View Full Version : Excel Spreadsheet Calculator for Impact Attenuator Performance - Please Critique
mrmustangman357
01-08-2010, 10:59 PM
I thought I should be looking at some sort of analytic model of the impact attenuator before I start buying stuff, so here is what I came up with:
Impact Attenuator Calculations (http://hilltop.bradley.edu/%7Ebjwilliams/Impact_Attenuator_Model.xls)
It uses the stress-strain data from your material and finds the impact absorbing qualities. It also finds G forces, speed, and elapsed time. This is my first stab at this type of model. Is there something off or does this really prove useful? I looked at the calculator by James Waltman and decided to try my hand at using the equations for each step. I can then vary the cross-section as it is being deformed (naturally the smaller crossection deforms first), which allows me to make a IA that fits the nose of the car and still be accurate.
What role does strain-rate sensitivity play in this? Also, can I vary the cross-section as described? Thanks for your help and hopefully other teams can use this spreadsheet if appropriate
Thanks,
Brian Williams
Bradley University
mrmustangman357
01-08-2010, 10:59 PM
I thought I should be looking at some sort of analytic model of the impact attenuator before I start buying stuff, so here is what I came up with:
Impact Attenuator Calculations (http://hilltop.bradley.edu/%7Ebjwilliams/Impact_Attenuator_Model.xls)
It uses the stress-strain data from your material and finds the impact absorbing qualities. It also finds G forces, speed, and elapsed time. This is my first stab at this type of model. Is there something off or does this really prove useful? I looked at the calculator by James Waltman and decided to try my hand at using the equations for each step. I can then vary the cross-section as it is being deformed (naturally the smaller crossection deforms first), which allows me to make a IA that fits the nose of the car and still be accurate.
What role does strain-rate sensitivity play in this? Also, can I vary the cross-section as described? Thanks for your help and hopefully other teams can use this spreadsheet if appropriate
Thanks,
Brian Williams
Bradley University
exFSAE
01-09-2010, 11:40 PM
Ehhhhh...
Not sure what you're trying to accomplish with this, to be honest. This is assuming some solid, regular object of constant material properties that's being smashed?
While it's not a terrible idea to take an analytical approach to this.. I honestly wouldn't bother. Impact attenuators are generally non-trivial structures.. layered honeycomb, weldments, etc. The impact action is highly nonlinear.. involves buckling.. structure self contact.. strain-rate sensitive properties.. it's very very difficult to get right even in FEA much less an excel spreadsheet.
Even scaled-model testing is essentially useless IMO. This is one item you really have to build and test, and test, and test.
Bcohen5055
01-10-2010, 12:10 AM
exFSAE, I totally agree with you, thank you for saying this.
My school has traditionally run one piece carbon fiber nose cones that are designed as impact attenuators. Me and one other student have been put in charge of designing and building the nosecone for this years car, we have a design and most of our materials ready to go but have been put on hold till we have some kind of analysis done first. I have spent most of break milling around looking for good information and have found a lot of graduate level research papers but very little that I can understand and apply to our specific part.
Sorry to thread jack but I'm wondering how many teams do this kind of analysis before building and smashing a nose cone. We have an impact pendulum we can use any time we need to, do you think our time would be better spent testing multiple construction methods and finding what works?
bmw.williams1
01-10-2010, 12:39 AM
If you have unrestricted access to a testing mechanism then do not even bother with looking at theory, especially with composites and a complex nose shape. That is of course, if you have near unlimited access to composite materials.
The judges are looking for physical test data regardless, so theory is only used to help you decide what to test. But as I said if you have unlimited access to testing, then just go crazy. Unless you have an entire year (depends what competition you are going to) then theory of a custom composite structure isn't worth it.
mrmustangman357
01-10-2010, 02:38 PM
What we are using is Dow Impaxx impact foam, which has a nice straight stress-strain curve up until about 60% strain, and it is highly strain-rate insensitive. You can use the spreadsheet to calculate really how much total energy is being absorbed and what strain level you are bringing the material to to achieve that strain level, and tries to show you in-between.
It does the "cellphone math" but just allows you to see what is in-between. Considering I HAVE to have some type of predictive model before we test anything, I feel this calculator is appropriate. There is an obvious downside to just testing and testing and testing. . . this material is super expensive and we just don't have budget or elaborate testing methods like some bigger schools do.
Adambomb
01-11-2010, 03:14 PM
I see. Was trying to figure out where you were getting the position dependent strain, etc. We used Hexcel for ours, which at least analytically "looks" to have more or less constant properties, therefore for this type of first look we only needed a couple simple calcs. And for that I agree, it is absolutely worthwhile to do this type of analysis, if nothing else you are already in the ballpark, which is very important when the stuff you're crushing is expensive, it takes time to test and analyze data, etc.
As for these calculations, off the top of my head the acceleration/energy distances look good, stresses looks comparable to what we've got, looks good to me.
As for strain-rate effects, aside from material strain rate effects you also have rotating and inertial effects of any little bits of crushing material, although with foam that may well be taken into account already with the manufacturer's info as long as it's some sort of "regular" geometry.
Looks like you're about ready to start smashing things! Always the funnest part.
Powered by vBulletin® Version 4.1.5 Copyright © 2024 vBulletin Solutions, Inc. All rights reserved.