New to Meteor Detecting

[SteveNickolls]

Hi Al,

Thanks for this very much. Guess what? Complete operator error/dumbness on my part.   When I was in Colorgramme every time I was trying to 'open' the RMOB data file in the meteorlog folder rather than clicking on the 'select' option. Clicking on 'select' makes everything visible in Colorgramme!   Yippee! You have made me very happy.

I will start again collating data using the Simon Dawes 2012-06-09 conditional actions in Spectrum Labs tonight as the last few days have been a muddle data collection-wise. I don't think the time this week has been completely wasted however as I have become aware of the sophistication of the conditional action script. I can also appreciate why different 'set ups' in terms of the conditional actions used can give rise to different 'counts'.

I will also try and see if I can  add my manually obtained counts for early September into Colorgramme.

Once I'm happy with things working properly I will make the step to send the backlog of data to the RMOB Organisation

Thanks once again Al.

Cheers,

Steve

[Al.Grant]

Hey Steve,

Glad you've got it sorted.  Colorgramme seems to work well once it's actually running ......  but instructions on exactly how to set it up are a bit thin on the ground.

I'm currently trying  another Conditional Actions script modified by Phil (Hykeham Observatory) that seems good at filtering out some of the small non-meteor detections that can get counted.

But only just loaded this and will take a couple of days to see exactly how the results work out ... but looking good so far.

I've done a couple of minor tweaks to it myself, but only to adjust the S/N ratio and to only take screen captures and audio recordings for the larger detections .... nothing clever, just easier to look/listen to the bigger ones, rather than every blip & ping.

My audio playback issue is now fully resolved and I'm getting nice screen captures with matching audio.  A few really good hits over last night and during this morning.  Calmed down a bit now though.

This was from about 9:45 this morning (my Spec Lab set to GMT), not the loudest but one of the longer ones.

Audio:   HERE

You can have multiple instances of Spec Lab running at the same time, both feeding off the same FCD.  

What I started looking at last night was having one running with the current setup and a second running with a much slower waterfall trace.

This might be interesting to see a longer period of activity during heavier meteor showers.

No need for meteor counts, RMOB data or to take screen grabs/audio of each detection for each event as the first one deals with that, but I can use SL's Periodic Actions to take occasional captures of the longer trace.  

I had it working last night with a waterfall trace of around 7-8 mins. But I'm still tweaking things to try for a longer trace whilst still showing up detections.

Al

[Al.Grant]

Steve,

Just seen your colorgramme and map locator on the RMOB site ....  seems to show you were also seeing decent levels of activity this morning    

Al

[Gasman]

Hi guys

Looks like you lot are really getting the hang of meteor observing :-) You'll have to do a 'How to' with Colourgramme/RMOB/Spec lab etc etc if you get the time !!. Wondering if anyone has done any decoding yet using the FCD pro+? I gather virtual audio cables are required? Will have to suss that out!

Cheers

Steve

[SteveNickolls]

Hi Al,

Yes I took the plunge and sent my first data to the RMOB site this morning. I'm pleased you have sorted the audio problem and I'm intrigued by you mentioning running multiple SL especially when you expect a larger number of meteors. While I was struggling with SL and Colorgramme I was struck by how sophisticated the modified Simon Dawes script was for determining whether a detection was actually a meteor compared to the original one I had been running. With the modified script I am sure I am getting fewer detections than before so my set up before was probably passing more spurious detections. The only thing I would like to change would be to screen print less information every detection, but I lack the knowledge to amend it.

That was a good trail you detected this morning Al, and I can confirm detecting it too. The detection lasted for 100 'durations' with my set up. With my previous set up I had physically measured on screen what a duration was, a duration was the equivalent of 1/7.52th of a second. Now I am using the longer script I will need to work it out again as I understand a 'duration' is one cycle of the script (a bit of elastic).

