ground!

Mike
Title: Re: Higher interference mode thresholds available!
Post by: dfroula on June 30, 2014, 01:59:33 PM
Yep...brain cells fading.

Thunking some more, I realize that losing both ground AND neutral would also be Very Bad, as two devices on separate legs will create a 220VAC series connection through the devices. Depending upon the load on each leg, power would be wildly imbalanced. Even worse, The current is just "looking" for a ground path - like through the cable TV ground (!) or, worse yet, a gas main (!!!).

Grounding the B. system to the third grounding terminal at the outlet is likely OK, as it carries no current, normally, and is there as a safety ground. In my case, the separate ground wire from each outlet is bonded to the neutral bus in the service panel, which is connected both to the transformer neutral and the water pipe ground.

There could be ground currents flowing between the service panel neutral/ground bus bars and the house ground, depending if there is a potential difference between the distribution transformer ground and the grounding point of the service panel. I guess this could introduce potential noise.

A separate grounding rod is starting to sound more attractive.

Don
WD9DMP
Title: Re: Higher interference mode thresholds available!
Post by: dfroula on June 30, 2014, 02:50:59 PM
Back on-topic, regarding the new interference thresholds......

This is driving me slightly nuts! I'm still looking for that perfect "set and forget" gain setup. I don't think I'll find it..... :mad:

The hysteresis is so extreme on the settings, they will happily allow you some extreme rates...until you tip over the edge. Then, you drop back to the basic, low interference threshold settings....for a very long time!

The only cure, as I mentioned, is simply to set the gains for a aggregate rate that will keep you out of interference under even the most stringent interference thresholds, then stick with it and enjoy the short-lived margin of high signal rates as a storm approaches.

Unfortunately, I'm not so patient.  I succumb to the temptation to boost those gains to take advantage of the highest thresholds that you are granted after a short period of compliance. Them WHAM, a storm locks you back into interference mode.....for a long time.

With my old setup and location, auto-noise adjust helped me some. Now I have almost no sustained noise issues, so "auto-adjust to noise" doesn't help me stay out of interference anymore. Running with auto-noise adjust is essentially like running in straight manual mode.

My own fault, I know. But those darned Blitzortung guys are holding out the stick and carrot, just to snatch it away again after a few nibbles.  :grin: I think they are trying to educate us.

Don
WD9DMP
Title: Re: Higher interference mode thresholds available!
Post by: dfroula on June 30, 2014, 02:54:17 PM
Like in my previous home, where some dimwit hadn't bothered to run a neutral wire to a ceiling fan and just used the hard conduit as a return path. Nice. Interesting charred marks on some of the thinwall conduit couplings.

D.

Yep...brain cells fading.

Thunking some more, I realize that losing both ground AND neutral would also be Very Bad, as two devices on separate legs will create a 220VAC series connection through the devices. Depending upon the load on each leg, power would be wildly imbalanced. Even worse, The current is just "looking" for a ground path - like through the cable TV ground (!) or, worse yet, a gas main (!!!).

Grounding the B. system to the third grounding terminal at the outlet is likely OK, as it carries no current, normally, and is there as a safety ground. In my case, the separate ground wire from each outlet is bonded to the neutral bus in the service panel, which is connected both to the transformer neutral and the water pipe ground.

There could be ground currents flowing between the service panel neutral/ground bus bars and the house ground, depending if there is a potential difference between the distribution transformer ground and the grounding point of the service panel. I guess this could introduce potential noise.

A separate grounding rod is starting to sound more attractive.

Don
WD9DMP

I agree with this.

Regardless, what I described was a very temporary arrangement, by way of determining the effect on signals received.  I don't recommend it nor is it still being employed.  And if one is unsure of one's power supply configuration in the first place, it may represent a hazard.  I hope that's clear.
Title: Re: Higher interference mode thresholds available!
Post by: Cutty Sark Sailor on June 30, 2014, 03:16:37 PM
Back on-topic, regarding the new interference thresholds......

