LocostBuilders - powered by XMB

MNR Cad Model

ffrgtm - 26/7/12 at 09:42 AM

Slowly making some progress....

loggyboy - 26/7/12 at 11:14 AM

Ooo copyright..... lol (joke)

designer - 26/7/12 at 12:35 PM

That's really good. But why?

loggyboy - 26/7/12 at 01:03 PM

quote:
Originally posted by designer
That's really good. But why?

To be fair I did exactly the same (well started) when I bought the striker, just to have as a record and as an 'exercise' for me.

D Beddows - 26/7/12 at 01:27 PM

quote:

That's really good. But why?

I think everyone who can use CAD has done it at least once (including me) - you come to realise it's a completely pointless exercise really as it takes so long to measure (or get measurements for) parts and then model them (apart from the chassis) you may as well just build the car

. Which probably explains why so few people ever bother finishing their CAD models properly!

There are notable exceptions however and if you enjoy it then why not

ffrgtm - 26/7/12 at 06:02 PM

Thanks guys, and I'll agree it can be a huge waste of time... this chassis took forever with all of the tube bends.

I did it simply because I'm running fea to check the results of some frame tweaks... also going to play with the geometry. I have some software than can iterate pickup points to optimize lap simulation times... may as well use it!

ffrgtm - 2/8/12 at 11:48 AM

I've gone a bit further with my cad model...

I've put together a very basic torsional rigidity test using the fixed and loading conditions shown in the wireframe example. I currently have the frame modeled only with beam elements which I don't feel give a very accurate representation. I did my best to model the unusual weld beads that join the bent tubes in an x-configuration... but even normal welds are difficult to simulate.

I have most of the suspension modeled and will eventually run a more realistic simulation... as is the loads aren't entering the frame very accurately... nor are the real suspension structures being tested. Before I get to that point I would like to have the frame modeled with shell elements so I can include the original gussets, suspension tab sheet metal, and test ideas.

The results shown here are displacement, URES is nothing more than sqrt(x^2+y^2+z^2). I loaded each of the two points with 500N in opposite directions, while the fixed conditions in the rear allowed individual rotation but no translation.

If anyone has any specific or ideas or requests I'd love to hear them. Like I said... this first test isn't very accurate but it suggests room for improvement.

loggyboy - 2/8/12 at 12:34 PM

Rollcage bars look very skinny, if they are being used as part of the calcs?

ffrgtm - 2/8/12 at 12:35 PM

Heres one where I tried some different loading conditions and requested a deformed (exaggerated) result. It helps you visualize how to improve the design.

ffrgtm - 2/8/12 at 12:40 PM

quote:
Originally posted by loggyboy
Rollcage bars look very skinny, if they are being used as part of the calcs?

I measured them to be 38mm in diameter. I found msa regs that state either 38 or 45mm so I just used the wall thickness from the 38 spec. I might have perspective mode on in that photo as well... that could make it look a little funny.

Anyways, I'm not using the cage in the calcs yet because I have having a hard time getting the mesh where the cage is welded to the gussets on the frame. I hope to get to that point eventually... although I feel like some things could be done to better take advantage of the cage.

loggyboy - 2/8/12 at 12:43 PM

Main rollbar and diagnols should 45mm min IIRC.

ffrgtm - 2/8/12 at 12:48 PM

Crap I'll have to check my frame again.... while we're on the subject, you wouldn't happen to know which 1" diameter tubes are the lighter 18 gauge rather than 16 gauge would you?

[Edited on 2/8/12 by ffrgtm]

loggyboy - 2/8/12 at 01:24 PM

16 gauge is 1.5mm
18 gauge is 1.2mm

So the 18 is lighter (no idea why the higher number is the thinner?!)

http://en.wikipedia.org/wiki/Sheet_metal#Gauge

[Edited on 2-8-12 by loggyboy]

ffrgtm - 2/8/12 at 01:30 PM

I mean which tubes in the MNR chassis are the 16 gauge and which are the 18 gauge. MNR uses the 18 gauge in less critical spots to save weight, I just cant tell exactly which spots those are since they're the same OD.

