Please discuss here all your remarks and pose your questions about all racing series, except Formula One. Both technical and other questions about GP2, Touring cars, IRL, LMS, ...
Arguably, unusual roll cage designs in saloon race cars are interesting, even if it is an aspect of a saloon racing car that probably doesn't make all that much difference to performance (probably with no "correct" solution and alternative designs probably having no real advantage over what one might call a "conventional" design?). It's interesting to see roll cage designs which are different than the norm IMO.
Here's the RML Chevrolet Cruzes from the WTCC with their unique not-door bars running to the top of the main hoop:
This 1995 Vectra super tourer has quite a substantial amount of reinforcement of the engine bay indeed! It seems like it would make installing the engine, albeit done from underneath, quite a pain!
Here's the extra triangulation around the C-pillar of Prodrive-built cars from the 90's and early 00's. The example is the Prodrive-built Super Tourer BMW but Prodrive-built Subaru WRCars used a similar design:
Larry Perkins adopted a somewhat similar double roll hoop design in a push for more rigidity in the early 00's, which was later dropped as being (probably) unnecessary. Note that this Perkins car does not seem to have much bar work and triangulation added to the floorpan however, unlike V8 touring cars built by rival teams around the same time.
The "Larry bar", a diagonal roll bar across the windscreen introduced by Perkins and permitted in Australian touring car racing at the time can also be seen above... It was banned in 2013 on the move of the category to a control chassis design.
Before that though, one team, Garry Rogers Motorsport, even ran a double diagonal (X) across the windscreen which rule makers soon deemed a hinderance to driver vision and banned!
By comparison to the Perkins car, this is the triangulation added to the floorpan of a Stone Brothers Falcon:
A roll cage in a saloon car can be a huge performance differentiator.
You are thinking about them only from a safety point of view, the car designer thinks about them from a structural reinforcement point of view.
Super Tourer's were in theory a monocoque construction, however most of the loads actually went through the roll cage, making them space framed, technically speaking.
It is rumoured, but I have never seen proof, that towards the end of the super touring era that the teams had even stopped taking complete shells to fit cages to, there we build 'roll cages' and then assembling the panels around the cage.
Super Tourer's were in theory a monocoque construction, however most of the loads actually went through the roll cage, making them space framed, technically speaking.
The bracing runs right through the middle of this Dodge Stratus Super Tourer!
Although it obviously makes it impossible to passenger rides with the car, and I'm not sure all sanctioning bodies allow such construction?
A lot of roll cages seem to omit that rearward mid-tier bracing from the horizontal cross-bar in the main hoop -- I guess because it makes it difficult to access things like the fuel pumps or driving cooling items located in that area?
It's a little difficult to see, but there is one rollbar running diagonally across the passenger side of this right-hand-drive car, and another running diagonally from the passenger side up to the centre of the roll hoop through the main cabin. There were asked to omit the latter from later builds, as it was felt it made it difficult for the driver to exit through the passenger side of the car (heck even the other bar is also a hinderance to doing passenger rides with the car!):
You can’t look at these without knowing the rule set they were built to. Attachmanent points, tubing type / diameters, door bar designs, etc. were all dictated by the rules.
So in addition to being safety items, they are peformance differentiators and the unibody designs of touring cars dictate what they want to do. As for the “ship in a bottle” they’ve been like that since the 1970s. IMSA and then SCCA in the 1970s went to tube chassis silohoutte classes because the cars in effect already were, just that the stock body was just drapped around it. Purposely allowing silohoutte tube chassis cars wasn’t a performance differentiator and the thinking was they’re easier / cheaper to repair then a “ship in a bottle” car built around a production chassis.
You can’t look at these without knowing the rule set they were built to. Attachmanent points, tubing type / diameters, door bar designs, etc. were all dictated by the rules. No different than F1 and it’s aero. It’s all highly prescribed by the rules and not always look that way if the rules were more free.
