john banks

With reports in the mags of the MY01 having 250% of the chassis stiffness of the MY00 saloon and 240% of 5dr then does that mean the current chassis is not very stiff given it is quite an old design? The Focus was said to be twice (100% again) as stiff as the Escort and the handling is in a different league along with ride quality/crash safety/NVH etc. The new Corsa is said to be only 30% stiffer than the old and a great improvement, the Astra was somewhere around 50% stiffer. I'm sure I read that most people don't seem to think strut braces help the Scooby as it is very stiff already. But if such a huge improvement can be made with MY01 are we thinking again? Does the current Scoob drive so well because of its AWD system rather than a particularly good chassis? The ride certainly does seem quite reasonable for the lack of body roll. Is it analagous to Stef's comments re good grip/average handling? Is the chassis TOTALLY different on WRC?
Any comments on my haverings?
Do I need to get a strut brace?
Do I need to get a life (my wife thinks so - she does not understand Prodrive settings/bumpsteer or subwoofers - she is not looking forward to the next few weekends of me installing ICE - oh well she can always read her storybooks)

chiark

Wotcha John,

People who have tried strut braces tend to fall into two camps:

PTMW! is very happy with his, as is Stef IIRC.

Regarding a life: they're optional, especially in the eyes of spouses.

You want to try explaining why you're fiddling with a soldering iron and bits of breadboard when the answer is to play MP3s in the car. "But you've got a CD changer!".

johnfelstead

Let me tell you a story, sitting comfortably??

When we first bought the Esprit racecar, it was a very unsuccessful car; it never finished a race, was usually on the last row of the grid and was about 12 seconds a lap off the fastest cars pace, an absolute nightmare.

We ran it for a few events just to see where we stood prior to taking a serious look over the winter at the car. The car was built for hill climbing so was on relatively soft suspension compared to race spec running about 200Lb springs all round.

The main problem with the car was major instability under brakes making it want to swap ends but the traction was awesome. It needed sorting in lots of other ways too but I will concentrate on chassis here.

We set about corner weighting the car and we could move the platforms on the shocks and the corner weights didn’t change, very strange. So we ran the last couple of races using ride height settings only, just to out all the bugs and get some data.

During the winter we stripped the car back to its space frame and started to do some torsional stiffness readings, We didn’t need any gauges, by loading up the front of the chassis we could see it flexing, as we loaded the front suspension points the steering column moved up and down, it was that bad!

We took a long hard look at the space frame design and could see that some vital links that would transfer the load through the chassis simply didn’t exist. My friend who I built the car with works as a stress engineer at BAE and so we had access to all the strain gauges and Computer Modeling programmes to help us design and test what we were changing.

We ended up replacing 90% of the space frame with our own design, using lots of triangulation and ensuring that the loads in the front of the chassis were fed all the way to the rear and visa vesa.

We now have a chassis that the flex registered is in thou, not inches. What this now meant is that when we set the corner weights up we can get the car to be within 1lb of weight difference per side, the smallest change on the platforms makes a difference.

What this meant was we had to start from scratch on the suspension settings, it took us a full year of racing to get the suspension settings spot on. Initially we had major problems with traction because the chassis wasn’t flexing anymore to give the mechanical grip, we had to tune the damping to do that for us instead.

We also altered the damping and spring rates enormously as we now had a chassis that could handle higher wheel rates. We eventually ended up with 1500Lb rear and 1200lb front springs and spent a great deal of time setting the suspension pickup points to have an ideal roll center for running F3000 sized tyres, we have zero bump steer front and rear also.

The car is now very stable in all conditions, has huge grip and is totally predictable. The car now runs top 3 at every event, and in the hands of Mark Hales who drove the car for a magazine article it smashed the lap record that had stood for 6 years on only his 5th lap. He said it was one of the best handling racecars he had ever driven, a perfect match of power/grip. This is a guy who has been racing since the 60’s and regularly drives cars worth millions in GT races and historics.

Without us having sorted the chassis rigidity the car would never have responded to changes, you need a very stiff chassis to allow you to exploit the suspension to the full, you do not want any flex in the chassis at all.

