Good news for this week! I finally have a good data-set of those pesky load-cells to start doing some analysis on to see if I can find a cure to their drifting
SPOILER: Haven't found a [decent] solution yet!
I have made some pretty graphs though:
These are some of the data-sets that I've gathered in order of appearance: Air temperature, Load-Cell temperature and Weight read by the cell.
I made these graphs by doing some wizardry in Octave (because I'm too poor for Matlab + Stallman would judge me) here's the function I wrote for it:
function plotTempWeight(tempFileName, weightFileName)
temp = csvread(tempFileName);
weight = csvread(weightFileName);
air = csvread('air.csv');
#gotta strip 0 values
temp = temp(temp~=0);
weight = weight(weight~=0);
normal_temp = (temp - min(temp)) / ( max(temp) - min(temp) );
normal_weight = (weight - min(weight)) / ( max(weight) - min(weight) );
normal_air = (air - min(air)) / ( max(air) - min(air) );
# now with the normalised values we must make some assumptions about length
# Temperature has a lot more readings so it makes sense to apply interpolation
# on the weighings
% create new time, with as many sampling times
t_new = linspace(1, numel(normal_weight), numel(temp));
t_new1 = linspace(1, numel(normal_air), numel(temp));
normal_weight = interp1(normal_weight, t_new);
normal_air = interp1(normal_air, t_new1);
I am however eager to gloss over this load-cell malarky as there's not much new since the last few rants but hopefully a solution will present itself soon and I can rant about that in detail!
I feel the need to point out that this pipe is cool! It can adjust height and well just in general be cool because Jason and I are doing it ourselves!
Still working on a few other ideas for this but it is nice to get the overhang out of the way! And let the plants praise the sun!
384 total views, 1 views today