Continuing from Part 1 and Part 2, this is the third and last post about how I reverse engineered a few off-the-shelf wireless temperature, humidity, and rain sensors, and used an Arduino to listen to and decode the sensor data. Update: RPi is also supported now! Check the provided programs at the end of this post.

Wireless Rain Sensor

Raw Waveform. There are several different wireless rain sensors on the market. My first target was an Acu-Rite 00875W rain gauge. I bought it two years ago and I don’t know if it’s still available now. This sensor turns out to be a huge pain. The main problem is that I can’t even register a consistent reading at 0 — every time I pop in the battery, I get a different signal that seems to have nothing to do with the previous readings. Also, the 0/1 bit patterns are completely unclear from the captured waveform — there are at least 4 or 5 different wavelengths. Here are two examples of the captured waveforms. They look nothing alike, even though both were captured right after the batteries are popped in and no rain has been detected.

After pulling my hair for a couple of days and finding no clue at all, I decided to give it up and try a different model. This time I bought an Acu-Rite 00899 wireless rain gauge. It’s probably worth explaining at first how the rain gauge works, because it’s quite a clever design. The outside of the transmitter unit looks like a bucket. Underneath the bucket is a plastic seesaw which swings left and right. Basically the rain water drains through the bucket hole onto the seesaw, and creates some motion to be detected.

At the bottom of the assembly is the battery compartment. Remove some tiny screws, the transmitter circuit is finally revealed. At the center is a reed switch, which is normally open but will close if there is a magnet nearby. It is very sensitive to magnetism. So where is the magnet? It’s on the bottom of the seesaw. The way this works is quite neat: the seesaw swings left and right, every time the magnet passes by the reed switch, it triggers a click. By detecting how many clicks there are within a given time, we know how heavy the rain is. Clever!

So let’s see if this rain gauge is easier to tackle. Following the same procedure as before, pop in the battery, power on the RF sniffing circuit, and launch the Audacity recording software, I got a waveform like the one shown in the image below.

Cool, this looks a lot better. I also made sure every time I pop in the battery I get the same waveform. So what patterns do I see here?

• Each transmission consists of 3 repetitions of the same signal.
• Every two repetitions are separated by a sync signal defined as 4 squarewaves of roughly 1.2ms wavelength.
• The bit patterns are: logic 1 is defined by a constant high of 400us followed by a constant low of 200us; and logic 0 is defined by a constant high of 225us followed by a constant low of 400us.

The sync signal is actually very similar to the humidity/temperature sensor I described in the Part 2. The difference is that there is no 2.25ms constant low preceding the 4 squarewaves.

Given the timing data, here is the Arduino program to convert this signal into bits:

• Download the Arduino Program for Sniffing Bits (rain gauge version)

Collect and Analyze Data. The transmitter sends a signal every two minutes, so it’s quite annoying that I had to wait two minutes for every reading. To create variations in data, I simply move the seesaw manually. This way I can precisely control how many clicks occurred between two readings. Here is a list of readings I captured. The numbers in the parentheses are the number of clicks since last reading, and the total number of clicks since the program started.

01110010 11111000 11110000 00000000 00000000 00000000 00000000 01011010 (0 / 0 total)
01110010 11111000 11110000 00000000 00000000 00000000 10000001 11011011 (1 / 1 total)
01110010 11111000 11110000 00000000 00000000 00000000 10000010 11011100 (1 / 2 total)
01110010 11111000 11110000 00000000 00000000 00000000 00000011 01011101 (1 / 3 total)
01110010 11111000 11110000 00000000 00000000 00000000 10000100 11011110 (1 / 4 total)
01110010 11111000 11110000 00000000 00000000 00000000 10000111 11100001 (3 / 7 total)
01110010 11111000 11110000 00000000 00000000 00000000 10001101 11100111 (6 / 13 total)

The encoding scheme is quite obvious: the first three bytes are the signature / device ID; the next 4 bytes record the total number of clicks since the transmitter is powered on. The leading bit of each byte is a parity bit (thanks to the lesson I learned from Part 2!) The last byte is for error checking.