Agriv8 - 2/8/12 at 02:38 PM

I should be able to get that info from MNR next time I am over.

I would also say the plates for the top front arms will strenthen the front, and the rear plate that the hoops bolt onto will also help.

but i am impressed you must have hours even days in the model.

ATB agriv8

ffrgtm - 2/8/12 at 03:33 PM

Thank you agrv8, and I appreciate the compliment! If you do find yourself over at their shop, could you please also snag the wall thickness of the smaller diameter (0.64 about) tube that they use for the front x?

And it's true that I have way too much time into this thing so far.... my CAD skills have gotten a lot better though so it's a "career skill investment"

I definitely do agree that I've left out some critical parts of the frame... to be honest I do have most of that stuff in my model, but getting it to mesh with beam elements is another story... I had to suppress them and just run it before I went insane. I pretty much had to start over from scratch and fully define all my sketches and treat everything as surfaces.

On a side note did you know that the $4000 cad workstation graphics cards ATI and Nvidia sell are exactly the same as their $200 models? The only difference is what drivers they allow you to install...

Solidworks is particularly stupid... I literally just changed the name of my graphics card in my system properties to the FireGL equivalent and models are running 10 times faster.

I need to sleep more often.

loggyboy - 2/8/12 at 03:45 PM

quote:
Originally posted by ffrgtm
I mean which tubes in the MNR chassis are the 16 gauge and which are the 18 gauge. MNR uses the 18 gauge in less critical spots to save weight, I just cant tell exactly which spots those are since they're the same OD.

Sorry I miss read your post! I thought you said which is lighter, not which ones are the lighter ones!

Agriv8 - 2/8/12 at 04:26 PM

Crap -hics drivers are the Whiches work, it will all be to do with Opengl support and other maths that the processor passes to the card to do rather than itself. but yes graphics cards and costs are a funny thing and confising matter.

The small tube at the front ( is it 12/13mm my chassis is to old ) is thick walled poss 2 mm maybe upto 4mm as I have used the tube for other things

I am sure Marc will be happy to assist as its always good to have indipendant maths backing up his own work. The Full Cage is also very structural if fitted !! regardless of the Double bend !!

Keep up the Good work and keep us updated with your analysis.

PS a couple of tube apears to have disapeared where the prop would travel to the Reverse box

ATB Agriv8

MK9R - 7/9/12 at 08:48 AM

Great work, very interesting

ffrgtm - 18/9/12 at 08:42 AM

Thank you

Unfortunately work is at a bit of a standstill right now while I develop the PDM for my team's FSAE car... I'm taking care to make sure I can print an extra pcb or two and use it for the MNR as well

ElmrPhD - 16/10/12 at 09:46 AM

Hey structural engineers...

My car-engined-car's RT+ chassis lacks the tube that runs from left to right under the gearbox, where the bell-housing ends (middle of transmission). Every other chassis I see on the 'net has something there. Should I be concerned?

Directly above that, where all others seem to have a permanent bar, I DO have a bar that I am to bolt in after the drive train is installed.

I do intend to beat the poo out of this car at the track, so I'm wondering if I need to add a brace (back) under the gearbox. Or does the presence of a factory full roll cage omit the need for that lower cross tube?

Thanks for any advice.

Cheers,
Steve, in the NLs

big_l - 31/10/12 at 12:38 AM

None of this takes into consideration the engine on a BEC us used as a stressed Member ?
Or the floor ?
Ps my 2012 chasis is slightly different to that also..
Very impressive though

big_l - 31/10/12 at 12:38 AM

None of this takes into consideration the engine on a BEC us used as a stressed Member ?
Or the floor ?
Ps my 2012 chasis is slightly different to that also..
Very impressive though

coyoteboy - 31/10/12 at 01:34 AM

Lack of translation on the rear elements will give a false stiffness, but it's a reasonable estimation. Nice work. Not sure I'd do it for a kit type car unless I were making mods, but who doesn't make mods?!

