LocostUSA.com

BIG MOD EDIT:

Table of Contents, Pictures with analysis:

Book frame with mods and results


Table of Contents, Models:

Book frame in Grape by Andrew Moore
Frames with cages in Grape by Andrew Moore
Early version of Car9 in Grape by Marcus Barrow
Kurtis Kraft 500 frame in Grape by Andrew Moore

Original first post below:
==================


After re-reading cymtriks' document, and reading Wesley Linton's thesis for the first time, both found here under kitcaranalysis and final thesis respectively: http://locost7.info/mirror/chassis.php I'm wondering why there aren't many builds, if any that incorporate their ideas.

In cymtriks doc, he shows how to achieve 2683 ftlbs/deg of twist and even cuts the weight of the chassis by 7 lbs with these mods. Overall it only adds two tubes to the cut list. He states that the chassis in book form has 1180 ftlbs/deg and weighs 181 lbs.

In Linton's thesis he is modifying a Luego chassis. He goes from 1352 Nm/deg (997 ftlbs/deg) at 120 Kg (264 lbs) to 6030 Nm/deg (4447 ftlbs/deg) at 127 Kg (280 lbs). He even goes on to analyze a simple roll cage and returns with 9152 Nm/deg (6750 ftlbs/deg) at 170 Kg (374 lbs).

Of course all of these figures are with an FEA program, so the real world will be slightly off, but the idea the same. In Linton's he even states that actual baseline measurement of a real Leugo chassis was 1330 Nm/deg, and that his FEA program made the chassis heavier than needed as the intersecting parts of tubes were not removed.

I'm not entirely sure why the difference in chassis weight is there, but it may be that cymtriks wasn't including all the panels of the chassis, or something. The other good thing about both of these is that they give numbers for each change made, so if you only can do one or two things, you can see what kind of affect it will have.

Anyway, I just think that given the amount of time and effort these two have put into their writings, and I'm sure there are others I've missed. It's a little disappointing that most people aren't actually using half their ideas. Sure they don't provide you with a cut list, but isn't that half the fun?

EDIT: This isn't aimed at anyone in particular, just a general thread.

If anybody has a chassis they'd like to get real data on--not that I think FEA is unreal, it's just that I could build the test equipment easier than I could buy and learn the FEA software and I have more confidence in physical testing--they can bring it to the Kinetic shop. Make an appoinment, in case something else is riding the torsion tester, but I'm always willing to do knowledge swaps so you needn't bring your checkbook.

_________________
Locost builder and adventurer, and owner/operator of http://www.kineticvehicles.com

Does that include non-locost chassis? I'd love to see what the rigidity is of my Legacy.

Sure, if you can pick it up and put it on the fixture. You'll need a way to lock the suspension, too--on the Locosts I just replace the coilovers with a steel tube.

_________________
Locost builder and adventurer, and owner/operator of http://www.kineticvehicles.com

I agree with Jack. Real world measurements are more valuable than computer simulations. Neither of the examples given were done by profesional engineers, (a thesis presentation is done by a student, not an experienced, profesional engineer). They don't appear to have been verified with real models, nor are any differences explained.

I've seen lots of things fail in real use even after lots of engineering and lab testing. When you put the pedal to the metal, or air under your ass, don't trust that computer or inexperience will save you.

CJ

FEA model vs. real world accuracy done BY students has been done before, see the FSAE papers by cornell and a few other universities on the subjects. The models aren't even that hard to setup given the right program and model. Rigid beam models are often used because they make modeling simpler as well as dramatically reduce calculation time over full solid models.

At any rate, the numbers they're posting aren't unreasonable.. at 374lbs that one chassis is incredibly heavy, you can get 5-6k rigidity out of a 200lb chassis if you design it right.

ANYWAY, my only theory on this subject is that many of the people build locosts are not engineers, so they do not see the advantages of adding extra tubing to chassis that "already works", or even changing the tubing design.

I don't think that Jack was saying that FEA is less important.

Also, cymtriks is an engineer. I think he does structural analysis or something similar.

I'm not saying that one should expect the exact same results as FEA when building a chassis, but similar results would be an improvement.

The other thing I found out recently was that some of the older caterhams, and original sevens used some of these designs before any FEA was done.

