I always assumed that the '72 ad was a typo ... that the actual distance the cg moved due to the oil tank move was 0.6 inches.

I am not so sure exchanging the engine and transaxle positions would move the cg more than about 6" ... the '72 engine was 401 lbs, and 915 was 119 lbs without starter or oil.

That moves the CG forward a little over 7/16ths of an inch. With the lines included, you're probably looking at half an inch.

JR

Sounds like Warren might be right then about the typo.

Yes -- what driving impressions do we have comparing cars otherwise similar?

Randy:

My notion for U shaped front suspension lockers was based on the safety supports used on race car on board jacks - you don't want to trust those buggers when crawling under the car, so you slip this thing around it.

For a torsion bar 911 I think just a tube of proper length would probably be as good, since you have to do some disassembly to get the dust cover off, and then put it back on, and will want to get the elastic bump stops out of the way I suppose, etc.

So that will have to be a tool of local manufacture, as they say. Especially as it will differ (unless one goes to the trouble of making something fancy with threads and such) depending on ride height and such.

See below for benefits of jacking up the front instead of the rear.

All:

I think we theory wonks (well, at least wannabes, in my case), have the same references. If JR says Smith doesn't have a formula, I won't look. But we all have Puhn, too, no?

Puhn gives calculations at page 159 for overall CG 3 dimensional location. Includes how to locate the x-y location. Tim's (which I see is an actual proof) is the same as Vanvalkenberg's (in which the trig is subsumed), and I suspect (were I not so lazy) is the same as what Puhn gives in a somewhat diffferent format.

Puhn then goes on to show how to calculate the CG of the UNSPRUNG weight, but all that requires in extra information are the unsprung weights. One supposes that it is the sprung weight that is the mass that contributes most to roll.

Staniforth just says CG is extremely difficult to measure (if I read him right), but goes on to talk some about the mass centroid, which Randy keeps bringing up. It is unclear to me whether it is sufficient to measure CG location to make useful calculations, or if one has to go on and deal with how the weight is distributed (as one would, for instance, when dealing with polar moments).

All the CG calculations seem to depend on a fairly straight forward proposition. The CG, fixed in relation to the car, is going to move relative to the ground as you tip the car. How far it moves depends on how high it is, and how far you tip the car. In doing suspension calculations this movement is often said to be small enough that it can be ignored, but for our purposes with exaggerated angle changes it lets us work backward.

Query for Tim - Valvalkenburg's sketch shows a formula car with scales under the rear wheels, and the front wheels sitting of stands. One can imagine a couple of guys just lifting the front end up with their hands and putting it on two bar stools.

In my failed attempt to do the measurements I jacked my 911 up from the rear (easy compared to the front). But the jacking was not from the axle centerline - it was from the rear motor mount. Is that going to lead to a different reading on the two front scales than jacking (or resting when jacked) the rear wheels on something? Maybe another thing a guy needs - a pair of very strong boxes a foot (or better, two feet) high, for the tires of the jacked up end to rest upon?

And since we have removed weight from the jacked up end, do we need to also lock the suspension at that end also? Since it will rise, maybe chain it so it can't uncompress?

Seems to me a guy could, with some mathematical effort and knowledge of effective spring rates, come up with a formula which would not require locking the suspension. You'd back out the angle change due to suspension change? If so, that would sure in hell speed up making this calculation for a bunch of cars at a club tech session, since the leverage effects on the spring are pretty much the same for all our cars (through 89 anyway), and everybody knows his bar size.

I may try jacking the front end next go around (whenever that may be) - over a certain angle my front spoiler hits the ground, so the front bodywork was off when I tried. The tail pipes are a lot higher, and could be removed.

Walt Fricke

Walt,

A couple of ideas come to mind. Obviously, the higher you raise one end of the car, the more accurate the data, up to a point. I have read that at least 10 to 12 inches is desirable, at the least. It seems to me an easy way to do it would be this:

Measure the front tire loads, for future reference.

Measure the centerline of the front axle relative to a reference point on the fender.

Stretch a piece of tape, or something similar, across the rear wheel opening and locate the rear axle centerline in two axis.

Using the front torsion bar adjusters, crank the front height up a couple of inches.

Raise the rear of the car the desired amount and check the measurement from the front axle centerline to the reference point. If it is the same as the static measurement, record the weight change. If it is higher or lower, re-adjust the torsion bar adjusters until you get the correct measurement.

Drop the car and reset the front height and weights.

I do think the car needs to be supported by the rear wheels, not a jack at some other point.

JR

Puhn says raise the risen wheels two feet. Obviously the exponential nature of the trig involved means that up to a point higher is better, as extraneous influences mean less.