This is driving me slightly nuts! I'm still looking for that perfect "set and forget" gain setup. I don't think I'll find it..... :mad:

The hysteresis is so extreme on the settings, they will happily allow you some extreme rates...until you tip over the edge. Then, you drop back to the basic, low interference threshold settings....for a very long time!

The only cure, as I mentioned, is simply to set the gains for a aggregate rate that will keep you out of interference under even the most stringent interference thresholds, then stick with it and enjoy the short-lived margin of high signal rates as a storm approaches.

Unfortunately, I'm not so patient.  I succumb to the temptation to boost those gains to take advantage of the highest thresholds that you are granted after a short period of compliance. Them WHAM, a storm locks you back into interference mode.....for a long time.

With my old setup and location, auto-noise adjust helped me some. Now I have almost no sustained noise issues, so "auto-adjust to noise" doesn't help me stay out of interference anymore. Running with auto-noise adjust is essentially like running in straight manual mode.

My own fault, I know. But those darned Blitzortung guys are holding out the stick and carrot, just to snatch it away again after a few nibbles.  :grin: I think they are trying to educate us.

Don
WD9DMP
Yep... educate us and push server to limits. No joke.
Don, honestly, I work better with the original 'normal' settings I've established... you and I both know that we were registering strikes far far away, back when we only needed a couple or three stations to register... What the new thresholds do, especially adding E field..., is essentially operate with the same settings, but detect more strokes, and weaker strokes.... So my gains generally are staying at previous levels.
It does offer 'more distant' stations a chance to join in the fun, since they can come up a point or two on gain without going interference.. but for us centrally located stations I think offers more of what I suggest above...
Tip... remember that E field CH C is mostly same data as H field signals... with a sharper cutoff >50Khz... if hitting interference, you could reduce that... one reason the preset gain is lower for EC. I wouldn't necessarily reduce HA or HB, I'd go with EC instead, for now. It also will respond to 'weaker' signals than H field, but the main info that Egon's looking for can be carried with EA and EB... for those running Dual... Stations running H or E only have to find their own hunting ground...
Asking us ain't gonna do it... they have to do what we did... find the average, set, learn how to respond if the situation changes....at least for now.
Title: Re: Higher interference mode thresholds available!
Post by: dfroula on June 30, 2014, 03:29:31 PM
Yep, I agree with all of that.

Hardest thing is to wait out a storm so that one can evaluate when things return to normal on their own, with the proper settings. It is also tempting to set gains low, reboot, wait for the higher thresholds, and boost the gains up again to get some counts in during the storm.  ](*,)


"Guily! Guilty! My evil self is at that door and I have no power to stop it!"

I found that E-field channel C trick as well. It helps to balance the performance of the two amps. The developers helped us there, as the final unadjustable gain stage of the E-field Channel C is lower than that of A and B.

Man, those guys are sure smart!  :grin:

Don
WD9DMP
Title: Re: Higher interference mode thresholds available!
Post by: Cutty Sark Sailor on June 30, 2014, 03:35:55 PM
Also remember you're the guy who dropped strong hints (translation: kicked my butt) last year about 'riding gains'.  :twisted:
Title: Re: Higher interference mode thresholds available!
Post by: dfroula on June 30, 2014, 03:38:00 PM
Oops....

Just call me Easy Rider  8-)
Title: Re: Higher interference mode thresholds available!
Post by: Cutty Sark Sailor on June 30, 2014, 03:42:57 PM
Oops....

Just call me Easy Rider  8-)
Fire all of your guns at once
 And explode into space
Title: Re: Higher interference mode thresholds available!
Post by: scarecrow93 on July 01, 2014, 09:51:59 PM
Has anyone noticed their loss of the higher gain settings the last couple of evenings?

Just logged into my controller which was running at 32 sig/min...which was fine under the new 80 burst/70 continuous settings.

It's been happening roughly the same time each evening and clears itself at the same time each night.  I  noticed a bunch of reds in interference yesterday as well.

