Receiving no signals!

[t...@quarendon.net]

So I've been playing with a realtek SDR dongle, with the aim of picking up some 433MHz signals with rtl_422, but am having trouble receiving anything at all.

So I've got a Foxnovo USB realtek  RTL2832U+R820T stick. One of these: http://www.amazon.co.uk/Foxnovo%C2%AE-RTL-SDR-Receiver-Realtek-RTL2832U/dp/B00FF32B5E

I'm running on Windows 10.

I can run rtl_power from the command line and get no errors, and data comes out, so I believe that the dongle and the driver are working at some level.

I can run rtl panorama (https://sourceforge.net/projects/guiforrtlpower/), and have used the "zadig" tool to get the right driver installed. 

When I run rtl panorama, and make it scan the normal FM radio range, and I get nothing, just a (basically) flat line (wavy, but I don't get any strong peaks as I'd expect). See the attached spectrum.

Interestingly I unplug the aerial from the dongle, and nothing changes!

I have the dongle directly into the computer, and the aerial (the one that came with the dongle) sitting on the windowsill.

Any ideas what I'm doing wrong? The fact that I can't pick up FM radio, and that it appears to make no difference whether I have the aerial plugged in or not seems to indicate something is bad, either in my brain (i.e I'm doing something wrong), or with the hardware?

Anyway, any help with this would be great.

Thanks!

[t...@quarendon.net]

Interestingly I changed the gain to the maximum of 49.6, and I now get peaks roughly where I might expect for the common FM radio signals in the area, so maybe the issue is that the default gain of 0 is meaning it doesn't receive anything. I don't really understand what gain is and why I wouldn't always want it set to maximum.

However, I run rtl_433 and I still don't appear to get any signals from my Apollo Watchman, but I don't actually know how frequent they are. Evidence elsewhere seemed to suggest roughly half hourly. 

It's possible I need a better source example, something that I can easily signal on demand. 

Not convinced that I'm fully working yet though, until I get a signal through from something.

[David Woodhouse]

On Fri, 2016-04-08 at 09:33 -0700, t...@quarendon.net wrote:
>
> However, I run rtl_433 and I still don't appear to get any signals
> from my Apollo Watchman, but I don't actually know how frequent they
> are. Evidence elsewhere seemed to suggest roughly half hourly. 
>
> It's possible I need a better source example, something that I can
> easily signal on demand. 

Stick a magnet to the dot on the watchman, and it'll send about three
readings a second. Bring it indoors and sit it near the receiver.

We've done a fair amount of work on improving the long-distance
reception, but perhaps you just aren't getting the signal from the
device when it's sat on the oil tank outside?

--
dwmw2

[t...@quarendon.net]

Stick a magnet to the dot on the watchman, and it'll send about three
readings a second. Bring it indoors and sit it near the receiver.

We've done a fair amount of work on improving the long-distance
reception, but perhaps you just aren't getting the signal from the
device when it's sat on the oil tank outside?

I have tried bringing it inside (not for long, the smell of oil permeates the house!) and putting the sender next to the aerial, and doing the thing with the magnet, but I didn't get any luck last time.

I'll try that again later, and see if I get anything. 

I've just found an old Efergy e2 that also seems to be supported, so I'll try that as another source, though nothing so far.

Thanks for the suggestion.

[t...@quarendon.net]

I've just found an old Efergy e2 that also seems to be supported, so I'll try that as another source, though nothing so far.

Thanks for the suggestion.

OK, so with the Efergy e2, with  rtl_433 set to a frequency of 433.5, and with -a on, I get output such as

*** signal_start = 16332122, signal_end = 16402752

signal_len = 70630,  pulses = 1

Distance coding: Pulse length 50629

Short distance: 1000000, long distance: 0, packet distance: 0

p_limit: 50629

bitbuffer:: Number of rows: 0

The output corresponds to the signalling pulse light on the Efergy sender unit. So some success.

However, if I use the -R 36 option to decode it, or indeed no arguments at all, I get nothing. 

I also get nothing if I leave the frequency at default, it has to be set to 433.5Mhz, which is what the spec for the Efergy says is broadcasts on.

So I'm getting something though.

My suspicion therefore is that my frequency might be wrong for the Apollo? The spec sheet (http://www.tekelekgroup.com/datasheets/DS-5029-01%20Tekelek-datasheet-TEK608.pdf) says it's 433.92, which is the default though.

I'll try again bringing the Apollo sender inside and setting it next to the aerial and see if I can get something.

[t...@quarendon.net]

with  rtl_433 set to a frequency of 433.5

So this raises a question for me. Assuming that I could make rtl_433 decode the Efergy, would I be able to read from the Apollo tank sender, and the Efergy, at the same time? My suspicion is not, as they are on different frequencies.

So I'd need TWO dongles?

[t...@quarendon.net]

I have tried bringing it inside (not for long, the smell of oil permeates the house!) and putting the sender next to the aerial, and doing the thing with the magnet, but I didn't get any luck last time.

I'll try that again later, and see if I get anything. 

OK, just brought the sender inside and put it next to the aerial and used a magnet on it. 

If I run rtl_433 -a, then I get a bunch of output, so something is being picked up.

If I run just rtl_433, or rtl_433 -R 42, I get nothing, presumably because it's failing to decode the signal in some way.

Sounds like the next thing I would need to do is to record some files with "rtl433 -a -t" and post them somewhere?

[David Woodhouse]

On Fri, 2016-04-08 at 11:20 -0700, t...@quarendon.net wrote:
> OK, just brought the sender inside and put it next to the aerial and
> used a magnet on it. 
> If I run rtl_433 -a, then I get a bunch of output, so something is
> being picked up.
>
> If I run just rtl_433, or rtl_433 -R 42, I get nothing, presumably
> because it's failing to decode the signal in some way.
>
> Sounds like the next thing I would need to do is to record some files
> with "rtl433 -a -t" and post them somewhere?

Indeed.

--
dwmw2

[t...@quarendon.net]

Indeed.

OK, now I'm confused.

Trying to capture some data, and I'm now getting loads of signals, where previously I was getting none (at least I thought I was, not so sure now), much too many to be just from the Apollo sender. 

Using rtl_433 with no arguments still doesn't recognise anything, so whatever these signals are, they aren't anything that rtl_433 currently recognises. 

There must be a new source of signals that didn't previously exist. I think I'll have to test again tomorrow.

How do I isolate things? How do I ensure that the signals the aerial is receiving are just from the Apollo sender? Put the sender and the aerial in something to act as a Faraday cage? I just naively put the aerial inside a metal bread bin that was just lying around in the kitchen, but it still picked up signals. So now I'm really confused. 

Any help welcome!

Thanks.

[t...@quarendon.net]

Right. I've pushed some data to the test repo. tests/oil_watchman/06

Don't know what was happening last night. Something must have been broadcasting lots of signals. All quiet this morning.

Let me know what I can do to try and diagnose these traces. I have no idea where to start!

Thanks.

[t...@quarendon.net]

Whatever data I've got in the files that I thought came from the Apollo watchman doesn't seem to look like the other test data. 

I'm getting 86 bit messages (actually one is 87), and looking at the bits, then do not appear to be Manchester encoded (there isn't a long sequence of pairs of bits with different values).

So either what I've got isn't data from an Apollo watchman, or this is a new form of protocol.

I'll have to collect some more data I think, leave it running for a while to collect some more data points.