0
$\begingroup$

I am looking to use some load cells (foil based type) to detect presence for home automation within in a room. I was looking to mount some loadcells to the joists for the floor (UK Based house so around 6x2" Joists around 13ft long on 400mm (1.3ft Centres) to try and measure deflection.

I'm not bothered for Exact measurements, but knowing the difference between a small dog and a human may be useful or determining more that one person.. but again not a real issue..

What size loadcell would people recommend, I've seen very small sensors used for deflection in steel components such as suspension components on cars.. this makes me think that I dont need large sensors on timber joists that 'should' move more?

For mounting I was thinking of bonding / gluing to the side of the joist, near the centre of the span and sanding it smooth first..

Thoughts, suggestions, help?

Thanks

$\endgroup$
0
$\begingroup$

Well, centre of the span is good as that is where the greatest change will be - but at the edge, because if you put it on the centre-line then the change in length will be minimal.

As for size - you need to estimate the actual change in length you are likely to see for the heaviest person or persons ( lifts always make me laugh as the average weight is about half my weight., but at over 6"...), then select a sensor which covers that range - usually best not at the extremes of its working range.

$\endgroup$
  • $\begingroup$ Thats just it I dont have a clue to work out the potential change in length.. as the floorboards would spread this load out between joists.... Nor do I have a clue how to actually size loadcells.. I bought a couple that are around 50mm long but wonder if that is far too big? $\endgroup$ – David Wallis Jul 30 '18 at 10:35
  • $\begingroup$ Well, are joists now meant to be 8”? And a search should give you the deflection calculations as specified by the UK building regs. $\endgroup$ – Solar Mike Jul 30 '18 at 10:56
0
$\begingroup$

As mentioned by Solar Mike, you'll need to calculate the deflection at midspan.

One's first instinct would be to make use of the fact that joist structures tend to be heavily standardized and there are handbooks around which give you all the information you need (I personally don't have any because I'm from Brazil, which doesn't have a tradition of joists).

However, this first instinct won't do you much good. As standardized as joists should be, reality isn't as kind. The problem isn't really the interaction with the flooring you've mentioned, but the simple fact that wood is a highly variable material, with significantly different Elastic Modulus for every specimen of the same species. Given that the Elastic Modulus is a key variable in calculating deflection, that makes the calculation very imprecise.

Then there's the fact that buildings aren't always built exactly to code and construction imperfections exist everywhere. And calculations are always made in a world of theory, where supports are perfectly pinned or fixed, when reality is much messier.

So if you really need to calculate the deflection, you'll need to take a core sample of the joist (even though removing a beam sample is perfectly fine, good luck telling the owner that!) and do a pretty detailed survey of the structural system. And you'll probably still be significantly wrong.

The easiest solution here is to forget about actually calculating the deflection, and just use the strain gauge results directly.

Since you've apparently already bought some, stick one to your house and then see what the strain gauge measures when you walk around the room, when your dog walks around, when you move furniture around, etc.

See if you can distinguish that signal. If you can, then that strain gauge is sufficient (feel free to buy cheaper ones and try them). If you can't, then you'll need to buy a more sensitive one.

Obviously this will require significant calibration, but you should probably get used to that anyways, since you'll probably need to calibrate the results for each of your customers' locations.

You'll then have to figure out how to handle moving furniture (which will alter the "empty room baseline") and long-term effects like creep.

$\endgroup$
  • $\begingroup$ Thats kind of the route I was going to go down.. I just wondered whether the more common smaller strain gauges would be sufficient - however like you say I should just try them.. I'm loathed to do it at the moment as we are moving house soon - so thinking in preperation for the new one.. and as for the customer... well thats me :) $\endgroup$ – David Wallis Jul 30 '18 at 13:12

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.