I've just finished to days of work measuring and calculating to get the long/vert CoG heights (not fun) prior to suspension design. Has anyone else done this and compared to the finished product?
No
You mean like this 
Sorry, couldn't resist
The model and real life so far are showing close correlation.
Rescued attachment Properties.jpg
yup, like that, though I used excel.
Now doing Bundorf analysis.
Must be useful to know the mass to the nearest milligram
When the CofG height gets close to the wheel center height it gets pretty tricky to measure it and confirm calculations.
I'd like to know how it can be managed to better than +- 10%.
Of course its much more accurate if the CofG is much higher than the wheel centers but that aint much of a sports car either I suppose.
A simple tilting table is the best tool for a finished car. You can get CoM's within a few percent.
I am doing CoG prior to suspension design and will do tilt analysis after build to see how close I was.
data set placed on website
You chaps must have much more accurate locost scales than I have.
Using the tilt method produces very small changes in front wheel weights and all I could determine was that the CofG was around the wheel center
height +- who knows? Certainly not better than 10mm.
See Milliken and Milliken section 18.2 for the maths.
Also note this maths is invalid if the wheels mave different rolling radii - can't remenber the SAE paper for it I lost my copy - wouldn't
mind another copy if anyone has one to hand.
Cheers
David
Hi dont forget to include the driver in the calculations. Yes i had to re do it the first time .
Cheers Matt
1) centre of mass is centre of mass - why would its height be different along or across the car?
2) When I was designing my braking system I tried to estimate height of COG by listing the big pieces of the car & their heights & combining
them in excel. Came to 40cm IIRC.
3) I intend to try to balance the car on the wheels of one side & then the other. Trouble I need quite a few helpers to manage this & I keep
forgetting... That should tell me the height of COG pretty accurately I reckon - I'll post on here if & when I remember to do it &
compare it to the theoretical 40cm figure!
cheers
Bob
I use a big spreasheet with the weights and x, y locations of every component. It's invaluable for calculating front/rear weight distribution and weight transfer. I really don't know how a car can be designed without.
centre of mass is only one figure, but each end of the car has components that can be placed in different positions.
The Sunday name is Polar Moment of Inertia.
Two cars can have the same CofG one with all the weight in the middle and one with half the weight over each axle. Although they have the same CofG
the second car will act more like a flywheel and resist turning. Once turning it will want to keep on turning.
Great for a family car giving a smooth ride but poor for a sports car that needs to make quick changes in direction.
As well as varying longitudinally the weights can vary vertically making for different responses at each end of the car in yaw, roll and pitch.
Confusingly this will lead to different requirements for damping and springing in yaw, roll and pitch.
Best just to try to keep all the heavy stuff in the middle and as low as possible.
If any compromises are required, as they always are, keeping it low has the biggest overall effect.
I both agree and disagree with aspects of the above post, but this is not the place or time for such a discussion as I need to get the next issue of
Race Magazine together.
I will post more base data following publication, which makes up the pre-kinematic design set.
Why disagree? I see little, if anything, that is incorrect in V8kid's statements????
Plain and simple physics and mechanics!
Cheers,
Syd.
quote:
Originally posted by Syd Bridge
Why disagree? I see little, if anything, that is incorrect in V8kid's statements????
Plain and simple physics and mechanics!
Cheers,
Syd.
my how we jump to conclusions.
I cannot see how PMI has anything to do with ride quality.
I am also increasingly unconvinced at the supposed low PMI benefit with limit handling given the feedback loop available to a driver, particularly one
such as all of us with relatively limited skills on a public road. Low PMI cars have low recoverability in yaw and only a relatively minor incremental
advantage in outright grip.
With recent developments in my own project I realise how limited text books, including Milliken, are at address all of the possible variables WRT
design.
[Edited on 30/4/08 by rpmagazine]
quote:
Originally posted by rpmagazine
I cannot see how PMI has anything to do with ride quality.
were I balancing a car on a fixed pivot and wobbling it by hand I might agree with you.
quote:
Originally posted by rpmagazine
my how we jump to conclusions.
I cannot see how PMI has anything to do with ride quality.