So in addition to being safety items, they are peformance differentiators and the unibody designs of touring cars dictate what they want to do. As for the “ship in a bottle” they’ve been like that since the 1970s. IMSA and then SCCA in the 1970s went to tube chassis silhouette classes because the cars in effect already were, just that the stock body was just drapped around it. Purposely allowing silhouette tube chassis cars wasn’t a performance differentiator and the thinking was they’re easier / cheaper to repair then a “ship in a bottle” car built around a production chassis. By the 1990s, production based unibody touring cars were back as viewers liked that they were watching real production cars (regardless of their ship in a bottle chassis) and not fiberglass / carbon fiber bodies over a purpose built racing chassis. Of course rules creep has brought us back to that in all major series.
Whenever I see bends in the middle of tubing, as in the off road vehicles above, all I can think is that's a good place for the tube to start failing. I sort of get the same idea when seeing all the parallel tubes arranged in a rectangular pattern; not very effective.
Whenever I see bends in the middle of tubing, as in the off road vehicles above, all I can think is that's a good place for the tube to start failing. I sort of get the same idea when seeing all the parallel tubes arranged in a rectangular pattern; not very effective.
Thanks for the insights!
So a bend that isn't supported by being a node with other tubes (or at least supported with a gusset like the RML Chevrolet) is a weak point?
Re: the parallel tubes & rectangles, parallel [square] tubes & rectangles is often how the original frame rails of a unibody car or frame rails separate chassis car are arranged though, aren't they? Triangulation is often minimal or even omitted as it gets in the way of driveshafts and other components?
[In some of the rock crawlers and US sanctioned cars, I think the side impact bars are mandated as being arranged parallel.]
Did Glenn Seton Racing, who built the below Ford Falcon, err when they arranged the seat mount & floor reinforcement tubes as parallel tubes & rectangles without triangulation? (To be fair, some other V8 Supercars of that era still had no roll bars reinforcing the floor pan at all. Note: the original main hoop cross bracing and rearwards triangulation was removed as the car was converted to a ride car with a third seat in the back!)
IIRC 888 Race Engineering, who were very successful, similarly used parallel and rectangular tubes in the floor and not triangulation unlike the Stone Brothers Falcon linked at the top.
Indeed when 888 built their first Ford Falcons in 2004, their roll cage was fairly insubstantial compared to other competitors (or even the Falcons from the year before from the team they purchased) -- maybe started from the point of view of minimising weight and would prefer to revise the design later if necessary?
Last edited by JordanMugen on 10 Dec 2024, 19:56, edited 2 times in total.
So in addition to being safety items, they are peformance differentiators and the unibody designs of touring cars dictate what they want to do.
Is minimising cost and weight also a factor? I.e., minimising fabrication costs for mass-produced cars like GT4 and TCR? Or even minimising fabrication costs for GT3 despite the higher price cap?
I see that GT3s often have very low door bars to make the cars more suitable for wealthy amateur drivers too, or in the case of the less expensive TCR just plain amateur drivers, lol (whereas many V8 Supercars had, and still have under the spec spaceframe era, very difficult entry and egress!).
Typical V8 Supercar high door bars (I note some sanctioning bodies only allow doors bars to be as high as 50% of the main hoop height?):
Typical GT3 or TCR low door bars: (Should JAS Motorsport have gusseted that sharp bend on the forward hoop of this TCR Civic? Or it's OK as the entire top front corner reinforcement bar is in effect gusseting it?)
As more modern original road car bodies get stiffer and use more high-strength steel, is it no longer necessary to construct such an elaborate roll cages to obtain the desired torsional stiffness?
I see that the Lexus GT3 and Multimatic Mustang GT3 use the most basic door bar permissible on the passenger side, for example.
Multimatic Mustang:
{AFAIK the Mustang is based on the production pressings, although I'm not sure that is mandatory in GT3 -- I think the Bentley Continental was a pure tubeframe car, please correct me if I'm wrong.]
Note also: just primer on the Mustang which is sensible surely, whereas JAS Motorsport (and many others too!) still use "look good" primer and metallic grey and clear coat to this day for some reason, in 90's Super Touring tradition -- despite the extra weight.
