summaryrefslogtreecommitdiff
path: root/trasmitter/README.rst.txt
blob: e231527c8bf2b8233f54ac9dd64557169993c660 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
Some links:
 
* capturing radio signals
  http://www.stevenhale.co.uk/main/2013/08/home-automation-reverse-engineering-a-worcester-bosch-dt10rf-wireless-thermostat/
  https://damn.technology/controlling-british-gas-wr1-receiver-arduino
  http://rossharper.net/2015/11/decoding-a-siemens-rcr10433-thermostat-signal-to-control-a-boiler-from-a-raspberry-pi/
 
IMPORTANT: the transmitter prefers 3.3V, it stops working after a
while on 5V.
 
Reveng
======
 
Sampled at 192k samples/second, found that the transmissions had 3
types of (presumably square) pulses:
 
======== ======= ========
name     samples duration
======== ======= ========
narrow       200  ~ 1ms
middle       300  ~ 1.5ms
wide         400  ~ 2ms
======== ======= ========
 
The two sequences are:
 
``nnnnnnWnWnnMnnMnnnnnWnnnnMnMWn``
``nnnnnnWnWnnMnnMnnnnnnnnnnWMMWn``
 
Actually no, the pulses are not ½ high and ½ low, they're shaped
weird. The source code has the correct widths.