I am also increasingly unconvinced at the supposed low PMI benefit with limit handling given the feedback loop available to a driver, particularly one such as all of us with relatively limited skills on a public road. Low PMI cars have low recoverability in yaw and only a relatively minor incremental advantage in outright grip.
With recent developments in my own project I realise how limited text books, including Milliken, are at address all of the possible variables WRT design.
[Edited on 30/4/08 by rpmagazine]
The response may be faster or rather feel faster, which is not always the same thing. However the controllability at grip limits and recovery is often poorer. So in short it feels great but when it 'goes' i.e. starts to slip etc then you need to be much much better to catch it and you cannot necessarily recover as well. Curiously this opinion is not mine but that of two test drivers for a local OEM. There was much more technical discussion of this, but as I said this is not the place.
quote:
Originally posted by rpmagazine
were I balancing a car on a fixed pivot and wobbling it by hand I might agree with you.
PMI is a very small contributor to many aspects of ride and control - relative to other issues. I suspect that unsprung mass is much more significant
in it's contribution to NVH and ride, though it is less significant to track performance assuming certain track conditions. If we were talking
about Australian tracks, which is what I know a bit about, I suspect that unsprung weight and dampers would have an additional importance to what I
have seen of in-car footage of overseas./ Local tracks can be quite bumpy in braking areas.
PMI is one of the aspects of design that we can 'optimise' (WRT our cars intended use) in order to offset the limitations of our design and
manufacturing capability and lack of development (after all we don't have the test rigs, prof engineers and test drivers and exhaustive damper
techs and tyre racks etc etc).
The one aspect that I feel we don't often acknowledge is that we effectively trade NVH performance for 'handling' and this is a very
big advantage we have over a manufacturer.
[Edited on 1/5/08 by rpmagazine]
Polar Moments not significant to handling??
Try driving a 911 hard in the wet and see if you change your mind.
Cheers,
Syd.
Chaps I'm surprised you don't think PMI is connected to ride quality. The acceleration of yaw and pitch are inversely proportional to PMI so
a high PMI car will have a low acceleration pitching moment which is more comfortable to us. Daniels gives a very good explanation without any maths
in his book "Car Suspension at Work" or if you like sums Milliken & Milliken cover it comprehensively if expensively. A free guide to
the Physics is on the net search "the physics of racing"
On reading rpm's post a few times I think he is saying that with very low PMI cars the driver does not have the reactions to match the response
time of the car.
Very valid for an everyday car where we let our concentration drift as we commute but surely a sports car is just that - a car for sport and hence it
should be challenging and rewarding? After all we do not have to worry about litigation like a mainstream manufacturer who will design his car to be
failsafe for the worst case driver
What I was, clumsily, trying to say in my previous posts is that we need to get the very basics right before drifting off into the esoterics of
suspendion design. Low weight, low down, slightly to the rear, sticky tryes with lots of horsepower and it will go like s**t off a shovel, brake well
and romp round corners even with very rudimentry suspension.
Cheers all
David
syd, if you read my posts you will see that I have said that I think the role of PMI is not as significant to other aspects.
When I interviewed Ron Tauranac just over a year ago he said his philosophy was that 'it is all about the tyres' and I can see the sense on
that basic statement. The are many many other contributing factors in addition to PMI and some of the interactions between them are complex in the
extreme.
V8kid your paraphrasing of my comments is mostly correct, however I would contend that driving on the road is inherently more challenging than on a
race track given the variables and that concentration is but one part of a person's ability to govern the attitude of a car, particularly in
transitions.
[Edited on 2/5/08 by rpmagazine]
Hmm I think we are discussing the motoring equilivent of "how many angels can dance on a pinhead".
I've never met this guru chappie you interviewed but for some reason he certainly is stating the bleeding obvious.
There must be an interim step between that and the incomprehensable surely?
On reflection quite probably not
Cheers all
David
Ron was the engineer behind all of the early Brabham and RALT cars.
He is still working and was designing a clubman for someone last I saw him at his place.
I don't know if you're fishing for a story for your magazine, but I'll continue anyway........If this all ends up repeated in print,
then at least I can point people to where it originated.
I would suspect, that Mr.Tauranac is being economical with the whole story. As most designers and engineers working in the racing industry are.