You can’t look at these without knowing the rule set they were built to. Attachmanent points, tubing type / diameters, door bar designs, etc. were all dictated by the rules. No different than F1 and it’s aero. It’s all highly prescribed by the rules and not always look that way if the rules were more free.
TCRs seem to vary a lot from what I've seen. I don't think there are that many restrictions on elaborate cage work or even lots of carbon-fibre bodywork or whatever (please correct me if I'm wrong) -- the restriction seems to be more that you have to deliver the car for the cap price without making a loss and that's what stops you from going to town and building a ÂŁ900,000 Super Tourer?
E.g., some of the TCRs have the front-end bodywork (like the wheel arches) cut away and replaced with lightweight carbon yet some don't, some have carbon-fibre tailgates yet some don't etc. They do seem to be allowed to run the roll cage through the firewall and up to the (mandatory) strut-type suspension tower tops, which not all production-based categories allow.
For instance some have bar work in the floor (like the Alfa Romeo TCR) while others like the Mk1 Peugeot 308 have none at all -- and very limited rearwards triangulation too in the Peugeot, unlike say the Audi. The Peugeot design doesn't seem like it can all be that good (there's a lot "missing" in the rear of the cage), yet the car is competitive so I guess PeugeotSport know what they doing, maybe the 308 is inherently stiffer than lesser rival hatchbacks!
Alfa Romeo TCR:
Audi TCR (Gen 1)"
Peugeot 308 TCR:
Hyundai have also joined in on the trend of minimising the passenger side door bars to the regulatory minimum!
I guess the analysis shows that it makes very little difference to the car stiffness and saving the weight is more important? [They'd do it on the driver's side too if there wasn't a driver who would probably like better side impact protection than a 1980's touring car?! Granted, a door panel with impact foam is now common.]
I guess the main hoop and forward hoop effectively have their base boxes boxed in to the sill of the unibody so running a roll bar between those points is not necessary, as the chassis sill itself forms that "bar" -- even though the V8 Supercar designs above did so.
------
Is it the case that the V8 Supercar designers got carried away? An ex-V8 Supercar designer went to the Prodrive Subaru World Rally team and redesigned the Impreza WRC with roughly 50% more roll bars circa 2004-2006, but it didn't make the Impreza WRC any more competitive! Was it all just dead weight (especially an unnecessary cross at the highest point of the rollcage?) and unnecessary fabrication time?!
Here's some of the V8 Supercar-inspired Impreza WRC cages... If anything it was a lot less competitive than earlier 2001-2002 Impreza WRCs (so actually floppy is good? less weight is good? -- or it's just a coincidence?).
You can see they've deleted the horizontal bar between the rear strut tower tops as unnecessary by the time of the 2006 car (it was still four-corner strut on that era Impreza, in theory great for rallying! although maybe that was already allowed to be changed from the road car under WRCar as opposed to Group A).
Is it a delicate balancing act of deleting unnecessary stiffening bars to save weight, while still achieving the desired torsional stiffness?
Indeed, also making sure that the stiffening bars are ideally lower in the cage -- like on the floor -- rather than higher up and also lowering the main hoop so that the whole roll cage is lower in the car to also lower the COG (assuming the rules allow for placing the driver low enough in the car for this)?
It's obviously easy to go overboard and triangulate everything like crazy adding hundreds of hours of fabrication time and unnecessary weight! (Albeit lighter chrome molybdenum, chrome manganese or Docol(R) steel unlike the heavy cold-drawn mild steel which an amateur club racer would likely be using in their home-built car.)
Yes, a bend in a nominally tension/compression structural member reduces its buckling strength, places additional (real, not simple elastic theory) moments on its terminations, and reduces its stiffness axially and probably in torsion. However in a crash it acts as a hinge and is a great way of absorbing energy plastically, and allows you to control where the deformation takes place.
It is rumoured, but I have never seen proof, that towards the end of the super touring era that the teams had even stopped taking complete shells to fit cages to, there we build 'roll cages' and then assembling the panels around the cage.