Another good example of this is the RAM formula1 car that won the BOSS series a couple of years ago. In its day it was the Minardi of the field, always trying hard but always slow and difficult to drive.

When the team bought the ram for the BOSS series they did exactly what we did, they bolted the chassis to a wall and applied load to its suspension pickups, they could see the chassis flex too.

What they discovered was that the alloy tub had been badly constructed, the rivet holes had not been fully debured and the alloy panels that made up the tub were sliding against each other allowing the chassis to flex.

They rebuilt the tub with all new rivets and made sure the holes were deburd properly, that’s all they did, nothing else.

That car went on to win many races and won the BOSS series outright, the original designer was in tears apparently as his design was spot on, sadly the tub was bodged. Without that RAM may now still have been in F1 and a front-runner, such a shame.

Finally, take a look at a modern WRC car, the roll cages primary job is to ensure the chassis pickup points don’t move in relation to each other, safety is its secondary job, not its first.

I had the opportunity to take a good look at a bare focus shell last year as we were considering building our own Group A focus running escort cosworth components. Compared to an escort shell there are some huge box sections in high stress areas that are there solely for rigidity, that’s one of the reasons the Focus struggled at first as its heavy and the roll cage made the production shell stiffeners redundant.

So in summary, the stiffer the chassis the better. However, the suspension has to be more sophisticated to give you the mechanical grip you would have received from body flex.

Trout

iTrader: (1)

So what about the cav.

I use a plumbline and a sextant to measure the flex, but with Koni suspension it's great. Even better than that Lotus.

****************

Sorry John, it was a muppet moment

It happens to the best of us

[This message has been edited by Rannoch (edited 06 October 2000).]

johnfelstead

i dont know

A sensible informative subject at last and it takes 4 posts to take the **** *sigh*

dont you think its time we dropped the bollox from the posts that have some technical merit?

EvilBevel

Interesting read, thanks for that JF.

It occured to me that, hypothetically, if the "old" Impreza has a flex somewhere of say 1 mm, and the new one would only flex 0.4 mm, you could in marketing speak say the car is 250 % stiffer ...

Of course, these numbers are just made up for arguments sake.

I do suspect the "old" Impreza is quite stiff : not comparable to the above story obviously, but the car does seem to react to slight changes in toe, camber, tyre pressure ..., which, if I understood the above, seems to point at a chassis without to many problems.

Am I correct in this logic ? (not sure I explained it well...)

jeremy

This is the million dollar question that nobody (no magazine editor, or manufacturer rep, or prodrive rep etc.) seems fully knowledgable enough to ask- How rigid should a road car be, so that it can both handle well and crash well? I once asked Roger Becker the chief of handling development at Lotus this question and he simply said to me that you can never make a car too rigid, "the more rigid the better". And continuing Audi recently claim to have taken apart rival BMW's to emulate their handling ability, and claim that the overall rigidity of the car was not BMW's secret, instead it was the ridigity between the axles.

So it would seem that there are many different ways of making a chassis rigid, and many different places on a chassis where rigidity can be individually increased- betwwen the wheels, between the front or rear suspension mounts, at the axle. And this tells me one thing- that maybe the overall rigidity of a cars chassis from front to end, might therefore not be as important as the manufacturers would lead you to believe. Total front-to-back rigidity might not play as large a role in handling, because 250% between front and back might only pan out to be a 25% difference at the wheels. Maybe not as integral as the ridigity between the actual wheels, which themselves might play a stronger role. Witness the old Audi Quattro, this car has all manner of 'homemade' strenthening on its roof, in its rear bulk-head, the strenth of the shell used in the unibody and in the strength of its axles and struts. So in many respects that old car might be more rigid in some ways than some of the newer Audis.

I suppose the most rigid road cars of all time must be the carbon tub-ed ones such as the Buggati EB110, Ferrari F50, Mclaren F1, Mercedes and Porsche mid-engine special. And herein lies another problem- none of these cars were really ever built in the vain of true Impreza-like road cars. All were purpose built road cars built for maximum grip along race-tracks- with the odd trip through town every now and again. Therefore no-one ever got to sense if the rigidity of these cars actually gave any benifit towards the qualities road users favor in their drives, namely progression, predicability, and overall feel for the road.

