Showing posts with label Temperature Sensor. Show all posts
Showing posts with label Temperature Sensor. Show all posts

8/05/2013

Power supply for remote sensors

Presentation: How to supply power to microcontrollers/ (remote) sensors with low dropout (LDO) voltage regulators and batteries


"Classic" 7805 Voltage regulator
On August 5 I had a short presentation at the Makerspace Attraktor in Hamburg on how to supply independent power to devices like standalone microcontrollers or (remote) sensors.

 The talk gave some comparison about different low dropout voltage regulators which are by design very well suited for battery powered devices. Decision finding on what parts are most suited and design considerations for independent sensors were explained detailing a remote temperature sensor project I'm working on for about two years now.

XBee temperature sensor


You can find the presentation (in english) here

If you want to stay in touch with new projects or blog entries you can follow me on twitter

6/18/2011

Temperature-Sensor Network with RTC and LCD Part III

We continue with the base station and start to implement the temperature sensor.

There are a lot of temperature sensors available. You could use simple thermistors, analog temperature ICs (like TMP36 or LM335) or digital ones like the Dallas DS18X20 (X stands for S/ B). They all have their pros and cons and I like to experiment with different sensors.

Analog Devices AD22100KT
This time I chose the Analog Devices AD22100KT in a TO-92 package. It's a "Voltage Output Temperature Sensor with Signal Conditioning" and has a large temperature span of 200°C (-50°C to +150°C) a nice linearity and accuracy (according to the datasheet). The datasheet declares an initial error of 0.5°C (at Ta = 25°C, AD22100KT) and a maximum error of 2°C. That sounds much better than the LM335 (at Ta = 25° it's 2°C inital error for LM335, max. 6°C) I worked with at some higher cost for the part and with 60°C more temperature range. Voltage supply is at least 4 (up to 6) volts and quiescent current ist 650µA maximum. As we connect the sensor directly to the Arduino both values are not that important this time (it will be if we connect a temperature sensor directly to the XBee - more on that later).

Datasheet Analog Devices AD22100
You will need a 1k ohm resistor and a 0.1µF capacitor between Vout and the analog input to drive the temperature sensor.
The datasheet also gives insight into the formula we later need to calculate the temperature from our analog readout: Vout = (V+/ 5V) * [1.375V + (22.5mV/ °C) * Ta]. It's the formula where we see the sensor has a temperature coefficient of 22.5mV/ °C (that's techspeak and means the voltage changes by 22.5 millivolts per degree celsius).




The next module ist the Real Time Clock (RTC) DS1307. It's one of Dallas Semiconductor (now Maxim Integrated Products) famous clock-ICs and it's quite easy to build and program. You connect the IC via analog inputs 4 and 5 to the Arduino. That are the I2C pins of the microcontroller and you can use one of the RTC libraries that are available (you can find a great tutorial here).You only need two resistors for the SDA and SCL lines, the 32,768kHz crystal and the 3 volts backup battery (like a Lithium 2032 3 volts coin cell).

Well so far we are almost done with the base station. Please be aware that the left and right power rails of the breadboard are not connected, because the left power rails have 3.3 volts and the right rails are supplied by the 5 volt output from the Arduino UNO.

Arduino UNO with XBee, RTC DS1307 and AD22100KT
In the next part we will finish the base station by connecting the LCD and having a look into the code for the Arduino UNO. In the meantime have fun!

6/13/2011

Temperature-Sensor Network with RTC and LCD Part II

XBee on perfboard
Part II of the temperature-sensor network starts with the Base Station.
The XBee Series 2 sitting on the base station connected to the Arduino UNO is configured in Coordinator API mode. I use the following settings (all other settings are the default Coordinator API setttings of the firmware) which are programmed via the X-CTU software from digi:
  •  AT ID 2001 (PAN ID) (see #1)
  •  AT DH 0 (see #2)
  •  AT DL 0 (see #2)
  •  AT AP 1 (see #3)
  •  AT SP AF0 (see #4)
  1. The extended PAN ID for my network (AT ID = 2001) is 2001 (I hope it will be no "Space Odyssey" though)  - but you can take any 64-bit value up to 0xFFFFFFFFFFFFFFFF or even leave it to 0 where the Coordinator would select a random PAN ID. Be sure to set the same ID to the other devices which should join the network.
  2. The Destination High and Low Address is both set to 0 (AT DH = 0, AT DL =  0) which defines the coordinator.
  3. API mode (AT AP) is set to enable (=1) because we want to receive and parse packets instead of just routing the traffic via the serial port like you would do in transparent mode.
  4. Perfboard back side
  5. The last setting is for the sleep period according to longest sleep period on a/ the End Device(s) (AT SP = AF0). This is to take care that the Coordinator (or any Router which is configured to this length of sleep period) will buffer the messages for the End Device long enough to get them transferred when the End Device awakens (no snooze allowed!). The sleep period is set to 28 seconds on my End Device because the sleep period is multiplied by 10 the value is 2800ms in Decimal or AF0 in Hex.


Unfortunately and for the sake of formality I have to say that my wiring seen in the following picture could be of risk to damage your XBee! As the Arduino UNO runs on 5 volts and the level on the TX pin (digital pin 1) is about 5 volts when HIGH, the DIN pin of the XBee (pin 3) could be exposed to more power than allowed. So I give no guarantee that your XBee is as robust as mine - take care of it and use a level shifter in any case!
XBee with level shifter



Update: For the last days I tested a very simple level shifter, which works for me. I have very low traffic though and there may be problems with higher data transmit rates, but for the time being I just added two resistors (22kohm and 33kohm) as a simple voltage divider and now have maximum roughly 3 volts on DIN of the XBee. So that may be a quick workaround.



Part III will continue with the base station.