Alternative Frame Rules Discussion

[Bill Riley]

This is a place for questions and discussion of the new for 2011 Alternative Frame Rules

[Bill Riley]

This is a place for questions and discussion of the new for 2011 Alternative Frame Rules

[Bill Riley]

Please see fsae online dot com for the sample frame and supporting documentation.

[oz_olly]

Hey Bill,

Thanks for starting a post on this topic. I tried to start some conversation on it in the 2011 Rule Book discussion but it seems there are no takers. I really like the idea of chassis design load cases. Can you comment on how the load cases were developed and why they are not dependant on vehicle mass? Do they have an inbuilt factor of safety?

Cheers

[KyleD]

I'd also like to see some input as to how these forces were decided upon. Specifically the front bulkhead and bulkhead support case which pulls on the front of the car with 150 kN. Pushing on the front of the car seems like a more realistic scenario so I'm curious to see how these forces and constraints came to be.

[Bill Riley]

KyleD: I revised the coordinate system in the final rules release and didn't catch the bulkhead loads were now pulling. That is fixed on the revised version on the fsaeonline dot com. I also fixed a typo or two. That will be the most current version as we work through the new rules.

oz_olly: That's a good question about where they came from. The basis comes from the impact attenuator rule which assumes a 300 kg car and driver combination. It didn't make sense to use actual vehicle mass since teams don't know what it will be when the they start out or at the time the SRCF is due. So there will be some mass effect depending on how heavy your actual car is. Next, the 40g peak decel force for the attentuator was used to create the peak load, and then a safety factor on that so the car doesn't fail behind the attenuator. The main hoop load vectors come from the FIA F1 rules. Having worked on F1 roll hoops, I was familiar with those design cases and scaled them down to a 4g vertical load for an FSAE car mass. Maybe I can pull this togeher and include it in the sample frame spreadsheet so people will have some reference.

I am really curious to see how teams designs will do with these loads and see how the cars perform. This will be a learning experience for all of us.

Bill Riley

[oz_olly]

Bill, the reason I ask about the basis of the load cases is that I am in the process of developing a load case document for our team. It mainly addresses safety critical design loads such as harness attachment loads, wheel loads, brake pedal load etc. I would like to reference to the new frame design rules for chassis design. I believe our team has sent a letter to the FSAE-A organisers asking if they will be using the new rules. We would really like to use them if they are. I imagine a key factor for each rules committee is having adequately competent volunteers to review the analysis submissions.

[SNasello]

Will teams that are not competing in Formula SAE in the USA also be able to use the alternative frame rules to design their frame? In particular teams participating at FSG/FSUK/FSItaly?

[Bill Riley]

Each competition makes their own supplemental rules and decides what to incorporate in them so I can't comment on whether other events will be using the Alternative Frame Rules.

The bad core properties was a copy/paste error on my part and I have revised and reposted the sample frame spreadsheet. I've included a change log so people don't have to hunt through it for changes.

[The AFX Master]

Bill, altough we are not planning to build an AF just now, i'm working in a composite or metal/hibrid (budget restraints will decide) monocoque chassis.

My available FEA Resources: SolidWorks Simulation, and Ansys (NASTRAN would be a hassle to obtain in my university since it's mainly Solidworks/ANSYS biased). I see that SAE requests the sample problem to be solved in whatever FEA package suits the team.

But, rules specify:

"AF5.2 A Nastran analysis deck must be submitted electronically with the SRCF and supporting documentation."

That means that NASTRAN is the only way to validate the AF frame?. What's the purpose of such duality on the rules?

[billywight]

AFX,

You can output a Nastran analysis deck from most solvers. Not certain about Ansys, but I know SolidWorks does (not that I'd recommend using SW for any advanced analysis though). If you're looking for a free (to FSAE teams) and very powerful FEM package, have a look at Altair Hyperworks.

