I'm working with Xbee Pro with wire antenna to transmit data from inside a refrigerator. The receptor outside is the Xbee with RPSMA antenna. I use an extern MCU to control the data transmission, transmitting every second.

When temperature is about 4 °C or higher, the reception is ok, every second I get the data. But when the temperature drops from that point, reception starts to be poor, starting to receive each 5 seconds or less.

The transmitter is reporting the Not Ack Received notification. The input voltages seem to be fine (3.3 V) and both the Xbee and MCU are working.

Does temperature affect the transmission or the antenna in any way? Should I change the antenna of the receiver or transmitter?

  • 1
    \$\begingroup\$ 1. What Olin says 2. Can you advise what is actually in the refrigerator. You say an "extern MCU" - is that external to 'fridge, or ??? . // As Olin says, battery is greatest suspect. Can you measure battery voltage while in 'fridge? - Solder on 2 wires and bring outside to a meter, or allow to get cold then open and measure rapidly. Bringing out wires is best as yoy can watch to see what voltage does with time and temperature. // Have you any sort of sanity check to ensure that the mcu is otherwise behaving OK at low temperatures? eg occasional beep etc. \$\endgroup\$ – Russell McMahon Sep 7 '11 at 19:26
  • \$\begingroup\$ @Russell I'm using a PCB with a Freescale QE128 reading ADC from a temperature sensor. Then sending that data to the Xbee that is on the PCB too. Sensor, Xbee and MCU are inside the fridge. I'm using a BDM to monitor what the MCU is doing in real time and it seems to be working. I measured the voltage rapidly and seemed fine, but I'll try soldering the wires to measure it from outside. \$\endgroup\$ – HzJavier Sep 7 '11 at 20:10

Temperature isn't going to matter to the antenna or the RF field it creates. The issue is most likely with your circuit or what is powering it. Are all components you are using rated for under 0C operation? Some are only 0-70C, although going down to -20C is fairly common. Don't assume. Check all the datasheets.

The most likely problem is the battery if that's what you're using to power this unit. Both output voltage and capacity drop with temperature for most batteries. What kind of battery are you using? What voltage range does your circuit need? Are your batteries fresh or fully charged?

You may have to rethink your power supply to be able to run from a lower battery voltage than anticipated.


One other possibility occurred to me, especially since you say everything is running from a 12V battery and you are only using 8V and 3.3V. A bad solder joint can work at some temperatures then fail at others as the metal expands or contracts. If this is a small device, try reflowing all the solder joints and see if that changes anything. Preferably add a tiny bit of solder to every joint, but definitely make sure the solder gets molten for a second or so at least. Then check each joint with a jeweler's loupe and redo anything that looks suspect.

Let us know if this changes anything.

  • \$\begingroup\$ I'm using a 12V battery to power the circuit. Then a 8V regulator to power the sensor. The MCU and the Xbee work with 3.3V and that voltages seem ok. Also both says that work under -45°C, but I don't know for the rest of the components. I'll check. Also I don't know if the current has to do something with it, even if the voltage is ok. \$\endgroup\$ – HzJavier Sep 7 '11 at 20:19
  • \$\begingroup\$ I tested the circuit separately. The MCU and Xbee work fine at -10°C. The problem is the switched power supply I'm using to get the 3.3V, but I still don't know the reason. I changed it to a linear one and the circuit is working fine now. Any ideas why the switched one doesn't work? Thanks for your help. \$\endgroup\$ – HzJavier Sep 9 '11 at 20:03
  • \$\begingroup\$ @Yonosoypol: There is no way to tell from the information you've given. However, all the same possible reasons apply to the power supply just as to the main circuit. \$\endgroup\$ – Olin Lathrop Sep 9 '11 at 23:03

I agree with what has been said as a very likely cause. A "long shot" occurred to me that maybe possible - is there condensation forming on the board at the lower temperatures? This might cause leakage current, etc that might upset the sensitive RF circuitry (or any oscillators on board)
I suspect not, but...

  • \$\begingroup\$ In fact there is some condensation, but is not too much. I thought that maybe affected in some way the circuit. It looks like it only affects the switched power source, but I still don't know in what way. \$\endgroup\$ – HzJavier Sep 9 '11 at 19:56

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.