Board logo

Chassis. Which is best and why?
rdodger - 22/11/12 at 07:07 PM

I have been reading the topic about removing a tube in the chassis to fit the engine and it got me thinking about the different chassis designs.

Lotus, Caterham, MK, MNR, MAC#1 etc and the book chassis. They all have similarities, but some are quite different in terms of bracing etc. Round or square tube?

So... which is best?
Why?
How can they be improved?
What have you done?


loggyboy - 22/11/12 at 07:16 PM

String - how long is it?


phelpsa - 22/11/12 at 07:20 PM

Popcorn at the ready

It depends what you want your car to do. None of them are very good off road.


RK - 22/11/12 at 08:04 PM

OK, I'll go! The real answer is "I don't know" but as usual, that won't stop me! Square is easier to weld (apparently), than round, and makes fitting body work easier. I am fairly certain that as long as the required welding is done, and there are the requisite amounts of bracing, round is no better than square. I think that the Aussie mods will give you a pretty stiff, safe chassis, so check what that is, exactly and go from there. I don't see how you can go wrong with it anyways.

I am not an engineer!


ReMan - 22/11/12 at 08:18 PM

MK is the best chassis and I should know I've got one
Delight
Delight


deezee - 22/11/12 at 08:33 PM

Lets all guess until the loudest person wins. The Haynes Roadster is the best because science made it with maths and then magic took place and it was promoted to the finest sports car. More torsional rigidity than a Mclaren F1 and aerodynamically better than a F14 Tomcat.


Hope this helps.


mookaloid - 22/11/12 at 08:54 PM

one day it's my ambition to build a live axle striker - or a caterham - either would be at the top of my list for "bestness"


big_l - 22/11/12 at 09:28 PM

Surely the MNR is right up there the quality is Awsome and very nicely disigned but quality comes with a price tag !!


Talon Motorsport - 22/11/12 at 09:55 PM

The best chassis is one that does the job that you ask of it at the price that you can afford. And that is probably the most sensible thing I ever said on here.


daniel mason - 22/11/12 at 09:58 PM

after driving loads of sevens. the caterham is by far the best balanced and well sorted. but very very small in comparison to the others mentioned.
could all be down to setup.but it really feels a well sorted chassis!


JekRankin - 22/11/12 at 10:08 PM

If I ever decide to own another kit, it'll be a used Caterham. I've discovered that I like well developed components and quality of fit and finish too much to bother with many of the cheaper kits!

[Edited on 23/11/12 by JekRankin]


daniel mason - 22/11/12 at 10:34 PM

my s3 is a tight squeeze. im only 5 10" and 12 1/2 stone.


hughpinder - 23/11/12 at 09:42 AM

Ok, I'll also light the blue touchpaper -
Sylva J15, I believe one passed the aussie test unmodified, which I don't think anything else has with a standard chassis (of what I consider to be a kit - cateringvans are too dear).
If you want the square tube vs round debate in summary - a 25mm *1.5mm square tube has the same nomimal bending strength as a 25mm*1.5 round tube, so round is better as its would be lighter for the same strength, but a square tube has (4/3.14159...) times the weld area at each join and is easier to attach panels to and also to make jigs to assemble it accurately, so square is better. A whole chassis in square comes in about 65kg painted, so I guess a round tube chassis would be 12-15kg less. I wouldn't worry whick it is unless I was racing at the top end where the weight reduction of round may be worthwhile. (next you should ask if aluminum tube is better or T45....)
DAX cars seem to be nicely made last time I looked properly.
Regards
Hugh


franky - 23/11/12 at 10:00 AM

I'd go for anything by Jeremy Phillips or caterham.


mcerd1 - 23/11/12 at 10:34 AM

quote:
Originally posted by hughpinder
DAX cars seem to be nicely made last time I looked properly.

the chassis is nice
and they are quite big/wide and the eingine sits a bit further back compaired to some, but they arn't quite as light
they also have quite alot of bracing:
Rush Chassis
Rush Chassis


maybe more bracing than it really needs but I don't think I'd be happy with a RH B2 chassis:




besides the dax must be the best as they are based on german engineering (started out as a Mohr Rush)
{runs for cover }


phelpsa - 23/11/12 at 11:33 AM

quote:
Originally posted by hughpinder
If you want the square tube vs round debate in summary - a 25mm *1.5mm square tube has the same nomimal bending strength as a 25mm*1.5 round tube, so round is better as its would be lighter for the same strength


IIRC last time I checked you needed 28mm round tube to get the same 2nd moment of area as an equivalent walled 25mm square tube, giving you a weight saving of 13% for round.

