What makes a good chassis?
Discussion
This is a very good topic, and the number of answers are along the lines of "what makes a good song?" i.e. everyone will have their own opinion/taste.
Here are some things to consider though...
1. Stiffer is generally better.
2. Handling comes not just from the chassis but also the suspension setup.
3. Weight distribution - not just front/rear but also height of COG.
4. Wheelbase to track ratio is actually very important.
Here are some things to consider though...
1. Stiffer is generally better.
2. Handling comes not just from the chassis but also the suspension setup.
3. Weight distribution - not just front/rear but also height of COG.
4. Wheelbase to track ratio is actually very important.
Stiffness is very important - you need the chassis to be as stiff as possible (in general) to mount the suspension to.
If the suspesion mountings can move, then the suspension cannot work effectively - it has to be rigidly mounted so the shocks are doing their job as they are designed, and also not taking up any twisting or other forces coming as a result of chassis deflection.
On a car designed to handle well have an optimum wheelbase/track ratio of (IIRC) 1.6:1?
This is obviously different cor cars with comfort as a priority, where a longer wheelbase smooths out bumps.
Theres lots to think about!
If the suspesion mountings can move, then the suspension cannot work effectively - it has to be rigidly mounted so the shocks are doing their job as they are designed, and also not taking up any twisting or other forces coming as a result of chassis deflection.
On a car designed to handle well have an optimum wheelbase/track ratio of (IIRC) 1.6:1?
This is obviously different cor cars with comfort as a priority, where a longer wheelbase smooths out bumps.
Theres lots to think about!
The phrase "a good chassis" has little relationship to the actual stiffness of the design. There is a complex interation between suspension, tyres, size and the structure of the chassis itself.
Many cars handle very well with low stiffness but these are usually very light weight cars.
Typically, and assuming equal standards of design, ladder frames are the least stiff, then spaceframes and finally monocoques which are the stiffest.
However, and this is very important, equal design standards are very rare and most spaceframes are designed on the basis of blind faith that they are actually half as good as they could be.
As a very (very) rough guide in ftlbs-
Spaceframes 400 (almost nothing!) to 5000 (the Locost is circa 1200 while the Ultima is circa 3300)
Ladder frames almost zero (if made with C channels) to 4000
Monocoques 4000 (Morris minor) to 20000 (a big estate car)
Note how much the ladder and spaceframe ranges overlap.
For equal stiffness a ladder, made of box sections and having some 3D features such as a welded on footwell structure, is roughly 5 to 10 percent heavier for the finished car. On a race car this would be a non starter, unless you could save the weight somewhere else or drop the stiffness requirement a bit and still win. For a road car the case is less clear cut as you have to consider ease of access, doors that you can get in and out of and long term durability. The Ladder is also simpler and cheaper to make and to modify. Most manufactures deal with these issues by using a backbone chassis (TVR and original Lotus Elan) which sacrifice elbow room over the transmission tunnel for doors and access elsewhere while still having a fairly simple overall concept. Again that is a very generalised statement.
Many cars handle very well with low stiffness but these are usually very light weight cars.
Typically, and assuming equal standards of design, ladder frames are the least stiff, then spaceframes and finally monocoques which are the stiffest.
However, and this is very important, equal design standards are very rare and most spaceframes are designed on the basis of blind faith that they are actually half as good as they could be.
As a very (very) rough guide in ftlbs-
Spaceframes 400 (almost nothing!) to 5000 (the Locost is circa 1200 while the Ultima is circa 3300)
Ladder frames almost zero (if made with C channels) to 4000
Monocoques 4000 (Morris minor) to 20000 (a big estate car)
Note how much the ladder and spaceframe ranges overlap.
For equal stiffness a ladder, made of box sections and having some 3D features such as a welded on footwell structure, is roughly 5 to 10 percent heavier for the finished car. On a race car this would be a non starter, unless you could save the weight somewhere else or drop the stiffness requirement a bit and still win. For a road car the case is less clear cut as you have to consider ease of access, doors that you can get in and out of and long term durability. The Ladder is also simpler and cheaper to make and to modify. Most manufactures deal with these issues by using a backbone chassis (TVR and original Lotus Elan) which sacrifice elbow room over the transmission tunnel for doors and access elsewhere while still having a fairly simple overall concept. Again that is a very generalised statement.
Alot of testing with a driver who enjoys an intimate drive is what makes a good chassis I'd say.
If it was purely down to mechanics/maths then you'd likely end up with a dull boring car. The reality is the whole setup is a big compromise on what you want. You might find 5% more grip due to optimising bump camber but in doing so you then get 25% more bump steer toe change.
So no real thing as a 'good' chassis. Just loads of different takes on it.
Simple things like adding gripper tighter rubber to a chassis/suspension setup that is good on road tyres might make it terrible due to the increased suspension deflections and camber/toe optimisations under deflection that are not optimised for the 'better' tyres different optimum slip ratio etc.
Too many variables to put it down to one thing.
