Tuesday, March 20, 2012

Mike Drew's Email About Early and Late Chassis

This is an email from the POCA Email Forum about chassis stiffness, by Mike Drew.

By and large the quality of the later-model Panteras was a mixed bag.   On the one hand, the quality of the materials used was far superior to those in the early Panteras (particularly the interior, leather vs. naugahyde).   But the switch from complex stampings to relatively simple jig construction also led to some engineering compromises that have bitten more than a few late-model owners in the years since. Rather than reinventing the wheel, I'll do some simple cut-and paste from the Aug 2001 PCNC newsletter, and May 2002 issue.   First, Charlie's experience:

A year and a half ago, I issued a warning about the need to check lower front suspension mounts for potential failure, after discovering a broken lower A-arm mount on Walter Villere’s car.   His car appeared to have suffered a shunt early in its life, and the front suspension may have taken a hit which weakened the mounting structure enough to inspire failure ten or twenty years later. Recently, Charlie McCall suffered a similar sort of failure on the rear suspension of his absolutely rust-free 1985 GT5-S.   During our recent drive across Europe, I had noted what appeared to be noticeable rear camber change on the passenger side of the car, and foolishly urged Charlie to have it looked at soon, instead of just diving down and looking at it NOW. The fact that his motor blew up the following day was perhaps a blessing in disguise, for when the car was towed to a repair shop, the workers there discovered incipient rear suspension failure; had the motor held together much longer, it’s quite possible he could have experienced full suspension separation and consequent loss of control. Although this was the first I’d heard of this sort of thing, I have subsequently heard from several other people who had similar failures; interestingly, they were all confined to late-model cars, i.e. GT5 and GT5-S.  
It’s unclear to me whether the failure can be attributed to the larger wheels and tires fitted to these cars, or to different manufacturing characteristics, but I believe the latter is more likely the cause. Apparently, the early Panteras feature a steel tube sandwiched in between the inner and outer frame rails, through which the lower control arm mounting yoke passes.   This transmits all forces to both the inner and outer frame rail.   In looking at these photos, it appears that this steel tube is not present on this particular car (it’s somewhat difficult to tell for sure.) That would mean that all pulling forces were felt only by the inner rail, and pushing forces by the outer rail. Eventually, the forces were too much for the mild Italian steel, and the washer and mounting nut pulled through the inner rail, and came to rest on the inside edge of the outer rail.   It’s also apparent that they then began to pull their way through the outer rail, and it wouldn’t have been much longer before they pulled through entirely!   The driver’s side of the car showed deformation and a crack along the bottom edge, indicating that failure was underway there as well. This knowledge would seem to demand an immediate inspection of this area on every Pantera.   I feel fairly confident that early cars will be in reasonable shape, but would caution owners of newer cars to be expecting some degree of failure here, and would urge anybody driving a GT5 or GT5-S to spend a few minutes peering underneath before their next drive, spirited or otherwise.
There are several potential solutions, some of which will come with additional benefits which makes them doubly worthwhile. Assuming that these later cars were manufactured without the aforementioned stress-relieving tube between the inner and outer rails, then external reinforcement is in order.   One solution would be to bend a plate of fairly thick steel into an “L” shape, drill an appropriate hole, and weld or screw it to the inside of the frame rail.   The idea here would be that the threaded portion of the yoke would pass not only through the inner and outer chassis rail, but also through this steel plate.   It should be at least three or four inches wide, enough to transmit the various forces over a larger area of the chassis.   Although the failures I’m aware of were concentrated on the rear mounts, I’d feel better having reinforcements on all four mounts. Alternately, the chassis braces sold by Hall Pantera and Precision Pro-Formance utilize all four A-arm mounts to affix the brace to the car (the Pantera Parts Connection brace only utilizes the rear a-arm mounts); purchasing and installing this brace would effectively result in a full-length steel reinforcement for the A-arms, as well as the primary advertised benefit of reduced chassis flex. One area of potential concern would be the parking brake mechanism.   Later Panteras utilized a completely different parking brake scheme, and the parking brake pulley mechanism is located at the rear of the car, between the a-arm mounts, instead of in front of the motor as it is on the early cars. Thus, there could be interference between this mechanism (or at least the mounting bracket) and the chassis brace.   However, the mechanism is thankfully mounted to a removable bracket, which presumably could be massaged to ensure adequate clearance.
In light of this potential problem, it should go without saying that it is important to not over-torque the A-arm mounts when installing them.   While it’s possible that failures of this type are caused by stresses imparted by the road, there’s no doubt that they can be accelerated by over-enthusiasm when tightening the nuts down (remember they have to be loosened and tightened each time the rear wheels are aligned, so one can assume that several hands have touched them since the cars were first put together.) We can never forget that we are driving cars which are approaching 30 years of age, and in most cases, have seen upgrades in terms of horsepower, braking and traction, sometimes dramatic ones.   While one shouldn’t expect these cars to simply fall apart like a soup sandwich, nevertheless we as owners can bear a certain responsibility for helping to induce failures of this type by performing these modifications.   It is only through vigilance that we can safely continue to aggressively drive these cars for the indefinite future.