Hi Gasman, I agree some sort of 'How To' would make it easier to get to grips with SL and Colorgramme. From what I have come across SL mostly delivers, at one level, what you need 'out of the box'. SL is so expansive in what it allows you to do it is daunting at first. There are however details specific to your pc that you will have to enter in the SL 'Configuration and Display Control' windows, for example your input device and output device, what audio sampling rate your audio card/chipset allows, system time etc. There's a handy circuit window that will tell you if all's ok or not. There's a big manual to read through which will help. There's a number of configurations you can try and when you get a bit adventurous there are extra conditional actions scripts on the Internet, but the two Al made available on this thread are good ones to try. I've only used the Simon Dawes script but Al has also tried one from Phil (Hykeham Observatory). These have amongst other things better discerning abilities on judging what is and what is not a meteor trail. If you go on http://cmhas.wikispaces.com/SLMeteor there's an excellent 'explanation' of the Simon Dawes script line by line. You will see that you can modify some lines to try out on your set up. There are also some 'differences' between scripts so be aware.

Cheers and good luck on your ventures,

Steve

P.S. If you want to use Colorgramme to forward your detections to the RMOB site (Radio Meteor Observatory organisation) I ran foul of how to 'open' the RMOB data file. Follow Al's advice and set up a C:\\SpectrumLabs\meteorlog folder in SL and in Colorlabs fill in ALL the Observer details :-)

[MarvinZ]

I have just received my dongle, I have gone down the none FUNcube Dongle and my chosen dongle arrived yesterday.

I have managed to get it installed on the laptop but now work interferes with any further progress for a few days. But rest assured I will be ready to ask a heap of questions soon.

So please all of you that are posting your efforts here continue to do so as I have found the assistance here excellent, and I thank you all for your efforts in passing on your knowledge and experience.

Thanks

[SteveNickolls]

Hi Peter,

Good luck with your installation.

Cheers,

Steve

[SteveNickolls]

Hi,

I happened to be looking at my computer screen early this morning and captured three interesting trails that I thought I would share. Checking on ISS Tracker showed the ISS over the France/Spain border region at 06.44 UTC so I'm wondering if the curved trail (middle screen shot)may have been the ISS? Annoyingly SL has reverted to printing the wrong frequency on the screen captures so I will have to rectify that. Did anyone else capture similar trails this morning?

Cheers,

Steve

[Al.Grant]

Hi Steve,

I definitely saw that lats 06:50 hit.

I think I got one of those earlier passes,  I'll have to have a closer look when I get home to confirm.

 Al

[Al.Grant]

Now using yet another script modified by Phil Norton (Hykeham Observatory), this allows me to upload certain screenshots to my new website.

http://www.ukstargazer.org/meteor-detection/   

The current waterfall trace is uploaded every 60 secs and the longest duration and brightest (highest S/N ratio) images are also uploaded.

I may drop the 60sec trace as it will often be pretty quiet.

My script was updated today, once it all settles in it will be good to be able to compare my detections to those on Phil's site.  http://www.lincolnastronomy.org/phil/

Steve:  Your location is pretty much in the middle of both of ours, so might be interesting to compare your detections.

Al

P.S.  Not much on my site at the moment other than the meteor info, but once I'm happy with that I'll start on the rest. And the template still needs some tweaks!

[SteveNickolls]

Hi Al,

Great to hear your update-good luck with the new script. I will have a good peek at your webpage too! I do like the idea of showing the longest and brightest images on your site. You have been very busy.

Yes, it's a good idea to compare detections and possibly even wider across the UK if we can. I haven't scrutinised the figures but the RMOB site shows a variance of data across the UK in terms of numbers of detections at the very least. I can imagine 'set ups' and conditional action scripts used have parts to play here.

I have always had my aerial mounted in the horizontal plane which could be a significant factor affecting compared counts. I have noticed though that since changing over to the modified Simon Dawes script (dated 9.6.2012) and sending data to the RMOB site the number of detections per hour on my set up have dropped considerably compared with when I was using the basic script that came with SL. For example, previously the lowest number detected per hour in July and August was usually around 10 (with the highest daily hourly counts being around 60 but once peaking at 140 during the Perseid shower) now even the highest number I have ever recorded is only 26 so there's some powerful filtering happening. I have attempted changing the secondary test values in Lines 32 and 33 of the script from the supplied 6 and 7 but to little apparent affect so I may well remove these two lines to see what happens next. 

