Jump to content


Photo

The first spaceframe chassis?


  • Please log in to reply
35 replies to this topic

#1 D-Type

D-Type
  • Member

  • 8,054 posts
  • Joined: February 03

Posted 10 March 2005 - 21:34

The thread about the first monocoque has set me thinking. When was the first true space frame?

I appreciate there's a big definition problem here. Clearly any car with a chassis based on two large longitudinal members , e.g. Bugatti, Alfetta, Auto Union doesn't count. The problem comes with cars like the W196 Mercedes or 250F Maserati with a frame made up of small diameter tubes which are generally, but not all, triangulated and where triangulated, not necessarily in three dimensions. But there are hardly any cars with tubular frames that are fully triangulated in 3 dimensions.

Over to the experts.

Advertisement

#2 Ray Bell

Ray Bell
  • Member

  • 53,966 posts
  • Joined: December 99

Posted 10 March 2005 - 22:54

Originally posted by D-Type
The thread about the first monocoque has set me thinking. When was the first true space frame?

I appreciate there's a big definition problem here. Clearly any car with a chassis based on two large longitudinal members , e.g. Bugatti, Alfetta, Auto Union doesn't count. The problem comes with cars like the W196 Mercedes or 250F maserati with a frame made up of small diameter tubes which aregenerally but not all triangulated and where triangulated, not necessarily in three dimensions. But there are hardly any cars with tubular frames that are fully triangulated in three dimensions.


I don't think that's a problem at all...

They're way too new to consider anyway. IIRC, the Chamberlain 8 had a spaceframe in 1928/29.

#3 Mike Lawrence

Mike Lawrence
  • Member

  • 288 posts
  • Joined: September 04

Posted 10 March 2005 - 23:52

That is a bloody good question, D-Type, I wish I had thought of it.

I think I am on safe ground when I say that the Cisitalia D46 of 1946 was the first series-built racing car with spaceframe construction, but I am not sure that meets the critera you have set. The first Buckler, the Mk V, was billed as having a spaceframe chassis but I am not sure that any Buckler car chassis meets your criteria. The Buckler kart which emerged during the winter of 1960/61 certainly did and it was very successful in Class IV (200cc with gearbox, that usually meant a 197cc Villiers engine and 'box) racing in 1961.

To gt the ball rollong, I am ready to go out on a limb, expecting to be proven wrong. I will say the Jaguar C-type.

The spaceframe came from aviation, as did the monocoque, my gut feeling is that the motor racing fraternity first went to the monocoque as with Alex Issigonis's Lightweight Special Malcolm Sayer, the man behind the C-type, came from aviation and is a hero whose praises have been sung, but not as loudly as I would like to hear. Spaceframe or monocoque, we are speaking Hurricane or Spitfire. There was little to choose between them in 1940 in terms of performance, but you could patch a Hurricane with glue, scissors, Irish linen and dope. Even I could do that. You needed a skilled fitter to patch a Spitfire and you certainly would not want me on the job.

Why were Brabhams so popular with privateers in the 1960s? If you buckled your spaceframe Brabham on Sunday in Denmatk and you were due to race the next Sunday in France, somewere on the route was a blacksmith who could fix your car, if you had a monocoque it had to go back to England. Alfieri wanted the Maserati Tipi T60/61 to have monocoque construction but he had no builders in Italy he could trust, hence the 'birdcage(s}' which imitated monocoque design.

I will say that the C-type was the first car to meet D-Type's critera, but I know that someone will know of a German special with a BMW engine. It's costing me nothing to make a fool of myself. Has anyone worked out how to use the Atlas spellcheck?

#4 Ray Bell

Ray Bell
  • Member

  • 53,966 posts
  • Joined: December 99

Posted 11 March 2005 - 00:03

I don't think the C-Type would predate the Chamberlain 8...

As for the spellcheck, I haven't even thought about it. I would have reckoned a writer can also spell.

#5 Bonde

Bonde
  • Member

  • 959 posts
  • Joined: December 04

Posted 11 March 2005 - 01:06

If memory serves me, I think I remember seeing a picture many years ago of the Monaco-Trossi (or Trossi-Monaco as it is sometimes written) radial-engined GP car of 1935, and as I recall it appeared to have a very aircraft-like and (nearly?) properly triangulated space frame. This wouldn't really surprise me because the rest of the car was so obviously aviation inspired.

