weather sensor: Plugin for RTL433
@axled Finally, I have found some time to come up with protocol implementation for pimatic-homeduino which I have tested with sample data you provided earlier. The code is on master now, but not yet released. The final commit is only contains the generated JS files
https://github.com/pimatic/rfcontroljs/commit/153b2f242184dbdc695178c21b10ecbc3e233a3c
https://github.com/pimatic/rfcontroljs/commit/38da6daf1baac8bcae37bf80fae372b6d5d2884b
"It always takes longer than you expect, even when you take into account Hofstadter's Law.", Hofstadter's Law
@mwittig
Thanks for your help, in the meantime i did some “development”.
Your code looks more professional than mine.
I found an issue with negative temperature, here is the code that worked for me:
if (helper.binaryToNumber(binary, 24, 27)==15){ //negativ
temperature = (helper.binaryToNumber(binary, 28, 35)-255)/10;
} else { //positive
temperature = helper.binaryToNumber(binary, 28, 35)/10;
}
Maybe this can be implemented before releasing “weather21” to master.
See my development attached (still little buggy and no clean code)
module.exports = function(helper) {
var protocolInfo, pulsesToBinaryMapping;
pulsesToBinaryMapping = {
/*'12': '0',
'22': '1',
'20': '1',
'23': ''*/
/*
'12': '0',
'13': '',
'20': '1',
'22': '1'
*/
//bessere Erkennung?
/*
'01': '0',//der Block geht nicht (sollte aber???)
'12': '1',
'111111': '111',//nach 01 und 12 setzen? (da sonst nicht eindeutig)
'13': '1'
*/
/*
'12': '0',//ab hier läufts stabil
'222222': '111',
'20': '1',
'13': '1'
*/
'12': '0',//ab hier läufts stabil - andere Reihenfolge
'20': '1',
'222222': '111',
'13': '1',
'03': '1',//wenn schlechte Übertragung
'21': '1',//wenn schlechte Übertragung
'02': '0',
'23': '1'//wenn schlechte Übertragung
};
return protocolInfo = {
name: 'weather97',
type: 'weather',
values: {
id: {
type: "number"
},
channel: {
type: "number"
},
temperature: {
type: "number"
},
humidity: {
type: "number"
},
/*debug: {
type: "string"
},*/
lowBattery: {
type: "boolean"
}
},
brands: ["Auriol"],
/*
Data layout:
X II HH F TTT CC
- X: preamble, sync
- I: id, 8 bit
- H: humidity, 8 bit
- F: flags, 4 bit (Batt, TX-Button, Chan, Chan)
- T: temperature, 12 bit, deg. C scale 10
- C: checksum, 8 bit
Pulse: 1222222220122012121212201212202012201220121212121212121220202012122012202020201212201213
mapped: 0111 1010 0001 0010 1110 0000 0000 1100 1110 0010 0001
XXXX IIII IIII HHHH HHHH FFFF TTTT TTTT TTTT CCCC CCCC
*/
//pulseLengths: [180, 308, 640, 60904],
//pulseLengths: [176, 308, 628, 60904],
pulseLengths: [192, 252, 640, 61372],
pulseCount: 88,
decodePulses: function(pulses) {
//var binary, h0, h1, humidity, lowBattery, result, t0, temperature;
var binary, temperature, lowBattery;
binary = helper.map(pulses, pulsesToBinaryMapping);
if (helper.binaryToNumber(binary, 8, 8)==0) {
lowBattery = false;
} else {
lowBattery = true;
}
//Formular for negative temperature
if (helper.binaryToNumber(binary, 24, 27)==15){ //negativ
temperature = (helper.binaryToNumber(binary, 28, 35)-255)/10;
} else { //positive
temperature = helper.binaryToNumber(binary, 28, 35)/10;
}
return result = {
id: helper.binaryToNumber(binary, 4, 11),
channel: helper.binaryToNumber(binary, 22, 23)+1,
//temperature: helper.binaryToNumber(binary, 28, 35)/10, //onlx last 8 bits because of negative value
temperature: temperature,
//debug: helper.binaryToNumber(binary, 24, 27),
//debug: debug,
humidity: helper.binaryToNumber(binary, 12, 19),
lowBattery: lowBattery
};
}
};
};
Is there something you need (further sample datas, etc)?