Bare - 31/10/12 at 02:46 AM

Good thing that contraption is "Screen Only"
The tube triangulation at the thing's front is sloppy/ foolish as it the top of the engine bay to upper dash hoop which in itself is in the wrong place to be useful structurally.
IF this is a model of some crap Kit Kar then the makers are Amateurs who really shouldn't be designing chassis.
Harsh? perhaps.. but this is real life and there are some serious precedents to copy.. if unknowing of the craft.
Caveat Emptor applies.. as always.

rost - 31/10/12 at 08:19 AM

quote:
Originally posted by Bare
Good thing that contraption is "Screen Only"
The tube triangulation at the thing's front is sloppy/ foolish as it the top of the engine bay to upper dash hoop which in itself is in the wrong place to be useful structurally.
IF this is a model of some crap Kit Kar then the makers are Amateurs who really shouldn't be designing chassis.
Harsh? perhaps.. but this is real life and there are some serious precedents to copy.. if unknowing of the craft.
Caveat Emptor applies.. as always.

Dude....

big_l - 31/10/12 at 10:56 AM

Tbh this thread is useless unless you know every inch of a Mnr chasis and also factor in engine is used for strength !

A lot if people know everything on the Internet if they were that good why arn't they doing it for a living ??

Not aimed at anyone but just need to take it with a pinch of salt !!

phelpsa - 31/10/12 at 11:14 AM

quote:
Originally posted by Bare
Good thing that contraption is "Screen Only"
The tube triangulation at the thing's front is sloppy/ foolish as it the top of the engine bay to upper dash hoop which in itself is in the wrong place to be useful structurally.
IF this is a model of some crap Kit Kar then the makers are Amateurs who really shouldn't be designing chassis.
Harsh? perhaps.. but this is real life and there are some serious precedents to copy.. if unknowing of the craft.
Caveat Emptor applies.. as always.

As with any form of engineering, it is all a compromise. Any fool can come on the internet and say it should be done differently without knowing the reasoning behind the compromises finally accepted.

Just wait until you see the rod ends in bending

marc n - 31/10/12 at 12:25 PM

quote:

Good thing that contraption is "Screen Only" The tube triangulation at the thing's front is sloppy/ foolish as it the top of the engine bay to upper dash hoop which in itself is in the wrong place to be useful structurally. IF this is a model of some crap Kit Kar then the makers are Amateurs who really shouldn't be designing chassis. Harsh? perhaps.. but this is real life and there are some serious precedents to copy.. if unknowing of the craft. Caveat Emptor applies.. as always.

an interesting excersise, but as already mentioned that frame isnt complete, there are quite a few tubes missing footwell diagonals tunnel bars rear suspension braces etc, as are all the gusset plates that are on a complete chassis i.e the steering rack which also braces the front end is missing, also the steel tunnel, the steering column mounts cage mount gussets etc etc, the engine (used as a stressed member ) is not in position with any of its mountings / tubes. i could carry on but the upshot is the screenshot represents a 75% complete chassis irrelavent of which cage / doorbars are fitted so if those where taken into account its about 60% complete !!

quote:

Just wait until you see the rod ends in bending

a compromise admittadley but thats why they are all specced bigger than necessary

if used in thier normal orientation i would be running 3/8ths all round

cheers
marc

ffrgtm - 1/12/12 at 08:42 AM

quote:
Originally posted by big_l
Tbh this thread is useless unless you know every inch of a Mnr chasis and also factor in engine is used for strength !

A lot if people know everything on the Internet if they were that good why arn't they doing it for a living ??

Not aimed at anyone but just need to take it with a pinch of salt !!

Oh don't worry I'm right there with you, FEA results in solidworks are generally garbage anyways. That's why you do "validation" afterwards. The one thing this method will allow you to do, is test different ideas and get real time feedback. The number won't match up, but the trends generally will.

That said it's been very difficult to model the areas where two sets of tubes are bent and have their contacting radii seam welded. Meshing was a nightmare. I still have difficulty imagining the reason that the chassis is done in this way... bending tubes and trying to make them line up like this is usually much harder than the vastly more common, and most likely structurally superior method of joining members at a single node with a single member bisecting another.

Marc, I'm sorry I don't have any good feedback for you yet... two engineering majors and an fsae car have put the MNR on the backburner