I think the server is yanking the rug out from under me.

Thoughts?

Thanks!
Lance
Title: Re: Higher interference mode thresholds available!
Post by: dfroula on July 01, 2014, 10:53:54 PM
In addition to my observations earlier in this thread, I believe the server reduces the interference thresholds if too high a percentage of signals are from noise, even if the limits have not been reached. No one on the development team has explained exactly how it works, however.

Don
WD9DMP
Title: Re: Higher interference mode thresholds available!
Post by: Cutty Sark Sailor on July 02, 2014, 06:56:37 AM
hmmm... I can't help but wonder if it's a hint to "go for network quality" not "station quantity" ...
especially with E's coming on line with 3 channels added...
Title: Re: Higher interference mode thresholds available!
Post by: Jumpin Joe on July 02, 2014, 07:28:18 AM
hmmm... I can't help but wonder if it's a hint to "go for network quality" not "station quantity" ...
especially with E's coming on line with 3 channels added...

I think the Sailor has just set his sail in the prevailing wind.....
Title: Re: Higher interference mode thresholds available!
Post by: W3DRM on July 02, 2014, 12:31:20 PM
hmmm... I can't help but wonder if it's a hint to "go for network quality" not "station quantity" ...
especially with E's coming on line with 3 channels added...

I think the Sailor has just set his sail in the prevailing wind.....

Yep, shouldn't we always be thinking of "network quality" and not our own "station quantity"? Seems as though that old human instinct of being first in everything comes into play and we forget that this is a "community first endeavor"...

The problem is, how do we get to that ideal state of "network quality"? Me thinks we still have  a lot to learn.
Title: Re: Higher interference mode thresholds available!
Post by: Dr Obbins on July 02, 2014, 12:48:23 PM
I figure if the station is giving 90% efficiency and 10% of the strikes are minimum stations, then the "station quantity" is the "network quality". This is until the network has better coverage or until the servers cry uncle as I posted in this thread that got a bit off topic. (http://www.wxforum.net/index.php?topic=23027.msg222027#msg222027) If the current top 3 stations went off line, over 10% of the entire networks strikes would not be recorded.
Title: Re: Higher interference mode thresholds available!
Post by: scarecrow93 on July 02, 2014, 01:27:08 PM
Yep, shouldn't we always be thinking of "network quality" and not our own "station quantity"? Seems as though that old human instinct of being first in everything comes into play and we forget that this is a "community first endeavor"...

The problem is, how do we get to that ideal state of "network quality"? Me thinks we still have  a lot to learn.

You're exactly right!!!

I got caught in a "Ooh shiny!" moment and jacked my gains up from my tried and true settings.  Just trying to stretch the range a bit and learned I sacrificed station resiliency in the process.    ](*,)

Backed the gains back down to the tried and true today and have been watching it pretty close with all the activity in PA and srn NY.  9000 strikes inside 500km and the detector still has some overhead before I have to worry about interference mode.  My locating ratio has taken a dive from all the static though.

Like you said.  It's all about network quality.  Need more stations.
Title: Re: Higher interference mode thresholds available!
Post by: JonathanW on July 02, 2014, 01:45:05 PM
I figure if the station is giving 90% efficiency and 10% of the strikes are minimum stations, then the "station quantity" is the "network quality". This is until the network has better coverage or until the servers cry uncle as I posted in this thread that got a bit off topic. (http://www.wxforum.net/index.php?topic=23027.msg222027#msg222027) If the current top 3 stations went off line, over 10% of the entire networks strikes would not be recorded.

That makes sense to me.
Title: Re: Higher interference mode thresholds available!
Post by: JonathanW on July 02, 2014, 01:47:48 PM
Yep, shouldn't we always be thinking of "network quality" and not our own "station quantity"? Seems as though that old human instinct of being first in everything comes into play and we forget that this is a "community first endeavor"...

The problem is, how do we get to that ideal state of "network quality"? Me thinks we still have  a lot to learn.

You're exactly right!!!

I got caught in a "Ooh shiny!" moment and jacked my gains up from my tried and true settings.  Just trying to stretch the range a bit and learned I sacrificed station resiliency in the process.    ](*,)

Backed the gains back down to the tried and true today and have been watching it pretty close with all the activity in PA and srn NY.  9000 strikes inside 500km and the detector still has some overhead before I have to worry about interference mode.  My locating ratio has taken a dive from all the static though.

Like you said.  It's all about network quality.  Need more stations.

I have an engineer interested in installing a station in Charlottesville, VA, but it will be after the break when Egon resumes sending kits.
Title: Re: Higher interference mode thresholds available!
Post by: Cutty Sark Sailor on July 02, 2014, 01:49:17 PM
A lot of the 'network quality' relates to work that must be done on the servers to get additional processing, newer algorithms, and features... which became delayed because of the unprecedented demand for kits recently. That's one reason for the hiatus in kit distribution, as well as parts availability.
Another new firmware for current boards should come down with the next week or so.

Now, regarding individual stations and network figures...

Assume 6 working stations, in our network, in a circle of 500 miles, surrounding a supercell to eliminate the question of station density.
The cell actually generates 5000 real strokes. There is no interference noise at any station. One perfect station sends 5000 signals. All 6 stations record 1000 strikes, with the other 5 stations sending less than 4000 signals. The network records 1000 strikes (20% of actual). The network goal is 80%. That's network quality for detection. 

Of the 1000 detected strikes 500 have a 9km deviation. The network goal for deviation is 1km. That's network Quality for location accuracy.

Where are the missing 4000 strikes? Why?  Especially when one station appears to have caught 100%.
Why is the deviation so high?

Title: Re: Higher interference mode thresholds available!
Post by: dfroula on July 02, 2014, 01:54:37 PM
It might be good to review the various basic "quality" metrics available at blitzortung.org and lightningmaps.org. As Egon has said, there is no good metric of station quality at the moment. It's one of his goals to come up with one when he returns from holiday:

lightningmaps.org (Statistics/Network)
============================

- Count(strokes/hour): The total number of calculated lightning strikes to which the station contributed.
- Ratio (strokes/hour): The percentage of total calculated strikes in the last hour to which the station contributed.
- Count (signals/hour): The total number of signals contributed by the station in the last hour, whether or not they contributed to a strike calculation. This includes noise and authentic strikes.
- Ratio(signals/hour): The percentage of signals sent by a station that resulted in a strike calculation.
- Efficiency: The percentage of total calculated system strikes over the last hour to the total number of signals reported by a station over the last hour.


blitzortung.org (Participants)
=====================
- Signals: Same as "Count (signals/hour)" above
- Effectivity S (first column): Same as Count(strokes/hour) and "Ratio (strokes/hour)" above, but limited to distances from the station from 0 to 50 km.
- Effectivity M (second column): Same as Count(strokes/hour) and "Ratio (strokes/hour)" above, but limited to distances from the station from 0 to 500 km.
- Effectivity L (third column): Same as Count(strokes/hour) and "Ratio (strokes/hour)" above, but limited to distances from the station from 0 to 5000 km.

Note that the numbers on each site do not align, likely because of differences in how the "last hour" is calculated.

Most people concentrate on the lightningmaps.org "Ratio (strokes/hour)" figure as the best estimate of station performance, since it gives an overall figure for the percentage of detected strikes to which the station contributed. This can be misleading, though, as high numbers may have to do with the geometry of current lightning activity in relation to station location, the number of surrounding stations that receive the same strike, and the intensity of lightning within the centers of lightning activity. A station can have very good RF sensitivity and gains correctly set for maximum detection, but still see lower than expected percentages.

The lightningmaps.org "Efficiency" percentage can be misleading, as the figure does not distinguish between reported noise signals and real detected strikes that an insufficient number of other stations have not detected. A station can be efficiently detecting strikes with very little noise, yet the "Efficiency" percentage will be low because stations are far apart. This number becomes more meaningful as station density increases.

