Physical Test Data

[helpseeker]

Hello all!
I'm wondering if one could find any desired test data on some database?!
of course there are infinite examples while there are only few already made files,so shall we use the available ones as the elements to make any composition out of them or the solution is to make some real tests and feed in the results in some fancy way or what?
im really new to this software and need it to optimize a for-bar mechanism regarding the bar lengths and the position of the joints.
thank you all for donating your time and patience.

 

[GregLocock]

No, there are very few bits of adequately documented correlation data that are freely available. The good news is that various configurations of 4 bar link are easily calculated by hand. The 'fancy' way of correlating data is to model the instruments on the real test item, put your model through the same excitation, and compare outputs.

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[helpseeker]

Hi Greg!
first of all let me express my gratefulness for your exact, quick and fruitful help.
but since i'm not that much into Adams, i ask some ambiguities of mine:
what do you mean of "to model the instruments on the real test item"?
does it mean i should make a real and physical model including buying some torque meters,bars etc and set up the mechanism and create the "Physical Test Data " worksheet for example by using the data acquisition method in matlab,any whre? or im false n it means im about to create some computer modeling for the four-bar link all happening in the computer?
may you receive help in some unexpected days of your life when you are needy for it when you are a helpseeker!

 

[GregLocock]

Yes, I mean that my models are not regarded as trustworthy unless we have correlated them to physical real world tests on a similar vehicle. Even then we regard them as useful indicators of likely behavior rather than absolutely reliable models.

So we instrument cars with 50 or 100 channels of measuring gear, do some maneuvers, then make the model do the same maneuvers, and see how well the instrument readings line up.

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[helpseeker]

Thank You Greg!
regards