Thanks
AxLED
@axled said in weather sensor: Plugin for RTL433:
Is there something you need (further sample datas, etc)?
If you have some sample data for negative temperatures this will be great. Assuming temperature is 12-bits signed I think using helper.binaryToSignedNumber() should work.
I am a little bit concerned about the signal variations which is probably caused by a too weak or to strong signal. Are you still using the setup you had initially with sender and receiver side-by-side with the RBX8 in a 80% shielded box?
Regarding the mapping to binary data I’ll take your findings into account.Note, the mapping dictionary pulsesToBinaryMapping does not play well for overlapping mapping as the resulting of keys may differ from the order of mappings declared. This even varies depending on the version of Node.js used. For this reason I have extended the mapping helper function to be able to define an array of mapping objects.
EDIT: Did you see the 03 pulse only at the end of the sequence or did also occur in between. If you have recorded sample data please send as much as possible. I can add the sequences to the protocol test suite and this test all the variations
"It always takes longer than you expect, even when you take into account Hofstadter's Law.", Hofstadter's Law
@mwittig
Thanks for your response and your time.
I will send you some sample datas with negative temperature within the next days.
Setup was a little different. I used a shielded box and a RBX6 (A. directly connected to GPIO; B. with arduino nano clone), but as the sensor is blinking at transmission, i can assign which entries in log are coming from my Auriol sensor.
Recognition of datas by my code works in principle, but it is limited up to 6-8m with 1x Windows/Wall in between. And as you wrote, the raw datas differ a little, so i added some mappings under pulsesToBinaryMapping in my code.
Would it be helpful, if i complete sample raw datas as follows?
Thanks and regards
AxLED
@axled said in weather sensor: Plugin for RTL433:
Would it be helpful, if i complete sample raw datas as follows?
Temperature, Humidity, Batterystate, Channel
and
connection setup of RBX6 (direct GPIO vs. Arduino)
distance of RBX6 to sensor and whats in between?
Well, it would be nice to have this detailed view on test data if you have in place already. Otherwise, it sounds to me like a lot effort to prepare the data. If you have recorded pimatic-homeduino debug log data from earlier tests please send this as a starting point. I can add the pulse sequences to the protocol test suite to test the variations of pulses as much as possible.
"It always takes longer than you expect, even when you take into account Hofstadter's Law.", Hofstadter's Law
I have revised and improved the protocol implementation
Regarding checksums I noticed two out of the four sample you posted earlier have a bad checksum. Right now, the checksum result is just populated. In the future it may be used to ignore samples with bad checksum.
"It always takes longer than you expect, even when you take into account Hofstadter's Law.", Hofstadter's Law
@mwittig
As the outside temperature is still positive where i live, i changed my “lab setup”.
I put the sensor in the freezer, so i read the values from the correspondending original weather station base.
Hardware:
Raspberry Pi 1 B+
Arduino Nano V3 Clone
433Mhz receiver: RXB6 (powered by 5V)
Software:
OS (cat /etc/os-release):
PRETTY_NAME="Raspbian GNU/Linux 8 (jessie)"
NAME=“Raspbian GNU/Linux”, VERSION_ID=“8”, VERSION="8 (jessie)"
ID=raspbian, ID_LIKE=debian
Linux raspberrypi 4.4.50+
pimatic version: 0.9.54
pimatic-homeduino: 0.9.18
Node: 4.6.2
Generall Sensor Datas:
ID: 161
Battery: not empty
Channel: 1
=========================================================================
Location:
RaspberryPi to sensor: 4m, door of freezer inbetween
Temp: -0,9°C, Humidity: 60%
19:12:38.357 [pimatic-homeduino] debug: data: "RF receive 272 644 204 61312 0 0 0 0 0111111112011201010101120101121212120101010101011212121012121212011210010110011001011003"
19:12:38.384 [pimatic-homeduino] debug: received: [ 204, 272, 644, 61312 ] 1222222220122012121212201212202020201212121212122020202120202020122021121221122112122113