It's important to specify which of these metrics one is referencing when tossing around numbers.

Regards,

Don
WD9DMP
Title: Re: Higher interference mode thresholds available!
Post by: Cutty Sark Sailor on July 02, 2014, 01:57:12 PM
A lot of the 'network quality' relates to work that must be done on the servers to get additional processing, newer algorithms, and features... which became delayed because of the unprecedented demand for kits recently. That's one reason for the hiatus in kit distribution, as well as parts availability.
Another new firmware for current boards should come down with the next week or so.

Now, regarding individual stations and network figures...

Assume 6 working stations, in our network, in a circle of 500 miles, surrounding a supercell to eliminate the question of station density.
The cell actually generates 5000 real strokes. There is no interference noise at any station. One perfect station sends 5000 signals. All 6 stations record 1000 strikes, with the other 5 stations sending less than 4000 signals. The network records 1000 strikes (20% of actual). The network goal is 80%. That's network quality for detection. 

Of the 1000 detected strikes 500 have a 9km deviation. The network goal for deviation is 1km. That's network Quality for location accuracy.

Where are the missing 4000 strikes? Why?  Especially when one station appears to have caught 100%.
Why is the deviation so high?


Hint... first quality indication will be signal to noise ....
Title: Re: Higher interference mode thresholds available!
Post by: Dr Obbins on July 02, 2014, 02:10:00 PM
Quote
The lightningmaps.org "Efficiency" percentage can be misleading, as the figure does not distinguish between reported noise signals and real detected strikes that an insufficient number of other stations have not detected. A station can be efficiently detecting strikes with very little noise, yet the "Efficiency" percentage will be low because stations are far apart. This number becomes more meaningful as station density increases.
Our friend in Hawaii is a perfect (and extreme) example of this - currently he has reported 11,574 signals in the last hour, and yet 0% are verified strikes.
Title: Re: Higher interference mode thresholds available!
Post by: Jumpin Joe on July 02, 2014, 02:51:39 PM
It might be good to review the various basic "quality" metrics available at blitzortung.org and lightningmaps.org. As Egon has said, there is no good metric of station quality at the moment. It's one of his goals to come up with one when he returns from holiday:

lightningmaps.org (Statistics/Network)
============================

- Count(strokes/hour): The total number of calculated lightning strikes to which the station contributed.
- Ratio (strokes/hour): The percentage of total calculated strikes in the last hour to which the station contributed.
- Count (signals/hour): The total number of signals contributed by the station in the last hour, whether or not they contributed to a strike calculation. This includes noise and authentic strikes.
- Ratio(signals/hour): The percentage of signals sent by a station that resulted in a strike calculation.
- Efficiency: The percentage of total calculated system strikes over the last hour to the total number of signals reported by a station over the last hour.


blitzortung.org (Participants)
=====================
- Signals: Same as "Count (signals/hour)" above
- Effectivity S (first column): Same as Count(strokes/hour) and "Ratio (strokes/hour)" above, but limited to distances from the station from 0 to 50 km.
- Effectivity M (second column): Same as Count(strokes/hour) and "Ratio (strokes/hour)" above, but limited to distances from the station from 0 to 500 km.
- Effectivity L (third column): Same as Count(strokes/hour) and "Ratio (strokes/hour)" above, but limited to distances from the station from 0 to 5000 km.

Note that the numbers on each site do not align, likely because of differences in how the "last hour" is calculated.

Most people concentrate on the lightningmaps.org "Ratio (strokes/hour)" figure as the best estimate of station performance, since it gives an overall figure for the percentage of detected strikes to which the station contributed. This can be misleading, though, as high numbers may have to do with the geometry of current lightning activity in relation to station location, the number of surrounding stations that receive the same strike, and the intensity of lightning within the centers of lightning activity. A station can have very good RF sensitivity and gains correctly set for maximum detection, but still see lower than expected percentages.

