Scuttle Shake

[Rich]

Ok, I've heard a lot about this on convertibles but hadn't actually experienced it until I took the roof off the Midas today, woooo, not a nice feeling....

Although I got used to the feeling after a little while I'd be interested in knowing how to reduce it. I suspect a conversion to coil springs would help by taking the 'choppyness' out of the hydrolastic ride, but as that isn't an option for me right now is there anywhere that could be braced to reduce the effect?

Rich

[manifold]

I had the coil springs in the Convertible (before transferring it to Coupe) and I found it fine. No weird closing door gaps etc. Certainly didnt worry me. Cant say any more at the moment as Convertible is lacking a front subframe as I'm building a special one.

I dont know if its a problem affecting GTM builds onwards or not. (Mine was bought from Pastiche, although I suspect it was possibly a Corby remnant).

Kind regards,

David

[Rich]

No problems with door gaps, at least not now I've re-hung the doors. I'm not sure the drivers door was ever on right looking at old pics of the car.

[manifold]

Yes theres a special black art to fitting the doors on the convertible thats for sure.

I believe Charlie got that one sussed though. By fitting the door on a twist, then rescrewing/riveting the frame.

I found the new frames sort that problem out a bit better though as they are steel. So are less susceptible to twisting out. Plus windows dont tip and bind as they are raised and lowered properly with two rollers, not one.

[Hans Efde]

I'd be interested to think of a way of measuring the stiffness of the monocoque so we can compare cars. Does anyone have an idea how to do a simple test? F.e. my idea is:
1) car level, no hardtop, doors closed and all wheels on the floor
2) measure the straight distance from the garage floor to the bottom of the car a1) just behind the front wheel and b1) just in front of the rear wheel.
3) position a jack in front of the rear wheel and jack the car until the front wheel becomes airborne.
4) similar measurements to 2), but now let's call them a2) and b2).
5) Calculate b2 - b1. This is the amount the vehicle is lifted at the rear.
6) Calculate a2 - a1. This is the amount the vehicle is lifted at the front.
7) Calculate (b2-b1) - (a2-a1) = c
7) The figure c tells us how much torsion is in the car, aka how weak it is.

I don't know how precise this is, but it would be interesting to get a few readings. I'll do mine when my car is ready again (at the moment jacked because of the speedo problem).
Any other ideas are appeciated.
Cheers, Hans

[Rich]

It would be an interesting exercise Hans, given the variety of suspension set ups I'm not sure results would be directly comparable between vehicles. That said, getting a baseline for a vehicle and measuring improvements in stiffness from specific modifications would be really useful.

I've been giving some thought to modifications, to my mind there are three obvious areas where I could make improvements.

1. Making the door lock-up more secure using tapered catch plates acting on door mounted strikers to lock the door in position when it's closed. If anyone has any idea where these could be obtained I'd really appreciate it.

2. Bracing the roll bar properly. My bar is currently a simple hoop, I'd like to improve the vertical mounting at the feet to improve the triangulation, run a bar across the base of the hoop picking up on the exhaust tunnel, and extend a flat mounting forward to pick up on the seatbelt mount.

3. Cross bracing the floor under the seats by doing away with the seat mounting frames and replacing them with triangulated frames that bolt between the sill and exhaust tunnel. The seats runners can them be bolted to the frames instead.

Any comments or suggestions welcome.

Rich

[ACourtney]

I would be interested to see the results of a survey of chassis stiffness. I suspect that the results would show a link between high mileage and loss of stiffness as simply the more duty cycles a laminate sees the more likely it is to delaminate.
A some point during their tenure GTM made a small change in the floor construction when a better polyurethane foam became available, but in theory the later shells should be more resistant to delamination.
The only polyurethane foam sheets available when Harold Dermott started making the convertible were polyurethane insulation sheets formed between two flat plates. These foam sheets have very smooth sides and Harold found that the only way to get a good bond between the glass layers and the foam was to use vacuum compaction.
In the mid-nineties polyurethane foam sheets became available that were cut from thick blocks of foam. The surface of these cut sheets are rough and open so they draw in the resin. When used on a wet laminate, the spongelike action of the foam will not only draw in resin but it will pull up the ends of the glass strands at the same time. Thus the cut sheet will form a stronger bond to the GRP skins, even without vacuum compaction, than the smooth moulded sheets. The down side is that the cut foam soaks up more resin so the resulting laminate is slightly heavier and more slightly expensive, but the gain in interlaminar shear strength and time saved made by eliminating the vacuum process more than counterbalance that. So that is why I say in theory the later shells should be stiffer.

In practice there are several other factors that can cause delamination, but the thickness of the GRP skins themselves is also critical.
The theory of composite sandwich construction is that it behaves much like an I beam. The GRP skins are the upper and lower flanges of the beam and the foam core is equivalent to the web. The overall stiffness of the I beam is defined by the thickness of the flanges, but also depends on the web tying them together, otherwise the two flanges act independently.
I'm told that the floors of Pastiche shells are generally thinner than the original Midas shells, whilst GTM shells varied a fair bit from car to car.

If you think that your floor is a bit thin then you can thicken it up by bonding on more layers. I would certainly recommend this course of action over adding additional bracing across the floor. If the bracing is just bolted in it is more likely to cause stress raisers which could lead to delamination. Obviously the down side of thickening up the floor is that you will need to take out the seats, carpets, roll- bar etc before starting working and then take a good deal of care to avoid covering the outer bodywork, pedals, gear lever etc (mask them up with plastic sheet). To get a good bond to the old laminate the whole surface will need to be roughed up with coarse grit - P40 - and washed down with acetone to remove any grease or old carpet adhesive. I would start with a layer of 300gm chopped strand mat and once that was fully wetted out I would lay on a layer of 800gm woven roving. This will add 2 to 3mm to the floor thickness. This layup can be repeated if you think the floor is really thin.

I hope this is of some help.

Regards,
Alistair

[Rich]

Thank you Alistair, very interesting stuff.

I've noticed on older pictures of the Convertible that there were rear braces coming off the main roll hoop, apparently bolting onto the rear tub panel. What's the current practice?

Rich

[manifold]

No ones mentioned about the amount of shake caused by the deflection of the inner wing, through lack of stiffness where the front shocker mounts fit, which eventually results in a shocker fisting its way through the panel..... Stick a video camera down there like I have in the past and you will see what I mean.

Most performance cars pay particular attention to keeping these areas rock solid. Theres non of this in the Mk3's. I dont know about Mk4's so cant comment.

Although the cars dont have struts its still important. Its certainly not the most impressive part of the Midas design and was poorly executed (ie non) in my opinion. The rear was a executed a lot better and its not having to carry any weight!

[wolfie]

I have to admit that scuttle shake is a problem I have with my 12 yr old convertible. I remember when my cousin was having a mooch around the tub while I was building it he said 'I bet you're going to get scuttle shake' when I asked him why he said he could see problems in the construction and probable strength because of the hollow section in the front arches with not enough cushioning effect between the bulkead and front outer shell. He also said he would have filled the sills if they were hollow in the sandwich construction. I thought nothing more of it but I may pick his brains again (he was a jaguar engineer for the mainstream division R&D before moving to the F1 team and now works for Prodrive so I guess he knows what he's talking about)