Friday, 5 December 2008
Temperature still falling slowly. I get the feeling that somehow the GSM is still enabled (due to the signal quality indicator) but that it is not attempting to 'lock on' to a network by sending out request packets / signals.
The phone setup in the prior experimental updates was:
Phone Mode: Normal
Data Connection: Bluetooth Serial Port Emulation- COM9
Battery Status: Discharging
I have just finished setting up the USB connection and the phone is now back outside.
The phone setup is now:
Phone Mode: Flight Mode (GSM Disabled)
Data Connection: USB Serial Port Emulation- COM3
Battery Status: Charging
Several questions come to mind:
1) Does charging the battery generate any heat? (Likely answer is yes).
2) Which services/peripherals are really disabled when the phone is put into flight-mode?
3) Will the steady state temperature of the two temperature sensors be lower than before? (Unsure as more than two of the test conditions have been changed i.e. not a strict manipulated/responding relationship between switching the phone from 'normal mode' to 'flight mode' and observing the effect on steady-state temperature).
I propose third round of experimentation (which suits my need to procrastinate more than anything else) in which I enable Bluetooth whilst in flight mode (removing the USB connection) so as to isolate the effect of disabling the GSM from the incidental heating which may result from the fact that phone is transmitting it's data via USB rather than bluetooth.
I just saw the phone temerature drop to 1*c whilst the battery temperature remains at 4*c. So whilst the battery temperature sensor was more quick to react to the environmental temerature (presumeably because it is on the periphery of the device unlike the internal phone sensor), the phone temperature sensor is giving a reading which is consistent with the area forecast for this area.
The elevated temperature of the battery could be explained in one of two ways:
1) The temperature sensor is out of range because 1*c is below the recommended operating temperature for the phone.
2) The power dissiapted across the battery is sufficient to maintain a steady-state temperature of 4*c.
I will bring the phone inside, turn off the GSM by putting it into flight mode in order to reduce the power consumption (and, hopefully, the power dissipation over the battery) and then sling it back out again . If this allows the steady state temperature of the battery to stabilise to a lower level then I am prepared to favour hypothesis 2 over hypothesis 1.
I think the experimentation is done for tonight. Not only because the temperature seems to have stabilised but also because it is currently my main phone... Ironically, despite just having written the above and uploading the screenshot, the phone temperature has just dropped to 2*c whilest the battery temperature remains at 4*c. Maybe I will leave it to dangle a while longer...
I did not think to put in my postcode first time round (which is odd for me).
It turns out that searching 'XXXXXX forecast' (where XXXXXX is my postcode) returned a much different result than searching 'Durham forecast'. I must reserve my judgement on which is the more accurate temperature data... One must also bear-in-mind that the temperautre of the phone is likely to be a funcation of: the ambient temperature, the thermal mass of the house, local heating/wind/shielding effects, dissipation of heat across components.
Further experiment may include: Test with GSM disabled. Test with Bluetooth disabled.
Also, I think it would also be interesting to compare these results to a high-precision Mecury thermometer.
No sooner had I posted and the temperature had fallen to that of the forecast! When I get a new phone (possibly at christmas). I can think about what use I might put my dear old K610. I think that sealing it in a water proof bag with silica gel bags and a long usb extension (rather than bluetooth) would make it a very good outdoor temperature probe. If I combine that with the time-lapse webcam data then I might have a reasonable, independant, indication of local weather conditions. Now all I need is my final year project to be finished and I can really get into sensor netwokring on a grander scale. Oh it will be sweet. For now, I am having evil thoughts about stripping back some of the form factor to see if that allows the temperature sensors to react more quickly.
The charger started whining so I unplugged it. As you can see the temperature is continuing to fall. I am interested to see if I can plot some graphs from this data. I would also love to know if the phone can log this data locally and if it would be possible to sync at the end of each day?
So I was messing around with MyPhoneExplorer- I established a Bluetooth link and am currently charging the phone (the chord of which also conveniently serves as a tether by which to dangle the phone out of the window). I don't like my phone that much anyway. But in the interest of science, seeing if the temperature readings can actually be reflective of the environment the phone is in, it is a small price to pay to have my phone dangle out of a window. After all, it is kind of in a protective casing. The idea for the test occurred when I was thinking of ways to put off doing my C coding assignment due in tomorrow. I was playing with Bluetooth in anticipation of receiving my wiimote (to make my own multi-point interactive display inspired by this work: http://www.cs.cmu.edu/~johnny/projects/wii/) and I came across the 'Monitor' function under the 'Others' group in the MyPhoneExplorer software. I googled the current forecast for temperature tonight and have been occasionally looking in at the temperature. Let's see if the phone temperature reading tends towards the forecast.
Currently, batt temp: 13*c and phone temp: 17*c. Looking good!