The lightningmaps.org "Efficiency" percentage can be misleading, as the figure does not distinguish between reported noise signals and real detected strikes that an insufficient number of other stations have not detected. A station can be efficiently detecting strikes with very little noise, yet the "Efficiency" percentage will be low because stations are far apart. This number becomes more meaningful as station density increases.

It's important to specify which of these metrics one is referencing when tossing around numbers.

Regards,

Don
WD9DMP

Excellent information Don.  =D>

I'm sure this will how many obtain a better understanding.

Joe
Title: Re: Higher interference mode thresholds available!
Post by: dfroula on July 02, 2014, 03:40:19 PM
It seems that one key component to coming up with a good quality metric would be the ability to distinguish between noise-triggered signals and real lightning.

The serve must certainly be able to do this. I would guess there is some kind of waveform qualification that takes place before attempting to correlate the timestamp of the signal with other stations. Otherwise, the server could not determine which signal from the various channels is "best" when a signal is sent. The server needs to compensate for the various filter delays and adjust the reported GPS timestamp.

Don
WD9DMP
Title: Re: Higher interference mode thresholds available!
Post by: JonathanW on July 02, 2014, 03:47:59 PM
It seems that one key component to coming up with a good quality metric would be the ability to distinguish between noise-triggered signals and real lightning.

The serve must certainly be able to do this. I would guess there is some kind of waveform qualification that takes place before attempting to correlate the timestamp of the signal with other stations. Otherwise, the server could not determine which signal from the various channels is "best" when a signal is sent. The server needs to compensate for the various filter delays and adjust the reported GPS timestamp.

Don
WD9DMP

Though signals processing wasn't my emphasis, I would imagine correlation techniques can be used to determine whether signals from more than one station are likely from the same source.  Signals that don't have a match can be discarded.  Perhaps it's also possible to derive from Fourier analysis which signals have energy distribution typical of lightning, and which don't (e.g. http://vlf.stanford.edu/research/global-lightning-geo-location (http://vlf.stanford.edu/research/global-lightning-geo-location)).

Most of my efforts have been toward getting my station set up in a permanent, low-noise manner.  Once that's finished, I plan on doing a little reading.  (e.g. http://wwlln.net/publications/dowden.toga.article.pdf (http://wwlln.net/publications/dowden.toga.article.pdf), others).
Title: Re: Higher interference mode thresholds available!
Post by: Cutty Sark Sailor on July 02, 2014, 03:55:21 PM
Nothing can happen with other stoke parameters until it's accurately located, first. So do expect, as Don says, a quality analysis to be implemented. Bad signals come from lots of things.. too much gain, (too many distorted skywaves), bad board, noise interfering , monkeying with board design and messing up delays... E field too much CAT5 and short coax... too long leads on H field... all kinds of things.

Note that as more western stations come online, weaker signals will be detected by those, that count as 'network' calculated, and the wonderful 'high' effitivities will become harder to obtain for stations in east... further stations will not sense the weaker strokes. Also look for signal quality analysis to be implemented,... experimentally and conditionally, on the server. Many of the distorted 'sky waves' from longer distances could be 'rejected' as unusable for refined analysis, and again the 'effectivity' we've grown fond of becomes more irrelevant. As you may have already noted, those who are running Dual E and H.. how many times the server will opt for a cleaner signal of E field. Also note how many more signals you may be getting running both.

Suggestion: and this is a bit trickier since the new 'auto thresholds' are being tested... and impossible with "Automatic mode"
H field - if you're running dual, turn off E field. (1x1 will work)  "don't send" is NOT the solution... leave it alone...
On a 'quiet day' with cells > 800 miles away, with your gains at your 'determined normal' set the thresholds about twice  your noise level. Watch it for some time, as storms move and stoke counts change. You may have to adjust it. Now, with numerous storms about 600-800 miles, set your gains (H field) so it goes into interference. Then raise threshold in 10mv jumps until you sense it will go out of interference. Leave it there for a bit. And watch it.