So does the new Impreza's 250% increase really mean the car is really 250% more rigid, I doubt it. Its probably just a lot better in a crash. I'm sure the guys from Prodrive could confirm little actual handling advantage in a back to back comparison of P1 and 01'.

I've only dented the surface of this conversation, and it would be enlightening for real experts to comment on the seeming uncertainty of all this manufacturer hyperbole.

Cheeky Jim

What a great bit of writing John, that is the first time I have read something about flexing and rigidity and finally understood the dynamics of the topic in reasonable layman's terms.

That should be cut out and made into an article and posted under 'Scoobynet Articles of Interest'.

Webbie - whadya fink??

Jamie

AlexM

Ok John - here's a serious question...

On the current Impreza, what sort of improvements in torsional rigidity can be expected by fitting front and rear strut braces?.

I can hear trim creaking when turning and reversing onto an upward slope i.e. introducing a deflection into the rear left suspension on it's own...

In the context of a road car with mildly modified suspension, how stiff is stiff enough?. How stiff IS the current car? Are strut braces simply a cosmetic accessory, or do they work?.

Also home come the Integrale is still rated as one of the all time best handling cars and the delta is definately NOT stiff?. I know this because I was helping a friend change a rear tyre and the passenger door was catching on the sill..

Just curious...

Cheers,

Alex

Edit - John, I've just re-read your message - wouldn't this statement contradict the principles you describe? What am I missing here?.

'So in summary, the stiffer the chassis the better. However, the suspension has to be more sophisticated to give you the mechanical grip you would have received from body flex.'

[This message has been edited by AlexM (edited 05 October 2000).]

johnfelstead

I can’t give you figures to answer that question properly as I don’t have access to any comparison data between chassis. I will try and expand a little on road car chassis design, as I understand it however.

I will have to use examples of cars I understand more, sadly I am not an Impreza expert but the theories are relevant.

Traditional strut braces stiffen the strut top mounts to remove the inward flexing of the strut mounting points. They do not give any additional torsional rigidity to the chassis in relation to front/rear. All road cars twist about the center of the chassis. If you have a perfectly still strut to strut configuration, you will still have front left to rear right twisting moments.

IF you look at a WRC car they have done away with the traditional strut braces and now triangulate the strut tops back into the main area of the cage, this helps to send the loads into the main chassis/cage structure to limit the front/rear flex.

Some cars are better than others straight out of the box. Take the sierra cosworth and escort cosworth as an example. The strut tops on the escort cosworth are significantly closer to the front bulkhead than the sierra. You gain less improvement on the escort by installing a strut brace. It is still worth doing however.

Also the sapphire cosworth has a stronger rear chassis than the escort cosworth or sierra cosworth as there is a bulkhead running across the back of the seats that acts like a rear strut brace.

One of the most effective ways of increasing the rigidity of a road car shell is to seam weld all the joints, this is almost unheard of in road car production as its very expensive and time consuming and is not possible to be carried out by robots, all production cars are spot welded together. The Impreza 22B is spot-welded and then seam welded I believe to give the chassis more torsional rigidity.

Another thing to bear in mind is the suspension configuration, cars that run separate coils and dampers tend to put less flex on the chassis because the mounting points are lower, look at a cosworth rear suspension, the coil that transfers all the loading into the chassis is very low down in the chassis compared to a McPherson strut and so the twisting moments are lower side to side at the point where the load is fed into the shell. You will still have the same front to rear twisting however.

Too many people confuse stiff suspension with better handling. In most road conditions, making the suspension stiff slows it down, it feels faster but it isn’t. This is amplified in the wet in particular as the mechanical grip limitations are more easily reached. My Impreza Sport in wet conditions is quicker through the corners than a 22B because I have better mechanical grip. It rolls a great deal as it has no anti-roll bars, in the wet this is a big help but in the dry the roll starts to cause problems and limits the grip across all 4 tyres.

The big issue with car handling is that what is good on a country lane in the wet will be useless on a smooth racing circuit. You say the Integrali is one of the best handling cars, that doesn’t mean anything, it may be great on cross country B roads because it has very good mechanical grip that will be helped by a flexible chassis, get it on a smooth fast flowing road and it will not perform so well as the flexing will be a disadvantage in that situation.

