ALFAGTV6.COM

la_strega_nera · Post by **la_strega_nera** » Tue Mar 14, 2006 7:56 pm

Ahh yes, our wonderfull ADRs.
My understanding on roll cages door bars to meet ADRs is that the bar must run below the level of the seat squabs, so the trick to getting a more effective diagonal may be to have it engineered with a set of seats with deep sides to the base cushion (deep race buckets mounted high?). Theres a good diagram on the Victorian RTA website, I'm reasonably certain this carries across most states.

As far as the angle on the diagonals, I had forgotten about that part of the CAMS regs, but between lowering the mountings of the diagonals to the verticals of the main hoop, and pushing the mounting points further back (through the firewall to the rear shock mount area) you should be able to get an acceptable angle? (I don't have a Conspiracy Against Motorsport manual infront of me, at work, and been a while since i looked at caging my GTV). My concern is that the top of the rear arch is not a particularly good structural point of the shell to feed the loads into.

I intend to half cage my ford with low-ish door diagonals to avoid the RTA hassles with head intrusion (And ease of ingress/egress. Had a ride in a v8 super and getting in sucks the big one), but its a 4 door and will have no back seat... mainly looking for improved torsional rigidity and some protection for sprinting/trackdays. Might tie the 3rd trailing arm into the rear cross braces too.

Ben

Daniel · Post by **Daniel** » Wed Mar 15, 2006 1:30 am

Well here's what I've done.
Like I said, some compromises to keep it street legal but definitely strong.

The A pillar is tagged to the cage all the way down including the kick panels, the original seat belt points are also tagged to the cage and the dash bar is tagged to the firewall.
Double cross in the main hoop and single diagonal in the rear and roof planes.
Upper side intrusion ties the front and rear legs together, lower one ties the bottom of the main hoop to the front legs and to the rear seat support cross member.

So here's the pics (unfinished obviously) -

MD · Post by MD » Wed Mar 15, 2006 1:45 am

Looks tough Daniel. Doing anything under the bonnet and around the firewall ?

Maurizio · Post by **Maurizio** » Wed Mar 15, 2006 2:56 am

MD for stiffness this is not an ideal sulotion.
I am not sure it can be done. But the pipe works better when it moves from for example left side junction @tower bar (on the shock) to right side junction @cross bar under the dash. You drawing makes a virtual rotation point which produces a twisting moment on the bar behind the dash.
When designing for stiffness you strive for pull and push forces.

la_strega_nera · Post by **la_strega_nera** » Wed Mar 15, 2006 5:49 am

Maurizio wrote:MD for stiffness this is not an ideal sulotion.
I am not sure it can be done. But the pipe works better when it moves from for example left side junction @tower bar (on the shock) to right side junction @cross bar under the dash. You drawing makes a virtual rotation point which produces a twisting moment on the bar behind the dash.
When designing for stiffness you strive for pull and push forces.

Thats why I suggested tieing it into the firewall. Between a strut top bar, the bars running from the front uprights and the shock mounts and the two "export brace" (google image "export brace" and "mustang" and you'll se why i call it that) you can use the firewall as a shear plane, and while the diagonals are not ideal (unless you then tie back into the tunnel) they do add stiffness and triangulation that is otherwise completely lacking in the front end apart from the diagonal box sections that run from the A pillar to the wishbone mounts.

Daniel, nice cage. You would pick up some more stiffness with a second diagonal in the rear window plane instead of the second diagonal in the main hoop (This is based on some analysis an aquaintance performed while designing a tube frame car), and a running a tube from the base of the main hoop to the base of the rear plane tubes that pick up off the arches should help stabilize that area a bit better.
You're in Brisbane? who did your cage fab?

{EDIT: Just noticed the tubes running from the door bars to the arch tops, works well enough

}

Daniel · Post by **Daniel** » Wed Mar 15, 2006 1:29 pm

Ben
The same bloke that does all of the work on this car did the cage - me

I don't know that I'd notice too much more stiffness than this will give me, I'm still running this thing on the street and won't be using the kind of spring rates that will shake it to pieces. Double cross in the main hoop was more for piece of mind in safety than chassis stiffness and since this car was planned around no cage at all, I think this one will give a substantial increase in chassis stiffness anyway.
Mike
I was not going to run through the firewall, just some well placed gusseting and reinforcement. I don't want to consume engine bay space at the moment cause I still have to stuff that 3.0 24v in there.

