Center of Gravity

[DPpatrol]

OK, this question is probably geared toward the engineers, but I'm sure a few of the fabricators will be able to answer this.

How does one calculate a cars center of gravity? Also how does one calculate the car's "mass centroid axis"? In what I have read so far, much of the varibles which describe and control how the suspension behave are based on these 2 measurements, but so far there has been no discussion of how these are calculated.

jason

[Kritter]

Unless you have software to do it(any good design software does it) or are willing to take use of calculus you can only estimate it. Most of the popel who refer to COG with reference to car design either know how to use the software and get the actual COG as in the material you are reading, design ars using 3D software that does it for them, or are blowing smoke up the arse with estimates or even worse flase assumptions.

Centroid, COM, and COG are synonymous.



An easy example of how to do it on a smaller scale. get an object and hook a string to it and two levels perpindicular to each other now keep moving the string until you find the location where the object has leveled out on both levels, now rotate the object, the string, and levels and do it again this will give you an estimate of the x,y,and z coordinates for the COG.



Kris

"A signature always reveals a man's character -- and sometimes even his name. "

[Dylan]

You can measure it empirically on vehicle scales. You can calculate the for/aft position by the front to rear weight split when the vehicle is level. The height of the CG is really what we want though to do this you take weight measurements front and rear when the vehicle is level and again when one end of the vehicle is elevated. Form the weight transfer due to the attitude change the CG height can be calculated. The weight transfer is not vary drastic so accurate measurements must be taken. In addition the suspension must be locked so that weight transfer does not change the suspension’s ride height. What you now have is the CG for the entire vehicle but what you want is the CG of the sprung mass. From here you can subtract out the individual unsprung masses from some given mass moment reference point (M*D^2)total -(M*D^2)unsprung. I have a work sheet on how to do this somewhere. If I can find it I will scan it and post it. It’s also described in the book by Milliken. I did this process on 3 different vehicles for a 4-link design project that I did at school.

If your designing a vehicle from scratch you may have the X,Y,Z points of your tube chassis. If you don’t have the whole thing in a solid model you can even set up an Excel spread sheet to sum up all the segments of the chassis and approximate the other components as point masses. I think you can even find info on the CG and radius of gyration of components like engines and such that you can put in you spread sheet.

I have even seen a huge pendulum that you could put a whole vehicle on and swing it to get the polar moment of inertia. My friend built a similar apparatus for measuring the moment of bicycles for single track vehicle analysis.
Dylan

[BradM]

Dylan is right. Do it with scales and a little trig. If you don't have a set of scales, many shops will scale a car for a nominal fee. Re-read his post and take note of the parameters he outlined.

[FABRICATOR]

There is another high-tech device for locating CG. It is accurate to within about 1.5", and is called a floooor jaaaack. Accuracy can be enhanced with another special tool called a-piece-of-square-or-rectangular-bar-stock. This will bring it down to within about .5".

Roll center in off-road racing is very dynamic in nature and is dependent on what the car is doing at that moment. Traditional (technical) roll center is applicable, at best, only to smooth turns. And even then it is affected by traction. Actual roll center is moving all the time and for all practical purposes has no "center".

<font color=orange>The best ideas are the ones that look obvious to the casual observer.</font color=orange>

[Kritter]

THe jack only gives 2 dimensions unless I am not seeing it right. Fore and aft dimension and side to side. What about the height dimiension?

Kris

"A signature always reveals a man's character -- and sometimes even his name. "

[BradM]

I may not have been paying attention all this time but I think <font color=yellow>roll center</font color=yellow> and <font color=yellow>center of gravity</font color=yellow> are two totally seperate ideas. LOL! Or were you just trying to induce a state of confusion?

<P ID="edit"><FONT SIZE=-1>Edited by BradM on 04/10/02 11:17 AM (server time).</FONT></P>

[Waldo]

Just go to the LBC and ask the D - O - DOUBLE G and he will give you the 411 on the COG or COM or whatever!

BRAAAAAAAAP!

[Kritter]

WORD

Kris

"A signature always reveals a man's character -- and sometimes even his name. "

[BradM]

Kris - It is a multi step process. You have to first consider the four corner weights on a level surface. That data can be used to locate the CG in two dimensions. Then do the same calculation for an inclined axis. If you raise one end and recalculate the distance to the CG. The change in distance (say from the rear axle to the CG) on an inclined plane, along with the angle of inclination can be used with the tanget function to solve for the height of the CG relative to a level ground plane.

[Kritter]

I understand that I jsut dont understand how simply a jack alone can do 3 dimension

Kris

"A signature always reveals a man's character -- and sometimes even his name. "

[frankh]

Do the scale have to be on a flat surface?

