Carbon Fibre Spaceframe

[D-Train]

I had a thought a while ago when I was talking to a friend about Carbon Fibre and how to lay it up etc. Anyway, the jist of the idea was running a space frame with carbon fibre tubes (i know it's been thought of before, but i can't find much of a discussion on it). I figure you'd have a light weight chassis that can handle the point loads of the suspension well, and a good torsional rigidity. Purchasing the tubes wouldn't be an issue either, there are plenty of companies that sell lengths of tube.

Rather than using carbon fibre to bond the tubes, you could have inserts (welded tubes or machined) so that you could in theory replace the members in case of an accident. I recall someone mentioning the CF tube failed before the adhesive used in their a-arms for bonding carbon to steel, so i figure that shouldn't be an issue. Apart from that, i can't see it being any more difficult to manufacture than a steel spaceframe, unless you need to bond brackets etc. along any of the tubes, but even that shouldn't cause too many issues.

People have told me carbon fibre's not so good in compression, so that might be a big factor (i'm not well-versed in the properties of carbon fibre). I imagine it'd be considerably more expensive than steel space frame, but would it be more than a monocoque?

Here's a quick list of pros and cons i could think of.

**CF spaceframe vs Steel spaceframe**
[PROS]
-lighter
-increased rigidity (?)
-stronger (in tension for members)
-little tube prepping req'd
-minor adjustment of tube lengths possible by how far down the inserts the tubes sit
[CONS]
-more machined parts (time consuming, costly)
-cost of tubes
-CF weaker than steel in compression (?)
-reliance on adhesive
-no bent members (assuming all purchased lengths)
-tubes shatter, not bend
-safety issues
-attatchment of components to tubes (i.e 1/2 way along tube)
-steel roll hoops still req'd

**CF spaceframe vs CF monocoque**
[PROS]
-more suited to point loads that suspension components deliver
-easier manufacture (?)
-easier stress analysis
-broken members replacable
-ease of mass production + less chance of major manufacturing defects
-no autoclave req'd
-safety
-body panels can be used for extra rigididy (stressed skin)
[CONS]
-weight
-panels still req'd
-rigidity
-cost (?)

Anyone got any thoughts on this? Or reasons why it's not possible?!? (i may have overlooked something here...)

[D-Train]

I had a thought a while ago when I was talking to a friend about Carbon Fibre and how to lay it up etc. Anyway, the jist of the idea was running a space frame with carbon fibre tubes (i know it's been thought of before, but i can't find much of a discussion on it). I figure you'd have a light weight chassis that can handle the point loads of the suspension well, and a good torsional rigidity. Purchasing the tubes wouldn't be an issue either, there are plenty of companies that sell lengths of tube.

Rather than using carbon fibre to bond the tubes, you could have inserts (welded tubes or machined) so that you could in theory replace the members in case of an accident. I recall someone mentioning the CF tube failed before the adhesive used in their a-arms for bonding carbon to steel, so i figure that shouldn't be an issue. Apart from that, i can't see it being any more difficult to manufacture than a steel spaceframe, unless you need to bond brackets etc. along any of the tubes, but even that shouldn't cause too many issues.

People have told me carbon fibre's not so good in compression, so that might be a big factor (i'm not well-versed in the properties of carbon fibre). I imagine it'd be considerably more expensive than steel space frame, but would it be more than a monocoque?

Here's a quick list of pros and cons i could think of.

**CF spaceframe vs Steel spaceframe**
[PROS]
-lighter
-increased rigidity (?)
-stronger (in tension for members)
-little tube prepping req'd
-minor adjustment of tube lengths possible by how far down the inserts the tubes sit
[CONS]
-more machined parts (time consuming, costly)
-cost of tubes
-CF weaker than steel in compression (?)
-reliance on adhesive
-no bent members (assuming all purchased lengths)
-tubes shatter, not bend
-safety issues
-attatchment of components to tubes (i.e 1/2 way along tube)
-steel roll hoops still req'd

**CF spaceframe vs CF monocoque**
[PROS]
-more suited to point loads that suspension components deliver
-easier manufacture (?)
-easier stress analysis
-broken members replacable
-ease of mass production + less chance of major manufacturing defects
-no autoclave req'd
-safety
-body panels can be used for extra rigididy (stressed skin)
[CONS]
-weight
-panels still req'd
-rigidity
-cost (?)

Anyone got any thoughts on this? Or reasons why it's not possible?!? (i may have overlooked something here...)

