Preliminary design for wheel scales - any thoughts?

[Sporqster]

This is a little home project to support my sportscar hobby.

I'm actually working on this with a 1400lb max working weight (per wheel)- so you could use this to set up your pickup truck if you wanted to. The load cell is basically a surplus industrial single-acting hydraulic cylinder with a gauge stuck in the port. That is then inside a piece of tubing with just a slightly larger ID than the OD of the cylinder. The weight of the wheel functions like a scissor jack to compress the cylinder, resulting in a pressure reading.

Based on what I've gathered from talking to the guys in the lab where I work, I should be able to get around +- 5% readings on the weight, most of the error in the cheap pressure gauge I'm looking at.

I'm eyeballing the total cost of this to be about $50 each, $200 for the set of 4, in materials. Not uber cheap, but heckuvalot cheaper than a set of longacre's.

Has anyone seen or attempted a rig like this? feedback? experiences? any reason you can think of that this is a really dumb idea? I'm sure someone has thought of doing something like this before, and I tend to think there must be a reason nothing like this is commercially available (to my knowlege).

[svenakela]

Looks like a good and simple design.
Make sure to not put the fingers inside it, though.

[harryn]

It is an interesting concept. My head does not work well on Fri, but it seems that you are running approx. 25:1 force reduction (combined mechanical and pneumatic) to obtain the readings. This implies that every imperfection or difference between units could cause a difference in indication vs actual of 25x as well. (I think)

I guess the important thing would be to do some kind of callibration on each one prior to use.

How long are you assuming the air will stay trapped in the cylinders at a time ?

I wonder if you could make something simple from an air ride shock or load leveler sort of setup.

[Geof]

One problem could be that the force acting on the cylinder is not linear with the vertical displacement of the load. This means if you calibrate it with a particular known load it will only be accurate when weighing that same load. Any other load with cause things to deflect slightly differently so the load will ride higher or lower and the non-linearity will cause an error. Also if there is any air whatsoever in the cylinder it will compress to a different extent and add to the non-linearity.

[svenakela]

Originally Posted by Geof One problem could be that the force acting on the cylinder is not linear with the vertical displacement of the load. This means if you calibrate it with a particular known load it will only be accurate when weighing that same load. Any other load with cause things to deflect slightly differently so the load will ride higher or lower and the non-linearity will cause an error. Also if there is any air whatsoever in the cylinder it will compress to a different extent and add to the non-linearity.

My first instinct thought was the same, but it's hydraulic and the movement should be minimal even with high loads. Oil is not "compressable" so if he makes sure there is nor air the deflection should be less than the reading error on the gauge. I think...

[mjrandle]

Hydraulic oil has a finite bulk modulus, is only "relatively" incompressible and will become noticeably compressible if any air is present in the oil.

The non-linerarity of the mechanism is a game breaker. Also, there will be a lot of friction that will affect the accuracy of the measurement: pivot bearings, piston seal, gland seal and gland bearing.

Hydraulic cylinder pressures are often measured on mobile cranes to determine the lifted load, but with pressure transducers rather than pressure gauges.

Best to avoid indirect measurement techniques whenever possible - particularly mechanical linkages. Maybe connect a pressure gauge to a bottle jack or pancake cylinder.

Hope this helps!


Cheers,

Mike

[svenakela]

Great info Mike.
The design is very similar to the bike lift my road racing friend uses, it scissors the bike up and down. It doesn't move a millimeter when a 250 kg load is put on it and it has a lot longer mechanism, so i still think that the non-linear movement is a small problem.

Sporqster, if you have the parts already, try it!
If you need the scales to check that you pass a scrutineering, 5% is a big loss anyway. But if you want the weight for chassi adjustments (as I suspect you want) I would give it a go. Make one and try it, compare it with another scale if you have that possibility.

[sc brian]

Your concept is clever in principle, but unpredictable friction might limit the performance more than your needs can tolerate.

I have made scales both simple and complicated for measuring all kinds of heavy things. A simple and (relatively) inexpensive solution is the use of air spings like the ones used in air ride suspension systems. Load versus pressure is well understood, they have lifting capability, and there is very little hysteresis and friction. Check the Firestone web site:

http://www.firestoneindustrial.com/

Good luck!

[NC Cams]

Complexity in search of a need.

Quite some time ago, a guy took a simple bathroom scale and a lever system and used it to measure the wheel weight of his race car. He actually had 4 scales and did all 4 wheels at once. He was selling them for a while but, I'm sure he quit as the idea was too simple to copy.

The lever was simply a way to load the car 1 foot from the pivot and the scale was placed 4-5 feet away from the pivot. Using simple moment diagram, he got his 250lb bathroom scale to read the much much heavier car at whatever scale factor he needed/wanted.

Example:
500 lb wheel weight times 1 foot lever arm = X lbs at bathroom scale x 5 foot lever arm. Do the math....

Until/unless they change the laws of physics, the cost/benefit/simplicity ratio of lever armed bathroom scales can't be beat.

[WeCheat]

NC Cams Is right bathroom scales work really well. I have been set for over fifteen years. They will get you too within 1% if you’re careful and do your setup exactly the same. I’ve checked the results with my Longacre's scales. Will they tell you if your car weighs say 2350lbs exactly, no but they will tell you your cross weights or rear bite to within 1%. I still use the bathroom scales when we go out of town and race because if they get stolen I would only to lose 150 bucks. The hydraulic scales that used to be made I believe were made by Rebcco if memory serves me right. They had a bad problem with binding from the side loads that the chassis put on the scales. I think you’re designed would have problems with binding from suspension movement and Camber. Your tires have to float on the scales to get accurate reading. Even Longacre's scales will give you a false reading if the tires don’t “float” and let the suspension settles naturally. Or you could look at it this way buy 4 electronic bathroom scales and you’ll be 21st century dude.

[svenakela]

Sometimes the solution is too simple to be seen.

[bentwings]

I use the bathroom scale and the 2x6 method to weigh my streetrods. I refined it by making pivots for the lift end and where the scale pushes on, and the actual load carring point. This gives quite accurate results. I think friction in the pivots is the biggest area. I just used 1/2 inch bolts right onthe wood. It's a simple calculation.

As for Sporqster's design, it will probably work.
As noted any air in the cylinder will cause problems. Also any side loads will too. But I think it's ok.
I think I would make spacers for the other wheels so the whole system is the same height when weighing. I did this with my 2x6 system so I could check each wheel.

Enerpac makes relatively flat cylinders that could be used directly with only a plate welded on the cylinder end to spread the load against the tire. You could use their hand pump with a gage and calculate a scale weight.