I guess the most important factor is that the test driver isn't a square, his engineers are good at interpreting his comments, and no one who doesn't know what they are talking about isn't allowed a say in it
(ie, manager/consultant types)
Many people say that the 205 GTi handled well, you could say it had a good chassis for the requirements of the car, but it was so basic and un-complicated. McP's up front, and a static toe/camber perpendicular to pivot trailing arm (torsion beam) rear. Just really simple, but the way those parts were setup was key!
I'm throwing suspension in with chassis here because the two are totally inter-related, ultimately it's all about controlling the tyre contact patches and tyre loadings.
Dave
If it was purely down to mechanics/maths then you'd likely end up with a dull boring car. The reality is the whole setup is a big compromise on what you want. You might find 5% more grip due to optimising bump camber but in doing so you then get 25% more bump steer toe change.
So no real thing as a 'good' chassis. Just loads of different takes on it.
Simple things like adding gripper tighter rubber to a chassis/suspension setup that is good on road tyres might make it terrible due to the increased suspension deflections and camber/toe optimisations under deflection that are not optimised for the 'better' tyres different optimum slip ratio etc.
Too many variables to put it down to one thing.
I guess the most important factor is that the test driver isn't a square, his engineers are good at interpreting his comments, and no one who doesn't know what they are talking about isn't allowed a say in it
(ie, manager/consultant types)Many people say that the 205 GTi handled well, you could say it had a good chassis for the requirements of the car, but it was so basic and un-complicated. McP's up front, and a static toe/camber perpendicular to pivot trailing arm (torsion beam) rear. Just really simple, but the way those parts were setup was key!
I'm throwing suspension in with chassis here because the two are totally inter-related, ultimately it's all about controlling the tyre contact patches and tyre loadings.
Dave
Edited by Mr Whippy on Friday 6th July 09:45
There are lots of different factors involved, but I'd say that the crux of the issue is that the grip available at the tyres should be consistent and linear in response under all circumstances. This ensures that the car handles predictably as you negotiate a corner, accelerate or brake.
Chassis stiffness is relatively unimportant these days, for road cars. The obsession with stiffness dates back to pre-WW2 vintage cars when the ladder frames flexed by an enormous amount and acted like undamped springs.
More important is that there should be no sudden changes in grip caused by changes in tyre camber and diagonal weight transfer when cornering should be linear with a slight understeer bias.
Dampers and steering feel are also incredibly important; the former to make sure that the tyre has consistent contact with the road and the latter to give you information about how much grip you have left.
How to actually achieve these characteristics is another matter altogether, though.
Chassis stiffness is relatively unimportant these days, for road cars. The obsession with stiffness dates back to pre-WW2 vintage cars when the ladder frames flexed by an enormous amount and acted like undamped springs.
More important is that there should be no sudden changes in grip caused by changes in tyre camber and diagonal weight transfer when cornering should be linear with a slight understeer bias.
Dampers and steering feel are also incredibly important; the former to make sure that the tyre has consistent contact with the road and the latter to give you information about how much grip you have left.
How to actually achieve these characteristics is another matter altogether, though.
cymtriks said:
Many cars handle very well with low stiffness but these are usually very light weight cars.
Too true. 60's Lotus Elan had a wonderful chassis. The chassis stiffness is pitiful by todays standards, but it was matched perfectly to the cars mass, power and suspension. Matching tyre size (diameter as well as width) to the overall chassis is also essential.Stiffness is usually key, because it gives you greater control of the parts which are flexing if that is pretty much limited to suspension components etc. There are some vinatge cars that handle well, with very low torsional rigidity by modern standards (and comparitively stiff suspension) but I suspect that is largely due to low weight, little grip and a good deal of fine tuning!
So, yeah stiffness firstly. Particularly on the front end around the steering!
Reasonable suspension geometry and plenty of adjustability. Good old double wishbones are probably the best way of doing this simply.
Light weight - still underestimated as a factor, whatsmore the mass there is should be as low as possible and (within reason) fairly central. The job the suspension has to do in order to control weight transfer is very much easier if the C of G isn't moving as far due to a low C of G and when the car reaches a steady(ish)-state one of the most fundamental factors affecting the weigh it handles is weight distribution.
So, yeah stiffness firstly. Particularly on the front end around the steering!
Reasonable suspension geometry and plenty of adjustability. Good old double wishbones are probably the best way of doing this simply.
Light weight - still underestimated as a factor, whatsmore the mass there is should be as low as possible and (within reason) fairly central. The job the suspension has to do in order to control weight transfer is very much easier if the C of G isn't moving as far due to a low C of G and when the car reaches a steady(ish)-state one of the most fundamental factors affecting the weigh it handles is weight distribution.
busta said:
So how is a chassis' stiffness measured?
In units? By degrees of chassis twist per ft.lb or Nm of torsion.Practically? By bolting the suspension mounts at one end to a bloody great mass, bolting the other end to a bloody great beam with a known weight hanging off it, and using a dial guage to measure the deflection.
The people who have actually done this will be the ones who tell you that it isn't as important as the other people make out. When you measure the actual deflection that you get at realistic torsional loads, you realise that they are pretty trivial compared to the deflection of the springs and those big, squishy rubber balloons that connect you to the road!
Gassing Station | General Gassing | Top of Page | What's New | My Stuff