This time I looked at the error checking byte more carefully, and noticed some interesting patterns. For example, everything else being the same, the change in its first bit matches the change of the first bit in the second to last byte. This suggest that perhaps the last byte is a parity byte — specifically, bit 0 is the parity of the first bits in all the preceding 7 bytes, and bit 1 is the parity of the second bit in all the preceding 7 bytes and so on. Just eyeballing the numbers, I believe this looks correct.

I felt quite happy that I made the right choice to abandon the first rain gauge which proved to be too difficult to solve. Well, dodging the challenge is not generally recommended, but in this particular case, I have no regret 🙂

Arduino Program and Validation. Here is the Arduino program that listens to the rain gauge and displays the number of clicks onto the serial monitor. The numbers have been validated with the display unit.

• Download the Arduino Program for Wireless Rain Gauge

Update: the code is adapted to RPi as well, using wiringPi. The code below uses wiringPi GPIO 2 (P1.13) for data pin.

• Download the Raspberry Pi Program for Wireless Rain Gauge

This concludes the three-part series. If you have comments / questions / suggestions, please feel free to leave them below. Thanks!

New: continue to Part 4, wireless soil moisture sensor.

Continuing from Part 1, this is the second post about how I reverse engineered a few off-the-shelf wireless temperature, humidity, and rain sensors, and used an Arduino to listen to and decode the sensor data. Update: RPi is also supported now! Check the provided programs at the end of this post.

Wireless Humidity / Temperature Sensor

Raw Waveform. The second sensor to tackle is the Acu-Rite 00592W3 humidity / temperature sensor. This one transmits not only temperature but also humidity values. The display unit is a lot larger and looks cooler than the previous (temperature-only) one. Cool, time to get hands dirty. Following the same procedure as before, pop in the battery, power on the RF sniffing circuit, and launch the Audacity recording software, I got a waveform like the one shown on the right image below.

This looks a whole lot longer than before. But the patterns are still quite clear:

• Each transmission consists of 3 repetitions of the same signal.
• Every two repetitions are separated by a sync signal defined as a constant low of 2.25ms (about 100 samples @ 44.1kHz sample rate). This is followed by 4 squarewaves of roughly 1.2ms wavelength.
• The bit patterns are also clear as there are only two types of patterns: logic 1 is defined by a constant high of 400us followed by a constant low of 225us; and logic 0 is defined by a constant high of 250us followed by a constant low of 400us.

Given the timing data, I quickly modified the Arduino program to convert this signal into bits. The only changes are how the sync signal is detected, and how the bit 1 and 0’s are defined.

• Download the Arduino Program for Sniffing Bits (humidity/temperature sensor version)

Collect and Analyze Data. I collected a few groups of data under different humidity and temperature conditions, and recorded the reference values shown on th

• Download the Arduino Program for Wireless Rain Gauge

e display unit. Given my experience with the first temperature sensor, I set the display unit to show temperature in Celsius, to save myself the trouble of doing the conversion. Here is a selected list of the data:

10010011 00000010 01000100 00100010 00001001 10111110 11000010 (21°C/34%)
10010011 00000010 01000100 00100010 00001001 11000101 11001001 (22°C/34%)
10010011 00000010 01000100 10100011 00001001 11011000 01011101 (24°C/35%)
10010011 00000010 01000100 00100010 00001001 01100101 01101001 (25°C/34%)
10010011 00000010 01000100 00100001 00001001 01101100 01101111 (26°C/33%)
10010011 00000010 01000100 10100000 00001001 11110011 01110101 (26°C/32%)
10010011 00000010 01000100 10100000 00001001 11111001 01111011 (27°C/32%)
10010011 00000010 01000100 10011111 00001010 10000001 00000011 (28°C/31%)
10010011 00000010 01000100 00011110 00001010 10010011 10010100 (29°C/30%)
10010011 00000010 01000100 00010100 00001010 11001001 11000000 (35°C/20%)
10010011 00000010 01000100 10010000 00001010 11010111 01001010 (36°C/16%)

