Tuesday, December 30, 2008

Results! and progress in the logging!

So finally I managed to write about the results of my logging experiment!

I didn't get to collect much data before a considerable portion of the attic was insulated, but nevertheless I guess it is enough to get a rough idea that the stuff is at least doing something to prevent heat from escaping - at least it prevents drafts in the attic, which I guess is half way to improvement!

So here is a rundown of the results:

before the insulation: average of 4.8 degrees C difference between inside and outside
more than half installed: average of 6.5 degrees C difference between inside and outside
almost all of it installed: average of 8 degrees C difference between inside and outside

The split in the graphic corresponds to the moment when more than half the stuff was installed, so before the split is before the insulation, after the split is more than half installed. One can see that the graphic for the inside temperature in the second part is smoother than in the first part, even as outside temperatures drop further down!

These results have to take into account that the house below the attic is warmed up regularly (although to a low setting of 18 degrees C), but nevertheless, for the same outside temperature before and after I do see that the inside temperature is higher now. Also the variance of the inside temperature is much lower than the variance of the outside temperature, which was not the case before the insulation was applied.

Overall, I think it was worth the money and work invested.

Furthermore, and as an additional proof of the worthiness of the investment, I verified that I spent an amount of gas in December 2008 identical to December of 2007, while the average month temperature was 1.4 degrees Celsius lower, and I spent the whole month of December 2008 at home, whereas in 2007 we took 4 days of holiday.
My estimate is that the extra insulation gives me an extra 10% savings in gas consumption compared to the situation without.
Next Spring - Insulate the floor of the ground floor of the house!

On another subject, I was finding it strange that the temperature logger was not going below 0 degrees, so I decided to investigate. I placed the sensor in the freezer, and it did go to -16 before I took it out. I then figured that due to rounding, the temperature would need to be below -1.45 before the sensor and the arduino code would register a negative temperature of -1. All this due to the quantification in the A/D converter and integer logic.
SO I have decided to adapt the code to include a true low pass software filter (and an efficient one also, from the computational point of view) and work with one decimal place, so I managed to get better results! Send me a note if you are interested in the arduino code to do it.
It also includes 2 keys - Select and Reset - which allow you to navigate in a small menu showing the maximum and minimum temperatures, reset the max min log, and see how full the inside EPROM is.

I also included a low pass hardware filter - RC - but somehow it prevented the sensor from registering values below 0, so I took it off for the outside sensor. Still must investigate what went wrong!

There are some more news, but I'll keep that for the next post!