Then, some destructive testing done on a 1971 Pantera to compare it with a
GT5-S:

Several months ago, I wrote a short article detailing the failure of Charlie McCall’s GT5-S lower rear a-arm mounts and cautioned that this appeared to be a fundamental design flaw in the later, hand-built Panteras.   While there is plenty of anecdotal evidence supporting the contention that late-model chassis is dramatically under-engineered, particularly in the rear lower control arm mounts, some owners of later cars are especially sensitive to the criticisms heaped upon these cars by the Pantera vendors who work on them every day.   They cite an overall improvement in development, comfort, and build quality and have big problems with people levelling accusations of structural inferiority upon their cars without any proof. So I decided to fly to Detroit and drive down to Kirk Evans' shop, do some destructive testing on early and late Pantera chassis rails and determine once and for all if there is a significant design problem. Well, here is your proof. Gary Roys' GT5-S (recently sold) was up on jackstands in Kirk's shop, minus any suspension, so it was a simple matter to bust out the Tool of Justice and ziz some sheetmetal away to peer inside and see what is going on.   Kirk happened to have a pair of 1971 chassis frame rails literally lying around which provided an excellent side-by-side comparison. A peek inside the absolutely rust-free GT5-S chassis was somewhat horrifying.   Kirk has spent more time than just about anybody dismantling and reassembling early Pantera chassis, but I believe that he hasn't really done too much with the late cars other than import them and sell them under the aegis of Amerisport during the late 1990’s.   He expressed considerable surprise at the comparative lack of spot welds on the chassis rail, and the hokey tack welds used to join the two halves together at the bottom.   Upon opening the rail up, there was no visible reinforcement except the tube, which he was able to move around by hand.   He agreed that this was grossly inadequate for the task at hand. A comparison with the 1971 chassis rail showed that somebody was on the ball back then—a second layer of sheet metal encompasses the inside of the frame rail and has a good 20 or so spot welds to keep it in place.   Inside the frame rail is an extremely complex reinforcement structure formed from sheet metal bent into an M shape and placed on end. Bottom line—in this area, early cars are relatively strong, and late cars are weak to the point of being potentially dangerous.   Evans plans to engineer a simple bolt-on or weld-on sheetmetal cover (similar to that used on the early cars) which he will sell at cost to any owner of a post-Ford Pantera. One side note—both structures feature internal reinforcement of one type or another that would be virtually impossible to deform simply by over-tightening the nuts which secure the yokes to the chassis rails.   There has long been speculation that the several failures reported on the GT5-era cars were caused by over-tightening, but that no longer seems to be a valid theory. Thanks to Kirk and Gary for helping settle this issue and hopefully solving this problem.

Photos from the destructive testing can be found here: Poca Gallery
Click the 'full size' link for a better view. While the problem is not terminal, it definitely needs to be addressed if it hasn't been already!