I will have a  look at the UK data in RMOB and see what similarities/differences I can find there such as number of detections and in daily detection patterns between observers.

Can I ask Al is your new script markedly different to the Simon Dawes modified one, particularly when discriminating potential detections?

This subject gets more fascinating with every step.

Cheers,

Steve

[Al.Grant]

Hi Steve,

I'll have a look and compare the original script with the latest one to see what differences there are.  I know the Dawes script introduced some tests to filter out non-meteors and also that Phil has added quite a few bits.

Apart from the auto upload screen captures, I can also now set a limit for what duration level actually initiates a screen or audio capture.

This doesn't affect the actual RMOB meteor count, only what gets captured.  This is REALLY useful as it means I only have to look at captures for big detection each morning rather

than sift through 40 or more screenshots to find any good ones!

If you want to try the latest one let me know.

The first thing I would check is the figure you have for THRESHOLD on line 5 of the script.

This is the s/n level above which a signal gets counted .... or at least passed along to the next test.

I think the original S@N script was set to 17.  I've tried settings from 17 - 20 but have settled on 18 at the moment.  At 20 I was definitely missing meteors that I could hear and see quite clearly!  

I think the antenna can be mounted either horizontal or vertical.  I've read that vertical is more sensitive .... but don't know how true that is.

I imagine there could be quite large differences depending on location and equipment (receiver/antenna etc) being used.  I've spoken to one guy (a radio ham)  further North from us and he gets

nothing from Graves and yet others in Scotland and Newcastle have no problems.

I suspect that the actual meteor detection numbers are less significant than the increase/decrease in detections during meteor showers.  But I think it's important to get the system to a stable state

you are happy with and then leave it.  That way you'll have a baseline level of detection when comparing rates before/during/after a shower.

I think I'm pretty much at the point where I can stop tinkering with the detection values, now it's more about what level to initiate a capture.

Cheers 

         Al

[Gasman]

Hi guys

Nice web site Al very useful!. Just got my new `Owl` 2m antenna (great build quality!) as recommended by Al and waiting to install it high up on my chimney stack. I`ll be pointing it due SE but really I`d like to get a more accurate bearing from my home town of York but can`t seem to get the actual location of Graves ??. Just a further point is there any software available to help tune an aerial? My new one can be finely tuned but needs an swr meter to do so!

cheers

Steve (gasman)

[SteveNickolls]

Hi Al,

Thank you so much for all this. Checking on SL the THRESHOLD value I'm currently using is 20 so I'll try 18 and see if this 'improves' things. I agree that when set to 20 SL does seem to miss obvious 'pings'. I can see how the changes to the script you have done will let you transfer data to your website and show the largest and brightest capture. I'll see how changing the THRESHOLD value goes before trying a new script Al but thanks for the offer of the newer script right now. I agree that my aerial polarisation may be contributing to a reduced count but I can live with that if it is not too significant and in fact it may be useful to have at least one aerial set up horizontally to be able to compare and contrast trails we detect. I've had a quick-ish look at the recent data on RMOB and the UK contributors seem to fall into three distinct groups dependant upon the number of detections. There is a very wide difference in detection numbers both overall and between 'groups'. None of the physical set ups are the same. It would be good though if we can get a decent similarity on the conditional action detection scripts were using at least for our area and I'm the observer that needs to change settings so I'll get onto it.

Thanks again,

Steve

[SteveNickolls]

Hi Steve (Gasman),

Glad to hear you now have your new aerial. The way I pointed my aerial was to go on Google Earth and use the 'ruler' icon to draw a line from your home to the Dijon site. The heading value will be useful if you intend putting your data on the RMOB site. On Google Earth look at any local feature (perhaps a nearby church or house that lies on the line between your house and GRAVES Dijon and line up your aerial by eye to that nearby marker. This should give an accurate point of direction for your aerial Steve. The GRAVES site can be found at 47 degrees 20 minutes and 52.65 seconds North and 5 degrees 30 minutes and 54.25 seconds East.

