What's the minimum to start meteor detecting?

[Stub Mandrel]

I have just got an RTL2832 based USB SDR and got it working with HDSR, of course all I hear is hiss as I haven't got an antenna!

I want to try meteor detection using the graves radar.

A visit to B&Q later and some swift DIY should get me a yagi sorted.

A few questions... (just a few!)

Is ordinary satellite dish cable the right impedance?

I hear stories of low pass filters and signal boosters ... are these needed?

I have a satellite dish SWR meter, can this be used to optimise the aerial in any way?

What are the choices for meteor detection software?

[SteveNickolls]

Hi Neil,

Welcome to meteor detecting.

I still use my S@N antenna for detecting meteor signals off GRAVES radar signal and B&Q was where I got a lot of my antenna parts from. ? It is a few years now since I had to look into the parts required and I'm not 'up' on the finer parts of detecting but my understanding is there are a number of types of cable suited to the job some being better than others because of their construction and component metals. You could try what you have in terms of cable at least for a first try.

As regards additional filters etc. I have not needed anything to complement my FUN cube Pro+ dongle and home built aerial, the necessary filters are built into the FUN cube, are they built into the RTL2832 based USB SDR?

To detect reflected signals from meteors using GRAVES your set up must be able to receive at around 143.050MHz (actually at 143.048MHz to hear the reflected 'ping'), does the software radio have that ability?

As regarding optimising the aerial that of course depends on the antenna design. Mine is a simple 3 element type and I have no issues having raised it about 12 foot off the ground from the shed and pointing towards Dijon (actually the area that GRAVES illuminates is quite wide). My set up is also horizontally polarised but either vertical or horizontal should be ok. Mine is also raised about 10 degrees upwards. Others have found having the antenna indoors reduces reception. Nearby housing in the way can also compromise signal. I'm lucky that I have a completely horizon free view to the SE.

I went along with the suggestion in the S@N article about using Spectrum Lab software and I got mine working straight up without any knowledge of radio etc. Spectrum Lab is very good but can be daunting but the detail allows you to tweak things once you get familiar with things. Some people are able to write their own code to work with Spectrum Lab, others use software written on the Internet. 

I take it you have begun to read the SGL thread-

 

Best of luck with your adventures. I hope others better to help on specialised matters will chip in.

Cheers,

Steve

[Stub Mandrel]

Serious thanks for that Steve.

I am lucky as my old weather station is about 6m high and I can point the aerial pretty clear south.

I found some claims that 75 vs 50 ohm feeder is not critical for a receiver. Hopefully,  have a coil of  Sky dish coax somewhere!

I think I will try and get it working in a basic way with the HDSR program and ask about Spectrum lab at the club (Rosliston seem to be one of the 'pioneers' with the Graves radar).

[SteveNickolls]

Cheers Neil,

Good luck with this. I hope others will be able to give you more assistance, if required on the technical side of things. The great beauty and satisfaction with meteor detection using radar is it can be done 24/7.

Good luck,

Steve

[The Admiral]

On 06/06/2018 at 14:13, Stub Mandrel said:

I have just got an RTL2832 based USB SDR and got it working with HDSR, of course all I hear is hiss as I haven't got an antenna!

I want to try meteor detection using the graves radar.

A visit to B&Q later and some swift DIY should get me a yagi sorted.

A few questions... (just a few!)

Is ordinary satellite dish cable the right impedance?

I hear stories of low pass filters and signal boosters ... are these needed?

I have a satellite dish SWR meter, can this be used to optimise the aerial in any way?

What are the choices for meteor detection software?

I think Steve has covered it really. I don't think the cable type is that critical unless you have a very long cable run or really are scraping around for signal. I guess sat cable has a pretty low loss. Signal level meters won't work because you won't have any signal to measure! The idea is not  to receive the ground wave from Graves at all. In any event, the alignment of the antenna isn't that critical as the volume of sky "illuminated" by the Graves radar is pretty extensive. I've put a map on the thread that Steve linked to. The angling of the antenna upwards I would contend is unnecessary as the sensitivity pattern in the vertical plane is largely dictated by affect of the earth.

I have an antenna in the loft, horizontally polarised,  and it works fine, though undoubtedly stuck on a pole outside and high up would always be better.

Keep us posted on your successes!

Ian

[Stub Mandrel]

Cheers Ian,

I have finished my Yagi, aside for waiting for the glue holding the rear element to set. I've used 6mm tube instead of 15mm, but my understanding is that the diameter of elements isn't critical. It's the length and spacing that matters. Tomorrow it will go up in the air and I will try and install the software on my old Vista laptop.

[Stub Mandrel]

In the hope of helping others who come later, you need the old version of Kadig 2.2 to install drivers in Vista

[Stub Mandrel]

Help!

I think I have HDSR working... but nothing seems to happen.