The difference of round vs square is negligible compared to actual chassis design and member placement.

[Edited on 23-11-12 by phelpsa]


Hellfire - 23/11/12 at 01:07 PM

Caterham.

Phil


Alfa145 - 23/11/12 at 01:43 PM

Roadrunner SR2


umgrybab - 25/11/12 at 01:48 PM

quote:
Originally posted by phelpsa
quote:
Originally posted by hughpinder
If you want the square tube vs round debate in summary - a 25mm *1.5mm square tube has the same nomimal bending strength as a 25mm*1.5 round tube, so round is better as its would be lighter for the same strength


IIRC last time I checked you needed 28mm round tube to get the same 2nd moment of area as an equivalent walled 25mm square tube, giving you a weight saving of 13% for round.

The difference of round vs square is negligible compared to actual chassis design and member placement.

[Edited on 23-11-12 by phelpsa]


I always knew a square tube was way stiffer in bending and I just ran the calcs on this, and for the 25mm square tube with a 1.5mm wall thickness, the equivalent round tube with a 1.5mm wall thickness would need to be 52.5mm in diameter. This does not answer everything as how often do you have a member in pure bending? Hopefully never is the correct answer. Like said above, chassis design and member placement is much more important.


phelpsa - 25/11/12 at 07:34 PM

quote:
Originally posted by umgrybab
quote:
Originally posted by phelpsa
quote:
Originally posted by hughpinder
If you want the square tube vs round debate in summary - a 25mm *1.5mm square tube has the same nomimal bending strength as a 25mm*1.5 round tube, so round is better as its would be lighter for the same strength


IIRC last time I checked you needed 28mm round tube to get the same 2nd moment of area as an equivalent walled 25mm square tube, giving you a weight saving of 13% for round.

The difference of round vs square is negligible compared to actual chassis design and member placement.

[Edited on 23-11-12 by phelpsa]


I always knew a square tube was way stiffer in bending and I just ran the calcs on this, and for the 25mm square tube with a 1.5mm wall thickness, the equivalent round tube with a 1.5mm wall thickness would need to be 52.5mm in diameter. This does not answer everything as how often do you have a member in pure bending? Hopefully never is the correct answer. Like said above, chassis design and member placement is much more important.


I've just done the calcs and got 30mm OD for equivalent wall thickness.


jossey - 25/11/12 at 07:49 PM

Best chassis is the one you build yourself. :-)

Tiger chassis's are good cos everyone is slightly different lol


garybee - 25/11/12 at 10:14 PM

I'm going to go the other way to everyone else and say...NOT a Ginetta G27

It is absolutely awful. It's so bad that I actually had to cut out the section between the rear suspension turrets and weld in something that I would be happy to bolt my seat belts to (original was simply not safe). As for the rest of it, one day I will give the car a full roll cage and a proper transmission tunnel to give it some shred of rigidity.


scootz - 27/8/13 at 09:44 PM

quote:
Originally posted by garybee
I'm going to go the other way to everyone else and say...NOT a Ginetta G27

It is absolutely awful. It's so bad that I actually had to cut out the section between the rear suspension turrets and weld in something that I would be happy to bolt my seat belts to (original was simply not safe). As for the rest of it, one day I will give the car a full roll cage and a proper transmission tunnel to give it some shred of rigidity.



Thread revival - I'm surprised to read that you think the Ginetta chassis is so sloppy as they have a strong heritage of building proper racecars!


scootz - 27/8/13 at 09:46 PM

PS - I found this thread when searching for 'torsional rigidity'.

