ADD
|
 posted on 14/6/05 at 02:07 PM |
|
|
Building in weakness
Hi All,
After many many months off I am now back in the garage building my IRS grenada based locost. I happened to see mention of the stiffness of the book
chasis around the place so have decided to have a go at building in a bit more rigidity. After reading many peoples comments on the subject I am
sorted, however I am now a bit concered that in the event of an accident I may be building my own coffin. Has any one got any ideas on how to build a
stiff chasis without compromising on the cars 'weak points' and therefore accident safety?
|
|
|
|
|
scoobyis2cool
|
| posted on 14/6/05 at 02:34 PM |
|
|
I'm not sure there's enough space around the driver to accommodate weak points - any buckling and the chassis rails will end up crushing
you. I suppose you might be able to allow the front to crumple but you don't want it to crumple enough to push the engine back into you. The
sides are so thin you wouldn't want them bendiing at all, and you'll have a fuel tank at the rear which you don't want getting
punctured. All in all I'd say you'd be better off building it as strong as possible.
I suppose the best safety tip is to take it easy and try not to get into a crash in the first place! 
Pete
It's not that I'm lazy, it's that I just don't care...
|
|
|
carnut
|
| posted on 14/6/05 at 02:54 PM |
|
|
I wouldnt be to worried from making the chassis stiffer.
If you had a side impact you are well within the wheel line. Suspension is likely to absorb a lot unless you hit a tree or get t boned by a bike.
If your realy concerned id recomend having a cage. If designed properly would also add a bit of stiffness.
|
|
|
niceperson709
|
| posted on 15/6/05 at 01:08 AM |
|
|
Hi there
down here in Oz we are obliged to make our Chassis stiffer than the book design and no one worries about this with regaurd to saftey for the occupants
your bigest worry will be from side impacts .........
Extra rigidity will give you pay offs in the way the car preforms and there will still be enough progressive deformation in the front structure for a
head on collision.
best wishes
Iain
Best wishes IAIN
life is not the rehearsal , it's the show so don't sit there thinking about it DO IT NOW
http://iainseven.wordpress.com/
|
|
|
ADD
|
| posted on 15/6/05 at 07:26 PM |
|
|
In the same vein, Does anyone know how to work out the tortional stifness of a chassis? There must be some software out there somewhere.
|
|
|
MikeRJ
|
| posted on 15/6/05 at 09:09 PM |
|
|
quote:
Originally posted by ADD
In the same vein, Does anyone know how to work out the tortional stifness of a chassis? There must be some software out there somewhere.
Yes, and very expensive it is too! The technique used is called finite element analysis (FEA) and you really need to know what you are doing to be
able to get meaningfull results. Have a search for Cymtriks posts, he uses FEA in his job and has done quite a lot of work on the Locost chassis and
ways of improving it. The stock chassis is not very torsionaly stiff at all compared to modern cars.
|
|
|
cymtriks
|
| posted on 15/6/05 at 09:15 PM |
|
|
quote:
Originally posted by ADD
In the same vein, Does anyone know how to work out the tortional stifness of a chassis? There must be some software out there somewhere.
I used Nastran for all my analysis which is posted on this forum and on the locost7.info (remember to type in the "7" as there is another
site by this name!) site in the files section.
MSC/Nastran is the works and costs a fortune. However there are much cheaper FE packages out there. You could try Framework at
http://home.wanadoo.nl/wolsink/ or SLFFEA at http://www.geocities.com/Athens/2099/slffea.html or FElt at http://felt.sourceforge.net/ as these
three are free!
Another package is Grape which I think you will need to pay for.
I can supply a Nastran input file for a Locost chassis if you want one.
|
|
|
ADD
|
| posted on 16/6/05 at 08:54 AM |
|
|
Hi cymtriks, I have read much about you in all the various threads about tortional analysis. I have also looked at your chassis mods. Can I ask, why
the Y braces? what kind of benifit do these have on the twisting of the chassis? I am thinking about beefing up the transmision tunnel some more any
thoughts?
I would also like to take you up on your offer for a Nastran file please. I have just downloaded your suggested free software so I will start having a
go.
Thanks for your help.
Adam
|
|
|
cymtriks
|
| posted on 16/6/05 at 08:37 PM |
|
|
Why the Y??
The double Y braces give a bit more room than the book tube R for the engine ancilliaries.
Using two Y braces is nearly as good as using two diagonals, R tubes, one on each side of the engine bay as in a Caterham and is much better than the
short diagonals as used on the Westfield V8 chassis.
If you can fit in a simple diagonal then this is best but maqny builders report that the engine ancilliaries get in the way towards the back of the
engine bay.
|
|
|
JonBowden
|
| posted on 17/6/05 at 08:26 AM |
|
|
quote:
Another package is Grape which I think you will need to pay for
You can download an evaluation copy free - I've downloaded it.
Trouble is I haven't figured out how to use it yet
Jon
|
|
|
AustinMTB
|
| posted on 30/6/05 at 04:38 AM |
|
|
quote:
Originally posted by cymtriks
...
MSC/Nastran is the works and costs a fortune. However there are much cheaper FE packages out there. You could try Framework at
http://home.wanadoo.nl/wolsink/ or SLFFEA at http://www.geocities.com/Athens/2099/slffea.html or FElt at http://felt.sourceforge.net/ as these
three are free!
...
I can supply a Nastran input file for a Locost chassis if you want one.
I'd appreciate a copy of your Nastran input file to learn from, please.
I compiled SLFFEA for windows and am working up a beam-based model that is "close" to the chassis. However, I'm not sure if everything
is set up correctly since FEA isn't my thing.