Again, I grouped the data into bits of 8 to make it easy to see the byte values. The first three bytes are the same. Those are probably the signature / channel ID; the fourth byte is clearly correlated with the humidity — if you pick two lines with the same humidity value, the fourth byte is always the same. On close examination, the lowest 7 bits of that byte converts exactly to the decimal value of the humidity. To make it clear, I picked out the fourth byte, and wrote them down together with the reference humidity below:
0 0100010 (34%)
0 0100010 (34%)
1 0100011 (35%)
0 0100010 (34%)
0 0100001 (33%)
1 0100000 (32%)
1 0100000 (32%)
1 0011111 (31%)
0 0011110 (30%)
0 0010100 (20%)
1 0010000 (16%)
Aha, one puzzle is now solved. Not sure what the leading bit is (this will be revealed later), but we can move on to the next puzzle — temperature. Clearly the temperature has to do with only the last three bytes. So let me re-list them below with the reference temperature:
00001001 10111110 11000010 (21°C)
00001001 11000101 11001001 (22°C)
00001001 11011000 01011101 (24°C)
00001001 01100101 01101001 (25°C)
00001001 01101100 01101111 (26°C)
00001001 11110011 01110101 (26°C)
00001001 11111001 01111011 (27°C)
00001010 10000001 00000011 (28°C)
00001010 10010011 10010100 (29°C)
00001010 11001001 11000000 (35°C)
00001010 11010111 01001010 (36°C)
Because the data was recorded as the temperature went up, the first byte is clearly showing the right trend: it goes up as the temperature rises. These are probably the most significant 4 bits. The next byte is not very clear at all: sometimes it goes up, sometimes it goes down. What’s happening here? The last byte is even more elusive: but given my experience with the first temperature sensor, this is probably some sort of error checking code, so let me put it aside for now.

To figure out what’s happening with the second byte, I needed more data. Using my old friends hair blower and fridge, I recorded a larger set of data, and here they are:
00000110 11101011 11011100 (-13°C)
00000110 11111001 11110000 (-12°C)
-------- -------- -------- --------
10000111 00000000 01111000 (-11°C)
10000111 00110101 10110010 (-6°C)
10000111 11010001 01001110 (-3°C)
10000111 11011011 01011000 (-2°C)
10000111 01011111 11011100 (-1°C)
10000111 01101111 11101101 ( 0°C)
10000111 11111001 11111000 ( 1°C)
-------- -------- -------- --------
10001000 00000011 00000011 ( 2°C)
10001000 00010100 10010011 ( 4°C)
10001000 01000001 00111110 ( 8°C)
10001000 11110011 11100100 (13°C)
10001000 01111101 01101110 (14°C)
-------- -------- -------- --------
00001001 00000110 01111000 (15°C)
00001001 00001111 10000001 (16°C)
00001001 00100010 10010100 (18°C)
00001001 11011110 01011111 (24°C)
00001001 11111100 11111001 (27°C)
-------- -------- -------- --------
00001010 00000011 00000001 (28°C)
00001010 00001111 00001101 (29°C)
00001010 11010001 01000100 (36°C)
00001010 01100110 11011001 (38°C)
00001010 11111111 01110010 (40°C)
-------- -------- -------- --------
10001011 10001110 10000010 (42°C)
10001011 00011101 00010001 (43°C)
10001011 01010011 01000111 (49°C)
10001011 01111000 01101100 (52°C)
-------- -------- -------- --------
00001100 10000001 11110110 (53°C)

The second byte still doesn’t show any clear trend; but the first byte is still well behaved: it consistently increments as the temperature rises. That’s re-assuring. I then noticed something very interesting, and that’s why I used dotted lines to separate the data: each dotted line marks where the first byte increments by 1. If you look at the temperature where these changes occur, there is an interesting pattern: the change occurs roughly every 12 degree Celsius.

From my experience with the first temperature sensor, I know that it’s likely the data reflects 10 times the Celsius, so that means the change occurs at about every 120. This is very close to 128. If this is the case, then the lowest bit of the first byte must be the 8-th bit of the actual data (that’s how one bit of increment corresponds to a change of 128), and thus there must be 7 least significant bits following it! Could that be from the second byte?