Holden Commodore built here by placing the sides and roof on top of the roll cage afterwards!
Supposedly quite the time saving without the need to delete brackets and with the better access, and the tolerances of the chassis jig might be tighter than the OE production line perhaps?
Whenever I see bends in the middle of tubing, as in the off road vehicles above, all I can think is that's a good place for the tube to start failing. I sort of get the same idea when seeing all the parallel tubes arranged in a rectangular pattern; not very effective.
Thanks for the insights!
So a bend that isn't supported by being a node with other tubes (or at least supported with a gusset like the RML Chevrolet) is a weak point?
Re: the parallel tubes & rectangles, parallel [square] tubes & rectangles is often how the original frame rails of a unibody car or frame rails separate chassis car are arranged though, aren't they? Triangulation is often minimal or even omitted as it gets in the way of driveshafts and other components? :?:
Did Glenn Seton Racing, who built the below Ford Falcon, err when they arranged the seat mount & floor reinforcement tubes as parallel tubes & rectangles without triangulation?
straight tubes will anyway fail by buckling eg if they are 'too long'
cross-bracing even without triangulation will increase a tube's resistance to buckling
strength isn't the same thing as stiffness
and regarding 'stronger' steel - stronger isn't stiffer ie different strength steels don't have different elastic moduli
the 'stronger' steel structure may only be stronger in a crash
Last edited by Tommy Cookers on 10 Dec 2024, 21:54, edited 1 time in total.
strength isn't the same thing as stiffness ....
and regarding 'stronger' steel - stronger isn't stiffer ie different strength steels don't have different elastic moduli
the 'stronger' structure may only be stronger in a crash
Thanks for clarifying that a Docol pressing, for example, isn't going to make a road car bodyshell stiffer, it will only affect the crash performance. I had the wrong idea there.
Also, so a chrome molybdenum or Docol rollcage (with thinner tubing than mild steel) needs the extra bracing to have the same stiffness? Or for the same weight it will still be stiffer as the lighter material allows for a lot more roll bars and triangulation for the same rollcage weight than a mild steel roll cage?
I guess it depends on the design and what the FEA simulation, or hand calculations of the spaceframe if you like trigonometry (nope, only stresses not elastic deformation), shows? Though FEA is perhaps better for taking into consideration the contribution of the complex-shape sheet metal pressings to the overall structure?
... Also, so a chrome molybdenum or Docol rollcage (with thinner tubing than mild steel) needs the extra bracing to have the same stiffness?
yes sort of
the strength specific to cross-sectional area is greater (aka it's a stronger steel) ....
but yes you have used this benefit to enable a reduced wall-thickness and/or diameter .... ie a lighter spaceframe
the direct strength of a your new tube is unchanged ....
but buckling strength has reduced - this due to its (reduced) dimensions (the elastic modulus hasn't changed)
this will demand closer-pitched bracing ie more & smaller bays (eg the 'birdcage' Maseratis)
or a material of low density and thicker-wall (eg magnesium alloy Porsches 917 etc or M-B W196)
in principle your frame will equally strong - and less stiff - but lighter
they always gave themselves a margin on buckling strength
(allowing afaik eg works Lotus 18s to use thin high-strength DOM steel and the customer 18s used cheap ERW stuff)
DOM = drawn over mandrel aka cold-drawn
ERW = strip rolled into tube with machine-welded seam
tight bends need steel to be annealed ie in the softest ie weak condition
ok in a rollover bar
Yes, a bend in a nominally tension/compression structural member reduces its buckling strength, places additional (real, not simple elastic theory) moments on its terminations, and reduces its stiffness axially and probably in torsion. However in a crash it acts as a hinge and is a great way of absorbing energy plastically, and allows you to control where the deformation takes place.
Good answer, I hadn't thought about controlled deformation on impact. My comment about rectangular type bracing was that lozenging could take place where the structure just folds over and collapses vs a triangulated joint; a gusset might be more useful and lighter.