VanV also suggests draining all oil and fuel. Someone points out that you can do this measurement with and without fuel, and thus figure how changes in fuel level will affect things!

Me, I'd like something ballpark but based on actually doing the measurement.

Walt

The weight at the rear (or front) end must be taken somewhere (anywhere) along the line of the reaction force. This means a vertical line that intersects the center of each wheel (i.e. the tire contact patch). This is the Principle of Transmissibility. You can apply a force to a body anywhere IN LINE with the force acting on it and it will react exactly the same way as it would at any other point along that line.

If you were checking the weight on the front wheels while jacking behind the rear wheels, you are artificially increasing the wheelbase without changing the center of gravity. As you move your rear jacking point away from the center of gravity, it is carrying less of the weight. As the total weight stays the same, this means that the front is now also supporting what the rear has lost. Now, if you measured the weight at your front wheels while supporting the rear of the car at the same point for your 'level' readings, this would be a proper comparison. However, you would have to ensure that this jacking point (point ‘A‘) was on a line that intersects both front and rear wheel centers so that it remains on an arc that shares its center with the rear wheel's arc (the front wheel center). We just use the wheels to do this because it's convenient.

http://forums.pelicanparts.com/uploa...1170136751.jpg

As you mentioned, the rear suspension would obviously droop unless restrained. This would hurt your results somewhat, but I guess most of its movement would be vertical, and since we are not jacking our cars 6 feet in the air, it may not amount to too much. You won't know how far off you are unless you also do it properly so that you can compare. In a pinch, I've used tie down straps to keep the elevated axle from extending, and wooden blocks and more tie-downs to keep the other end from compressing. Not elegant by any means. I'm still waiting for a good solution for our McPherson front struts! Walt, you may be on to something, but I don't want to take anything apart if I don't have to.

I think these are what you are referring to:

http://www.hrpworld.com/index.cfm?form_pic_id=1673_1&action=product_pictur e&form_cat_id=579,212,363

Walt, as for your suggestion regarding the use of the car's ride rate and change in ride height to determine the weight change at an axle: I think that there is too much friction in the suspension, and not enough “resolution“ for this measurement. Assuming that you knew the total weight, the front ride rate (not even considering the tire rate), wheelbase and height raised, using some typical numbers, you could not expect the suspension to compress even an inch.

eg: raising rear wheels and assuming C of G height to be 18 inches above the wheel center line (HIGH!).

Wheelbase: 90 in
Height raised: 24 in
Weight: 2400lbs
Weight distn: 60% rear

The difference in front axle weight would be 130 lbs. Assuming a ride rate of 200lbs per inch: http://www.instant-g.com/Data/911CoilConv.html

This amounts to about only 5/8 inch. Because of the friction in the suspension system, I don‘t even like to leave the dampers attached when I‘m corner balancing. The rubber bushings are another problem.

Hope this helps

Tim K

In side view, the Cg is about where the “rear emergency seat is located” Frere, 1997, PORSCHE 911 STORY (6th ed.), p. 257.

Vintage thread bump!

If the Higgs boson really was discovered in 2012, it means that gravity is not what we were taught in elementary school. Gravity would not be a force of attraction between masses but an external "pressure" created by the curvature of spacetime.

A simple analysis of why it is important:

Automobile handling - Wikipedia, the free encyclopedia
http://en.wikipedia.org/wiki/Automobile_handling#Center_of_mass_height

For more info, you need to read words written on actual paper:

Fundamentals of Vehicle Dynamics by Thomas Gillespie is one source.

Race Car Vehicle Dynamics by Milliken is pretty comprehensive but ignored most of the European work.

My bet is that the Cg on air cooled 911s is higher than 18" since both the Cayman and the Nissan GT-R have a Cg that is higher than 18.1" (this was in an article on the new 2013 Scion FR-S which stated that they had beat those two worthy sports cars, with a Cg of only 18.1" - Road & Track, March 2012, p. 34).

I'd be surprised if the Cg was higher on newer cars...

I don't know about that. On the older cars, there aren't all that many heavy things up high. The wiper motor/mechanism is about the only thing that comes to mind. If you strip a 911, most of the heavy bits are down low and the structure that is in the upper half of the car is pretty light. The newer cars are packed with all sorts of crap and a lot of it looks to be higher, at least to me. The cowl area is full of all sorts of junk and the battery sits way up high.

JR

Your tax dollar at work. CG heights are more towards the end.
Unfortunately no Porsches in the list ...

http://www.eng.auburn.edu/~dmbevly/mech4420/vehicle_params.pdf


Note the error bounds given for the CG height.