After staring at the data for a while, I see a path now: if you put together the lowest 4 bits of the first byte and the lowest 7 bits of the second byte, into a 11-bit binary number, this number consistently rises with the temperature. Aha, let me wrote them down here:
0110 1101011 (-13°C) -> 875 => -14.9
0110 1111001 (-12°C) -> 889 => -13.5
0111 0000000 (-11°C) -> 896 => -12.8
0111 0110101 (-6°C) -> 949 => -7.5
0111 1010001 (-3°C) -> 977 => -4.7
0111 1011011 (-2°C) -> 987 => -3.7
0111 1011111 (-1°C) -> 991 => -3.3
0111 1101111 ( 0°C) -> 1007 => -1.7
0111 1111001 ( 1°C) -> 1017 => -0.7
1000 0000011 ( 2°C) -> 1027 => 0.3
1000 0010100 ( 4°C) -> 1044 => 2
1000 1000001 ( 8°C) -> 1089 => 6.5
1000 1110011 (13°C) -> 1139 => 11.5
1000 1111101 (14°C) -> 1149 => 12.5
1001 0000110 (15°C) -> 1158 => 13.4
1001 0001111 (16°C) -> 1167 => 14.3
1001 0100010 (18°C) -> 1186 => 16.2
1001 1011110 (24°C) -> 1246 => 22.2
1001 1111100 (27°C) -> 1276 => 25.2
1010 0000011 (28°C) -> 1283 => 25.9
1010 0001111 (29°C) -> 1295 => 27.1
1010 1010001 (36°C) -> 1361 => 33.7
1010 1100110 (38°C) -> 1382 => 35.8
1010 1111111 (40°C) -> 1407 => 38.3
1011 0001110 (42°C) -> 1422 => 39.8
1011 0011101 (43°C) -> 1437 => 41.3
1011 1010011 (49°C) -> 1491 => 46.7
1011 1111000 (52°C) -> 1528 => 50.4
1100 0000001 (53°C) -> 1537 => 51.3
The single arrow points to the decimal value of the 11-bit number. There is obviously an offset of 1024, so I subtracted the value by 1024 and then divided it by 10. The result is shown following the double arrow.

Now we are almost done. If you compare the result with the reference temperature, there is still some difference, like a 1.6 to 1.9 degree constant shift. This is probably due to calibration. The shift seems consistent, so I picked 1.9 as an empirical number. Putting everything together, the temperature is calculated as: [(11-bit number – 1024) / 10 + 1.9] rounded to the nearest integer.

Parity Checking Bit. At this point you have probably already figured out about the mysterious leading bit (which I ignored above). It is a parity bit — it shows if there are an odd or even number of 1’s in the remaining 7 bits. It’s the simplest form of error detection. Because the first temperature sensor I worked with did not have a parity bit, I didn’t think about it immediately. It’s only after I figured out about the temperature encoding scheme that I came to realize the leading bit must be used for error checking. Here is a summary of the encoding scheme:

At this point, the last byte is still a mystery — as I said, it’s likely some sort of CRC checking code. Perhaps the experts can shed some light here?

Arduino Program and Validation. It’s now time for the final test. Here is the Arduino program that listens to the transmitter and displays the humidity / temperature value to the serial monitor.

• Download the Arduino Program for Wireless Humidity/Temperature Sensor

Update: the code is adapted to RPi as well, using wiringPi. The code below uses wiringPi GPIO 2 (P1.13) for data pin.

• Download the Raspberry Pi Program for Wireless Humidity/Temperature Sensor

The result matches the display unit quite well. Cool. Mission accomplished!

Continue to Part 3 and Part 4, or Back to Part 1.

