complexx

So the largest deflection is 0.147?

kurt M

This is most defiantly not the GR thread. Not that the GR thread was not many hours of guilty entertainment...

If I am reading things right 12Klb load deformed the flat X .8xx and the pyramid X .1xx. If this is true then this simulation would point to the pyramid X being a bit better in a broadside. Granted this is not a large amount of movment but with the pyramid X starting further away from you along with better numbersl. One thing to note it the largest deformation of this pyramid X is right near the drivers side. Might be worth it to build with this tube being thicker than the other 3.

bruinbro

Yep. I should note that the deflection numbers are for total magnitude, not just in the Z direction. It's a snap to generate a chart with delections in a particular direction, just time consuming to port them into this forum. Do let me know if there is something particular you would like to see.

Bro

bruinbro

Actually, the flat X had a max deflection of 0.817" and the pyramid X had a max deflection of 0.147".

Good thinking about multiple thickness tubes. Again, I wouldn't put any credence in the absolute numbers as there is a lot of other factors involved in replicating a real crash. I think these numbers are fine, though, for comparison between different designs.

Another note, you will notice that the load is distributed over a fairly large portion of the X. I did this as I didn't know how much of a door typically gets caved in. If you guys want a different loading scenario, please let me know. I'm unemployed right now and have some time to devote to this in between scrambling to find a job and trying to survive. A lot of my time is spent job hunting and related activities on the web, so I will usually be available throughout the day.

Bro

complexx

Would it be possible to runt he analysis with some sheet metal gussets like in some of the pictures previously posted in this thread? A gusset in each of the corners of the center of the X.

If I had to guess what the thickness of the sheet is, I'd say, maybe .050

bruinbro

Check back later this evening (maybe in an hour or so).

Bro

complexx

Yay! I'll check in the morning, its time for bed

bruinbro

Here's the analysis for the gusseted basic X design. I'm still scratching my head...

Basic model



Broadside crash



Broadside crash side view



Head-on crash point load



Head-on crash distributed load



Thoughts?

Bro

Bill935K3

Very interesting!!
I agree that distributed load is basicly a waste of time. In a front end crash bars from crush zone and suspension pickup points will transfer to points.
One thing I think you might be missing in your model is bottom box beam? (see photo) At least in my 75 911S tub this is a massive if fairly thin walled structure.

Geo

I quite agree. I was just making note that there is no post limit to a thread. In fact, once could argue the GR thread proves there are no limits of any sort to a thread.

kurt M

Was this the flat X? If so there is little change over the non reinforced model. If not then
we got some head scratching for sure. Was this a single plate reinforcment? I wonder what a Taco type install would modle like. I would expect it to move the strain and its result of out to the ends of the tacos. End up with a truncated cone shape.

Geo

I would also run the scenarios with a relatively small load distribution at an angle for the different cages. What this would represent is in an angled crash the corner of a car would be what would contact the structure. Also, try this with the impact being off-center of the X.

complexx

Try the gusseted X brace, but bend the X brace outward like in one of the previous analysis. I think it was the second one you posted. if your feeling like a special person, you could also try to the same analysis with a .100 gusset. This would be similar to what a taco gusset would do as opposed to a plane sheet gusset.

M758

Interesting indeed.

Part of the reason for the plate can be seen in the stress numbers. All that displacement causes a lot of stress. Could infact yeild things. The plate helps take the load some.

I also agree that to be effective you need to figure in some thing for the car structure. The head on impacts are not realistic since if the impact gets to those forward bars my feet & legs are NOT going to like it anyway. Also the two verticals really should be tied in at the top. That will give the upper points a place to be.

Nice work by the way. Hey just like in the real world once the model is done all the real work begins.

complexx

I would think that in this scenario the deformation would be reduced compared to with just a plain X bracing. Perhaps check your settings again to make sure the proper thickness materials were used.

Perhaps its just the sheet that is deforming to that extent?