Welded steel tube space frames for vehicle use were, I think, essentially the ca. WWI-era creation of Anthony Fokker and Reinhold Platz. As I recall the story, Platz happened to be very enthusiastic about and good at the the then fairly new method of oxy-acetylene welding and he suggested to Fokker that it was a much quicker and cheaper way of producing aircraft fuselages in series relative to the wooden girders with all their fittings and fasteners. In fairness, Fokker also used wire diagonals - but naught wrong with that - it's efficient, even if the 'compression wire' is just along for the ride.

I may need to do some Googling to refresh my memory...on both topics...

Mike,

I recall reading in a post somewhere (might do a 'Search BB') that the 'Check Spelling' function is currently inoperative. As an alternative, it might be worth considering writing (longish) posts first in Word (or whatever), spell check there, and the cut and paste the text to the post. Don't forget to leave any automatic word division function switched off if you cut and paste from your word processor!

#6 Ray Bell

Ray Bell
  • Member

  • 53,966 posts
  • Joined: December 99

Posted 11 March 2005 - 01:29

The Chamberlain was described in more detail before...

http://forums.atlasf...=&postid=269140

I'll see if I can come up with pictures. But being from 1928, they might be a bit yellowed.

#7 David Birchall

David Birchall
  • Member

  • 3,004 posts
  • Joined: March 03

Posted 11 March 2005 - 01:56

The Gobron-Brillie' of 1904 would seem to have a space frame, or have ai been sniffing the cork too much again? :blush:

#8 Tim Murray

Tim Murray
  • Member

  • 14,404 posts
  • Joined: May 02

Posted 11 March 2005 - 12:00

As D-Type pointed out in his opening post, there is a major difference between a spaceframe and a tubular chassis. The Gobron was not a space frame - it merely replaced the standard channel-section longitudinal chassis members with similar items constructed using tubes.

The Trossi-Monaco definitely had a spaceframe-type chassis but, as Ray has pointed out, it came after the Chamberlain. So, was the Chamberlain a true spaceframe, or did it just have a chassis constructed using tubes?

#9 David Beard

David Beard
  • Member

  • 4,887 posts
  • Joined: July 02

Posted 11 March 2005 - 12:38

Originally posted by D-Type
The thread about the first monocoque has set me thinking. When was the first true space frame?


When did the first electricity pylon appear in Britain?

#10 D-Type

D-Type
  • Member

  • 8,054 posts
  • Joined: February 03

Posted 11 March 2005 - 13:12

Originally posted by David Beard


When did the first electricity pylon appear in Britain?

:evil: :mad: Or for that matter the first pithead winding gear. Or what about the chastity belt - that was a 3-d metal structure. Or the framework for a Mongolian yurt. Or even the framework supporting the dinosaur skins that Ug invented to keep Rachel Welch warm!

OK, maybe I should have said 'spaceframe with a wheel at each corner' :rotfl: :rotfl:

#11 uechtel

uechtel
  • Member

  • 1,515 posts
  • Joined: April 01

Posted 11 March 2005 - 17:53

Originally posted by Mike Lawrence
I think I am on safe ground when I say that the Cisitalia D46 of 1946 was the first series-built racing car with spaceframe construction, but I am not sure that meets the critera you have set.


Very interesting subject. Was the Cisitalia a pure spaceframe or was it based upon a conventional "supporting" tubular frame?

#12 fines

fines
  • Member

  • 9,647 posts
  • Joined: September 00

Posted 11 March 2005 - 17:59

Originally posted by David Birchall
The Gobron-Brillie' of 1904 would seem to have a space frame, or have ai been sniffing the cork too much again? :blush:

If the Gobron counts then you have to include the 1891 Peugeot T3 as well! :lol:

#13 fines

fines
  • Member

  • 9,647 posts
  • Joined: September 00

Posted 11 March 2005 - 18:02

Originally posted by uechtel


Very interesting subject. Was the Cisitalia a pure spaceframe or was it based upon a conventional "supporting" tubular frame?