The ultimate aim is to have the suspension controlling the tyres; this is the only way to get he most out of the car in all conditions. In order to achieve that very careful attention needs to be placed on wheel rates and damping characteristics in low/high speed bump rebound. All the F1 cars for example now use 5 way damping. They have low/high speed bump adjust, low/high speed rebound adjust and also blow off valves that allow full deflection on high impact occurrences like hitting a curb. That’s why you see the modern F1 cars mounting the cubs all the time and why McLaren stuffed everything with its first Adrian Newey Chassis, he sussed out blow off valve suspension before anyone else. It wasn’t the aerodynamics as the press kept spouting off.

Maybe what I have written has given you more questions than answers, its an incredibly complex question to answer but as long as you have a suspension package that can react quickly enough, a stiff chassis is what will give you the ultimate handling that is reproducible.

CharliePsycho

In my youth we used to prepare minis and mk1/2 escorts for club events, I learnt an awful lot about handling there...

The crucial point that johnfelstead is making is that a lot of suspension movement in road cars is in the body flex. For which it does an admirable job. Just like the seat plays an important role in absorbing bumps. Believe it or not a lot of thought goes into seat design for absorbing bumps transmitted though the body. Equally the body is tuned for *road* use.

Some flex is actually tuned in, notably flex between the front and rear axle is often tuned in to give an element of 'rear' steer (at least it was.)

On a track car you have completely different requirements. Importantly those requirements are much more specific, and they change on a track-track basis. You could in theory tune the body for each event, it is far better however to have a fully predictable platform on which to then hang far more tuneble components, i.e. tuneble suspension...

In the old days we used to rehang most of the suspension in a mini off the roll cage itself. Escorts used to bolt through to the mounting points of the cage (depending on peoples interesting 'live' axle games ) You could then very much ignore what the body was doing to a certain extent, which was why cars that looked like they were undriveable were still in fact competitve after a crash

Now the Impreza... It ain't broke... would you really notice the difference if I fixed it?

Andy Banks

A few things occur to me when reading all this

1) the 250% figure

it is based on the UK spec car presumably. Is the STi any stiffer? Wonder what the ratio would be between old STi : new STi...

2) stiff chassis, road spec suspension

Even if you have a chassis that is as stiff as a race car, if the suspension is designed for "road use" will the increased chassis rigitity provide much benefit? Isnt the benefit limited by the suspension setup aswell.

3) marketing

to have 250% more of anything is great for marketing people.

4) Focus/Corsa

It was mentioned that the Focus/Corsa are much stiffer than the Escort/Astra and the handling is in a different league. However, with those cars many other components will have been changed. The handling improvement is not down to the increase in chassis rigidity alone. It's easy for marketing people to quantify an increase in chassis rigidity, but less easy to explain how the designers have changed the suspension setup aswell. Basically, I wouldnt expect the new Impreza to show the same improvement as the Focus/Corsa did as the car is already well sorted.

Just my thoughts...thanks for the interesting posts!

Andy


[This message has been edited by Andy Banks (edited 05 October 2000).]

[This message has been edited by Andy Banks (edited 05 October 2000).]

jeremy

John,
It was great to hear your story about how a real person makes a real car better. #@$% knows that no manufacturer would ever so much as reveal the slightest construction technique used on their own.

So John, here is the thing- if you made the Esprit so much better through the specific modifications you made to the car, then why cannot the more adventurous from amoungst this net do the same to their Impreza's???

Might not the Impreza stand to make similiarly great strides in handling ability? Yes, you may say that only would it for the track. But I say why not give it a try, and see how ridigity might make Impreza's better in all conditions including the unfountunate crash.

Might this new form of 'mod' be more effective than all the other bits and pieces so many of us fit?

Triggaaar

John,
Thanks for all of that - a complex subject, well explained.

Although you don't concider yourslelf an impreza expert, can you give your opinions on the current range (how stiff each impreza is), and what gains can be had by adding strut braces (inc there effect on track and cross country).
Cheers