Set up for Radio 4, I get a bright yellow streak up the middle, but just a hiss from the audio.

With no antenna, the yagi or 50cm of wire I get apparently the same signal.

Set up for Graves, I get nothing but an even fainter signal.

Is there an easy test to see if the SDR works?

[The Admiral]

Hmm, seems a bit odd Neil, unless you have a very strong R4 signal perhaps. What mode are you in, USB, LSB, CW, FM? 

42 minutes ago, Stub Mandrel said:

Set up for Graves, I get nothing but an even fainter signal.

Do you mean a fainter R4, or another signal? You shouldn't pick up anything there.

As a start, I'd suggest what I always recommend, tuning to the 2m beacon GB3VHF on 144.430 Mhz (http://www.gb3vhf.co.uk).

See if you can detect that, at least it's near the Graves frequency.

Ian

 

[Stub Mandrel]

I'm on  92.7 FM, setting and gan't see a bean. with a 30cm wire I should get something.

For Graves, upper sideband, there's nothing

I will try 144.430 and see what happens.

[Stub Mandrel]

1 hour ago, The Admiral said:

tuning to the 2m beacon GB3VHF on 144.430 Mhz

Aha!

I get a yellow stripe! And the bottom display shows a peak at about -90dB above a noise floor at around -150db.

I assume that means all is well?

?

[The Admiral]

Short answer Neil, I don't know! I'm not familiar with HDSDR and it's screen appearance. I originally used SDRConsole. Perhaps a screen dump?

My initial reaction is that you have a very low signal level, but then again, your noise floor is very, very low. I would have thought it ought to be higher than that if you are genuinely connected to the antenna. Do you get any sound though? If you tune a kHz or so either side of 144.430, can you hear a tone? This tone should change periodically as the beacon cycles through its program. If not, perhaps it isn't configured properly. I'm assuming that the 'stripe' is a signal, but this should step around a bit periodically as it is the carrier has periods when it is modulated, either frequency or on/off (morse).

Here's what I get:

Ian

[Stub Mandrel]

Hmm, I suspect signal to noise ratio is what matters - my signal trace looks like yours but with a clearer peak, but the 'waterfall' wasn't shifting - but I din't watch for long.

HDSR seems to have the same controls and general features.

I want to switch to spectrum lab as it does automatic recording - I'm unlikely to want to spend hours looking at a blank screen waiting for meteors - perhaps during the perseids!

I want to be sure everything works first though, it seems that even with configuration files this is not straightforward stuff.

[Stub Mandrel]

Here's a trace of the beacon:

And graves:

 

 

[The Admiral]

56 minutes ago, Stub Mandrel said:

I'm unlikely to want to spend hours looking at a blank screen waiting for meteors

Indeed, Neil, that's why I moved to Spectrum Lab! You might wish to wait for a meteor shower so that you can check that your set up is seeing the meteors, as then you wouldn't have long to wait before you can spot one. A lot of people seem to find trouble with setting up for meteor detecting, not entirely sure why, but I guess being an avid amateur radio enthusiast in my youth did help . Not so with Spectrum Lab, though, it almost caused my ageing brain to catch fire. There is a lot of detailed documentation, but it can be rather dense. I think if you can get hold of the script that others use you should be well on your way, though if I cast my mind back a few years to when I set mine up, there are some gotchas in terms of some of the settings which aren't incorporated in the .ini file, and making sure that your sound card can communicate with the software. IIRC your sound card needs to be able to sample at 192kHz.

Looking at your latest post you seem to have a decent signal on GB3VHF. I can't work out what your bandwidth is set to though. Mine's 2.8kHz in my image above. You will obviously want it wide enough to encompass any Doppler shift on the scatter signal. I see figures in the Hz range on your screen, I don't know exactly what they represent, but indeed if your BW is that narrow, it would account for the very low noise level. Then again, if the light blue band to the right of your tuned frequency is your bandwidth, that looks reasonable.

Did you let the GB3VHF signal go through its full sequence, CW, clock signals, Morse, and frequency shift? The latter should show up as a stepped frequency signal like mine. You don't say if you can hear the sound.

Anyway, your Graves trace looks OK. Just a question of, as you say, sitting and watching (or listening)! The top pane looks as though it's the waterfall, which must be working.

Good luck.

Ian

[Stub Mandrel]

2 minutes ago, The Admiral said:

Looking at your latest post you seem to have a decent signal on GB3VHF.

Actually, its 144.030 not 144.430... but I returned and get a similar result, there must be signals at both frequencies.

I've tried bandwidths from narrow to wide with similar results.

The big problem with Spectrumlab is that everyone's instructions are for funcube which pretends to be an audio card, and I want to use it with a generic SR dongle.