[TG]

We're going to be doing a carbon fiber space frame. The thing is with carbon tubes is that carbon makes horrible joints. That's why we will have metal tubes on the ends of the carbon tubes to weld to one another. Basically we will have a car version of this bike (http://www.titusti.com/ultralite.html). So there are no blind bonds with the Bi/Fusion (http://www.vyatek.com/VYATEK/dox/Bi-FusionIntroduction.ppt) and all of the joints are welded so there are no worries about the adhesives. The guy that developed this tech used to be in ASU SAE a long time ago. I didn't even know that until he called me back after I had inquired. But yeah, we're laying it up at his place cause of all of the proprietary stuff in it.

[ss_collins]

Ease and speed of contruction - build a tent or Gazebo just with triangulation so minimal flex.

What am I on about - go to a garden centre and see how tents and gazebos are built - or go to the woods and see how a 'bender' is built

[jack]

merlin bikes:

http://www.merlinbike.com/bikes/2005/cielo_dtls.aspx#

http://www.merlinbike.com/bikes/2005/cielo_dtls.aspx#


the "inserts" are usually called lugs. they would need some really intense jigs, because any small mistakes will make the other end of the tubes way off.

monocoques can be really cheap to make if you get everything donated, which isnt too hard. the same could be said for carbon tubes, but if you end up having to pay for them, they may be really expensive.

[fsae_alum]

They're not that expensive. Look at Composite Tubes (http://www.macqc.com/tubing.php) If you buy tubes from Canada it's even cheaper!

Yep...lugs wouldn't be that hard to make using CNC. Aluminum or even 4130 would be nice. Heck even SS would be nice (no painting and no corrosion).

[Denny Trimble]

http://www.merlinbike.com/images/2005/cielo_lg.jpg
http://www.merlinbike.com/images/2005/cielo_stays_lg.jpg

[jack]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Yep...lugs wouldn't be that hard to make using CNC. </div></BLOCKQUOTE>

uhhh, we are talking something like a 7 OP part here, with lots of surfacing....

thanks for fixing the links denny.

[TG]

With the Merlin, they first create the lugs and then glue in the tubes. This creates blind bonds so you don't know the quality of the bonds. With the Titus, they first co-cure the tubes with the metal ends and then weld the ends together. The co-curing also creates a bond stronger than the tube itself, so you know that it won't fail at the bond (lots of empirical testing has shown this), unlike the bonded in carbon tubes. Also, you don't need the accurate lugs.

[Cement Legs]

What is the effect on the tubes of the heat from welding?

[TG]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Cement Legs:
What is the effect on the tubes of the heat from welding? </div></BLOCKQUOTE>

That would be one of the proprietary steps http://fsae.com/groupee_common/emoticons/icon_wink.gif But we can't use aluminum tubes because the thermal conductivity is too great, unless the sections were very large, which would obviously be pointless.

[Travis Garrison]

Very cool...I've wondered about a CF spaceframe myself, you'll certianly turn some heads if you can pull it off.

You might even be a hit in the design tent, check out Race Car Engineering and Mechanics (http://www.amazon.com/exec/obidos/tg/detail/-/1557883661/qid=1125445066/sr=8-1/ref=pd_bbs_1/104-7533949-8245543?v=glance&s=books&n=507846) pg 162. Of course that is if Van Valkenburgh isn't too disgusted with the rest of us to come back (his last Racecar engineering article sounded like he was getting bored, not that I blame him)...

[fsae_alum]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by jack:
<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Yep...lugs wouldn't be that hard to make using CNC. </div></BLOCKQUOTE>

uhhh, we are talking something like a 7 OP part here, with lots of surfacing....

thanks for fixing the links denny. </div></BLOCKQUOTE>

If you'd quit trying to make them yourself on a $999 Harbor Freight desk top mill and have a machine shop do them on their 5 axis machine this wouldn't be a problem. http://fsae.com/groupee_common/emoticons/icon_wink.gif

Seriously though....it's not THAT hard. Design for manufacturability IS a good idea ya know! http://fsae.com/groupee_common/emoticons/icon_wink.gif

[Z]

So you've got maybe 30 tubes carefully cut to size, and maybe 20 x 3, 4, or 5+ pointed lugs accurately machined or fabricated.

Now you start gluing the tubes into the lugs one at a time... Oops! Got two sides of the triangle together, but can't get the third side in...

Hmmm, start again. Mix up a big pot of glue, get maybe 50 helpers - one for each tube and lug - and try to assemble all at once??? Would work for a bicycle frame, but 50 pairs of hands???

What am I missing??? And how would you repair a single shattered tube in an otherwise quite rigid spaceframe?

Maybe the lugs could be two pressed metal halves that clamp around the tubes, one from the inside of the chassis and the other from the outside. The inner side halves would be fixed in a jig, and in turn would act as a jig for the tubes. Then glue and possibly "lockseam" (or weld?) the two lug-halves together?

Z

(Edit: Above assumes you don't use TG's method of welding (expensive?) metal-ended-carbon-tubes together like normal spaceframe construction.)

[fsae_alum]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Z:
So you've got maybe 30 tubes carefully cut to size, and maybe 20 x 3, 4, or 5+ pointed lugs accurately machined or fabricated.

Now you start gluing the tubes into the lugs one at a time... Oops! Got two sides of the triangle together, but can't get the third side in...

Hmmm, start again. Mix up a big pot of glue, get maybe 50 helpers - one for each tube and lug - and try to assemble all at once??? Would work for a bicycle frame, but 50 pairs of hands???
</div></BLOCKQUOTE>

T-H-I-N-K ...It's NOT an impossible task. You're an engineer...you're suppossed to come up with the solution to the problem, not just throw your hands up in the air and say that it can't be done before even attempting it.

There are ways to do this type of spaceframe...you just have to think outside the box and design it that way from the beginning. Challenge yourself. Sit down and think out the problem and I bet you'll come to a solution.

[Cement Legs]

You are right Z it would be very hard... maybe delicate would be a better word, but it should would be right. I bet a CF tubular spaceframe would be even more stylish than bellcranks http://fsae.com/groupee_common/emoticons/icon_wink.gifhttp://fsae.com/groupee_common/emoticons/icon_biggrin.gif

Something like that I would really enjoy because of the problems and learning curves involved. Prob not a good idea for a first year team or a low budget team.... too many what ifs, but I hope to see something like that in the next couple of years.

[Z]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by fsae_alum:
...and design it that way from the beginning. </div></BLOCKQUOTE>
That's where I see the handicap of this method - you have to layout the tubes to suit the manufacturing method. Ie the requirement of being able to slide the tubes into the lugs constrains your design freedom. I am sure there are workarounds - I suggested one way (the "split lugs") and TG's method is another way.

Come to think of it, WWU's "twin-tube" chassis is a very simple and elegant form of this type of construction (though not really a "spaceframe").

Anyway, the rules require a floor and some bodywork. So a monocoque with reinforcing ribs at the rule-required bulkheads and side impact areas, to take the point loads, might be a bit easier???

Z

(Edit) Cement, Just saw your post. Yeah, but you would have to put the bodywork INSIDE the spaceframe http://fsae.com/groupee_common/emoticons/icon_wink.gif.

[jack]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Come to think of it, WWU's "twin-tube" chassis is a very simple and elegant form of this type of construction (though not really a "spaceframe"). </div></BLOCKQUOTE>

...yeah, and they are a complete pain in the ass to build.

the CNC stuff takes FOREVER too, and the stuff we do is way simpler than a 4, 5+ tube lug (and yes we use a machine that cost a little more than $999).

basically, me and a few others here at wwu have kicked the idea around, and it just seems like way too much to pull off. but dont let that stop you from proving me wrong...

[Jetser]

Why would you build a CF spaceframe if a steel one is much easier to build and probably lighter. And if you can save weight, how much will it be and how much extra time will it cost to build.

The way of design should be:
- What kind of functions do I need to fullfill?
- Which material is the best to use to get this done the lightest/cheapest/easiest to manufacture.

Last year we've build a chassis with a monocoque at the front and a steel/aluminium spaceframe at the back. The spaceframe was then glued together, which wasn't easy to do. Next year I would advise the team to build a steel spaceframe at the rear. It can easily be lighter than the one we had this year.

[D-Train]

Wow, someone from Delft speaking AGAINST carbon fibre?!?

Nah, seriously, why did you guys decide to glue the aluminium/steel space frame together? And is the possibility of weight reduction due to better design, or better use of materials?

[Jetser]

Our 03 car had an aluminium spaceframe at the back (also aluminium mainhoop and bracing) but we had a lot of trouble with the heat-treatment after the welding.
The 05 car did have a steel mainhoop and bracing due to the rules. Our rear-bulkhead would become too heavy if we've made it out of steel. Most connections are steel to aluminium, so glueing would be the best option.

The possibility of weight reduction would be a combination of both, since the use of materials is a designparameter.
But I must agree that our 05 car could be a lot lighter (+/- 2kg, for us thats a lot) with the same use of materials, mainly due better design of the rear-bulkhead.