[Bill Riley]

billywight is correct. I picked Nastran decks because it was the closest thing to 'universal' I could come up with. Some of the neutral formats were logical, but based what most packages can output and what we can read in Nastran was the clear winner. Also, Nastran is straight forward enough and well documented enough that even if your code didn't output a Nastran deck, which would be strange, you could manually create a Nastran deck for almost any geometry in a text editor of your choice. So that is why we need a Nastran deck but you can use whatever you want (that gives good correlated answers). One last thing, it doesn't have to be a commercial code because I wrote a Matlab FEA code in school, and if you wanted to use your own code for the analysis you could. That's why we have the sample frame, to make sure you're not being led astray by the results.

[df_fsmb]

Unless our initial analysis is completely flawed, the structural requirements of the AF rules are not the same as the structural requirements of the basic set (SEF).

We FEA-d a frame which meets the requirements of the basic rules (superior alternative tube material for all kinds of tubes + all the required triangulations etc.) and it just failed completely, especially under the front bulkhead and impact attenuator tests.

So, it could be us, make no mistake, but it looks like we are sticking with the basic set as the AF rules based frame would have to be much heavier. It might be different at monocoque design, but we are not there yet http://fsae.com/groupee_common/emoticons/icon_smile.gif

[SHARP_SAAHIL]

Hi
everyone, Actually I just wondering what I have to send on 1st of November regarding AF.
Can anyone reply with less complex, easy to understand what actually I have to send them.?

[SHARP_SAAHIL]

If I fails to send that AF thing before 1st November, what will happen...... will they disqualify my team...??

[M3Shark]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by df_fsmb:
Unless our initial analysis is completely flawed, the structural requirements of the AF rules are not the same as the structural requirements of the basic set (SEF).

We FEA-d a frame which meets the requirements of the basic rules (superior alternative tube material for all kinds of tubes + all the required triangulations etc.) and it just failed completely, especially under the front bulkhead and impact attenuator tests.

So, it could be us, make no mistake, but it looks like we are sticking with the basic set as the AF rules based frame would have to be much heavier. It might be different at monocoque design, but we are not there yet http://fsae.com/groupee_common/emoticons/icon_smile.gif </div></BLOCKQUOTE>

We Came to the Same conclusion after testing multiple designs of an all steel(4130) chassis through the AF Rules. The designs meet the Standard rule book, but failed to meet the AF rules.

[Tickers]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by SNasello:
Will teams that are not competing in Formula SAE in the USA also be able to use the alternative frame rules to design their frame? In particular teams participating at FSG/FSUK/FSItaly? </div></BLOCKQUOTE>

Supplementary rules for FSUK state that cars built to the AF rules will not be allowed in class 1 for 2011. This should be something for any US team to consider if they plan on attending Silverstone.

[df_fsmb]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Tickers:
<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by SNasello:
Will teams that are not competing in Formula SAE in the USA also be able to use the alternative frame rules to design their frame? In particular teams participating at FSG/FSUK/FSItaly? </div></BLOCKQUOTE>

Supplementary rules for FSUK state that cars built to the AF rules will not be allowed in class 1 for 2011. This should be something for any US team to consider if they plan on attending Silverstone. </div></BLOCKQUOTE>

There are no 2011 supp rules for FSUK on www.formulastudent.com, (http://www.formulastudent.com,) so where did you find that out?

[Tickers]

Sorry for the extremely late response. The supplementary rules were given out at Learn to Win.

[Felippe Vieira - Fórmula FEI]

Anybody knows how to export a Nastran analysis deck from Ansys? We are in version 12.1.

[BeaverGuy]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Felippe Vieira - Fórmula FEI:
Anybody knows how to export a Nastran analysis deck from Ansys? We are in version 12.1. </div></BLOCKQUOTE>

You have to use FE modeler to export from ANSYS and must be done through the workbench Application. Instructions are in the Help file ir you search for Nastran. However,it only supports the mesh and doesn't take across boundary conditions or constraints. It also does not like exporting composite materials if you have used layered elements. Other programs like Hypermesh can convert ANSYS CDB files into Nastran format with some exceptions on the conversion of APDL commands. The best bet will probably be to find a person familiar with both programs and have him write a converter or find a converter online.