After some norm established for H field, turn H off, (1x1), turn on E.... do something similar, assuming most storms still ±600 mi.

Now bring H field back to the gains settings you'd determined before... if go interference, either raise H field thresholds, or lower H field gains till clears interference...

What I think this will do is begin establish a baseline for 'Normal" for you, at a sacrifice of signals >1500km, but use the E field more efficiently, with it's cleaner signals and three channels. When full server automatic is implemented, the server will likely not go beyond your overall settings normally, but will be able to 'down gain' etc as needed. As...
When storms get closer than 600Mi, just reduce H field gains.... 1x1
Then closer than 400-500 bring E fields down if need be... depending on your probe, you may have to go 1x1 or 1x2 when cells <50-100mi...
Title: Re: Higher interference mode thresholds available!
Post by: cbh3 on March 31, 2017, 11:25:36 PM
On a 'quiet day' with cells > 800 miles away, with your gains at your 'determined normal' set the thresholds about twice  your noise level. Watch it for some time, as storms move and stoke counts change. You may have to adjust it. Now, with numerous storms about 600-800 miles, set your gains (H field) so it goes into interference. Then raise threshold in 10mv jumps until you sense it will go out of interference. Leave it there for a bit. And watch it.

After some norm established for H field, turn H off, (1x1), turn on E.... do something similar, assuming most storms still ±600 mi.

Now bring H field back to the gains settings you'd determined before... if go interference, either raise H field thresholds, or lower H field gains till clears interference...

What I think this will do is begin establish a baseline for 'Normal" for you, at a sacrifice of signals >1500km, but use the E field more efficiently, with it's cleaner signals and three channels. When full server automatic is implemented, the server will likely not go beyond your overall settings normally, but will be able to 'down gain' etc as needed. As...
When storms get closer than 600Mi, just reduce H field gains.... 1x1
Then closer than 400-500 bring E fields down if need be... depending on your probe, you may have to go 1x1 or 1x2 when cells <50-100mi...

Mike,

As a new blue system owner and knowing things change over time with hardware and software, do you still subscribe to using this method/process of adjusting the antennae systems on the unit today?
Title: Re: Higher interference mode thresholds available!
Post by: Cutty Sark Sailor on April 01, 2017, 12:31:45 AM
Absolutely.  If you have noise, intermittent or consistent, running any Autos will cause great consternation and irritation. Learn your environment, and adjust to it.  There is such inconsistency across the network in station settings and operations that the 'auto thresholds' are likely never to function properly, in my opinion... and I'm not sure they're even pursuing further exploration along those lines at the moment. I'm a believer in manual for my location, for sure... and it's become worse as new noisemakers have appeared. Autos would flat cycle me in and out to the point I'd be useless.  On both my Red and Blue systems... Few of us should be operating >800dat miles at this stage of network growth, anyway... data beyond the 2nd  skywave reflection is pretty much useless for accuracy and quality.
Title: Re: Higher interference mode thresholds available!
Post by: ke9lz on April 01, 2017, 02:13:40 AM
I totally agree with cutty on this.  You kind of missed the best time to adjust your system during the winter because it gives you a period of very little lightning so that you can find the that point of man made noise and lightning when adjusting in manual mode but there will still be more time for you to find that point yet.  Many make the mistake to try to make it to the top of the list looking for counting every strike even if it is 2000 miles away but in reality your looking for that strike a lot closer in because it is a team effort with all stations.  I have 2 stations, a blue and a red spaced 70 miles apart and they both stay real close together when viewing the station stats and both were set in manual mode using winter (very few lightning strikes) conditions to set my gains and thresholds. When the lightning is within 1000 miles from my stations you will see both of them right there at the top of the list and to me that is where you should be shooting for.

