DIY corner weighing scales?

[andrew_mx83]

wouldnt you just stand on each set of bathroom scales to check their accuracy and if they are out, mark the percentage on each one with a texta and take this into account in ur spreadsheet?

[oldcorollas]

that's very doable, but it will get old having to bounce the analogue scales all the time to overcome their stiction, and then still have to convert the numbers afterwards..

ideally you would have a fat and skinny (or two skinny) friends to help to calibrate the differences in all the analogue scales

[big_zop]

Not sure what shipping will be like from the states, but definitely a lot cheaper than the the one you found in the first post.

http://www.speedwaymotors.com/Econom...stem,8818.html - $699US

[twentyEight]

:facepalm:

Problems with your statement:

1. Analogue scales are plenty accurate, definitely within a kg Analogue (spring) scales are inaccurate regardless of cost. Go home, step on the scales and get someone to zero them when you are on them. Now step off and on several times, it won't go back to zero 5 times out of 10. The weight difference shown will be over a range of more than a kg difference (ie weight displayed +/- .7kg = range of 1.4kg difference) and when you weigh 70kg (for example only) that's ok, but when your car is putting 300kg on each corner (more in most cases, less in the case of my KE25), the difference isn't +/- .7kg, it's +/- 2.8kg as it's a 2% inaccuracy... That's up to a 5.6kg inaccuracy!
   
   
2. A kg difference over the weight of a car wont mean anything anyway 1kg no, but 5kg difference (as above) on each corner does make a difference when you're trying to get the corner weight to within 1kg difference for track purposes.
   
   
3. Any significant inaccuracy across 8 sets of scales will cancel itself out They will all have the same inaccuracy, so therefore it will not cancel itself out.
   
   
4. Digital scales wont work anyway for other reasons (as you said)

How will it be harder to check the difference on analogue scales anyway? It is difficult to accurately tell the difference between 72kg and 73kg on an analogue scale.

..........

[MWP]

Plenty of things wrong with your reply there twentyEight...

Analogue (spring) scales are inaccurate regardless of cost. Go home, step on the scales and get someone to zero them when you are on them. Now step off and on several times, it won't go back to zero 5 times out of 10. The weight difference shown will be over a range of more than a kg difference (ie weight displayed +/- .7kg = range of 1.4kg difference) and when you weigh 70kg (for example only) that's ok, but when your car is putting 300kg on each corner (more in most cases, less in the case of my KE25), the difference isn't +/- .7kg, it's +/- 2.8kg as it's a 2% inaccuracy... That's up to a 5.6kg inaccuracy!

*NO* set of scales will have an accuracy better than 2% unless they get calibrated at least yearly.
Hell, even a $500 set of digital scales probably wont be better than 5% without frequent calibration (and i dont just mean zeroing).

1kg no, but 5kg difference (as above) on each corner does make a difference when you're trying to get the corner weight to within 1kg difference for track purposes.

For an F1 team sure...
For the average track warrior 5Kg will be fine.

They will all have the same inaccuracy, so therefore it will not cancel itself out.

Yes they do... law of averages.
Reading wont be perfect, but still better than one set.

It is difficult to accurately tell the difference between 72kg and 73kg on an analogue scale.

No set of scales will be that accurate anyway.

[takai]

Something like this would be the easiest:
http://cgi.ebay.com.au/Load-Cell-hig...item5193acc8b0

Pair with an arduino for readouts.

Edit: just checked for combined shipping on 4 of them and it comes to AU$250 delivered. Add a bit of steel and an arduino and you have a VERY good set of corner scales.

[LeeRoy]

Any significant inaccuracy across 8 sets of scales will cancel itself out They will all have the same inaccuracy, so therefore it will not cancel itself out.

Actually, if you had two scales that read lower on one corner, and two scales that read higher on another corner the inaccuracy would be much greater.

[Hiro]

Actually, if you had two scales that read lower on one corner, and two scales that read higher on another corner the inaccuracy would be much greater.

^this
Inaccuracies would only even-out over large sample quantities or repeated applications

[oldcorollas]

Something like this would be the easiest:
http://cgi.ebay.com.au/Load-Cell-hig...item5193acc8b0

Pair with an arduino for readouts.

Edit: just checked for combined shipping on 4 of them and it comes to AU$250 delivered. Add a bit of steel and an arduino and you have a VERY good set of corner scales.

good find!!!!!!

[takai]

Yeah, so who wants to buy them. I dont have the cashola right at this moment.
Thinking of getting a small load cell to get it sorted out and work out how to make it work.

[ed]

what about:

http://cgi.ebay.com/ebaymotors/Manua...item1e5cb314a8

replicate that setup with normal digital bathroom scales (wire in switch to avoid 'turn off' issues)

with the metal plates on rollers, and the correct length lever arm ratio you could easily bring the load on each single scale to under 1-150kg. you lose resolution, sure, but thats a $60 outlay on 4 scales, plus metal, wheels, wleding and time

cheers
ed

["Z" UTE]

takai, a little bit of fabrication, and you could make a nice engine weighing scale with that $45.00 load cell.

cheers Chuck.

[takai]

Nah, would want an S shaped tension cell, rather than the plate shear cells.

[oldcorollas]

they have a lot of nice looking stuff
http://www.elane.net/index.php?go=lo...nd_instruments

[takai]

Crossposted from other forums:

For a while now ive been toying with the idea of using commercial load cells and a simple Arduino for the corner weighting of cars. Its a relatively simple concept, basically taking the measurements from 4 scales placed under the tyres of the car, and allowing for cross weights etc.
This is a commercial unit from GMP Products:


The main motivation for making my own is quite simply that the commercial ones are ridiculously priced. The one above is around $2000 for what is essentially 4 load cells and a bit of circuitry.
This has led to lots of people doing some relatively bodgyarse solutions, such as weighing your car with bathroom scales and chunks of wood:


So for this positively high tech project ill be using a simple Freetronics TwentyTen 'duino and a handful of load cells from Elane in China (http://www.elane.net/ and http://stores.ebay.com.au/Best-Digit...-Postal-Scales).
In addition to those two main components, im going to be using an INA125P op-amp for converting the wheatstone bridge circuitry to a simple 0-5V analog signal that the Arduino can read in.
For testing ill simply output the data to the PC, but eventually ill get a 16x2 or similar LCD for display of the data.

Courtesy of Google and the fact that the Arduino has to be one of the best platforms for collaborative development there are tons of people out there using the 'duino with load cells, including this guy: http://cerulean.dk/words/?page_id=42 which should get me started along the way.

At the moment i have just bought a little 1kg compact load cell (http://cgi.ebay.com.au/ws/eBayISAPI....m=350370902765) for testing, and im waiting on my INA125Ps to arrive from Farnell, and im hoping in a few days that ill have some good stuff to report