I haven't gotten around to doing the whole tilting thing, but my estimate is that it's at about crankshaft height, or about 15 or 16 inches depending on the car. It also is noticeabley higher (just based on the driving feel) in Targas.

My guess for the z axis of polar moment/cg is what Bill V had posted. It should also correspond with the stock jacking points.

You can't say "nope" -- Porsche claimed this in an ad - I think the one for the 1972 cars. I assume they would not lie in print! If they did, the penalties in the US would be quite high (and note this is quantifiable and easy to check). I have an old post on this where I posted the ad. Before I saw it, I was sceptical that the oil tank placement did much.

BTW, where is the "20 lb." figure from?

I would also love to see where the vertical CG is located. The usefulness is that we could [1] try to lower it even further, or compare lowered ones (say on cars w/lexan instead of glass) to get better ideas on how to dial in vehicle dynamics, and [2] it would be useful to amuse ourselves with by comparing it to other cars -- more so than arguments about oil I think...

Tim - How accurate is the equation:
h= [ L^2 / WH ( Wr - Wri) ] + r

and where did you get it? Just by inspection it looks like an estimate.

I don't think the angle will be too difficult to measure accurately if one measures the heights of some known points and uses that. Modern computers and calculators should render decimal place inaccuracies ("round-off" error) a nullity. I'm not sure what that comment was based on. If needed, 14 digits can be had by running FORTRAN.

Well, I can and I did. It doesn't matter to me how much the oil tank weighs, or how much it got moved. You dont' need that information to disprove the claim. If you start with a measured fore/aft weight distribution, it's really easy math to calculate the location of the center of gravity. If you take that point and then move it 6 inches forward, you can recaculate the new weight distribution, which I did. If you then compare that weight distribution with a selection of samples from a range of cars, you can satisfy yourself pretty easily that it didn't move six inches. It's easy math. Look again at the result I came up with and ask yourself if you've ever heard of a 911 from that era, in street trim, with over 48% of it's weight on the front wheels.

By the way, the ad would have been done by an advertising firm, on behalf of the importer. Ask yourself if you think the guys in Stuttgart proofed all the ads. I bet they didn't.

I can't explain why they claimed what they did but I can prove it's not correct.

Cheers,
JR

As an aside, if anybody has ever measured the height of the C of G on a 914/6, I'd love to hear about it. I am planning a track car and would like to have the number when sizing the brakes and a few other things...

JR

I suspect that it's about the same, maybe a fraction of an inch lower because it doesn't have as large of a greenhouse, and the occupents sit lower.

CGH...center of gravity height, is difficult to calulate because is dependent on many parameters, such as tire diameter, height from ground, frontvs.rearheight, left to right weights and of course front and rear weights among other things.

Normally these items have little variation with other brands, but within the Porsche community they are very much owner dependent and will vary significantly. That means a different answer for each set-up.

With a few measurements and a little math you can get pretty close with the following version.

http://www.racerpartswholesale.com/longtech6.htm

Have fun

Jerry

Knowing where the CG is three dimensionally is important if you want to calculate what various suspension changes (e.g - torsion/spring rates, sway bar diameters or lever arm lengths, and raising or lowering the car) will do . You can calculate various things like camber change or roll center movement per degree of roll just from the geometry of the suspension, but the CG is at the end of the lever that starts at some geometric point.

Basically a racing/track/autox thing.

Walt

It does seem like a lot, but I will go with what the factory said until I really do see some calculations or measurements, much less "proof".


also, the home office will typically approve all ads

I hear you Walt. With all of the discussion and no real definitive answer it appears no one hear is using the exact CG for any suspension changes or what ever.

We all know it is heavy to the rear so it goes with out saying that any weight transfer that we can do will be to the front and as low as possible or taken out completely.

No offense but around here it seems like it is more a subject for guys with slide rules to try to calculate than racing types trying to figure out what the next suspension changes are.:D

Okay, let's use a car with a 42% front/58% rear weight distribution and an 89.5 inch wheel base. To find the distance between the center of the front axle and the CG, multiply 89.5 by .58. This yields a measurement of 51.91 inches.

Now, let's assume we have moved the oil tank and the CG moved six inches forward. This means our new CG is located 51.91-6=45.91 inches beind the front axle. Using the reverse of our first calculation, we calculate what the fore/aft weight distribution MUST be, for the CG to now be where it is. Thus, 45.91/89.5=51.3% is the weight on the rear tires so the new distribution is 48.7%f/51.3%r.

Since we know that no early 911 had a weight distribution like that (feel free to review the measured distributions of various cars from 1968 to 1975, that I listed earlier) the only logical conclusion that can be reached is that the CG moved somewhat less than 6 inches. I would be happy to calculate just how far it did move but I can't find enough comparable data to do so. There were greater variations from differences in body styles, options and other features that changed from year to year, to do it accurately. If somebody had a both '71 and '72 911 that were optioned exactly the same, we could do it easily with a set of scales, in about 10 minutes. Anybody want to volunteer, for the sake of posterity?