Many years ago I claimed of the HH49 to have a space-frame chassis - youthful exuberance!

#14 David Birchall

David Birchall
  • Member

  • 3,004 posts
  • Joined: March 03

Posted 12 March 2005 - 01:39

How about a cyclecar? Post WW1, inspired by aircraft, built on a wing-and-a-prayer?
The Layatte was wonderful - an early aircraft without the wings :cool:

#15 Peter Morley

Peter Morley
  • Member

  • 1,900 posts
  • Joined: October 02

Posted 12 March 2005 - 10:09

When I studied space-frame structures, many years ago as an engineering student, the definition of a space frame was something like a structure made from rods that were connected with flexible joints in the corners - the strength/stiffness of the structure being derived from the shape and the material used with the joints having no acknowledged effect.
e.g. in the case of the simplest shape - a pyramid - you could take a bunch of straight rods and connect them with something like rose joints at each corner, the shape stops the structure moving and the rods give it the required strength.

Was it not the case that the first (and possibly only) true spaceframe chassis was that used by Colin Chapman on his own Lotus mark 8.
Access problems caused by the structure meant that all subsequent chassis were compromised rather than theoretically perfect.

#16 Ray Bell

Ray Bell
  • Member

  • 53,966 posts
  • Joined: December 99

Posted 12 March 2005 - 11:03

Surely it's a given that perfection in spaceframe construction is too difficult to achieve with a chassis? And that, for the purpose of this discussion, indeed, for the purpose of the description, something that is triangulated in every bay in which it can be without having to penetrate other components (radiators, drivers, engines...) is close enough to be counted?

#17 Doug Nye

Doug Nye
  • Member

  • 8,346 posts
  • Joined: February 02

Posted 12 March 2005 - 11:24

Immediately postwar, I think, Rudi Uhlenhaut designed a multi-tubular chassis of his own in which - even if every tube junction had been a strip of cloth - inter-action between the member tubes would have enabled the frame to hold its shape. In other words every single component reacted its fellows in either tension or compression to create a self-supporting, self-sustaining structure - the definitive spaceframe. The subsequent Mercedes-Benz 300SL of 1952 came close to this ideal, with inevitable fudgery around the door openings, the deep sill frames there dictating the use of 'gullwings' as Karl's great work has told us all. For what it's worth I have always regarded the Trossi-Monaco of 1935 as being the first significant quasi-spaceframe developed for mainstream motor sport. But don't forget Robert Waddy's 'Fuzzi' - short for 'fuzzi-lage' as in aeroplanes...

DCN

#18 Ray Bell

Ray Bell
  • Member

  • 53,966 posts
  • Joined: December 99

Posted 12 March 2005 - 11:27

So still there's nothing solid to pre-date the Chamberlain?

#19 Tim Murray

Tim Murray
  • Member

  • 14,404 posts
  • Joined: May 02

Posted 12 March 2005 - 12:17

No-one has yet answered my earlier question. :)

Originally posted by Tim Murray
So, was the Chamberlain a true spaceframe, or did it just have a chassis constructed using tubes?



Advertisement

#20 Ray Bell

Ray Bell
  • Member

  • 53,966 posts
  • Joined: December 99

Posted 12 March 2005 - 12:48

I don't have pictures here, but as I recall it was as good a spaceframe as most early sixties openwheelers.

#21 Bonde

Bonde
  • Member

  • 959 posts
  • Joined: December 04

Posted 12 March 2005 - 14:59

Sorry about dragging structural semanthics into this thread also, but the term 'space frame' has always puzzled me a bit. As others have rightly pointed out, the ideal structural 'cell' of this kind is one consisting of rod members arranged in a tetrahedron, i.e. a full spatial triangulation. Any number of the basic 'cells' ca be joined up to form the complete structure. If the members are very slender (high ratio of length relative to cross-sectional properties), all forces will be in the direction of the members, and they can just as well be joined with ball-and-socket joints - there is no transfer of moments between, indeed no moments in, the individual members - assuming loads are introduced and reacted only where the members join. Even if the joints are, say, welded, and thus capable of transferring moments, in the geometry with high slenderness ratio members secondary moments at the nodes (joints) will be minor - unless we start loading the members far away from the nodes.