Also, if your OEM test drivers told the full story, or at least you reprinted it, then all would know that low PMI gives a faster car overall with
better cornering power.
The high PMI is difficult to get to turn in, then difficult to straighten out. But, has softer reactions at the limit, and that limit is much lower
than a low PMI car. It also needs different damping for the bumps, as the weight is closer to the axles. The dampers need to deal with the weight more
directly, instead of the leverage arms that low pmi gives.
The low PMI car will corner much better, albeit with sharper responses which need sharper reflexes. It will have sharper turn in, and sharper exit
behaviour. Downside is that at the limit in cornering, (and this is a limit higher than the high pmi car), the car will not have the progressive
breakaway that the high pmi does. Breakaway will be more off/on, but with sharp reflexes is no less a disadvantage than the the high pmi. You've
just got to have good feel in your backside. The limit, and response at the limit, can be made progressive with damper adjustment and
Mr.Tauranac's tyre adjustments, and proper suspension dynamic characteristics.
If all the above is not so, then why do all the Locost racers, touring cars, V8 supercars et al strive to get the engines back as far as possible in
the car?
Why do open wheelers and LM sportscars have the engine betwen the axles and as close to centre as possible?
Why do all knowledgeable racecar engineers strive to centralise all weight, and lighten the ends? Some are near to being obsessive about lightening
the extremities.
Geez, even the dinghies I raced as a youngster, down at St.George Sailing Club, benefited from light weight at the ends.
There must me something of effect and truth .
Cheers,
Syd.
[Edited on 2/5/08 by Syd Bridge]
fishing for a story? no I do not want for material and there is little in this forum that is print-worthy in a technical magazine.
The paraphrase of Ron is my own. There was of course much more to the conversation.
However I think you have mis-understood my point. As I have previously said PMI may well be overshadowed by many other aspects of design.
Dampers 'see' two masses, the unsprung and sprung, please explain the PMI effect on damper performance.
Rereading your post for the third time I can see that you have not understood what Ron was saying. Virtually ALL dynamic aspects are about making the
tyres function to the best of their ability.
[Edited on 2/5/08 by rpmagazine]
[Edited on 2/5/08 by rpmagazine]
[Edited on 2/5/08 by rpmagazine]
Steady on rpm I thought Syd stated the case precicely and succincitly - as well as correctly in my opinion. You must admit it would be very readable
in a Mag.
It sounds cheap to trash others opinions we are all here to exchange views based on our experiences and deductive powers afterall.
Can you elaborate on your points regarding the damping? I assume you were alluding to the load transfer on the contact patches from the PMI via the
dampers which kind of reinforces my point.
Cheers all
David
I am not here to trash anyone's view but I am very short on time and have an allotted 2 minutes or so as a break in layout. I cannot devote much
attention to this ATM so excuse the point format:
Simply put *everything* you adjust on the car to affect handling has its effect through the tyres.
The control issues re recovery etc are in Milliken and Dixon and I cannot explain them better
Dampers do not in any way deal with 'weight', the springs do. The dampers deal with mass/velocity, and unless you have some very fast
transfers and some very advanced dampers high low speed adjustment) you will not be able to address weight transfer. I disagree with some of the other
points as described as they need covering statements such as 'all other things being equal'...which is kind of a moot point in vehicle
design and suspension design.
[Edited on 2/5/08 by rpmagazine]
Mr Tauranac is not wrong in saying that it all happens through the tyres.
BUT, how you control the tyres is what it is all about. Has Mr.Tauranac told you how to do this yet, explicitly and simply?
I doubt it, and I also doubt he ever will, as I've previously commented about designers and engineers. They won't tell you something you
can't find somewhere else, they will tell you what is easily found out though.
Oh, and forgive me for mixing weight and mass, but most people understand weight, but get lost when you start talking about masses. This really
isn't the place for a discussion on mass dynamics and how they effect PMI, and vice versa.
This is supposed to be a forum for amateur builders, and they don't need to get embroiled in a lot of unnecessary theory, and application. They
just want to do it, simply and easily.
Just like Locost racecars!!
Cheers,
Syd.