I see the Quantum Xtreme (of which I have a renewed vested interest in) has a torsional stiffness in excess of 4000nm. What would be the norm for a typical 7-style spaceframe?


britishtrident - 27/8/13 at 09:59 PM

quote:
Originally posted by umgrybab

I always knew a square tube was way stiffer in bending and I just ran the calcs on this, and for the 25mm square tube with a 1.5mm wall thickness, the equivalent round tube with a 1.5mm wall thickness would need to be 52.5mm in diameter. This does not answer everything as how often do you have a member in pure bending? Hopefully never is the correct answer. Like said above, chassis design and member placement is much more important.


I think you need to check the maths again


rodgling - 27/8/13 at 10:06 PM

quote:
Originally posted by garybee
I'm going to go the other way to everyone else and say...NOT a Ginetta G27

It is absolutely awful. It's so bad that I actually had to cut out the section between the rear suspension turrets and weld in something that I would be happy to bolt my seat belts to (original was simply not safe). As for the rest of it, one day I will give the car a full roll cage and a proper transmission tunnel to give it some shred of rigidity.


Interesting, I think the GKD chassis is loosely derived from a Ginetta and it looks pretty over-engineered to me, if anything. Not that I know very much about chassis design mind.


britishtrident - 27/8/13 at 10:08 PM

One of the problems with square tubes is when under torsion you get nasty stress concentrations at corners where the tube meets another tube at welded joints, however in any decently design space frame this should not be a problem.

[Edited on 27/8/13 by britishtrident]


phelpsa - 28/8/13 at 12:20 AM

quote:
Originally posted by scootz
PS - I found this thread when searching for 'torsional rigidity'.

I see the Quantum Xtreme (of which I have a renewed vested interest in) has a torsional stiffness in excess of 4000nm. What would be the norm for a typical 7-style spaceframe?


Probably nowhere near that, but why do you need that sort of stiffness?


MikeRJ - 28/8/13 at 07:13 AM

quote:
Originally posted by phelpsa
quote:
Originally posted by scootz
PS - I found this thread when searching for 'torsional rigidity'.

I see the Quantum Xtreme (of which I have a renewed vested interest in) has a torsional stiffness in excess of 4000nm. What would be the norm for a typical 7-style spaceframe?


Probably nowhere near that, but why do you need that sort of stiffness?


To allow the suspension to do it's job.


mcerd1 - 28/8/13 at 07:49 AM

quote:
Originally posted by MikeRJ
quote:
Originally posted by phelpsa
quote:
Originally posted by scootz
PS - I found this thread when searching for 'torsional rigidity'.

I see the Quantum Xtreme (of which I have a renewed vested interest in) has a torsional stiffness in excess of 4000nm. What would be the norm for a typical 7-style spaceframe?


Probably nowhere near that, but why do you need that sort of stiffness?


To allow the suspension to do it's job.


or to get on the road down under with a 4cyl engine
(I think its 6000Nm/deg for V8's )

if you google for folk who have imported to / built cars in Oz you'll see a lot of talk about torsional stiffness

[Edited on 28/8/2013 by mcerd1]


nick205 - 28/8/13 at 08:42 AM

Without resorting to Google, what sort of torsional stiffness do modern tin tops and their cabriolet versions achieve?


A quick Google suggests a figure of 35,000Nm/degree for the Lamborghini Aventador. Obviously the high end of the scale

[Edited on 28/8/13 by nick205]


motorcycle_mayhem - 28/8/13 at 11:07 AM

The chassis on my series landrover (109) is pretty strong, and that's only 2.5mm box section. There's no round tube on it, just a simple ladder frame. Really needs the suspension upgrading, with pickup points for wishbones (wishbones will have to be fabricated as well, etc.). With a good Pinto engine, still the ultimate powerplant after all these decades, should be all you would want.

Tube size and strength:

If the wall thickness of a tube is increased then the ability to support bending loads increases by the same amount. If the wall thickness is doubled then the ability to support bending loads is also doubled.

If the size of a tube is increased then the ability to support bending loads is increased by the square of the increase in size and the tube stiffness is improved by the cube of the increase in size. If the diameter of a tube is doubled then the load carrying ability for bending loads is improved four times and the stiffness under bending loads is increased by eight times.

