• 10 dec 2017: forum version update. In case of issues use this topic.
  • 30 nov 2017: pilight moved servers. In case of issues use this topic.
Hello There, Guest! Login Register


[Partially Supported] TFA / Conrad Weather
#31
Ok i will do new measurements directly from sensor itself instead of using the main unit

New values (directly read on sensor itself), format if
Temp : Col5_4_3 (full raw) Decimal value

Code:
21.4 : 011001000110 (0010110001001001100100011000110011100110007582) 1606
21.5 : 011001000111 (0010110001001001110100011000100011100110117582) 1607
21.5 : 011001000111 (0010110001001001110100011000100011100110117514) 1607
21.6 : 011001001000 (0010110001001010000100011000100011100111117582) 1608
21.7 : 011001001011 (0010110001001010110100011000100011100101017582) 1611
21.9 : 011001001111 (0010110001001011110100011000010011100100017)    1615
22.1 : 011001010010 (0010110001001000100101011000010011100110117582) 1618
22.2 : 011001010011 (0010110001001000110101011000010011100111017548) 1619
22.2 : 011001010100 (0010110001001001000101011000010011100111007582) 1620
22.2 : 011001010100 (0010110001001001000101011000010011100111007582) 1620
22.2 : 011001010100 (0010110001001001000101011000010011100111007582) 1620
24.4 : 011001010111 (0010110001011001110101011010100011100110017514) 1623
22.5 : 011001011001 (0010110001011010010101011010100011100110117582) 1625
22.6 : 011001011010 (0010110001011010100101011001110011100111017582) 1626
22.6 : 011001011010 (0010110001101010100101011001010011100101107)    1626
22.7 : 011001011100 (0010110001101011000101011001000011100101007548) 1628
22.7 : 011001011100 (0010110001101011000101011000110011100111007582) 1628
22.8 : 011001011110 (0010110001101011100101011000100011100101017)    1630
23.1 : 011001100100 (0010110001101001000110011000110011100111117548) 1636
23.2 : 011001100110 (0010110001101001100110011000110011100100117548) 1638
23.3 : 011001101000 (0010110001100010000110011000100011100100117)    1640

I still think that what is transmitted and what is displayed is not exactly in sync (look at 22.2°)
 
Reply
#32
My guess would be

~ -70 + (3 * col3) + (1.5 * col4) + (0.1 * col5)

The correct starting value of -70 is of course what we are looking for.
 
Reply
#33
Ok i have nothing to make a test for -70° (now looking for liquid nitrogen :-) )
 
Reply
#34
The sensors don't have a range until -70, only -20. But i think the theoretical lowest value of the protocol is around that value.
 
Reply
#35
Yves, you need "Dry ice" to reach those temperatures, is possible :-)

Website link:Dry ice

Greetings,

Mike
 
Reply
#36
I would go for a "dry ice bomb" then :-)
 
Reply
#37
ok so the formula : ~ -70 + (3 * col3) + (1.5 * col4) + (0.1 * col5) is not working and i don't think that it will work (including with other values than -70).
so still stuck to decrypt the temperature...
 
Reply
#38
Why not?
 
Reply
#39
Blush here the solution (with temp now!!!)

T17.8 H64 :0000001111001001000000011000000100000000117582

0000001111 : Station ID

00 : don't know/care

10 : channel (00 means Ch 1, 01 Ch 2, 10 Ch3)

0100 0000 0110 : temperature is sent by the sensor in °F (and not °C), the lowest value that the protocol support is 90°F (000000000000). Again you have to swap (321) the order of the column so in our example it give us 0110 0000 0100 which is 1540 in decimal. So the rule is 1540/10 - 90 = 64 °F (17.78 °C) (this rule works fine for all temp that i tested positive/negative)

0000 0100 : humidity swap the 2 col and it give the bin value of the HUM (ie 100 000 = 64)

