433MHz weather-station antenna - need something better than helical

[floobydust]

I'm trying to read 433MHz OOK/ASK data from a Radio Shack remote temperature transmitter.

I've tried a bunch of chinese receiver modules, best so far is superhet with a 10.7MHz IF and crystal filter, uses PT4303 IC in a module called "SRX-882".

It has 32mm helical spring antenna which works OK but the noise floor gets terrible once I am near a PC or connect a logic analyzer to it. Ferrite beads and clamp-ons help to a point- but I think I need more signal. The spring antenna is rated 2.15dBi

The battery-powered Radio Shack LCD display (receiver) uses a dipole antenna inside.
But the PT4303 is single-ended. I didn't think a dipole works when ground leg has a bunch of stuff connected to it? i.e. a PC.  I could make a balun.

Just looking for a few more dB gain antenna that isn't too hard to build so I can analyze the data stream.

[MiDi]

Did you tried CC1101 or SX1231 (RFM69)?
If you have only one device to receive from, then a simple to build and good antenna is a bi-quad, but it is strongly directional.

[cdev]

Bi quad might work but be physically much larger than one designed for wifi bands, which are compact.

Another option would be a little yagi antenna with a gamma match.. How far away are we talking about here?

You could scope out the situation and get a better visual idea of what the signals are dealing with - with an RTLSDR.

There is rtl_433 software which demodulates all sorts of signals, plus you can use a generic SDR program like CubicSDR or SDRsharp HDSDR, etc, to see the traffic - very useful for antenna adjustment.

The cheap "RTL2832U+R820T" DVB-T TV dongles you can get on ebay for ~$10- pair them with the open source osmocom.org driver software.

[floobydust]

It's about 60 feet distance and through a few stucco walls. The problem is not so much range, it's increasing S/N so the PC or test gear doesn't drown it all out.

I did look at the CC1101 but you really need to know the TX packet format to configure the demodulator and decoder. It might be easier with their SmartRF Studio software, but still over 60 registers to program.

Here, the Radio Shack protocol is unknown. I've tried decoding using Oregon Scientific or Thermor but no luck yet.
I did see a slim Jim and J-pole antenna but never seen them before.
The bi-quad antenna looks awkward at 18.5" across? I wasn't sure how to support the structure. https://www.changpuak.ch/electronics/bi_quad_antenna_designer.php
Where is the reflector for this, it got me confused.

edit: found vid

[cdev]

Did you try rtl_433? ( https://github.com/merbanan/rtl_433 )

Id be surprised if it didn't have some clues it could offer. It has a mode where it tries to help you figure out unknown formats.

A j-pole or a variant of it might be the best omnidirectional option. You could make a simple J-pole out of 300 ohm twin lead and enclose it inside of a thin walled piece of PVC pipe (it has to be thin walled or you'll have too much loss) Or hang it indoors from a hook on the ceiling. (not right against the wall) If you can find a continuous signal you can adjust the gamma match (sliding where the coax connects to where you get the maximum received signal strength) and then solder it there.

[MiDi]

Sorry, I meant CC1101, CC1151 is TX only, corrected my post.
There are ready to use sticks (e.g. cul stick, moteino, jeelink) and couple of libraries and examples to kickstart.

FHEM has build in many modules to decode this type of stuff.

But you are right, if you do not know bitrate and bandwidth it is hard to go.
For debugging and decoding rtlsdr might be a good starting point, but it is not so good for continuous logging.

Edit: if you have access to transmitter you could hook up a logic analyzer there to figure out protocol.

[cdev]

I would be surprised if rtl_433 didn't decode a Radio Shack sensor, which by simple virtue of the source, is likely to be common.

There is also a program called URH that is designed to help in demodulating unknown protocols.