inertia test rig(help needed)

[marcush.]

There is another way - but only for the brave.

I guess it would require all liquids to be removed, rigid links to replace the dampers, a steering lock, and rigid "rigging" wheels. Even that might not be enough for the blancmange that is a road car.

this was exactly the method teached by Claude Rouelle in the 3 day seminar for race car dynmics.worked always well .You have to persuade the car owner ...or wait till he´s surely not visiting the workshop ..but you can always horrify him showing the pics in a convenient moment...

[mep]

Reading the question carefully the user wants to know how to measure the inertia (MoIx, MoIy, MoIz). So far most of the answers focus on finding CoG location, which is only part of the story. I have heart once that it is possible to find that by hanging the complete car on a robe of know length and let it swing. By measuring the periodic time it is apparently possible to calculate the MoI, even though I have a few problems to understand that. First of all because this basically generates a pendulum where I think it is primary affected by the mass of the hanging body rather than its MoI. Then it has to be guaranteed that the vehicle always has the same angle relative to the rope in order make the vehicle rotate and not just swing. The contribution of the vehicle rotation to the periodic time has to be calculated then. Whereas, I wonder if that procedure gives you the MoI around the pivot point of the robe rather than the MoI around the CoG of the vehicle, which is the required MoI.

[Tim.Wright]

The periodic time depends on the mass AND the MOI. If you know the mass and CG location, you can seperate that effect out and solve for the MOI. As I said in my first post, the final calculations can be very sensitive to the CG location so you need to know it accurately.

The first calculation will give you the MOI about the swing point. Then you transform it back to the CG using the parallel axis therom which say
MOI_at_cg = MOI_at_swing_axis - mr².

Where m is the mass, r is the distance from the swing axis to the CG. You can see the final term is squared, so if your CG loc estimate is wrong, it can cause large errors in the results.

Best to keep the "r" short I think.

Tim

[mep]

The periodic time depends on the mass AND the MOI.
Tim

Generally speaking yes, but only when the body actually tilts while being in the pendulum movement. A body with a high MoI could theoretically swing as a pendulum but not tilt with it. In reality it will do something in between but it will bring a mistake into the result.



Uploaded with ImageShack.us

[Tim.Wright]

But thats not how the test is supposed to be executed. The whole point is to rotate the car about an axis parallel to the one you are measuring.

You dont simply swing it with a single rope because of the reasons you mentioned. Also (as we found) they stretch too much making the CG location move. We used two ropes from the front and the back of the chassis to force it to rotate about a point.

Realistically, like with any measurement equipment, you should be making things out of rope. You need a stiff platform which doesn't deflect. Like I said at the beginning, you won't get out of this with decent results without a pretty significant structure.

Tim

[gato azul]

If you are interested in the "pendulum" test, then try to google "trifilar pendulum"

which looks something like this:


alternatively, you can read up some stuff here:

http://www.ima.org.uk/_db/_documents/ma ... ms_huw.pdf
http://www.dtic.mil/dtic/tr/fulltext/u2/287534.pdf
http://www.iftomm.org/iftomm/proceeding ... s/A464.pdf
http://sem-proceedings.com/27i/sem.org- ... ystems.pdf

or look for a commercial service, which seems to be a sensible option for a "one off" test, such as Cranefield or REGINS/INPROP (InTenso-Rig/Milan-Italy), depending from where you go that will cost something between 1000-1500US$, but maybe higher at some places.



http://www.regins.org/en/content/projec ... inprop.pdf