0000 : battery, if > 0100 then BAT is failing

0011 : cksum ?? don't know/care

Another important point : I have 3 sensors all of them emit different pulses:
Code:
Station 1
494 1976 494 1976 494 1976 494 4199 494 4199 494 2223 494 4199 494 4199 741 1976 494 4199 494 4199 494 1976 494 1976 494 2223 494 1976 494 4199 494 4199 494 2223 494 1976 494 1976 494 1976 494 2223 494 1976 494 4199 494 4199 494 2223 494 1976 494 1976 494 4199 494 4199 494 1976 494 1976 494 4199 494 4199 494 1976 494 1976 494 4199 494 2223 494 1976 494 1976 494 1976 494 4446 494 8398

Station 2
464 2088 464 2088 464 3944 464 3944 464 2088 464 4176 464 1856 464 2088 464 2088 464 2088 464 2088 464 2088 464 1856 464 4176 464 3944 464 3944 464 3944 464 2088 464 3944 464 3944 464 3944 464 4176 464 2088 464 3944 464 1856 464 4176 464 2088 464 3944 464 2088 464 4176 464 2088 464 2088 464 3944 464 4176 464 2088 464 2088 464 2088 464 2088 464 2088 464 3944 464 2088 464 4176 464 7888

Station 3
669 2007 669 1784 669 3791 669 2007 669 3791 669 3791 669 1784 669 1784 669 1784 669 3791 669 2007 669 3791 669 3791 669 2007 669 3791 669 1784 669 3791 669 2007 669 2007 669 2007 669 1784 669 3791 669 1784 669 3791 669 3791 669 2007 669 3791 669 3791 669 2007 669 2007 669 2007 669 2007 669 3791 669 2007 669 1784 669 2007 669 3791 669 1784 669 3791 669 3791 669 2007 669 2007 669 7582

A number on 3 digits is the repeating schema, and after :
if number < 2500 : it's a 0
elsif number < 6000 : it's a 1
else number is the high pulse to define the end of the message
 
Reply
#40
We did a good job here Dodgy. So i was right about the absolute minimal value Smile. Also though about Fahrenheit but didn't check that in the last step.

2 comments:
1. The first yellow column is part of the ID.
2. The last yellow column is the checksum. Going to look for that logic as well Smile

Can you try to change this function in debug.c:
Code:
int normalize(int i) {
    double x;
    x=(double)i/pulselen;

    return (int)(round(x));
}
To this:
Code:
int normalize(int i) {
    return i;
}
This disables normalization and gives you the raw pulses. Then post a pilight-debug output for all three sensors.

Creating a protocol should be easy now:
http://www.pilight.org/development/protocols/

Check the alecto.c and threecan.c as a reference.

The cool thing is that i believe this device also emits the same protocol:
http://www.conrad.nl/ce/nl/product/39324...archDetail
 
Reply
  


Possibly Related Threads...
Thread Author Replies Views Last Post
  Bresser 3Ch weather sensors Harald654 13 4,091 02-15-2019, 01:06 PM
Last Post: saak2820
  [Fully Supported] DHT22 IcedEarth 86 30,142 11-18-2018, 09:33 AM
Last Post: curlymo
  Weather Station Globaltronics GT-WT-01 Prutsky 13 6,582 04-09-2018, 07:34 PM
Last Post: NevelS
  [Fully Supported] Arctech Old Martin 111 33,759 03-27-2018, 04:43 PM
Last Post: genfersee
  [Fully Supported] Clarus Switches Marcin 69 23,092 01-30-2018, 07:10 PM
Last Post: Niek
  [Fully Supported] Remote Control Socket (RC101-U/RC201) Sean 18 9,942 01-04-2018, 06:18 AM
Last Post: ettman8
  [Fully Supported] LM75 and LM76 temperature sensor horst_dieter 64 24,195 11-19-2017, 08:54 PM
Last Post: edepi
  Weather Underground "wunderground" humidity wimThoelke 6 1,307 10-23-2017, 09:44 PM
Last Post: wimThoelke
  [Fully Supported] No-brand temp/humidity sensor (alecto_ws1700) meloen 57 33,952 12-25-2016, 09:53 PM
Last Post: creamers
Lightbulb [Fully Supported] Kaku Door sensor (AMST-606) geerttttt 50 23,213 11-29-2016, 06:52 AM
Last Post: curlymo

Forum Jump:


Browsing: 1 Guest(s)