[DPpatrol]

Thank you everybody for your answers.

jason

[geoff]

how to find the height of the CG with a floor jack? weve got the side to side and front to back idea down pat... thanks

"We are the music makers, and we are the dreamers of the dreams." -- Willy Wonka

[Kritter]

Here is an easy formulae I use to calculate CG of a vehicle
1. The front wheels are set on scales and the rear wheels are set at the same height using a spacer.
2. A reading of the weight is taken from the scales
3. The rear axle is raised lifting the vehicle to an exact height
4. A new reading is taken from the scales and the weight difference is calculated

Centre of gravity height = HCOG

Wheelbase distance = dWB

Weight difference = WDIFF

Weight total = WTOTAL
http://www.animalhousedesigns.net/cogformula.gif

Kris
"Buy American before it's too late"

[FABRICATOR]

It simply involves suspending the car from a single point. A single jack can be used to lift one end of a beam placed through the car, and spaning between two benches or stands. This has a chain or other hookup to a cross beam (piece of tubing) attached across the inside of the chassis. The height of the cross beam and where it hangs when the car becomes neutral (no longer adds to or resists tilting) will be the CG point. It is important that the cross beam and all attachments be symetrical. If they are, they will have no effect on the reading. A hoist or forklift is even better, and you won't need the large beam at all.

All fluids must be drained or filled so there is no weight transfer. The result will be within millimeters, depending on how careful you are. Suspending a full bodied car could be a nightmare, but suspending a tube framed vehicle is quick and accurate.

<font color=orange>The best ideas are the ones that look obvious to the casual observer.</font color=orange>

[FABRICATOR]

The simple floor jack under the car (or hoist from above) will give you "2 dimensions" right away, including a balance plane across the car. The cross tube simply goes on that plane, also across the car. All you need to find is the height.

<font color=orange>The best ideas are the ones that look obvious to the casual observer.</font color=orange>

[geoff]

right, im just not following how to go about finding the height, without using corner weights on an incline. thanks.

"We are the music makers, and we are the dreamers of the dreams." -- Willy Wonka

[FABRICATOR]

You should be able to guess the height within 4 to 6 inches. Then place the cross tube through the chassis. One way is to use a couple of flat plates of metal or even wood clamped vertically to each side of the car on two longitudinal frame tubes. Each plate needs a hole in the middle of it for the cross tube. The plates can be placed on the inside or outside, but must be equal in shape and size. They also must be placed symmetrically upon the car so they do not affect the CG. Once the car is raised, you can guestimate at how far off you are by tilting the chassis. If the lift point is high the car will try to stay level. If the lift point is low the car will try to stay tilted. Adjust the plates and try again. A tripod arrangement coming up from the floor is even faster, but you must factor in the weight of the tripod.



<font color=orange>The best ideas are the ones that look obvious to the casual observer.</font color=orange>

[Porterrace]

This shop tech thread is cool, but let's talk about something that I think is the most important thing in off road racing. Good old fashion common sense. I truly do not mean to make light of any engineering education some of you obviously have, but don't loose sight of the goal. Off raod cars are not Indy cars and thank god they are not nascars, they never see the same direction, hole , speed, or anything more than once. The driver can't even see where to point the car most of the time. 95% of the cars I see in the desert have had tons of money put into them and a whole hell of a lot of hard work. But it just seems a few things can't see the forrest for the trees. My point is every once in while it's o.k. to put down the keyboard, look at something, think of where the stress and impacts are coming from and use your brain, and find a strong way to solve the problem. Just a thought.

[ntsqd]

You make a very good point. Unfortunately common sense isn't all that common.
FWIW, In my own experience knowing at least little about the engineering behind something makes the judgement calls much easier to make. I'm still on the steep end of the learning curve in knowing when calcs are needed and when the "eyeballic" approach is more appropriate. As such I tend to over-analyze things. I see that as the better side to lean towards, but is easy to fall into the trap of always doing the calcs.
The converse is to build something, have it fail, and not know why or be able to figure out why. That, to me, is the worst.

Knowing where the CG is allows a finessed eyeballic approach, no need for linkage software or a CAD package unless you want to use it. I've seen a lot of linkages laid out in chalk on the shop floor, and most of them worked fine.

TS

I used swerve around my halucinations, now I drive right thru them.

[hoeker]

"I've seen a lot of linkages laid out in chalk on the shop floor, and most of them worked fine"

this is the most amazing thing about most off-road fabricators. they can do this (chalk lines) and be succesful. off road may not be indy, but in many ways is more complex. take a 4-5,000 lbs car and turn it at 120 then add in a big **BAN ME****BAN ME****BAN ME****BAN ME****BAN ME** unsuspected bump. you just threw the indy car into the fence, but the chalkline off road car is still kickin' - i'm convinced off road is starting to come around to the high tech and agree with ntsqd that a little bit of real engineering knowlege will help develop the best off road fabricators in the world. the next generation fabricator, will probably fail most of the time on the PC, not the track.

my .02$

<A target="_blank" HREF=http://www.rosshoek.com>www.rosshoek.com</A>

"to be successful, you must first learn to disagree without being disagreeable."

