Hi Everybody!
I wanted to publish this in a blog but unfortunately it is not available in the GR-Sakura group. I ask for the help of the Admin to transfer it into the Blog area (and make it available), if possible.
The Adafruit's DHT library for the DHT11, DHT21 and DHT22 temperature and relative humidity sensors has been adopted for the GR-Sakura board. These sensors use a non-standard 1-wire communication protocol, which is handled by bit-banging and sofware timing. The '1'-s and the '0' can be distinguised by measuring the data pulse lenght. The biggest difference between the Sakura and Arduino cards is the speed factor: the overhead time spent in the main cyle (when 40 data bits from the sensor are read) is much bigger in the case of the Arduino, which should be taken into account. This means, that the default parameter value used as "discrimination level" is too low for the Sakura board.In order to determine the optimal value of the "discrimination level" the debug features of the original test program were extended.
The gr_DHT_sensor library consists of two files: DHT.h and DHT.cpp
This program can be used to test one of the above mentioned sensors. I used an Aosong Am2302 (aka DHT22) sensor. The schematic is shown below:
The DHTtest sketch has several debug features, but to activate all of these you have to define the DEBUG symbol.
The DEBUG symbol can be defined in two ways:
a. You can define it by adding a -DDEBUG compiler option in the makefile (this method is preferred in the case of command line compiling)
b. You can define it in the DHT.h header file by adding the following line:
#define DEBUG
1. LED1 is ON when the board is waiting for some input from the serial (or virtual serial) port. (note the usage of the SCI_AUTO option in the sketch)
2. LED2 is ON while the relative humidity is read
3. LED3 is ON when the temperature value is read
4. LED4 is ON if there was a communication error with the sensor
5. Current time and Last time are printed out through the serial port (this kind of sensors are rather slow, so you should wait 2 seconds between readouts)
6. The cycle counts for each of the 40 data bits are outputted to see their statistical sacattering. This helps to determine the optimal "discriminator level", i. e. to safely distinguish the '1' bits form the '0' data bits.
7. Number of detected data bits and the read data byte values are outputted. The fifth byte (the checksum) is shown against the value calculated as the seum of the first four bytes.
A typical output is shown in the above figure. As one can see, the count calues can be divided into two groups:
a. lower values are between 10 and 12. These values correspond to '0' bits.
b. larger values are between 31 and 32. These values correspond to '1' bits.
This means, that an intermediate value, say 20 will be a good candidate for the discriminator level value.
When you don't define the DEBUG symbol, the only the relative humidity and temperature values are outputted. See the figure below.
This example application is the adoptation of the DHT11 + ChipKit UNO32 project from Embedded Lab. It consists of two parts:
a. Firmware running in the Sakura board, which in each 2 seconds measures the temperature and the relative humidity, and sends data to the PC (through the USB, as virtual serial port).
b. PC application witten in Processing/Java which automatically connects to the first serial port and displays the received data in a graphic window.
The hardware schematic is the same as in the firts figure above, i.e. the data wire is connencted to pin 2.
Changes in the software:
- Due to the higher resolution of the DHT22 sensors, we modified both the firmware and the PC application (in our version floating values are sent...).
- Due to the extra decimals the graphical display of the relative humidity was slightly reorganized.
You can see a typical output in the next figure.
As you see, there is also a "Start Log" button. When you click on it, the measured values will be logged, together with a date/time stamp. When you click again on that button, the data logging will stop. (The sudden increase in the humidity happened due to my breathe...)
MM DD YYYY HH MM SS C F RH(%)04 21 2013 20 43 35 26,2 79,2 32,504 21 2013 20 43 37 26,2 79,2 32,504 21 2013 20 43 40 26,2 79,2 32,504 21 2013 20 43 42 26,2 79,2 32,504 21 2013 20 43 45 26,2 79,2 32,504 21 2013 20 43 47 26,2 79,2 32,504 21 2013 20 43 50 26,2 79,2 32,504 21 2013 20 43 52 26,2 79,2 32,504 21 2013 20 43 55 26,1 79,0 32,504 21 2013 20 43 57 26,2 79,2 32,504 21 2013 20 44 00 26,2 79,2 32,504 21 2013 20 44 02 26,2 79,2 32,504 21 2013 20 44 05 26,2 79,2 32,604 21 2013 20 44 07 26,2 79,2 32,604 21 2013 20 44 10 26,2 79,2 32,604 21 2013 20 44 12 26,1 79,0 40 <-------04 21 2013 20 44 15 26,3 79,3 44,804 21 2013 20 44 18 26,3 79,3 42,604 21 2013 20 44 20 26,2 79,2 40,104 21 2013 20 44 23 26,2 79,2 38,304 21 2013 20 44 25 26,2 79,2 38,304 21 2013 20 44 25 26,2 79,2 36,904 21 2013 20 44 28 26,2 79,2 3604 21 2013 20 44 30 26,2 79,2 3604 21 2013 20 44 30 26,2 79,2 35,2
DHTxx Sensors (tutorial): http://learn.adafruit.com/dht
DHT22 (AM2302) datasheet: http://www.adafruit.com/datasheets/DHT22.pdf
Adafruit DHT-sensor-library at GitHub: https://github.com/adafruit/DHT-sensor-library
ChipKIT Project, Temperature and relative humidity logger: http://embedded-lab.com/blog/?p=5453
Processing homepage: http://processing.org
Fritzing 用のパーツ（GR-SAKURA）: http://japan.renesasrulz.com/gr_user_forum_japanese/m/mediagallery/19.aspx
Fritzing homepage: http://fritzing.org
Source code and example PC application: http://renesasrulz.com/app_kits_and_demo_boards/gadget_renesas_user_forum/m/mediagallery/511.aspx
Fantastic!
Awesome. Perfect example for my students.
Hello Istvan Cserny !
First of all, Good job !
Then I'd like to test your code on my Sakura board but I am new to all this.
So I have a question about how to import on Renesas Web Compiler the two files (the library) DHT.cpp and DHT.h ?
Thank you for your help.
Hassan.