I really don't want to take anything away from either person or their work, I really do think they've made a great contribution to the community. They've given food for thought, change this or that and what difference is present compared to the base model.
I guess, to get back to the original question, I agree why fix something if its not broke. Comparing the original Lotus, Westfield, Locost, DSK, the Aussie Clubmans, and even the newest Caterhams and CSR, there's allot of differences, and even more of choices.
Given the the Locost is by name and spirit, simple and cheap, changes don't seem to be made unless needed. Its works why make it harder except perhaps for racing to gain some advantage (against a Lotus 23 anyone?) or for higher powered engines. Another case seems to be when someone has the resources to make more work to meet some other goal...they want something fancier, stronger, more corrosion resistant, or just better than everyone else just because they can.
Variety is the spice of life, fancy stuff is nice so are potatos and beef, my car is kinda like ..um..stew or canadian stir fry... light, simple and aircooled with just a little rice made from whatever leftovers or nice things I happened upon or had.
CJ

From http://www.pistonheads.com/gassing/prof ... 5&t=173635
Nickname: cymtriks
Name: Chris Greatrix
Member Since: 10th November 2003
Total Posts: 1,777
Occupation: Engineer
Region: Gloucester
Country: United Kingdom
Notes: I've progressed from a Skoda Rapid, via another skoda and a Ford KA to a new Focus.
I'd like to build my own car some day.

I'm starting to think that this is going to be the best answer. It makes sense.

The best answer for whom? If you follow the "if it's not broke then don't fix it" mindset then you are basically accepting that whatever plans you are using are "good enough." What if car companies took this approach? How long could a car be sold if the model was never updated. If everyone had to design thier own car I'm sure many people would still be driving super old vehicles because they were simple, surely were not broke and handled "good enough." I took it from your first post that your current model might not be "good enough" for you. Maybe you want something that handles even better or just something different. I say feel free to step outside the box but do enough research, be it FEA or real models, to make sure it's safe. It will definetly take lots of time and more money but it may be worth it. It's great to see the advancement of locost and similar chassis design, so if you have the resources, go for it! I guess my point of the post was don't get stuck with a "good enough" mentality.

What I mean is that it's probably the best answer for my question of why there are very few locosts that deviate from book plans.

I definately don't have a "good enough" attitude.

I'm a compulsive improver; sometime to make things stronger/stiffer/faster/etceterer, and sometimes to make things not significantly worse while making them easier/cheaper. However, there are a lot of Locosts out there driving while mine's still improving. If you do significant mods, you'll be spending significant time doing them.

We got to the moon 'cause the head honcho (Werner von Braun) wouldn't let folks improve past what was needed to get the job done. One of his mottos was "Better is the enemy of good."

This hobby is satisfying to different folks for different reasons.

_________________
Locost builder and adventurer, and owner/operator of http://www.kineticvehicles.com

Their particular anysis may not be an earth shattering contribution to the Locost community, but I feel I do have to defend FEA for just a moment. FEA is employed by every major manufacterer in the world because it saves time and money.
We make rocket motors and spend months and months on FEA models and then validate them with test coupons. That way we get real-world data without the expense of a full-scale build. On a new rocket motor design we will only build 1 or sometimes 2 development cases before they are ready for flight. That is a multi million dollar cost savings!
I doubt anyone wants to spend 3 weeks+ building a chassis just to have Jack torque it in half because something went wrong and then turn around and add one or two more tubes and torque it again.
FEA allows you to modify a little bit at a time and hone in on your ideal design without even touching the welder.
For those that have the time, resources and ability I say analyize away and share with the rest of us! I'm always looking for more information to make informed decisions.
- If I ever see the light of day after my grad program I intend on doing a full FEA model of my chassis. (Of course by then I'lll probably have the majority of the chassis built anyway. I guess when it comes right down to it, engineers are just junk-yard warriors too, with bigger egos and a few more tools in the box (even if we don't always use them.))

For now I've taken the triangluation approach and made sure that every single member in the chassis has at least one triangle in it as well as installing a few out-of-plane braces. This will cost me a few pounds in the long run, but I'm looking at it like this...
It's a miata on a half-ton diet. - It's going to blow my socks off whether it's 900lbs lighter or 1100lbs lighter.

_________________
In God We Trust... All Others, Bring Data.

Miata based, custom chassis build in progress. First Drive!!! http://locostusa.com/forums/viewtopic.php?f=35&t=2269&start=375

If I knew how to do FEA and had the tools, I'd do FEA first and physical tests second. FEA could save me a huge amount of time because it would confirm that most of my ideas won't work without having to build and test. I'm not a Luddite; I'm merely an old dog with a finite set of tricks. Thus my Miata-based car is likely to weigh more than Rust_Bucket's Miata-based car, and may take me longer as well. Maybe when FEA for Dummies comes out I'll give it a go.

_________________
Locost builder and adventurer, and owner/operator of http://www.kineticvehicles.com