Here's the SLFFEA model so far with a torque applied to the front top beam:
[Edited on 30/6/05 by AustinMTB]
 
Rescued attachment fea_model1.png
Rebuilding a 1983 Jeep CJ-7 currently; engine swap to EFI.
Playing with the idea of a Locost.
|
|
|
stressy
|
| posted on 30/6/05 at 06:38 AM |
|
|
A really good challenge is to try and do the calcs in MS EXCEL or similar, after all The lambo spaceframes were all done by hand without the advances
of computer aided enineering at all ................
WHO DARES SPINS
|
|
|
ADD
|
| posted on 30/6/05 at 09:12 AM |
|
|
In all the stress images I have seen noone seems to include rear suspension hangers in their pics, is there a reason for this? I would have thought
that there was a lot of 'twist' occuring in this part of the chassis. Am I right or wrong????
|
|
|
niceperson709
|
| posted on 30/6/05 at 09:27 AM |
|
|
The only valid way to analize the rigidity of the chassis must include all of the suspension components . Here in Oz our torsion test replaces the
shocks wit solid struts and feeds the force into the chassis via a beam with a fulcrum in its center and conection to the front hubs any other aproch
would not replicate the forces the chassis will face in the real world see the attached picture .you will see the rear secured to the floor and the
front beam with force applied at the end dial gauges along the sides measure deflection . any computer simulation should be aiming to replicate
aspects of an actual torsion test to be any value in development of the design.
Best wishes
IainImage deleted by owner
Best wishes IAIN
life is not the rehearsal , it's the show so don't sit there thinking about it DO IT NOW
http://iainseven.wordpress.com/
|
|
|
NS Dev
|
| posted on 6/7/05 at 03:28 PM |
|
|
was going to add, bu**er the computer simulation, first build a jig to verify the stiffness of the chassis as std (just as pictured above) only then
can you relate this back to the theoretical stiffness and then base modifications from here.
|
|
|
niceperson709
|
| posted on 6/7/05 at 09:52 PM |
|
|
NS DEV SAID
"was going to add, bu**er the computer simulation, first build a jig to verify the stiffness of the chassis as std (just as pictured above) only
then can you relate this back to the theoretical stiffness and then base modifications from here."
To right Mate we get quite a few keyboard jockeys who keep trying to re invent the virtual wheel here I've lost count of the number of 3D
models of a seven posted here or sugested , If they spent more time building rather than tinkering in cyberspace ...........
best wishes
Iain   
Best wishes IAIN
life is not the rehearsal , it's the show so don't sit there thinking about it DO IT NOW
http://iainseven.wordpress.com/
|
|
|
rick q
|
| posted on 7/7/05 at 12:41 AM |
|
|
niceperson709 - where do you hail from? I'm in Buderim
|
|
|
AustinMTB
|
| posted on 7/7/05 at 06:15 AM |
|
|
quote:
Originally posted by niceperson709
NS DEV SAID
"was going to add, bu**er the computer simulation, first build a jig to verify the stiffness of the chassis as std (just as pictured above) only
then can you relate this back to the theoretical stiffness and then base modifications from here."
To right Mate we get quite a few keyboard jockeys who keep trying to re invent the virtual wheel here I've lost count of the number of 3D
models of a seven posted here or sugested , If they spent more time building rather than tinkering in cyberspace ...........
best wishes
Iain
Looks like I touched a nerve in what appears to be the resident Luddites.
But, good points on how the chassis is tested. I'll spare you yet another 3d model picture and just say that the model is taking into
account loads through the suspension now. You'd be proud if not for the whole computer thing.
I can understand how hastily grabbing the keyboard and sketching some ideas can only lead to a better understanding of how something works, or in this
case, how something reacts to a force. Surely the designers -- scratch that, the builders of the new Airbus didn't waste any time with
planning or virtual tinkering. Such folly! Quit wasting time and just weld it up, Bub!
cheers,
-Brent
Rebuilding a 1983 Jeep CJ-7 currently; engine swap to EFI.
Playing with the idea of a Locost.
|
|
|
TheGecko
|
| posted on 7/7/05 at 10:42 AM |
|
|
Without wanting to insert myself into an argument, I'll have an each way bet
I can see valid arguments for both the virtual and physical approaches. As Iain says, at the end of the day, testing the actual chassis is the only
real test. However, if the test proves that it's 'wrong' then there's a lot of time and materials to throw in a skip. A few
iterations with a computer model before picking up the welding torch could save a lot in the end.
Brent, you mentioned in your first message that you had a copy of SLFFEA compiled for Windows. Are you interested in sharing it or is it too big to
distribute easily? I have Grape (free version) but SLFFEA looks fairly good and I don't have a Linux environment at present.
Best regards to all,
Dominic
(also in Brisbane, Australia)
|
|
|
AustinMTB
|
| posted on 8/7/05 at 04:57 AM |
|
|
quote:
Originally posted by TheGecko
Without wanting to insert myself into an argument, I'll have an each way bet
...
As with everything, balance is key.
quote:
Brent, you mentioned in your first message that you had a copy of SLFFEA compiled for Windows. Are you interested in sharing it or is it too big to
distribute easily? I have Grape (free version) but SLFFEA looks fairly good and I don't have a Linux environment at present.
Sure thing, but no guarantees. The entire SLFFEA package with executables in the bin directory is 2.0MB large, and can be fetched at
http://relro.net/slffea-13.zip
Funny thing is, that was the whole point of my first post although looking at it now, I didn't make that very clear.
cheers,
-Brent
Rebuilding a 1983 Jeep CJ-7 currently; engine swap to EFI.
Playing with the idea of a Locost.
|
|
|