[FABRICATOR]

I was going to mention that but I like to stay on track with the original questions. You could say; balance the car fore and aft, side to side, put the CG 36 inches ahead of the rear axle, keep it as low as possible, and leave it at that. You could also probably assume that no one here knows what to do with the information anyway. That’s where current art is. These techheads are looking at this stuff from the other side of the fence. That fence is getting lower all the time. They know that CAD is capable of designing and stressing parts, comparing materials, tracking statistics and changes, and can be upgraded and expanded. Cars like Cart, Indy, and F1 are nearly 100% computer designed. And they are no less spectacular in their performance than a TT. The main thing missing is tons and tons of information. Computers can gobble this up and analyze it in milliseconds. Just knowing the CG of all the latest cars and trucks would be a very valuable bit of information. IMHO, statistical information alone, would surpass all the speed secrets of all the top fabricators combined. Then throw in analysis, instant modifications, re-analysis, and stand back.

Fabricators and thinkers have brought us this far. And even F1 certainly came most of the way by hard work, trial and error, and common sense. But computers have the ability to take the ball and run with it. That can happen to off-road too just as soon as someone wants to do it.

What computers do not have is imagination. It will still be up those who have it to ask the questions, figure out how to make what the computer says is best work, and keep things safe. A computer would just as easily and willingly design something that could be fatal. To assume that computers have no knowledge or experience is not accurate. It’s as strong and effective as the person using it. Common sense will still prevail. A saying that is still in its infancy says; "The smart man knows of computers, the educated man uses computers, the wise man knows when to use them."


<font color=orange>The best ideas are the ones that look obvious to the casual observer.</font color=orange>

[Bob_Sheaves]

I like the comment, but would change one thing "...the wise man knows when NOT to use them."

Seriously, what is being discussed here is about 12 to 15 years behind the automotive industry as a whole (knowing the whys and wherefores of what makes a vehicle do what it does). Computers and CAD systems are just another tool for the engineer to use. Simply saying that something doesn't mean anything or is not worth worrying about, especially when that person is NOT familiar with vehicle dynamics, simply is an accident (at best- at worst, a death due to incompetance) waiting to happen. That is not to say that I don't use "experience" in an effort to save time on a project, I do. HOWEVER, I have been there and done that for 28 years and know what my limits are, when it comes to "glossing over" calculations. If I EVER have any doubt, I run the calcs and judge the results as either valid or invalid.

Best as always.

Bob Sheaves

<P ID="edit"><FONT SIZE=-1>Edited by Bob_Sheaves on 01/24/03 12:03 PM (server time).</FONT></P>

[Dylan]

I have used the method that Kritter described in practice a couple of times.
It works pretty well and isn’t that hard to do, but you have to take very precise measurements to get decent accuracy. It is also a good idea to “freeze” the suspension by putting solid rods in place of the shocks. Also this method gives you the CG of the total vehicle not the sprung mass, which is what you want for a lot of your vehicle dynamics calcs. You can however use the total CG and calc the unsprung out of it, or you could remove the suspension first


“Those of you who think you know so much really annoy those of us who do!”

[Porterrace]

I didn't mean to stray off the origional question, if you were talking to me for changing it. I was however trying to pose a question that seems more practical to the standard off road world, and those who generally try to build their own stuff. As far as not knowing how to use the information even if we had it believe me Porter race cars has it and uses it, so does Mike Smith, and so does all or Robby's fabricators. We have All the best Engineering software available and a mechanical engineer on our payroll to turn our thoughts into proper design, but for most of the guys trying to build a cool little prerunner it is mega overkill. Computers only can tell you what you already know, if you don't know it they can't tell you what is right and what is wrong. This thread is labled tech, I was just trying to give some honest advice to help some people out.

[Kritter]

"It is also a good idea to “freeze” the suspension by putting solid rods in place of the shocks. "

Save your self the time and "fix" the object you dont want to move by right clickin on it and clicking "fix"

Kris
<A target="_blank" HREF=http://www.barneysprecision.com/fabproducts.htm> Fab Parts</A>

[FABRICATOR]

Bob, there are virtually two different technologies at work. Stability over the rough at speed, and handling. These vehicles must switch from one to the other all the time. Accurate information on the principles at work are elusive and limited, even to the best designers. And basically non-existent in computers.
Dylan, this goes to the heart of the topic. And, as usual, it boils down to application. What is the information needed for? Handling in the turns, high speed over the rough?? That and other things determine how to factor in the unsprung weight, etc.
Porter is absolutely right in that without a very experienced human element, the computer is useless. But think of what it could do with that element AND plenty of the right information. But we have a ways to go.

<font color=orange>The best ideas are the ones that look obvious to the casual observer.</font color=orange>

[zooper]

Maybe we should rename this thread the philosphy behind building a off road racer. My mini-baja buddies and I recently took a vist to the Team Industries factory, if you are into snowmobiling you probably recaginze the name. Well one of the engineers sat down and talked to us about philisophical stuff. He was saying that while his company puts out really good products its only because they have worked out all the bugs in a trial and error method, which takes tons of people and a whole bunch of man hours. How many times did Edison try to make the light bulb, like a 1000. Even though I am in school for engineering, i am realizing that most of the stuff that is taught is there so when I get out into the real world I can make decisions without consulting a calculator or computor. I am trying to say I belive trial and error with a critical eye is the byfar the best approach to a project. Amen!
Brady

one mynd