Steve 
Title: Re: Higher interference mode thresholds available!
Post by: cbh3 on April 01, 2017, 09:36:21 AM
Absolutely.  If you have noise, intermittent or consistent, running any Autos will cause great consternation and irritation. Learn your environment, and adjust to it.  There is such inconsistency across the network in station settings and operations that the 'auto thresholds' are likely never to function properly, in my opinion... and I'm not sure they're even pursuing further exploration along those lines at the moment. I'm a believer in manual for my location, for sure... and it's become worse as new noisemakers have appeared. Autos would flat cycle me in and out to the point I'd be useless.  On both my Red and Blue systems... Few of us should be operating >800dat miles at this stage of network growth, anyway... data beyond the 2nd  skywave reflection is pretty much useless for accuracy and quality.

Great, thanks for confirming that. If the weather cooperates I'll try that in the coming days.

One of the first things I learned was taking the system out of full Auto, and putting it into Manual. I found a good YouTube video that gave me a crash course on recommended setting after not finding much elsewhere. Then I discovered I had intermittent interference, and nothing to sniff it out with since I was still learning. And of course it couldn't possibly be coming from my house... but it was. I connected the unit up to battery (wifi and network too) and killed power to the house. Yep, interference gone! I ended up finding 6 dimmer switches in the house that were the cause (all swapped out now with normal switches). So far so good, but I expect I might have one more source somewhere (neighbor).

Like you mentioned I'm learning what my noise environment is like, and will be for the next few months to a year. I'm currently enjoying a somewhat low noise floor (15-25mVpp) with the loops mounted above the garage. I expect things will come and go; just hoping I'll have some control over them when they do. I may play with the 'Auto Adapt to Noise' setting to let me know what's capable, but you make a good point about not overreaching with the station.

I really wish the unit had was some sort of timestamped log file or a way to let you know when it goes into interference outside of seeing your stats fall on the websites. I understand that without a local disk that is not possible, and would be an exponential increase in server load and bandwidth for a once in awhile review of that data. Just something that would be nice.

Thanks for your time.

Chuck
Title: Re: Higher interference mode thresholds available!
Post by: cbh3 on April 01, 2017, 09:50:41 AM
I totally agree with cutty on this.  You kind of missed the best time to adjust your system during the winter because it gives you a period of very little lightning so that you can find the that point of man made noise and lightning when adjusting in manual mode but there will still be more time for you to find that point yet.  Many make the mistake to try to make it to the top of the list looking for counting every strike even if it is 2000 miles away but in reality your looking for that strike a lot closer in because it is a team effort with all stations.  I have 2 stations, a blue and a red spaced 70 miles apart and they both stay real close together when viewing the station stats and both were set in manual mode using winter (very few lightning strikes) conditions to set my gains and thresholds. When the lightning is within 1000 miles from my stations you will see both of them right there at the top of the list and to me that is where you should be shooting for.

Yeah, I agree that I might have missed my chance for awhile, being in West Michigan and in a kind of centralized area. Maybe I'll catch a break. I would have done it over winter, but I didn't get my golden ticket email until the top of March.  :grin:

I admit is was pretty awesome to hear those first clicks last Saturday after the build of the unit and my loops. When I wasn't swamped by unknown (at the time) interference my station was locating strikes 3-4k away once in awhile. Yesterday I was even at the top of the Blitzortung Effectivity L list for some time.

Now that I'm mostly interference free, I plan to begin the process of reading and tweaking for more accurate reporting, so I'm not flooding the servers with signals all of the time. I'll be looking to increase my Effectivity M list percentage perhaps, or improve other stats shown on LM.org. I also noticed yesterday that my station was "seeing" strikes off the East Coast at the same time as other stations, but wasn't being given credit for the detection by the servers. Yep, I'm still learning the whys and why nots, but I'm a quick study.

Chuck
Title: Re: Higher interference mode thresholds available!
Post by: cbh3 on April 08, 2017, 08:46:31 AM
It's a nice and quiet morning for most of the USA to tweak unit settings in relation to noise floors, or to collect information about the surrounding environment without strikes being detected.  \:D/