This thread got mentioned, so it may as well get bumped up again.

Reminder: we are talking about mods to a street car (for racing a cage is pretty obvious).

Sherwood had said above "Some of the above can be accomplished on a finished tub."

My question is which items (seam welding and adding stiffening plates) will damage the paint or require removal of undercoating first on a finished tub?

Randy,
From the information I received about the sill plate reinforcements, the sheet metal is added under the carpeting up to the door weatherstripping. Thus, an installation would conceivably be invisible once carpet and rubber are in place. A call to Eurotek would confirm that.

Seam or stitch welding, on the other hand, requires prepping down to bare metal, then refinishing.

Sherwood

Re torsional stiffness data and CG heights (which, indeed, are very hard to come by):

Ludvigsen, p1381:

"Finite-element modeling (FEM) was vital to tha comprehension of the structure that saved materials and weight while at the same time increasing the body's torsional stiffness by more than 30 percent over the 993; the FEM computer model of the 996 body had a staggering 180,000 separate analysis elements. An important conribution to strength was made by bonded-in front and rear glazing, which accounted for 21 percent of the torsional stiffness increase. Making a similar contribution were the underbody diagonals that bridged the baps in the cody where the suspension subframes were inserted. The complete body's torsional stiffness was 14,800 lb-ft per degree of twist."

From that one can work backward to the 993. In notes I have from something authoritative - possibly the factory website when the 996 was new - I wrote 22,000 lb-ft per degree for the 996, but alas I didn't note the source. Quite a difference. Maybe that was for the 997?

Also from somewhere: the 936 was 3,165 lbs-ft/degree (which would be in line with the Donahue quote of 2,000).

One of the standard texts shows a picture similar to the one posted, with a bunch of guys standing on a long beam attached across the front of a car. Might not be too hard to do on a 911 (pull the front suspension, bolt the beam into the suspension mounts) , but how would you hold the opposite rear side down? Need to do it on a frame bench? It would be fun to try this. Once you got things set up you would measure at a bunch of points along the side to see where it was twisting.

Ludvigsen gives the CG of a 956 as 17.3". He gives that of a 914 ass 19.7. My notes give (911=~20.7?), for what that is worth.

No one I know has actually measured the CG height on their 911. It is kind of complicated (lock the front suspension, for instance). When I had my car on scales I jacked the rear up as high as it would go (but didn't lock the suspension) and wrote down the height increase and what the front weighed. When I ran the results through a program I grabbed off the web a couple of years later I got a garbage result, which I attributed to my garbage in.

Walt Fricke

Walt a picture would be helpful or some type of explanation of "bonded-in glazing". I think a more ridgid body has advantages over just stiffer bars and sway bars...lighter too.

Glued windshield and rear window I guess.

"Underbody diagonals"? Does anybody have a pic of these and can they be retrofitted (i.e. McGyver'd) to an earlier chassis?

Tyson Schmidt's photo, Page 6 of this thread. "X" bracing welded to bridge the void/hump at the steering rack. bolt-in

Bracing/plates between the rear torsion tube and the hump (between the seat dimples) above the transmission. Weld-in

"A" pillar kick panel bridging - ala targa reinforcements. Weld-in

Torsional Stiffness Data:
Ford GT-40 Mark I – 12,500 lb-ft/degree
VW 1500 – 7,000 lb-ft/degree
Rolls-Royce Silver Shadow – 10,800 lb-ft/degree (wheelbase almost 10 ft long)
-- from p. 137 of Campbell, Colin. 1973. DESIGN OF RACING SPORTS CARS. Robert Bentley, Cambridge, MA.

Randy,
Thanks for the additional info. Here's an updated chassis stiffness chart:

http://forums.pelicanparts.com/uploa...1165817778.jpg

Sherwood

Nice.

Add a column for year of manf. - then make a scatter chart -- might be revealing...

This looks great, but al least on my car the master brake cylinder is in the way making it impossible.
Also the under body swaybay is somewhat in the way.
Maybe it would be possible to strengthen the steped design belly pan instead.

Edit: my car has the aluminum crossmember and no brake servo.

Sherwood, according to your list the VW is one of the most ridgid cars made...my experience with a '84 GTI indicates otherwise. The car had a real body twist that allowed the rear wheel to rise on hard cornering. I have seen VWs with a complete cage eliminate this problem.

I would have to believe that pre 80s cars with seam rain gutters would have to be less ridgid.

The reason this is happening is because your inner rocker panels are rusted.. almost certain. You need to replace them by removing the outer rockers. Cut all welds and remove.. its real easy! (that was a joke).

I did a brief search. Here's a compilation from Japanese Car Life:
http://www.carlife.net/board/read.html?listcode=42797&page=4&brdTable=brd_free

Rolls Royce Phantom : 40,500 nm/deg
Koenigsseg CC : 28,100 nm/deg
Lamborghini Gallardo: 23,000 Nm/deg
Lamborghini Murcielago : 20,000 nm/deg
McLaren F1 : 13,500 Nm/deg
Pagani Zonda S : 26,300 nm/deg
Ferrari 360 Modena : 23,000 nm/deg
Ferrari 360 Spider : 8,500 Nm/deg
Ferrari 430 : 27600 nm/deg (supposedly 20% higher than 360)
Ferrari F50 : 34,600 Nm/deg
Aston Martin DB9 Coupe : 27,000 Nm/deg
Aston Martin DB9 Convertible : 15,500 Nm/deg
Lotus SID : 16,000 nm/deg
Lotus Elise 111S : 11,000 nm/deg
Lotus Elise S2 Exige (2004): 10,500 Nm/deg
Porsche 959 : 16,000 nm/deg
Porsche 911 Turbo(2000) : 13,500 Nm/deg
Porsche 911 Turbo 996 : 27,000 Nm/deg
Porsche 911 Turbo 996 Convertible: 11,600 Nm/deg
Jaguar XJ : 21,700 nm/deg
Jaguar X-type : 16,319 nm/deg
Chrysler Crossfire : 20,140 nm/deg
Dodge Viper Coupe : 7,600 Nm/deg
Volvo S60 : 20,000 nm/deg
Volvo S80 : 18,600 Nm/deg
Audi A2 : 11900 Nm/deg
Audi A8 : 25,000 Nm/deg
Audi A8 (2004) : 36,000 Nm/deg
Audi TT : 10,000 Nm/deg (22Hz)
Audi TT Coupe : 19,000 nm/deg
BMW Mini : 24,500 nm/deg
BMW E36 Z3 : 5,600 Nm/deg
BMW E46 : 18,000 nm/deg
BMW E90 : 22,500 nm/deg
BMW Z4: 14,500 Nm/deg
BMW Z4 M Coupe : 32,000 Nm/deg (BMW AG's press)
Volkswagen Fox: 17,941 Nm/deg
Volkswagen Golf V GTI : 25,000 Nm/deg
Volkswagen Phaeton : 37,000 Nm/deg
Ford GT40 MK1 : 29,000 nm/deg
Ford GT: 27,100 Nm/deg
Ford Mustang 2003 : 16,000 Nm/deg
Ford Mustang 2005 : 21,000 Nm/deg
Ford Mustang Convertible (2003) : 4,800 Nm/deg
Ford Mustang Convertible (2005) : 9,500 Nm/deg
Mazda Rx-8: 30,000 Nm/deg (hard to believe)
Mazda Rx-7: ~15,000 Nm/deg
Alfa Romeo 159 : 31,400 nm/deg
Renault Sport Spider: 10,000 Nm/degree
saab 9-3 22,000Nm/degree

I'll compare with the numbers on my chart and revise if needed.

Sherwood

The Mustang Conv. is worst with 4,800.

Audi A8 (2004), Volkswagen Phaeton, Rolls Royce Phantom are best, all beating F and P cars...


PS - I modified that Car Life data to put the numbers into Excel as numbers, not text. Let me know if you want the modified file to run stats on or characterize the unit bodies by steel, AL, CF, etc.

It would interesting to compare several varieites of internal bracing such as Harness bar, roll bar and cage. Obviously the cage is much stronger but also adds eight and hurts street usability...I would thik something as simple as a harness bar would compensate for a sunroof or older seam gutters. Also measuring camber change with and without suspension support welding and cross braces would be very informative.

Randy,
That would be great, but what are the advantages in converting the data to Excel? I trust you. Go for it.

Sherwood

For 914's Engman makes a kit for the inner long to stiffen up the chassis. Here is a couple pics of my installation. Weighs 18lbs and is undetectable under the carpets. Dramatic improvment once installed.

http://forums.pelicanparts.com/uploa...1168469703.jpg
http://forums.pelicanparts.com/uploa...1168469713.jpg
http://forums.pelicanparts.com/uploa...1168469722.jpg

Travis, is there a "Kit" for the 911 also...I guess a couple of planels could be made but a precut kit would be much easier.

There are no kits for 911's. This particular kit was 18G steel and could be made DIY. I will send PM to the guy who makes the kit to see what he thinks.