Sorry I can't help with aerial tuning though Steve but perhaps someone else can help here or it may not be necessary to tune it at all.

Let us know how you get on with your system :-)

Cheers,

Steve

[Al.Grant]

Hi guys

Nice web site Al very useful!. Just got my new `Owl` 2m antenna (great build quality!) as recommended by Al and waiting to install it high up on my chimney stack. I`ll be pointing it due SE but really I`d like to get a more accurate bearing from my home town of York but can`t seem to get the actual location of Graves ??. Just a further point is there any software available to help tune an aerial? My new one can be finely tuned but needs an swr meter to do so!

cheers

Steve (gasman)

Hi,

I used this site to work out my bearing to Graves (Just East of Dijon)  http://www.sunearthtools.com/tools/distance.php

On the map simply drag 'A' to your location and 'B' to the destination and it will give you a range and bearing.

From York it looks to be about 148-150 degrees.

I wouldn't worry too much about tuning your antenna with a swr meter, as far as I'm aware this only a major concern if you are intending to transmit.

Cheers

          Al

[Gasman]

Ta for that Al, Just installed my antenna tonight & hope to try it out later. How come there are always family do's when you want to try out stuff? ;-)

Steve

[Gasman]

Hi guys

Really pleased with my new antenna and got this within 10mins of connecting up. I`m using Simon`s actions but not sure which parameter to alter as the screen grab shows it as a `Test` and I heard a definite whistle.

and this one too!

cheers

Steve

[SteveNickolls]

Hi Steve, congratulations! Good screenshots of the events. I have gone through my records and did get two detections around the time of yours though both were only of duration 1 for me :-( I believe that 'Test' is meant to be printed on the screen as there as its a test SL runs whenever timer3 expires after a detection, I think this is the line in question- sp.print(M_TIM,"S=",M_SIG,"dBm N=",M_NOI,"dBm F=",M_FRQ,"Hz t=",M_DUR,"S:N=",M_SIG-M_NOI,"Test",TEST_A,TEST_ :M_DUR=0

Good luck with deciding on your settings, I'm still tinkering with mine as I seem to be getting fewer hits than other observers (it may be the horizontal position the aerial is set up at rather than being vertical as whatever changes I make to the conditional action script I don't see much change in the detection numbers. Still interesting though.

Do let us all know how you get on.

Cheers,

Steve

Cheers

[Al.Grant]

Hi Steve [Gasman]

Yep looks like you're getting some meteors.

As Steve N says that 'Test' on the screen is part of the measurements in the script to try distinguish between meteors and other hits from non-meteors.

If I remember that script correctly, you can edit out the TEST_A & TEST_B parts of that line so it doesn't print on screen.  But make a backup first just to be sure or save with a different name!

Notepad++  is a good tool for this and also worth getting the 'Compare' plugin for it (both free).

It's worth checking the THRESHOLD value on about line 3.  I tried a few values and have settled on 18 at the moment, 19 was also OK but at 20 I was definitely missing valid meteors.

When you get settings that you are happy with you can save them as a .USR file so it's easy to go back to a working config if you tweak other things.

With that script (or modified version of it) There are a couple of quick & easy buttons you can add.  Pick one of the spare ones on the left side and enter the info below.

This gives you a detection count for the current hr - It doesn't save the data anywhere, but is useful if you don't want to constantly watch the screen!

This one is also useful for turning the audio recording on/off - Just be aware that leaving it running all the time will result in some HUGE files!

I really wouldn't leave it running for more than a couple of mins .... and definitely not overnight!

You'll find the info at the bottom of this page:   http://cmhas.wikispaces.com/slmeteor

It looks like you are using the S@N script as a base setup ..... same as I did, and it's still the base of my config, you are also seeing detection at around the same 22khz freq that I get them.  

If you want to take a quick screen shot you can click that 'Capture Now' button.