MD · Post by MD » Wed Mar 15, 2006 1:34 pm

Maurizio

Don't know when it was the last time anyone has had a close look at the firewal / dashboard area but it would have to be one of the strongest combination of "panels" on these cars. Couple that with a cross bar between the A pillars so that the bar is welded to the "dashboard" panel, and effectively becomes part of the cage, this makes for a very rigid area.
If you then back brace as I have illustrated, you have a very strong triangulation with a compromise between fouling engine bay componets and a useful location.

Volvo is a renowned company for building tough cars. If it is good enough for Volvo to back brace to the firewall, it's good enough for me.

As for taking the A pillar bars right to the castor mount, well, it does no harm but it makes no material gain either as this section is well reinforced in that plane and is adequte for track use. So if it is adequate, it's adequate. It doesn't need more weight and overkill.

Ben,

As the saying goes, a picture is worth a 1,000 words. I am not getting a clear enough picture of what you are saying and so I hesitate to make a comment but as a generalisation, I agree where I beleive you are coming from. Perhaps a scribble sketch off a scanner?

Cheers,

MD

Mats · Post by **Mats** » Wed Mar 15, 2006 1:51 pm

MD wrote: Volvo is a renowned company for building tough cars. If it is good enough for Volvo to back brace to the firewall, it's good enough for me.

Huh?

MD · Post by MD » Wed Mar 15, 2006 2:10 pm

OK matts I think I know what you mean by huh?

Using the name Volvo was too generic. I am actually not acquainted with Volvo's global production variants. What I should have said was my wife's Australian import spec 1983 240 GL sedan had back braces to the firewall. I assumed that this was Volvo general policy but I concede that assumption was just not thought through. Maybe this was just a thing they did for the Oz market for the condition of our roads back then, who knows.

As for the claim of toughness, well, I have never seen a passenger sedan before or since that model that was better built (mechanical's aside)

I hope that clarifies it. If it doesn't, I am sure you will let me know.

Mats · Post by **Mats** » Thu Mar 16, 2006 1:18 am

Never seen it on a 240 but both 7 series and 9 series had them.

To me it's a signal they didn't manage to do it right in the sheet metal design and it's a band-aid solution.

You never see tham on production cars nowadays, too costly to mount on the assebly line methinks and probably a bitch to simulate in the crash calculations.

la_strega_nera · Post by **la_strega_nera** » Thu Mar 16, 2006 1:55 pm

MD wrote:Ben,

As the saying goes, a picture is worth a 1,000 words. I am not getting a clear enough picture of what you are saying and so I hesitate to make a comment but as a generalisation, I agree where I beleive you are coming from. Perhaps a scribble sketch off a scanner?

Cheers,

MD

Which bit do you want me to clarify?

Micke · Post by **Micke** » Thu Mar 16, 2006 2:21 pm

What do you do when you want to optimize the roll cage but don't have access to a good FEM program?

Answer here:

la_strega_nera · Post by **la_strega_nera** » Thu Mar 16, 2006 8:48 pm

Micke wrote:What do you do when you want to optimize the roll cage but don't have access to a good FEM program?

Answer here:

Cool
I've not used this software, have access to better tools, but i know of at least one guy who's used it, and its freeware.
http://www.grapesoftware.mb.ca/

Mats · Post by **Mats** » Fri Mar 17, 2006 1:36 am

Usually the problems with FEM is that you need to know how to mesh the model and how to interpret the data. You can get shitloads of data out of it but if you don't know how to read it it's useless.

Mickes hands-on approach is very attractive because you can get a good feel for how the parts react when you do changes. Difficult to get strasslevels though but if you design to Rp 0.2 in a cage I'd say you need to think things over.

Edit: ?Syntax Error

Maurizio · Post by **Maurizio** » Fri Mar 17, 2006 2:07 am

@Micke

just plain brilliant

@Mats, so true, do alot of fem work, mostly very thin sheet (<1 mm) and it is very difficult to get the right/reliable data out of fem.
Standard hand calculations (roarks) is much more reliable.
But fem creates nice pictures to convince the project leader

To max the cage to Rp0.2 is the best cage in stifness/weight ratio