[Edited on 3/5/08 by Syd Bridge]
yes, this above all places is the place for a discussion about such things, otherwise we perpetuate poor understanding and application of a limited
range of theory.
How do we progress a hobby without progressing understanding and skill?
You had discussed weight transfer, PMI and the effect on dampers had you not?
[Edited on 3/5/08 by rpmagazine]
i must admit , i find some of these threads going from one extreem to another, starting on PMI etc is fine if your building a formula 1, which most of
us are not,
i think we all no, or should no the basics as far as PMI goes and that is the more central you can have the bulk of the weght mass in your car the
better it will handle, the main reason why mid engined cars were designed and built and handle so well.
but a front engined car or for that matter a rear engined car can be made better by transposing as much auxillery weight as posible to offset the
engine mass
ie set the engine back as far as you can pracitibly then place things like battery fuel tank etc in areas to counter ,
i dont wish to rain on anyones parade here but ive been building and racing all sorts of cars for to many years to remember, and although i'm
learning new tricks every day quantum physics just isnt one of them, and personaly i dont think the guys on this forum need to go to that extreem, i
no i dont
JMTBW
cheers Kaspa
There are frequently too many generalisations of theory, particularly theories that are taken out of their original context, which in reality is where
they have meaning. None of what we are discussing is close to complicated in the sense of quantum mechanics and not complicated in the sense of
suspension design by a suspension engineer. Many builders I converse with know this sort of stuff backwards and the conversations are such that I have
difficulty following it. These are the guys who win and build good cars and I would contend they are the people we should be learning from.
I recently had the sobering experience of speaking to a suspension engineer who described what he saw as the process for designing the kinematics of
my cars suspension. It was sobering as it indicated that what I previously saw as best practice was deeply flawed.
Or should we continue to discuss design of suspension and handling using terminology and understand as per what Staniforth wrote 30 years ago?
The irony of this comment above is that Bundorf released his paper in 1976 or so...so it has been around for quite a long time.
I should add that I feel that the principle advantage that a manufacturer such as Lotus has is its relationship with the tyre manufacturer and
it's development team.
[Edited on 4/5/08 by rpmagazine]
[Edited on 4/5/08 by rpmagazine]
Err you've lost me - what does Bundorf say then?
http://en.wikipedia.org/wiki/Bundorf_analysis
http://en.wikipedia.org/wiki/Automotive_suspension_design
Note the engineer I spoke to said that the wiki bundorf section needs a lot of work, but it gives the basics.
quote:
Originally posted by rpmagazine
Dampers 'see' two masses, the unsprung and sprung, please explain the PMI effect on damper performance.
I was actually thinking more of mass/vel as in acceleration, but you are correct, a damper is as its names suggests a damper of mvt.
The ratio of damper valving is partly set/influenced by the two masses in that the bump valving reflects the vehicle mass and the rebound the unsprung
mass, or at least this is how I have reasoned it.
[Edited on 4/5/08 by rpmagazine]
CoG heights
These heights, at the front and rear axle lines, define the preferential axis about which the total suspended mass will want to rotate about.
There is of course only one true (as in the strict definition of CoG)CoG height but the term gets used in defining the two points through which the
axis mentioned above must pass.
The true CoG lies somewhere on the axis defined by the front and rear CoGs, the position depending on the weight distribution.
Faults with Staniforth etc
Well they can't be that serious as a great many very good cars have been made using these principles.
Perhaps the biggest "obvious" critisism is that the method assums a solid tyre. This can be compensated for in a spread sheet by adding a
tyre displacement versus load function to wheel movement. The effect is to make the effective roll centre a bit higher.
Another critisism is that the method assumes freely suspended masses. They're not! the springs and dampers make sure of that. As the springs and
dampers get harder the whole system gets closer to four tyres bolted to a rigid block. Again the effect is to slightly increase the "true"
roll centre. A completely rigid suspension would place the roll centre at the CoG, it couldn't be anywhere else.
quote:
Originally posted by cymtriks
define the preferential axis about which the total suspended mass will want to rotate about.
Oh sure we could start using forced based RC's etc, but as you note there is no cheap way of working these numbers as yet. No doubt we will get there, probably about the same time we are forced to drive 1lt diesels!