In this and the next two three blog posts (Part 2, Part 3, and Part 4), I will describe how I reverse engineered a few off-the-shelf wireless temperature, humidity, and rain sensors, and used an Arduino (Update: RPi is also supported now!) to listen to and decode the sensor data. This has been a really fun journey and I will document the process as thoroughly as I can. Because there are lots of details, I don’t want to jam everything into a single post, so I have split the whole story into three posts. The Arduino and RPi programs are provided at the end of each post.

Introduction

The first question to ask is always: why am I doing this? Well, for good reasons: these off-the-shelf sensors are cheap, well-built, outdoor-proof, battery-driven and power efficient. If your project needs local weather data and you don’t want to spend time building your own transmitter units (which would bring up a whole bunch of engineering issues), these cheap sensors are the way to go. My original plan was to make use of the sensor data for sprinkler control systems. I actually set this as my next challenge to tackle in a blog post I wrote two years ago. It’s a shame that I lost track of it since then. But hey, two years later, I finally finished it. Better late than never!

Here are the three sensors that I gathered and will use as examples in the following. They all work in the 433MHz frequency band.

Disclaimer: I am not associated with Acu-Rite in any ways, I just picked these sensors because they are common in retail stores.

Preparation

The tools involved are quite simple: I used an Arduino and a 433MHz receiver. You should use the superheterodyne type of receiver as it has very good signal-to-noise-ratio. The super-regenerative type is too noisy and will only work in short range.

Also, to obtain an initial waveform in order to bootstrap the process, I used a RF sniffing circuit from my previous blog post (about interfacing with remote power sockets), a PC with line-in port, a 3.5mm audio cable, and the free Audacity software. If you don’t have a PC with line-in port (most laptops these days don’t have), you can buy a USB sound card which has line-in port.

Wireless Temperature Sensor

Raw Waveform. The temperature sensor is the simplest, so let’s take it down first. What I have at hand is an Acu-Rite 00782W3 indoor/outdoor temperature sensor. The package includes a receiver display unit, and a transmitter which sends a temperature reading every 1 to 2 minutes. Pop in the battery, power on the RF sniffing circuit, and launch the Audacity recording software, I got a waveform like the one shown on the right image below.

By carefully looking at the waveform, I found the following patterns:

• Each transmission consists of 8 repetitions of the same signal.
• Every two repetitions are separated by a constant low sync signal that’s roughly 400 samples (in Audacity you can select a region and see the sample count). Given that the sampling rate is 44.1kHz, this is roughly 9.0ms (400 / 44.1 = 9.07).
• The bit patterns are pretty clear: logic 1 is a constant low of about 180 samples (4.1ms), and logic 0 is a constant low of about 95 samples (2.1ms). Every two bit is separated by a constant high of about 24 samples (0.54ms).

arduino
Given the patterns, I then manually wrote down entire sequence of bits. For example, the image above shows a signal that’s:

11100110 10000000 11111010 01001011

I grouped them into bits of 8 so it’s easy to see the byte values. I also recorded the reference temperature displayed on the receiver unit at the time of capture, which is 77 degree Fahrenheit. At this point we don’t know yet how the bits are encoded (i.e. how they translate to 77). None of the bytes is directly equal to 77. But this is ok — if it was that easy, it wouldn’t have been fun any more 🙂

Create Temperature Variations. In the next step, I will create temperature variations so I can get a lot of different signals and reference readings. By checking how the signal changes, hopefully I can decipher the coding pattern. How do I vary the temperature? Simple: use a hair blower to increase the temperature, and throw the sensor into a fridge to decrease the temperature.

But I am not in a hurry to do that just yet — manually translating the waveform into bits is very tedious, and I worry that if I make a mistake that can compromise the analysis. So I need a way to automate the signal capturing process. Since I already know the signal timings, I can create an Arduino program to automatically translate the waveform into bits. This will make the capturing process a lot faster.

An Arduino Program for Bits Conversion. To get started, I connected the VCC, GND, and DATA pins of the RF receiver to Arduino’s 5V, GND, and D3 (interrupt 1). Then I wrote an interrupt handler to process the captured signal and convert it to bits. Earlier I’ve studied the RCSwitch library, which gave me a good idea of implementation.