Due to these effects larger sized tubes with thinner walls are usually better than smaller thick walled tubes of the same overall weight.


mcerd1 - 28/8/13 at 12:17 PM

quote:
Originally posted by motorcycle_mayhem
Due to these effects larger sized tubes with thinner walls are usually better than smaller thick walled tubes of the same overall weight.

but if you go too thin the tubes can be easily damaged - and a bent/dented tube is not much use to anyone


PhillipM - 29/8/13 at 03:06 PM

quote:
Originally posted by scootz
PS - I found this thread when searching for 'torsional rigidity'.

I see the Quantum Xtreme (of which I have a renewed vested interest in) has a torsional stiffness in excess of 4000nm. What would be the norm for a typical 7-style spaceframe?


Not sure, but they all seem kinda low, our buggy tested out at well over 36,000nm/degree when it was first built.


scootz - 29/8/13 at 08:32 PM

quote:
Originally posted by PhillipM
quote:
Originally posted by scootz
PS - I found this thread when searching for 'torsional rigidity'.

I see the Quantum Xtreme (of which I have a renewed vested interest in) has a torsional stiffness in excess of 4000nm. What would be the norm for a typical 7-style spaceframe?


Not sure, but they all seem kinda low, our buggy tested out at well over 36,000nm/degree when it was first built.



Was it machined from titanium billet!


PhillipM - 29/8/13 at 09:17 PM

Titanium isn't very stiff, just fyi


phelpsa - 29/8/13 at 09:45 PM

quote:
Originally posted by MikeRJ
quote:
Originally posted by phelpsa
quote:
Originally posted by scootz
PS - I found this thread when searching for 'torsional rigidity'.

I see the Quantum Xtreme (of which I have a renewed vested interest in) has a torsional stiffness in excess of 4000nm. What would be the norm for a typical 7-style spaceframe?


Probably nowhere near that, but why do you need that sort of stiffness?


To allow the suspension to do it's job.


No you don't. You need a certain level of stiffness to allow weight transfer balance control, but the suspension will work quite nicely with very little torsional stiffness. I'm just asking the question, what makes you think that that level of stiffness is required? Not saying it isn't, just asking why!

In reply to the question about modern tin tops, numbers in the region of 10,000Nm/deg are common these days for the chassis, which is very good. This mostly to reduce NVH though, and all the compliance in suspension bushes means that hub to hub torsional stiffness isn't that brilliant.


phelpsa - 29/8/13 at 09:48 PM

quote:
Originally posted by PhillipM
Titanium isn't very stiff, just fyi


Depends how you use it...


PhillipM - 29/8/13 at 10:07 PM

With my budget?
About the only place I can use it is in my razor


MikeRJ - 30/8/13 at 11:38 AM

quote:
Originally posted by phelpsa


No you don't. You need a certain level of stiffness to allow weight transfer balance control, but the suspension will work quite nicely with very little torsional stiffness. I


I strongly disagree. How do you damp the motion where the chassis is twisting rather than springs compressing/extending?


TheGiantTribble - 30/8/13 at 02:40 PM

Raises head above the parapit...

You choose your suspension settings to work with how stiff or not your chassis is

In older racing cars, the drivers would 'read' the twisting of the cars as they went round corners.
The tyre pressure and suspension settings/stiffness were all softer so as to work with (or because of) the lack of chassis stiffness.

eg

In a modern formula one car, 50 percent of movement is in the tyres...you need a very stiff chassis to make this work.
In say a 1960's Formula one car, much less stiff chassis in fact almost jelly....lots of suspension movement (and a lot more fun to watch it all working IMHO)

Drops rapidly back behind shelter.


Camber Dave - 10/9/13 at 08:52 PM

I am building mine with 25mm square main tubes as per the book. So easy to fit floor and sides.
But for the side diagonals I am using 19mm round tube flush with the outer face.
I used 2 diagonals for each 'box' with a tricky joint in the middle.
This gives a bigger weld area into the corners to fully triangulate the sides.
I will just Sikaflex the panels to the round tubes as they will be riveted and glued around their edges.


Bumble - 10/9/13 at 09:58 PM

Dutton is the way to go!!!!!!!!!!