9 minutes ago, The Admiral said:

I see figures in the Hz range on your screen, I don't know exactly what they represent

I think these are the bandwidth of the vertical slices of the waterfall display, as they get bigger the display gets chunkier. I think the blue band is the bandwidth for the audio display (bottom right)

[Stub Mandrel]

12 minutes ago, The Admiral said:

Did you let the GB3VHF signal go through its full sequence, CW, clock signals, Morse, and frequency shift? The latter should show up as a stepped frequency signal like mine. You don't say if you can hear the sound.

How long does that take?

I have done something that seems to have made sound go very quiet...

[The Admiral]

18 minutes ago, Stub Mandrel said:

How long does that take?

I have done something that seems to have made sound go very quiet...

On the GB3VHF website there is this link to an audio file of what you should hear.: http://www.gb3vhf.co.uk/gb3vhf.wav

Can't help you with the quiet issue.

A bit odd that you get the same result on the wrong frequency!

OK, your current bandwidth looks a good enough starting point. I see your problem, I was using the Funcube. But there should be plenty of experience with using SL with other dongles.

Ian

[The Admiral]

Just to add to the above, here is a screenshot when tuned to 144.428MHz, just as it's starting it's FM mode.

 

Now here's a screenshot when tuned to 144.030MHz. There's nothing there but background radio noise!

There's quite a bit of background bubbling and burbling to be heard when tuned to these regions, from weak transmissions. Here's an audio clip of what I get in the 'barren' region:

ZOOM0004.WAV

Your display looks to be devoid of such background radio noise. So I can't help but think that you are not getting a proper signal into your device. Your very low background measurement would be consistent with this. What happens if you disconnect the antenna and shield the input socket?

Ian

[Tomatobro]

It took me quite a few attempts to get a station up and running so its a bit of a steep learning curve.

Three suggestions.

1. get the USB dongle close to the antenna and use a 10 mtr powered USB extension lead. This eliminates the losses of long Coax leads.

2. Use Airspy software for first system evaluations.

3. On the dongle setup screen ensure that the preamp is set to on and that the IF gain value is set to 25 or more.

Also it's worth remembering that the figure entered for the frequency will not be the actual frequency until the dongle frequency offset has been calibrated. The uncalibrated offset is not usually large but be aware of it.

The attached screenshot is from a new antenna under test for the first time.

 

[The Admiral]

9 minutes ago, Tomatobro said:

1. get the USB dongle close to the antenna and use a 10 mtr powered USB extension lead. This eliminates the losses of long Coax leads.

Useful comments , though I'm a bit surprised that you needed to resort to doing this, unless you had a very long run and/or you were using lossy co-ax. The following table from the WH Westlakes site shows typical losses, and unless one opted for the poorer cables the losses are pretty trivial. I've got a short run anyway, so it doesn't affect me, but I don't think Steve Nicholls does this, and his antenna is in his garden. Neil is using satellite cable which is intended for rather higher frequencies, so should be well capable of handling 144MHz.

Ian

[The Admiral]

Neil, I just thought I'd check out the meteor activity, and I didn't have to wait too long before I got a strike, and they are not infrequent. So you should be able to see if its working. Mind you, it follows the statistics of small numbers. Think of buses!

Note that I'm using a newer version of SDR Console, so the page looks a little different.

Ian

 

[Tomatobro]

I agree with Admiral's comments about coax runs but losses due to inappropriate coax selection can be a factor in why a systems performance is poor. Mounting the Dongle close to the antenna removes one of the variables. I think many give up because they cannot get the system to work which is a shame. A long run of RG58 is an absolute signal killer @ 144mhz. But 2 mtrs works just fine.

The good thing is these posts make folks aware of the issues which is what SGL is all about.

[Stub Mandrel]

Sorted!

I had a generic dll which was finding the dongle but not working it properly.

All my searches for places to find ExtIO_RTL2832U.dll had just turned up this other DLL.

I searched out a PDF that explained SDR from the ground up, and a section on HDSDR linked to ExtIO_RTL2832U.dll.

Lo and behold, I could then get radio 4 (very badly!) and then the beacon, amongst some noise.

I tuned to graves and almost immediately there was a sort of 'dong' sound and little skinny horizontal yellow tadpole about 9mm long started moving up the waterfall!

Of course nothing else in the next five minutes...

I have about 10-12m of coax, but I can easily reduce this to five metres and keep the USB  receiver indoors.

I think the fact I haven't got a proper connector isn't helping the signal - poking the centre conductor into a BNC adaptor and insulation taping the shield to the outside is going to reflect a lot of signal back up where it came from... but amazingly I cant find a rewireable BNC plug anywhere (I have ordered one from the bay). My other option is to find a satellite lead joiner and attach a short end of BNC cable as a stop gap.

I seem to have every form of coax connector or adaptor on the planet except the ones I need.

 

[Stub Mandrel]

1 hour ago, The Admiral said:

 

 

That's pretty much what I saw, only it was nearly horizontal ?

I think it must be working.