Technically, the interrupt function is triggered every time the signal changes from high to low (falling edge) or low to high (rising edge). A ring buffer is used to track the timing between every two triggers. It then checks the timing to see if a sync signal (roughly 9.0ms) is present. Because the signal may still be noisy the first time the sync signal is received, we will wait till the second or third time it’s received to actually do the bits conversion.

• Download the Arduino Program for Sniffing Bits (temperature sensor version)

Collect and Analyze Data. Now let the fun begin. With the Arduino program, I gathered a lot of data under various temperatures. Make sure to also gather some low temperature reading by putting the sensor in a fridge: I was surprised that the signal can actually go through a fridge door! Here is a selected list. The number at the end of each line is the reference temperature value shown on the display receiver.

10001011 10000001 00111110 00111101 (89°F)
10001011 10000000 11100100 01010001 (73°F)
10001011 10000000 10101101 00011011 (63°F)
10001011 10000000 00000111 00001010 (33°F)
10001011 10000000 00000001 01010000 (32°F)
10001011 10001111 11111011 00010111 (31°F)
10001011 10001111 11100010 10111010 (26°F)
10001011 10001111 11001011 10100101 (22°F)
10001011 10001111 01111100 10011010 ( 8°F)

Now the coding pattern is a lot more clear. The first byte is always the same, so it’s probably a signature or channel byte. This is used to avoid interference among multiple transmitters. Note that this byte is different from the one I manually wrote down, so I suspect the transmitter changes the signature every time it’s powered on. The second and third bytes show a clear trend as the temperature goes down. The last byte has no clear trend. It’s possibly some sort of CRC checking byte.

So how do the middle two bytes translate to the temperature values? Well, if you look at the 32°F line: the middle two bytes are very close to 0 (ignoring the leading 1). Since 32°F is roughly 0°C (Celsius), is it possible that the middle two bytes give the temperature in Celsius? After converting the reference temperature values to Celsius, the puzzle is instantly solved (the arrow below points to the decimal value of 12 bits shown in blue):

10001011 10000001 00111110 00111101 (32°C) -> 318
10001011 10000000 11100100 01010001 (23°C) -> 228
10001011 10000000 10101101 00011011 (17°C) -> 173
10001011 10000000 00000111 00001010 ( 1°C) -> 7
10001011 10000000 00000001 01010000 ( 0°C) -> 1
10001011 10001111 11111011 00010111 (-1°C) -> -5 (two's complement)
10001011 10001111 11100010 10111010 (-3°C) -> -30 (two's complement)
10001011 10001111 11001011 10100101 (-5°C) -> -53 (two's complement)
10001011 10001111 01111100 10011010(-13°F) ->-132 (two's complement)

So the temperature, in Celsius, is given by the 12 bits (shown in blue), divided by 10, and rounded to the nearest integer. Aha, that’s it! Now I can modify the Arduino program to not only print out the bits, but also decode the signal and get the real temperature values displayed onto the serial monitor.

• Download the Arduino Program for Wireless Temperature Sensor

Update: the code is adapted to RPi as well, using wiringPi. The code below uses wiringPi GPIO 2 (P1.13) for data pin.

• Download the Raspberry Pi Program for Wireless Temperature Sensor

Note that the program uses pretty tight margins (1ms to 2ms) for screening the signal timings. Depending on the quality of your 433MHz RF receiver, you may have to increase the margin to improve error tolerance.

I did some quick testing by hanging the temperature sensor outside on a tree. Then I compared the temperature value reported on the serial monitor and the reference temperature displayed on the receiver unit. The two matches very well. Cool, mission accomplished! 🙂

Summary of the Process

• Use the RF sniffing circuit and Audacity to capture an example signal; examine the signal and estimate timing information.
• Write an Arduino program to automatically capture signals and convert them into bits.
• Collect a lot of data, record the bits and write down the reference temperatures.
• Examine the changes in the bits as the reference temperature changes, identify those bits that matter, and reason about how they translate to the reference value.
• Modify the Arduino program to decode the bits into temperature value.

Continue to Part 2, Part 3, and Part 4.