[JohnSmithy]

Would Bill Riley or anyone else like to elaborate more on the defintion of Failure, as listed in the AFR under AF3 definitions.

[Bill Riley]

First, let me clarify that teams that are approved and compete in the US events (Michigan/California) using the alternative frame rule will be eligible in other competitions. The other series may or may not allow you to design to the AFR without going to US events first.

The question on clarifying failure is a good one. The idea is there are different ways the structure could fail. For example, the main hoop braces could buckle, and this would be considered failure. Tensile, compressive or shear failure of any and all tubes is another situation that needs to be analyzed.

[Bill Riley]

If somebody will send me their Ansys model (via katklauz at aol.com) I will see if I can open it. If I can then you won't have to send a Nastran deck. If I can't you'll have to find a way to make one. Nastran is a pretty standard language and frankly you could manually convert your Ansys deck to Nastran if you had to in text editor, it would just be a big pain.

[Bill Riley]

Has anybody looked analyzed a monocoque using the AFR? I would like some feedback on how that is going...

[KeWLKaT]

Bill, I have been working around the clock on anaylizing our CF monocoque. Sadly, I cannot get it to pass requirements on one load case. I have the last loadcase with the off-axis "diagonal" front impact that I just cannot get to pass.

And I'm only looking at displacement for now, as I still have a couple of mechanical tests to do in order to find the failure values for my criterion.

I am using beam elements for the hoops and plates for the 2.5D model of the monocoque. For the plates, they are PCOMP and reflect the type of laminate we have been using regularly.

For instance, if I analyze our chassis from 2010 which passes the SEF from last year, it passes everything in the AFR, with the exception of that last loadcase with the off-axis load.

I mean, our strait up frontal load case passes with flying colors and very little deflection. But the fact that the last loadcase is off-axis, it creates a situation where the chassis deflects right in the cockpit area. And that area, no matter how much thickness of core (doubled, tripled the usual quantity), and no matter how many layers of UD fiber, I still get a minimum of around 70-80mm deflection.

I just find that last load case a bit too hardcore. I mean, it's a diagonal load to the front, and the boundary condition is all the way to the back of the main hoop. That creates about 60 to 80 inches of leverage with a very big amount of force.

Way too much in my opinion. If the load case's boundary conditions would include the bottom nodes of the front hoop as well, it would make a bit more logical sense to me.

Unless the rules change, our team will keep using the SEF, because there is no way around that off-axis load. I would be curious to see a team out there that can manage that load case, I just cannot picture how the geometry of a monocoque could be able to pass that.

[Bill Riley]

That case may be overly harsh. Can you take the 150 kN straight case and vary the angle and tell me at what angle your design passes? So for example to decompose the force for 10 deg apply Fx=147kN and Fy=26kN. If that works keep increasing the angle until you equal the displacement target. This way I can gauge what type of requirement would be acceptable. Stress will also need to be looked at.

[KeWLKaT]

Bill, I just ran a few tests, and I can get usually between 11 to 14 degrees before busting 25mm displacement, depending if I go from our normal laminate disposal (which includes a lot of different core thicknesses and ply number variations) to extreme reinforcement in the cockpit area. And as I said previously, I am not even looking at the stress here, just displacement.

The cockpit area is the problem area. Can you tell me why we do not fix the displacement at the bottom nodes of the front hoop as well?

If this is supposed to mimic an impact, why are we fixing the displacement on the main hoop only?

[Pete Marsh]

Would the harness mounts make a good boundary for these tests?
The driver is the bit we are most concerned about looking after, and the heaviest component. Wouldn't it make sense to bring loads into or out of the chassis from where the driver is attached?


Pete

[KeWLKaT]

The car's weight is technically normally around 50/50 on front and back with the driver. Given that the location of the suspension hardpoints vary a lot with each car, fixing the displacements for the front and main roll hoops would be the best option in my opinion for this certification program.

[KeWLKaT]

Bill, any chance to take a look?

[KeWLKaT]

So... What teams ended up using the AFR last year and had a good experience doing it? Did you run into any problems with particular load cases?