
Re: Vehicle Corner Weights
Brian Putnam Dec 10, 2012 10:07 PM (in response to Garrett Dipietro)Unless you want to model in the spring rate from each corner, also taking into account the stiffness of the frame, your best bet is it to approximate the reaction at each wheel. Assuming the tires are lined up frontback and side to side, first use sum of moments to figure how much weight is on the front pair and rear pair. Then do the same thing to detrmine how much is on the left side and right side. Say for example that you come up with 75% front and 25% rear, and 60% right and 40% left. Multiply out the values, so that the right front (RF) is .75*.60=45%; similarly LF=.75*.40=30%, RR=.25*.60=15%, LR=.25*.40=10%.

Re: Vehicle Corner Weights
Garrett Dipietro Dec 11, 2012 9:11 AM (in response to Brian Putnam)Brian multiplying out the values only works if the cross weight percentage of the vehicle is 50%. If we knew the percentage from 0100% we would have the 4th equation, LR+RF=TW*CW%.

Re: Vehicle Corner Weights
Ryan Brumund Dec 11, 2012 1:18 PM (in response to Garrett Dipietro)Brain's example, which is what I would do, does not assume / need a cross weight of 50%. In fact his example shows a cross weight of 55% RF+LR and 45% LF+RR. Assuming the original data, your calculations, and their final check are all going to be done with the vehicle on the bump stops, then I would think that what he did would work (unless I am completly missing something).
You essentially get your additional equations via LeftWeight = COGx relative to vehicle width, Front weight = COGy relative to vehicle length. Things get messy if things are measured and calculated in different states / positions because the hieght of the COG will start to play a role.
On another note, if you have simulation, you could place your parts, apply gravity and measure the reaction forces at the “wheels”.

Re: Vehicle Corner Weights
Garrett Dipietro Dec 11, 2012 4:09 PM (in response to Ryan Brumund)This is an excel sheet I made to show that multiplying the front/rear left/right percentages calculated from the cg will not give you the right answer. You are correct in that you don't have to have a 50% CW%. The left cells are from the base vehicle we are working on before we add anything to it. Green and maroon cells are things we will eventually not know, and all we will have is blue, yellow and grey. The cells on the right are just to give an example of how different the actual weights could be with everything else but the CW% being the same.

Corner Weight Example.xlsx 12.8 KB

Re: Vehicle Corner Weights
Brian Putnam Dec 11, 2012 6:04 PM (in response to Garrett Dipietro)Garrett,
I looked at your spreadsheet. Thank you for putting that together. I have never had the chance to have actual measurements of all 4 wheels.
It looks like this is where the spring rate, chassis deflection, etc. comes into play.
Being a guy who likes data points, what happens if you overinflate or underinflate one of the tires? How does that affect the actual measured weight at each wheel?
Please let me know what you end up doing.
Brian

Re: Vehicle Corner Weights
Garrett Dipietro Dec 12, 2012 10:27 AM (in response to Brian Putnam)We aren't worried about different cases like that. The vehicle has to be loaded onto another vehicle and they have a maximum load per wheel. We will be sitting with the airbags deflated on the bump stops and each tire deflated to the same point. When we measured those weights of the base vehicle it was in the exact state that it would be when it was loaded. We would only be adding more parts to it all across the vehicle. The reason we have to be closer than the estimates is that we are close to the maximum on the left front tire and can't have an estimate that is +20 lbs.
If the basic motion in Solidworks actually followed real world physics it would work perfectly. I would do a motion study with linear springs attaching a "floor" and the middle of each wheel. Deflection of the spring and known spring rate would give me the weight over that tire. However in a test with a 10lb weight and a spring, the spring deflects nowhere near where it should. Different weights and spring rates all were off by varying amounts. I can't think of a way to simplify it but if it wont work in a singular case like that then it wont work with a full vehicle and 3 other scales.

Re: Vehicle Corner Weights
Brian Putnam Dec 12, 2012 10:53 PM (in response to Garrett Dipietro)Not that this solves any solidWorks issues, but maybe you should consider some ballast that you can move around when the vehicle is complete and setting on the scales.







Re: Vehicle Corner Weights
Per Engberg Dec 11, 2012 5:14 AM (in response to Garrett Dipietro)You can get additional equations by calculating at the moment about the diagonal line from one corner to the opposite (or any other arbitrarily chosen line). Of course, as Brian said, this depends on spring rates, the supporting surface being approximately flat and and frame stiffness.

Re: Vehicle Corner Weights
Garrett Dipietro Dec 11, 2012 9:17 AM (in response to Per Engberg)The vehicle will be lowered on airbags to the bump stops so no springs. We can assume an infinitely stiff chassis to simplify things and get at least a close approximation. And after all is said and done the vehicle will be weighed on a laser leveled scale system. I will try calculating at the moment about the diagonal see what I come up with.

Re: Vehicle Corner Weights
Garrett Dipietro Dec 11, 2012 11:15 AM (in response to Garrett Dipietro)Tried it and it doesn't work. Realized after that the equations for the x and y of the COG are essentially moment equations about x and y. You can't have a 4th equation from the diagonal as its just the combination of the other 2.



Re: Vehicle Corner Weights
Jerry Steiger Dec 19, 2012 2:41 PM (in response to Garrett Dipietro)Garrett,
You've got a statically indeterminate problem. You can solve for the summation of vertical forces and moments about x and y if you remove one of the wheels, allowing you to solve for the loads on the other three wheels. It seems reasonable that if you do this for each of the four wheels and then average the three values for each wheel that you ought to get the right answer.
Jerry S.