You're quite right, I think we need to recognize though is that the state of the art in calculation and proper understanding of the problem has
moved on from what an amateur can reasonably expect to wrap their brains around.
I for one would struggle to get going doing a force based analysis even if I did have a copy Adams to hand. As a result I've flogged the death
out of Susprog3d instead hoping the approximation of what we think is happening and whats actually happening are the roughly the same thing.
After all not yet understanding the fundamental nature of matter hasn't yet stopped us making things from it.
Greg the guy I have doing my suspension uses Adams every day, but he will still use his own software for Godiva. It would be possible to do a susprog with force based kinematics...but is the market there for the costs of a complicated redesign? I suspect not so we get what the market allows and we must accept that we cannot take the geometric kinematics too seriously.
quote:
Originally posted by Doug68
quote:
Originally posted by cymtriks
define the preferential axis about which the total suspended mass will want to rotate about.
I think this may be true providing you've just gone over a cliff. The rest of the time an object will want to rotate about the points at which the input force is being reacted against.
To demonstrate this take an object on your desk and push it over.
The other thing is the method which we all use and swear by is not the same as that the SAE accepts as the correct method. Unfortunately unless you are a pro or a PHD uni student you won't have access to the tools to work that method.
You people would be surprised at how much, or in reality how little, of all that theory actually goes into a top end racecar.
The numbers may be crunched as an after thought, but the basics are all that are needed. Keep everything as light as possible, as low as possible, and
as central as possible.
Suspension is a simple geometrical exercise. Springs and dampers need a little maths, but certainly not a supercomputer!
From then on it's about packaging. Same goes for roadcars, but costs are the major factor in roadcar design. And speed of build. Every minute
saved on the production line can be, and is, costed.
Cheers,
Syd.
By the way, if suspension was rigid, then the car would probably roll about the outer contact patch in cornering. Maybe???
[Edited on 11/5/08 by Syd Bridge]
Sure Syd, you CAN make it as simple as you like, but things are usually constrained by the rules and available time.
However OEM's do a lot of work on suspension and they do take in all of the variables if they can, which is why cars have progressed so much in
the past 15 years.
I recall an article with the designer of the Juno racecar, certainly an effective car, stating that the inner suspension mounting points came with the
gearbox and there was little advantage in redesigning the box casting just to get the mounting pints where Juno wanted them.
This certainly seems to bear out Syd's view. I'm not inferring the mounting points are not important just that they were probably right
enough in Juno's case, by evolution if nothing else, and that the packaging considerations were more important.
Extending this arguement surely it is valid to say that if there is only a small deviation between calculation methods - does it matter.
Course that opens up the question of what is small!
At the F1 prctice yesterday We saw Lewis Hamilton, with all the computing power available to them, having to make a decision, quite a big decision, on
the tyres to use based on his intuition - the fact he got it wrong is neither here nor there the issue is his opinion was the deciding factor. Most of
us are not in that league and the issue for us is if they can't get it right with thier computing power we have no chance outside the basic setup
to get it right using calculation alone..
So if we get the basics right we can muck about on the road/track with springs bars dampers etc to get the ballance right but one thing it is really
difficult to change significantly after the car is built is where we have put the weight. I guess I have taken the long way round to restate low and
central is the most important thing.
Finally if you need more convincing most rally cars used to mount the communications box on the roll bar cos it was convenient to plug into it. Works
cars don't they put it on the deck - guess why.
Cheers All
David
I think we should differentiate between a racecar which only sees a track and a car that also see road use. Having spoken to a designer Ford's
most recent sportscar, I was surprised at the months of effort that went into designing the suspension in Adams, which was then backed up with months
of testing, particularly of different tyres. Again we get back to 'fit for purpose'. In designing and building our own cars we trade some
aspects of performance, most notably NVH and safety, for those aspects we want more of.
I think too that the amount of effort that goes into designing also reflects the skills and resources that the racecar manufacturer has at its
disposal. In this respect we should be cautious in interpreting the magazine comments from Juno.
[Edited on 12/5/08 by rpmagazine]
Interesting
http://www.racecar-engineering.com/allarticles/257234/sportscar-suspension-design.html