On the semanthics:

The planar structural 'equivalent' of the terahedron is the plane truss - the basic 'cell' of which is a triangle of members joined at their ends. Loads in the ideal variant are in the direction of the members only (axial).

A planar frame is a structure that, like the truss, is rigid, able to resist the applied loads without undue deflection (i.e. not a 'machine' or mechanism), but one where the loads are not in the (axial) direction of the members only - a frame will thus be subjected to bending moments in its members in addition to axial (tension, compression) loads.

To me, then, a 'space frame' needs actually not be triangulated - it's the spatial 'outgrowth' of the of the planar frame. Examples of this would be such as the typical Cooper chassis, the Ferrari 156 Shark Nose, and to an extent the structure that Audi today calls a space frame - moments are transferred in and between members.

What we in racing typically call and define as a space frame should, according to my take on the semanthics, be termed a 'space truss if the intention of triangulating every bay is succesfully accomplished. But I will bow to the every-day usage in racing and stick to the 'space frame misnomer for the tetrahedron-of-tube-members-based structure since it is, and has been, used so prevalently.

In Real World structures, not least in racing car chassis, we typically see a combination of both space frame and space truss. In a typical three-bay mid-engine single-seater chassis, where the forward bay is bounded by the front bulkhead and the dash-board bulkhead, the center bay is the open-topped cockpit bay, and the aft bay is the engine bay, bounded by the seat-back bulkhead and a rear bulkhead, the front bay is more often than not a true space truss, the center or cockpit bay a space frame, and the rear (engine) bay may be either, depending on whether its upper opening is triangulated or not. But even the cockpit and engine bay openings may be edged with trusses, thus actually making the complete structure a true space truss structure - an example being the Mercedes chassis of Uhlenhaut that Doug mentioned - albeit that had the engine in front.

Another point to cloud the semanthics is that in Real World structures it is often not practical, or even desirable, to introduce all loads at the nodes (joints between members) only, which means that the Real World structure more often than not will encounter and need to be able to accommodate moments (typically bending, but also torsion) as well as forces in its members and joints. Also, in many cases, some sides of bays may be 'filled' with riveted-on sheet metal, typically a floor pan and/or bulkhead webs, this web then often able to povide shear resistance in place of a diagonal (rod or tube) member - and quite often diagonals were retained anyway in order to enable the web to be of large area and thin gauge without the bay becoming too easily prone to web buckling.

In many instances, bulkheads were, and are, plane frames in order to enable the driver's legs and mechanical components to pass through their planes - and many more fairly recent single-seater chassis (for instance Baldwin's Formula Ford VanDiemens from RF77 through RF83 and some Lolas) had cockpit bays that were essentially 'framed' by rectangular section upper longeron tubes capable of resisting significant bending moments in the plane of the opening - a quite neat and simple solution to the problem. Bulkheads with open web planes can and were also constructed as plane trusses forming full hoops, or, again, as frames manufactured from bending-resistant members, such as rectangular tubes. For the purist, bending is anathema, axial load providing the highest degree of structural efficiency, but then again we live in a Real World where paractical considerations and cost pay no small part.

Installing gussets at the joints may be used to enhance the structural performance of a frame, but gussets may also introduce unwanted secondary bending moments if used indescriminately at the nodes of trusses - and a truss should have the centroidal axes of each member intersect in a single, common point (no off-sets).

Many early 1950s (and later) front-engined US Champ Cars, such as Kurtis Krafts, had tall (deep) and (almost) properly triangulated trusses as side members of what were essentially 'ladder' frames. They weren't very resistant to pure torsion, but worked by being resistant against differential bending of the side members, which is the typical loading mode of most open-cockpit single seater chassis, even on current F1 chassis. Big, (virtually) tubular side members, like on a Lotus 25, will resist torsional loads, but still need to be joined by substantial members to do so - even the large diameter tubular 'rails' of some 'ladder frames' such as the Auto Union and some immediate post-WWI designs (ERA G-type comes to mind) will accept torsion individually, but when joined by the necessary cross members will essentially be loaded in differential bending. It all pretty much depends on the geometry of the structure.

So, basically, what this all boils down to is that it's actually quite difficult to make a clear and well-defined 'cut' of what Real World structures should rightfully be called. Perhaps the term 'welded tubular steel chassis' is sufficiently generic and non-descript to cover most combinations of structural configuration! As Coopers proved in their day, the important thing is whether it does its intended job well or not...the rest of us can then sit in our Ivory Towers and play with words and definitions 'till the cows come home...


Tim,

Sorry I can't answer your question either - I'd also really like to see the chassis structure of the Chamberlain!


Doug,

I've always been fascinated by specials such as the Fuzzi - but unfortunately I've never seen pictures of it that reveal anything about its chassis structural configuration, so I can't comment on it at all. What was the fate of Fuzzi, BTW?

#22 David Beard

David Beard
  • Member

  • 4,887 posts
  • Joined: July 02

Posted 12 March 2005 - 15:07

Another brilliant post from Mr Bonde :clap:

#23 Tim Murray

Tim Murray
  • Member

  • 14,404 posts
  • Joined: May 02

Posted 12 March 2005 - 16:04

Absolutely - thank you Anders. I just wish my old structures lecturers had been as clear and lucid in their explanations (structural analysis was never my strong point).

Originally posted by Bonde
They weren't very resistant to pure torsion, but worked by being resistant against differential bending of the side members, which is the typical loading mode of most open-cockpit single seater chassis, even on current F1 chassis.

I'd never really considered before that anything other than absolute torsional rigidity was any sort of virtue.

Originally posted by Bonde
Big, (virtually) tubular side members, like on a Lotus 25, will resist torsional loads, but still need to be joined by substantial members to do so - even the large diameter tubular 'rails' of some 'ladder frames' such as the Auto Union and some immediate post-WWI designs (ERA G-type comes to mind) will accept torsion individually, but when joined by the necessary cross members will essentially be loaded in differential bending. It all pretty much depends on the geometry of the structure.

So is it fair to say that the Lotus 25 type of 'monocoque' can be thought of as basically just a ladder chassis with large cross-section longitudinal members? I don't know quite where this leaves the 'first monocoque' thread. :drunk:

#24 Ray Bell

Ray Bell
  • Member

  • 53,966 posts
  • Joined: December 99

Posted 12 March 2005 - 20:56

And to translate this to the real world, this is how one chassis was made to fit around a driver and an engine...

Posted Image

#25 Bonde

Bonde
  • Member

  • 959 posts
  • Joined: December 04

Posted 13 March 2005 - 00:32

Thanks, gentlemen :blush: Structures are Fun! (for those who think structures are fun...) - I do actually provide occassional courses for 'green' structural designers in Danish arospace industry, and I enjoy it - I hope they do, too!

Although this bit really belongs under the monocoque thread, as I alluded to it in this thread, here is a little elaboration on 'torsion' versus 'differential bending' - I get the two mixed up often myself!

Imagine a tube, closed at both ends by bulkheads. Twisting the two ends relative to one another, puts a torsional load on this structure, inducing shear stresses which will flow all around the circumference in a closed loop. Remove all of the top quadrant of the tube, to get the driver in, and perhaps also a lot of the lower quadrant to provide room for a prop shaft, engine cooling, ground clearance or whatever, and the shear flow is interrupted - the structure will no longer resist torsion. Instead, the side quadrants will will now be forced to act as two indvidual, parrallel beams joined by a few cross-members, with one bending up at one end, the other bending down at the same end as we keep twisting at the bulkheads. It is basically the ability of the structure to provide a continuous path for the shear flow that determines whether it will accommodate torsional load or resist load in differential bending. However, in a true, 'pure' space truss, obviously no shear flow is generated - all internal stresses are axial, even when we load it by twisting its ends. In a stressed skin structure, or a tube, shear flow is generated, and the smaller the 'obstructions' to the shear flow, i.e. cutouts, the more efficient the structure remains. The equivalent on the space truss is the removal of diagonals - this induces bending in the members.

Try this little experiment: Take a (cylindrical) cardboard tube and, without destroying it, try to compress, bend and twist it. Now, interrupt the circumferential shear flow - i.e. slit the tube lengthwise and repeat the experiment. There is still exactly the same mass of material in the structure, and it fills up exactly the same volume, but there is no doubt whether the structure is more efficient before or after the slit...

Another interesting experiment requires a longish rectangular box that can be closed at both ends - the box that contains a tube of toothpaste is suitable. Try twisting it with the ends open and the ends closed - note the difference. Now glue or tape in some more intermediate partitioning walls along its length - repeat the experiment and note the difference. Do the same experiment with a cylindrical tube and note the differences again. You will notice that the cylindrical tube will resist torsion perfectly, even without intermediate partitions, whereas the rectangular tube will perform a lot better as intermediate partions are added. Both will perform better in bending and compression with intermediate partitioning walls (called webs, diaphragms, frames or bulkheads, depending on thier other functions). When I compare torsion box bay and gauge dimensions as typically used on aircraft structures and on elderly racing cars, (and when I look at the detail execution of each) I am not surprised that the aircraft ones will fly for ages and the racing car ones sometimes needed to be replaced during the season. If it weren't for the need to accommodate big bag tanks...

Continuity and support of thin members makes a structure efficient, so although, say a 50's vintage USAC car, did not make optimum use of the weight spent on the chassis structure (cross members were few, and diagonals the vertical planes and in the upper and lower planes were conspicuoulsy rare), they were still a lot more efficient than the old channel-section ladder frames they replaced.

By the sixties welded tubular steel chassis (okay, space frames) were getting quite efficient - high stiffness with little weight, but they weren't terribly strong beyond normal operating loads, as pictures of for instance the Moss Goodwood wreckage show: Once a critical member (usually the upper longerons in the cockpit) bows...

Modern space frames tend to be designed so much by 'crashworthyness' requirements that, provided the detailing is good, adequate stiffness is almost difficult to avoid, but optimum stiffness still worth chasing. Designing to the rule book virtually guarantees adequate global strength, but the good designs get more stiffness and local strength from the samme weight. As mentioned in my previous post, the unavoidable mix of frame and truss elements make FEM calculation even of relatively simple spaceframe chassis a virtual 'must' these days for optimum efficiency.


Ray,

I've seen worse! What car is that? And does the engine provide a structural function also?


Tim,

Structure of virtually any shape can be made to work - provided you just throw enough material at it - I don't have the impression that, for instance, the old USAC 'roadsters' were particularly light. Now, making a structure efficient is much more challenging - especially, to my mind, if it isn't allowed to cost much and still be made to last and yet be light...The modern F1 structures teams turn out the most beautiful and efficient structures on (or even above) this planet, within the constraints of the rule book - but ask them to do a chassis that will sell for 5000 € and you won't see which way they went to avoid you, I'm sure. Racing did, and probably still does, profit from ex-aerospace engineers - I wonder what happens to the traffic going in the opposite direction? I may well be totally wrong, but I have the impression that aerospace cost concerns would be a rude awakening to some F1 design team members. In many ways a shame really, that there are no longer companies that do both F1 and junior formulae production cars - but then again, I'd prefer the name of a Formula Ford to be Lotus rather than, Say, Red Bull...

As for your take on the Lotus 25 chassis being a ladder frame with enormous tubular longerons I can't really argue against it - the wall of a tube is also a stressed skin and has to obey the same laws of physics as any other piece of structure... :)

#26 dolomite

dolomite
  • Member

  • 951 posts
  • Joined: March 01

Posted 13 March 2005 - 10:38

I'm sure I read somewhere recently that Chapman did not regard the Lotus 25 chassis as a monocoque, he originally referred to it as 'stressed skin twin-tube ladder frame' or something similar. It was others who started calling it a monocoque and the name stuck.

Edit: it is in Andrew Ferguson's 'Team Lotus - the Indianapolis Years'. Ferguson refers to Chapman editing a draft of the press release for the launch of the Lotus 29 to replace 'monocoque' with his preferred term 'twin-tube ladder frame'.

#27 Ray Bell

Ray Bell
  • Member

  • 53,966 posts
  • Joined: December 99

Posted 13 March 2005 - 11:17

Originally posted by Bonde
.....I've seen worse! What car is that? And does the engine provide a structural function also?


I don't think so... in fact, judging by the mounting point on the chassis, I'd say definitely not...

The chassis is the Brabham BT23 altered to fit the Alfa Romeo 2.5 V8 for the Alec Mildren team.

#28 Ray Bell

Ray Bell
  • Member

  • 53,966 posts
  • Joined: December 99

Posted 29 March 2005 - 01:54

Thanks to Brian Lear... and the book The Chamberlain - An Australian Story written by current owner John Hazelden, here's some 1930 pictures of a racing car with a space frame...

Posted Image

Posted Image

#29 Ray Bell

Ray Bell
  • Member

  • 53,966 posts
  • Joined: December 99

Posted 29 March 2005 - 12:15

And a little further... from the same page in the book, some more detail:

Posted Image

I certainly hope that resolves the issue of the first racing car to use a spaceframe chassis... but as always, you never know when you will be surprised by something else...

#30 Tim Murray

Tim Murray
  • Member

  • 14,404 posts
  • Joined: May 02

Posted 29 March 2005 - 15:15

A perfectionist would probably say that there was not enough triangulation in the horizontal plane for the Chamberlain to be considered a true space frame. But as Ray pointed out earlier, it is just as worthy of being called a space frame as many '50s and '60s designs - and it appeared in the '20s. In my view Ray has proved his point.

#31 oldtimer

oldtimer
  • Member

  • 1,291 posts
  • Joined: October 00

Posted 29 March 2005 - 19:06

Enough of this fluff!

#32 Ray Bell

Ray Bell
  • Member

  • 53,966 posts
  • Joined: December 99

Posted 29 March 2005 - 20:03

It's not old enough for you?

#33 oldtimer

oldtimer
  • Member

  • 1,291 posts
  • Joined: October 00

Posted 29 March 2005 - 22:43

Actually Ray, the Chamberlain Special pre-dates both me and my interest in racing cars. But, many thanks for the fascinating information and photos on the Chamberlain Special. Interesting that the weight is only a 100lbs or so less than the Vanwall being discussed in another thread.

#34 Bonde

Bonde
  • Member

  • 959 posts
  • Joined: December 04

Posted 29 March 2005 - 23:17

I agree with Tim Murray, and I'll accept use of the nomenclature 'space frame' in the context applicable to later designs (i.e. - with the caveat that 'space frame' does not necessarily imply 'space truss' ;) ).

The Chamberlain chassis is not unlike those of Frank Kurtis of the early 1950's - so for its day it is certainly quite advanced.

Either way, the stiffness-to-wight ratio of the Chambelain Special chassis must have been order(s) of magnitude higher than contemporary channel-section ladder frames. The transverse members are also quite short due to the narrowness of the chassis, and they form full portal hoops, so that all helps, too. It's also interesting to see the lack of total confidence in the method of construction employed, as evidenced by the sizeable gussets at the nodes!

A remarkably interesting, neat and fore-sighted design. Thanks for sharing those brilliant picture, Ray! :up:

I think we'll have to search very hard to find documentation on any deep-section, welded tubular truss-type chassis earlier than this one...especially if we are looking for diagonals in the horizontal plane.

#35 Ray Bell

Ray Bell
  • Member

  • 53,966 posts
  • Joined: December 99

Posted 29 March 2005 - 23:28

Originally posted by Bonde
.....Thanks for sharing those brilliant picture, Ray!


Oh, I'm blaming Brian Lear... he did it, not me. I just posted them.

.....I think we'll have to search very hard to find documentation on any deep-section, welded tubular truss-type chassis earlier than this one...especially if we are looking for diagonals in the horizontal plane.


Must find out if there's any shots that show the horizontals better... watch this space...

As for the gussets in the corners, I was thinking that they'd have accounted for a little of that weight.

#36 D-Type

D-Type
  • Member

  • 8,054 posts
  • Joined: February 03

Posted 30 March 2005 - 09:22

I think the gussets are just a simple way of joining three members coaxially using fillet welds.

Joining the first two tubes at right angles is straightforward it's simply a shaped tube and an all round fillet weld. To make the diagonal's axis coincident with the other two would necessitate a complex double cut of the tube and welds on welds. The gusset allows the diagonal to be welded with two simple fillets - one each side.

I have found that structural steel fabricators will go to great lengths to use fillet welds in preference to butts. I don't fully understand the difficulties in welding but I know they exist.