Temp: -1,8°C, Humidity: 58%
19:14:10.684 [pimatic-homeduino] debug: data: "RF receive 272 644 200 60948 0 0 0 0 0111111112011201010101120101121212011201010101011210121210121201121201100112011210100113"
19:14:10.706 [pimatic-homeduino] debug: received: [ 200, 272, 644, 60948 ] 1222222220122012121212201212202020122012121212122021202021202012202012211220122021211223
Temp: -2,6°C, Humidity: 56%
19:15:43.667 [pimatic-homeduino] debug: data: "RF receive 272 640 200 61316 00 0 0111111112011201010101120101121212010101010101011212121212121001011001101010101101003"
19:15:43.691 [pimatic-homeduino] debug: received: [ 200, 272, 640, 61316 ] 1222220122012121212201212202020121212121212122020202020202112122112212112121212113
Termp: -3,8°C, Humidity: 55%
19:18:17.562 [pimatic-homeduino] debug: data: "RF receive 276 644 208 61332 0 0 0 0 0111111112011201010101120101121201121212010101011212121212120112120101120101011001010103"
19:18:17.588 [pimatic-homeduino] debug: received: [ 208, 276, 644, 61332 ] 1222222220122012121212201212202012202020121212122020202020201220201212201212122112121213
=========================================================================
Location:
RaspberryPi to sensor: 7m, nothing inbetween
Temp: +1,3°C, Humidity: 80%
19:45:10.666 [pimatic-homeduino] debug: data: "RF receive 268 632 192 61328 0 0 0 0 0111111112011201010101120112011201010101010101010101010101010101101001101010011001011003"
19:45:10.686 [pimatic-homeduino] debug: received: [ 192, 268, 632, 61328 ] 1222222220122012121212201220122012121212121212121212121212121212212112212121122112122113
Temp: +2,9°C, Humidity: 81%
19:47:14.629 [pimatic-homeduino] debug: data: "RF receive 272 632 196 61316 0 0 0 0 0111111112011201010101120112011201010112010101010101010101010110101201101001100101010103"
19:47:14.646 [pimatic-homeduino] debug: received: [ 196, 272, 632, 61316 ] 1222222220122012121212201220122012121220121212121212121212121221212012212112211212121213
=========================================================================
Location:
RaspberryPi to sensor: 9,5m, nothing inbetween
Temp: +5,3°C, Humidity: 83%
19:50:20.703 [pimatic-homeduino] debug: data: "RF receive 252 636 196 60952 0 0 0 0 0111111112011001010101100110011001011010010101010101010101011010011001101010101010100113"
19:50:20.729 [pimatic-homeduino] debug: received: [ 196, 252, 636, 60952 ] 1222222220122112121212211221122112122121121212121212121212122121122112212121212121211223
Further information: I coud be the weather station is rounding in ± 0,1°C steps. As i dont see through the freezer a can’t tell you the temperatur the sensor display itsself shows.
AxLED
@axled said in weather sensor: Plugin for RTL433:
As the outside temperature is still positive where i live, i changed my “lab setup”.
I put the sensor in the freezer, so i read the values from the corresponding original weather station base.
Excellent. Thank you very much for providing the detailed test data which is very helpful. I quickly test the first three values. Negative temperature seem to rendered correctly, but the values decoded are 0,1°C lower than the indicated values which is probably OK as you have indicated. CRC are a bit of a mixed bag. Every second sample seem to fail even though the data looks ok. It is probably a glitch in the calculation. The guy who implemented for rtl_433 was also struggling with this. Anyhow, CRC checks are just informative right now and I may even strip them off the code in the end.
I’ll be travelling tomorrow and most of next week and I won’t be able to do much work on pimatic before 2nd week in February.
"It always takes longer than you expect, even when you take into account Hofstadter's Law.", Hofstadter's Law
@axled Finally, I have found some time to come up with protocol implementation for pimatic-homeduino which I have tested with sample data you provided earlier. The code is on master now, but not yet released. The final commit is only contains the generated JS files
https://github.com/pimatic/rfcontroljs/commit/153b2f242184dbdc695178c21b10ecbc3e233a3c
https://github.com/pimatic/rfcontroljs/commit/38da6daf1baac8bcae37bf80fae372b6d5d2884b
is there a timeline, when the protocol will be implemented?
Regards
AxLED
