Sunday, 31 May 2009

a little inclination

The table on my patio is on a sloping patio, so it slopes itself, and anything I put onto it seems to be about to slide all the time. It's enough to be noticable, and enough to be annoying. Dad and I have been laying a new patio (building some walls, turfing a new lawn - normal lightweight garden maintenance for someone in my family...), so today we were playing with a digital inclinometer. A patio should slope at roughly 3 degrees, and the table seemed so slanted that I estimated 15 degrees. The reading? 1.5 degrees; 2 degrees at the worst.

Now when you look at it on the scale of, say, 0 to 90 degrees, 1.5 degrees seems to be tiny. Piffling. Not even noticable. Lost in the noise. But in reality it's both very noticable and appears to make a difference (I can point to exactly the direction that each patio slab slopes in, and none of them are over 2 degrees sloped). So I had a look at my spirit level - something I'd never really thought to do before. It's marked at 0.5 degree per mm and there is 1mm from the bubble (if it's completely straight and still) to the marker lines on either side of it. Now I think I'm getting sloppy if the bubble is half that distance away from the line, i.e. if it's 0.5mm off true. Which means that theoretically (although not, on a too-hot day, in practice) my wall should be within 0.25 degrees off true in all directions. Now I personally think this is really quite an achievement - and not something that we ever really think about beyond "is the bubble inbetween the lines". We DIY types and shelf-putter-uppers really should congratulate ourselves for that...

...but there's an experiment here to be done. If I can spot a 2 degree slope on my table, I'm wondering just how small that slope has to be before we humans really can't tell which direction it's going in. I'm guessing somewhere around half a degree, but I think some more time with the patio and inclinometer (and possibly some friends and some beer to do some more advanced "what is the effect of alcohol on..." perception tests) is called for here.

No comments: