Structural Reinforcement

[911pcars]

When jacking up the car from the jackpost or from one corner, you may have noticed the door opening distorts (i.e. the window frame rubs on the top door sill - mine does anyway). Computer aided chassis design wasn't even a remote possibility when the basic 911 tub was created 40 years ago. Compared to current platforms, the torsional strength of a 911 could be improved. Upgrading to larger displacement and more powerful engines would be more effective if chassis stiffness was commensurate with the added torque and cornering forces created by ever-stickier rubber compounds.

If this can be done efficiently and effectively, the vehicle should handle a lot better as equipped even without the myriad number of aftermarket suspension upgrades.

Short of installing a welded cage structure, does anybody have any real world ideas on what it would take to increase the rigidity of the chassis w/o adding weight above and beyond what cage tubes and bracketry would weigh (~100 lbs?). I'm mindful of the reinforcement used by the factory in their racing RS and RSR chassis, but that was in conjunction with many welded tubes. I'm thinking more along the lines of the reinforcement pieces used to strengthen the 914 chassis and pieces added to 911 cabriolet conversions - improvements that don't necessarily compromise interior or storage space (e.g. hidden). Any thoughts?

Sherwood

Note: Moderators helped separate this from a related thread. Thanks

[RallyJon]

Sherwood, that should be a thread all it's own.

[Jim Sims]

"does anybody have any real world ideas on what it would take to increase the rigidity of the chassis w/o adding weight above and beyond what cage tubes and bracketry would weigh (~100 lbs?)."

If one didn't care about appearances, then high modulus prepreg carbon fiber sheet could be added at strategic locations onto the steel body panels (floor, roof, panel (below rear parcel shelf) ahead of engine, front firewall, etc.). It would be important to get the fiber oriented in optimum directions. Redesign/change of materials of the rear side windows and their mountings could turn them into shear panels, Systems of corner locking points (with appropriate local reinforcement) could also turn the doors, hood and engine hatch into structural shear panels.

None of this would look pretty except perhaps to a structural engineer but would utilize the "floating" body panels and be a better use of CF compared to the typical CF "Bling" one sees: steering wheel horn pads, dash fascia, electrical fuse panel/chassis covers, etc.

[Bill Verburg]

How's that?

My feeling on the subject is that if you need greater chassis stiffness then you also need a rollbar,

I think that even a bolted in bar, like the DAS in my 993, stiffens the car up. But have no hard evideance of that "feeling"

A welded in cage w/ gussets at the suspension pickups and stitched seams would certaiinly go much further in this regard w/ minimal weight consequences.

[jluetjen]

First off -- do you have a Targa or a Cab? Or Not? As far as I can tell the coupes are pretty stiff. The open topped cars on the other hand...

Mark Donohue quotes in his book "The Unfair Advantage" that the "early" 911 coupes were about 2000 ft-lbs/degree. Some of this was discussed on this earlier thread.

[davidppp]

Hmmmm.

To be ruthlessly truthful. I would strongly suspect bad repair work or rust if a coupe 911 shell distrted noticably when jacked.

Unfortuantely many cars I have seen have had new outer rockers tacked onto severely weakened inner sill structures which have either been poorly repaired, or simply left part-destroyed.

I have no twist data ( note, does anyone please), but have never seen noticable distortion in a nweish shell.

So far as Carbon is concerned, it will work, but its modulus is the same as steel of the same thickness, it is one tenth the weight, and hundreds of times the cost.

I'd use steel.

The weakest points IMHo are the front inner wings which not only tend to bend, but also are often rusted inside the boxes, the crossmember mounting bushes which are not very well welded into the structure IMHO, and the rear chassis box sections where the hot air tubes enter..

That said the 911 is avery strong shell provided all rust is replced propwerly....

And heavy too ..

Kind regards
David

[Cory M]

You could always seam weld the chassis.

[304065]

Sherwood, did you read the Up-fixen article by Bruce Anderson about vibration of the 914? They took the car to Honeywell and put a vibration transducer under the wheel, and then shook the car at various low frequencies. With a transducer on the car itself, they could measure how the structure resonated, and then relocated roll cage tubes to strategic points in order to dampen the vibrations. So the result was not merely stiffness, but harmonic "tuning" of the chassis itself, which also resulted in a quantitative improvement-- I think it lowered the lap times.

[Wil Ferch]

The 911, torsion bar coupe chassis is (still) remarkably stiff..on whatever design basis it was created from.

Like Bill V.... I would guess that even a bolt-in DAS roll bar ( rear section) is enough to increase torsional and maybe even longitudinal rigidity by a meaningful amount. I have a DAS in my 85 coupe . I've been toying with the idea of adding small tabs to the upper roll hoop near the B-pillars, to take advantage of the seat belt D-ring bolt ...and bolt the bar there too, as an additonal hard -mount point. The widely spaced places this bar bolts into the chassis must contribute to the 3D stiffness enhancement.

I don't have the numbers readily at hand about the torsional stiffness of the 911 body...but the value quoted here of 2000 ft- lbs/ degree seems artificially low. My recollection is that the number is closer to 6000-8000 ft-lbs / degree...maybe even 10,000 ft-lbs/degree. I'll continue to check my notes and resources and report back.

- Wil

[Jim Sims]

"it would take to increase the rigidity of the chassis w/o adding weight above and beyond what cage tubes and bracketry would weigh (~100 lbs?)."

"So far as Carbon is concerned, it will work, but its modulus is the same as steel of the same thickness, it is one tenth the weight, and hundreds of times the cost.

I'd use steel."

The question was a stiffness increase within a weight budget (100lbs). HM modulus carbon has much better specific stiffness (stiffness gained per pound of material) than steel. It doesn't cost 100's of times more than steel (except maybe in the government side of the aerospace industry where there is little cost pressure to be efficient).

[masraum]

I'll take a guess at what may add some stiffness through the middle of the car.

Take the center "tunnel" run in up really high in the cabin like in the CGT and with your newly modified really short shifter (a requirement after the previous mod) have at it.

[Wil Ferch]

Steve:
That might strengthen the longitudinal beam strength more than torsional stiffness, though...
- Wil

[Zeke]

Originally posted by davidppp
Hmmmm.

To be ruthlessly truthful. I would strongly suspect bad repair work or rust if a coupe 911 shell distrted noticably when jacked.

Not so. And Sherwood has a coupe, to answer the other question. Also, he is an engineer, so I don't take this question lightly. I shouldn't even reply. But.......

So, since I will be out classed, out trig'ed, out molecular'd, out everything here against youz guys, I will only offer that I think the area of the gas tank needs to be treated more like the 914 with a bulkhead from tower to tower and connected to the pan. The tub from the cowl to the rear window bulkhead looks good to me.

10 lbs. 90 left. Go for it.

[island911]

I'll throw in a couple pence worth:

1) using CF would be a mistake due to the large thermal expansion rate difference. (ever see a bi-metalic strip bend from a few degrees temp difference? ( common thermostats). . .IIRC CF/Steel will be much larger. Note: A common bimetallic pair consists of invar(iron-nickel alloy) and brass)

2) geometry will give the most bang per pound. The shell defines the outer most limits (best mechanical advantage) of any stiffening elements. . .. find those softer spots.

3) it seems that Ruf did some flush reinforcements on the A & B- pillars. (makes a "cage" look completely Fred Flintstone)

[randywebb]

I've never seen numbers for the 911 (nor on where the CG is either), so I hope you can find something, Wil.

Here is a tidbit I collected somewhere at some time...


Mark Donahue mentions in his book "The Unfair Advantage" that the 911's that they worked on flexed at a rate 2000 ft-lbs/degree, which he considered plenty stiff. Keep in mind what this means to your alignment. For every 2000 ft-lbs of torque you apply to the chassis, it will twist 1 degree. Imagaine that you have 500 lb springs on the front and that they are about 2 feet from the car's centerline. If you corner such that you compress the outside spring by 1 inch, this is 500 lbs at 2 feet or 2000 ft-lbs or torque. Add on a stiff roll bar and you most likely have even more torque being applied to the frame. Now your rear suspension is rolled 1+ degrees further then your front. This additional roll at the rear basically dilutes the effects of your sway bars. As a rule of thumb, your chassis stiffness should be about 20x stiffer then your spring stiffness if you want to keep it's contribution to the cars total movement to less then 5%.

[911pcars]

"........Also, he is an engineer, so I don't take this question lightly. I shouldn't even reply. But......"

Zeke,
Not so, but I stayed at a Howard Johnsons the other night.

Randy quotes Donahue mentioning a figure of 2000 ft.lbs/degree of flex. Let's use that as one reference point. Donahue was probably quoted in the mid-70s when he and Penske were setting up the RSRs for the IROC series. Compared to today's chassis, 2,000 ft.lbs/deg. may be relatively stiff.... in 1970. Compared with todays passenger cars, it isn't.

Donahue was probably referring to a factory stock 911 rather than one of the full cage RSRs which are impressively stiffer. I also remember Bobby Allison from a C&D mag road test complaining that the 911 chassis flexed so much at speed he couldn't find the gears using the infamous 915 shifter mechanism. I guess he was looking for something a little more positive.

By way of comparison, I did some research and came up with the following chart.



Using Mark Donahue's number of 2000 ft.lb/deg., one can see that 30 years marks a long continuum in terms of chassis design.

Stiffening the suspension to compensate for a relatively flexible chassis has its limits. Past a certain point, chassis flex and fatique will increase as more stiffness is added to the suspension and/or more loads are placed on the chassis via higher cornering forces, higher output engines and stickier tires.

I have my doubts whether a standard roll bar will have the desired effect. A full cage might be the quickest route to maximize torsional rigidity. Not sure it's the most cost effective path as extensive chassis prep is required. There's also a compromise in passenger and storage space and access - not to mention the loss of an as-delivered factory appearance. I think most owners would be very satisfied with a 25-50% increase in torsional rigidity while maintaining a stock factory appearance in exchange for a 100 pound penalty.

Let's hope for a yet-to-be-developed chassis stiffening solution so owners can achieve a higher level of handling without compromising ride qualities or to achieve even higher levels of car control with the addition of standard suspension upgrades. The chart alludes to steps that might achieve those objectives.

You engineering types out there. Get to work.

Sherwood

[rob justice]

This post got me thnking.
Before jacking my car this morning, using a set of digital calipers I took a couple of sizes round the door/ door frame.
Using the jacking point to lift the car - once in the air - the difference in measurements was 1/100th of a millimeter at one point only. (Top of 'B' pillar/door frame). A size almost accounting for nothing.
Placing the car on stands using rear torsion bar end covers. Again no difference in sizes.
As my car is an 85, is there a possibilty there could be different shell strengths over the years? Or an older shell has weakened with age?

[AtomicDog]

Somebody had pictures of gussets that they had installed to strengthen the area around the shock towers... I remember someone asking why the orientation was not vertical, and the reply dealt with the loading presented by torsion bars vice coil-overs.

I don't even know where to start a search, since this was not the main focus of the post.

Anybody know where/who was responsible? I'd love to see some test results, as the mods were very well thought-out and implemented.

Chris

[davidppp]

I syand corrected on the price of CF...

The raw cost of steel around here is about 30c per lb..fixed would be more of course..

What is the cost of CF now?

Many years since I bought any..

100lb of steel put into the right places would be expected to make a big difference I'm sure..

Sorry, but I am firm on the observation that the 911 coupe should NOT flex noticably when jacked.

none of mine do..

A change in door gaps etc means there is a structural fault for sure..

Kind regards
David

[jluetjen]

Originally posted by 911pcars
I also remember Bobby Allison from a C&D mag road test complaining that the 911 chassis flexed so much at speed he couldn't find the gears using the infamous 915 shifter mechanism. I guess he was looking for something a little more positive.

In the case of the 911, I think that this has less to do with the chassis stiffness then it does with the bushings and hardware that hold the engine/transaxle assembly into the car. Bobby didn't like the 911 from the get-go and is about as meaningful and enlightening as me saying that all stock-car drivers are stupid because they're inbred. Neither comment is particularly ensightful, nor is the cause analysis particularly well researched.

Anyway, his analysis of the car has been beat to death on other threads. Let's just not go there right now since the comment doesn't really add any data to the discussion at hand.