It saves the screen shot straight into the /Spectrum dir.

That script is great and got me detecting, but I did find it difficult trawling through dozens and dozens of screenshots each morning looking for the good ones.

The modifications Phil has added allow a bit of control over what gets grabbed.  So smaller meteors still get counted (for RMOB) but no screenshots taken.  

Then each morning I might have a dozen or so captures, but I know they're all decent ones.

Al

[Gasman]

Hi Steve and Al

Thanks for the info. I`ve installed my 2m Owl yagi vertically as per the instructions, not too sure if it would make any difference mounting it horizontally? -

I`ve since found Phil`s modified actions which seem to monitor things a little better although I watched the screen and heard a whistle from the pc speaker indicating a hit and also showing it on screen but no text shown??. Maybe the actions are a little hit or miss depending on the intensity? only been trying it for a short time so still finding me feet .

Couple of Spec lab queries if poss!

Wondering why my hits seem to be seen at 2.2kHz on screen and what does that indicate exactly if we are monitoring at 143048kHz please?

I`d like to eventually leave the pc to monitor 24hrs, is it possible for Spec lab to only save screens containing hits and not generate huge files of nothing?

I have an old desktop running Linux Mint 16 it would good to make use of this for meteor detection but although I can run Spec Lab through Wine I can`t find anywhere to set the frequency as the Linux FunCube controller (Qthid) is only for Funcube dongle and not the pro+ model, can I just enter the frequency (143048kHz) in the top left box in Spec lab indicated `vfo`?

many questions to follow

best

Steve (gasman)

[SteveNickolls]

Hi Steve,

Nice aerial you have there. Good luck with bedding in your set up. Yes sometimes you will get what appears and sounds like a 'ping' but the script you are using and its settings may determine otherwise. There are parts of the script you can tweak as you know and in the end it is down to you being satisfied. I'm still at that stage :-)

The GRAVES signal is unmodulated (no sound) and as your system is tuned to slightly below 143.050 MHz (at 143.048 MHz) the meteor trail reflections produce an audible 'ping'. Your set up will have, as part of the threshold settings, a range of frequencies that you are telling SL to scan (for me it is 1740 to 2780 Hz) with 2000Hz being the centre frequency. The reflected meteor trails will often impart a Doppler shift effect due to the velocity of the moving plasma trail. When you see a signal above 2000Hz the meteor will be moving someways towards you and if below 2000Hz away from you. If you look at a larger trail on your monitor it often shows a slight frequency shift on its path into the atmosphere. So the trails you have seen were made by meteors ablating while moving someways towards you from the GRAVES signal.

Al will be able to help regarding saving screenshots as he uses a further modified script that records only the larger 'pings' as JPEG's saving disk space. He still records all the pings just saving the larger ones.

I'm not able to help re Linux-sorry. I must admit I would like to have a small dedicated Win 7 pc for solely running meteor stuff, a small ITX system would be a low cost, cheap to run set up.

Good luck in your work!

Cheers,

Steve

[Al.Grant]

Hi,

Been a while since I had a box running linux.

There is a cross platform freq controller claims to be for the PRO and PRO+  http://www.oz9aec.net/index.php/funcube-dongle/qthid-fcd-controller

Not tried it  .......  so don't blame me if your FCD+ goes up in smoke! 

Al

[Al.Grant]

Wondering why my hits seem to be seen at 2.2kHz on screen and what does that indicate exactly if we are monitoring at 143048kHz please?

Yesterday I started adding a 3D waterfall image to the files uploaded to my webpage.  http://www.ukstargazer.org/meteor-detection/   

The 'almost live' waterfall & 3d get updated every 60 secs so may not show anything !

What I have noticed is that the freq shown on the normal trace is around 2200hz whilst the same detection on the 3D capture is much closer to 2000.

Not had a chance to look closer at that and though I'd like to find out why, as long as I'm getting the detections I'm not too worried.

Al

[Gasman]

Hi Al

Which actions are you using or have you modified your own and are those actions capable of saving only screen shots of hits please?

cheers

Steve