I have no first hand experience here, since I didn't work for either Porsche AG or VWOA in 1972. Still everything I have read over the last thirty years has led me to form the opinion that Porsche was interested in building the cars and was happy to let the distributors do the rest. In addition, if you look at Porsche sponsorship of racing in this country in the early seventies, it was VWOA calling the shots, not Stuttgart. I really don't think Porsche AG paid much attention to what VWOA did. Maybe somebody here has a more complete understanding, from having been directly involved.

Cheers,
JR

Jerry

Within limits, I can adjust the roll centers on my track car. I can make the front A arm longer, and raise (or lower) its pivot. Not much, of course. And I can raise or lower both rear pivots. Some of this I can just do, and some requires fabrication and whatnot but seems within reasonable reach. And raise or lower the car on its coilovers really easily. These affect suspension arm angles. The angles (and lengths) determine where in the suspension arc (or curve) the wheels are. If I want to keep the loaded tire as flat as possible (I run 9" and 10" bias ply slicks), matching camber change to roll seems useful, and I can do at least a little of that with my adjustments. But I need to know the roll rate.

At present I just guess and make changes and see if things feel this way or that, or if lap times improve (but seldom in decent experimental conditions), and especially try to move the tire wear toward the center of the tire. And one would do these things (or better) in any case. But I'd like to have a better understanding.

For instance, I started with 315/375 springs. Went to 315/400. Then to 400/500. And then to 400/600. This last seemed just a bit tail happy, so I disconnected my rear sway bar (both F&R are 19mm and front has always been full tight and rear less so). But I'd really like to understand in advance how changes of this sort affect things, since I am both driver and crew chief.

Porsche made changes in the fabled RSR (RS also?) to better match camber change with roll. You can't figure things like that without CG height. Ditto if you are fiddling with anti-dive or anti-squat.

We all think of the 911 as inherently tail happy. But some fast torsion bar cars run relatively small front T bars - 19? Anyway, 22s instead of 23s - to go along with the biggest possible rear bars (33? Seems larger than the spline size). Seems counter intuitive, but perhaps this better matches front and rear roll moments. Different from the old chart in the Automotion catalog in any case.

A guy with a torsion bar car (in a stock type race class) can't change as many things. But he could so some calculations on how ride height affects things. That might lead to the conclusion that scraping the ground almost may not be the optimal height, at least at a handling track. Or it might not.

But it is slide rule stuff. Or, more accurately in most cases, suspension program stuff.

Walt

Well put.

________________________

Here is a diagram that might be interesting - I forget whose book it is from:

http://forums.pelicanparts.com/uploa...1169935638.jpg

You're right Randy. I goofed a little bit. I took another look at it, and noticed that in my haste I missed a subscript and it propagated though my calculations. When finding the angle (theta), I used 'L' instead of 'Li" (see below). Anyways, for a typical car raised a couple of feet at one end, the error amounted to only a few percent. I've made the appropriate changes.

Here is what I've got.

http://forums.pelicanparts.com/uploa...1170054201.jpg
http://forums.pelicanparts.com/uploa...1170055038.jpg
http://forums.pelicanparts.com/uploa...1170055472.jpg

The page for the link that Cgarr posted doesn't seem to be working. Anyone else have any luck with it? I'd like to compare.
Also, I believe that Carroll Smith had a similar equation in one of his books. Does anyone here have them?

Tim

I have all of the Carroll Smith books but I don't think he gives the formula for calculating the center of gravity height. He has a few eqations in Tune To Win, but the mostly deal with load transfer and some suspension-related isses.

Paul Van Valkenburgh's book Race Car Engineering and Mechanics has it. The equation he gives is:


Edit; The following formula was jacked up by the inability of the forum software to keep my characters the same. It also like to screw with my spacing for reasons unknown. Read the next post for the formula.....
______
2 2
(ƒ¢W)(L) �ãL - x
h= ______________

(W) (x)

Where W is the vehicle weight
L is the wheelbase
x is the change in height for one axle


That's the best I could do with writing the formula on the computer. I hope when I push the submit reply button all the spacing stays the same. I have my doubts...


I have tried the link from Cgarr and it worked for me.



Randy,

Was that enough of an explantion for you on the calculation of the fore/aft CG?

JR

Well,

As I feared, that didn't work. I don't understand why.

For the formula, you multiply the change in weight at the static axle by the wheelbase and by the square root of the difference of the wheelbase length squared and the change in height of the raised axle, squared. You then divide the result by the product of the weight and the change in height for one axle.

I hope this makes sense,

JR

There is also an equation in Race Car Vehicle Dynamics, IIRC.

Walt - are you inspired to make up some suspension locking devices??

This might also be of interest -- it shows the location of the major masses on a 911. The actual wts. are from my car.

http://forums.pelicanparts.com/uploa...1170098495.jpg

FWIW, regarding oil tank location, '72 style versus '73-'89 style.

The oil tank plus 4 quarts I weighed at 19#.

The distance moved is 3' 5".


But you also have to factor in the gact that with the weight behind the rear wheels, it is removing weight from the front wheels. With it in front of the rear wheels, it is adding to the front weight.

On the CB scales, it made a difference of .5% in FR/RR weight distribution.

That's also not taking into account the weight of the oil lines and the oil in them.