Carbon Fiber Half Shafts

[Eric Liu]

Does anyone know:
1. Where to find carbon fiber tubes that can take torsional loading?
2. Where to find a manufacturer of carbon fiber half shafts?
3. How to make your own carbon fiber half shafts given that you have the tubes?

Thanks.

[TG]

I think you're asking the wrong questions here. The hard part about carbon half-shafts, torque-tubes, and drive-shafts is the sheer transfer of torsional loading from the carbon tube to the metal fitments that attach to the flexible couplings at the end of the shaft. You either need an adhesive bonding agent that has sufficient sheer strength or you need to design a tube that has mechanical leverage to transmit the torque (square, triangular, sinusoidal spline, or any other non-circular profile at the ends).

As far as sourcing these from a manufacturer, the only FSAE instance that I've seen is the drive-shaft the University of Guelph purchased from CTG a couple years ago for their AWD system. I believe they said it cost them around $800 if I remember correctly... not cheap stuff. Everything else I've seen has been made by students specifically for their application.

If you still want to go ahead and find some carbon tubes somewhere for the application, you will want a tube with a large amount of fibers oriented at an angle more than 0 degrees and less than 90 degrees from the principle axis of the tube (this angle really is design dependent and you will have to do some calculations to figure out what best suits you). Also, you will need to make sure the matrix used is compliant with the adhesives you intend to bond the couplings to. You may or may not be able to find information on this and you might have to perform some tests to find out.

After that, all I can say is good luck.

[benjo]

Here's a company that may be able to help you out with some advice or products. Found them in a RE magazine from a while ago.

http://www.torqline.com/

[vreihen]

This 2003 forum thread about Composite Half Shafts (http://fsae.com/eve/forums/a/tpc/f/125607348/m/8536003534/p/1) might be of interest to you.

If you're looking for CF tubes, it doesn't look overly complex to make them yourself after reading this Composite Tube Fabrication Tutorial (http://www.aerosleeves.com/Articles.asp?ID=108). Seems easy enough, and doesn't require any expensive equipment like an autoclave. Do you think that half shafts will fit in a kitchen oven? http://fsae.com/groupee_common/emoticons/icon_smile.gif

If you'd rather buy pre-made CF tubing, Carbon Fiber Tube Shop (http://www.carbonfibertubeshop.com/small%20tubing.html) is worth checking out. This page link even provides details on their production technique (which sounds awfully similar to the DIY link above), and gives warnings about using "common" pultruded CF tubing in a torsional application like a driveshaft. They will do custom orders to your specs, and the owner also operates the racingcomposites.net forums which you might find of interest for fabricating tips.

Hope this info is helpful to your decision.....

[UoP Simon]

Right, lets revive this thread

I'm working on Carbon Fibre half shafts for the University of Portsmouth Formula Student team.

Currently the most likely method of manufacture I will be following is using biaxial carbon tubular sleeving, hand layed up over a mandrel - as described in this tutorial: making shafts (http://www.solarcomposites.com/MakingShafts.html)
or here (http://www.rocketryplanet.com/content/view/2189/38/)
with pictures.

Does anybody have any particular insights or advicea on the construction of the shafts?

In particular information on which epoxies to use, both for laying up and for bonding the metal (aluminium alloy?) inserts into the end

Additionally does anyone know of suitable composites companies within the UK, for sourcing either the fibre sleeving or for the epoxy?

Soller Composites do seem to have good prices for the products I need and they have offered me some advice, however it would mean shipping from the US

Any help or information would be welcome

[lithiumdeuteride]

For a minimally constrained half shaft (and yours had better be), the only significant load the middle of the shaft sees is shear. But shear is of course identical to biaxial tension and compression, at +/- 45 degrees to the direction of shear.

Therefore, the easiest layup (though not necessarily the most weight-efficient) is simply many plies of 45-degree plain weave fabric. As the shaft is put into shear, half of the fibers will be in tension, and half will be in compression.

In reality, the tensile and compressive strengths of carbon fiber are not the same, and the half shaft is typically loaded in one direction (accelerating) more highly than the other (engine braking). Therefore, there is some ideal layup which has more fibers at +45 degrees, and fewer at -45 degrees, for example.

But I suspect it is not worth trying to achieve that level of optimization. Just use enough +/- 45 degree plain weave fabric, and you'll be fine.

[exFSAE]

Could just make a lightweight steel halfshaft...

[DanVaan]

One of our professors has a mandrel for making composite tubes for pipeline research and let us use that last year. Basically it involves winding the fibers onto the tube in whatever orientation you desire. Then it is cured in an oven where it rotates, and voila, you now have a carbon tube.
The more difficult part is connecting the ends. One of our members did a senior research project to find the best method of joining the aluminum ends to the shaft. He used some kind of 3M 2-part resin (i'm not sure exactly which one) and did a pile of tests to make sure that the resin layer wouldn't fail at the loads and fatigue that we expected.
The shafts worked great on last year's car, and they have definately seen some abuse.

[UoP Simon]

yep the plan so far is to use biaxial sleeving at +/- 45 degrees.

Anyone with experience with which epoxies work best in this application?

[Nivek]

Sorry to bump this but I was doing a little research on the same topic and found this.

The question is what is the best resin to join aluminum to carbon fiber

[RANeff]

JB Weld

[rrobb]

You need to do a bunch more research. Aluminum bonding is no joke. I'd recomend starting your search with BAC 5555.

[cvargas]

Originally posted by Nivek:
Sorry to bump this but I was doing a little research on the same topic and found this.

The question is what is the best resin to join aluminum to carbon fiber

One thing that you will run into is galvanic corrosion when bonding aluminum to carbon fiber. You'll need to make sure that the aluminum is completely sealed off from the carbon.

[Nivek]

Yes, I'm glad you mentioned that. This can be reduced/eliminated through a sheet of fiberglass insulated between the aluminum and carbon fiber correct?

[kapps]

Not sure what you mean by sheet of fiberglass but common practice is to make sure the first ply of the layup in the area that's going to touch aluminum is fiberglass.

[Thrainer]

I'm not sure if this is really a problem in FSAE. We don't run in salt water and haven't had a problem in two years with CFRP suspension arms and driveshafts. In our case, the CF does contact the metal inserts, so the entire part is electrically conductive.

[PBR-Simon]

Like people have said before making a carbon tube with an astronomical FOS is the easy part. Bonding to it isn't so. When I was designing carbon fibre A-arms I found this thread to be very helpful.

http://fsae.com/eve/forums/a/t...48/m/15110476641/p/1 (http://fsae.com/eve/forums/a/tpc/f/125607348/m/15110476641/p/1)

It has a tremendous amount of information about how to bond ally to carbon for shear loading due to tension and compression (it can also be applied to torsional loading). One thing that is left out of this thread is the use of tapered joints. Tapering of the substrates can eliminate the bathtub like stress distribution that afflicts simple lap joints. (If you don't understand this then read the thread)

[murpia]

Originally posted by Thrainer:
I'm not sure if this is really a problem in FSAE. We don't run in salt water and haven't had a problem in two years with CFRP suspension arms and driveshafts. In our case, the CF does contact the metal inserts, so the entire part is electrically conductive.

An epoxy primer is a good solution to preparing aluminium to bond with carbon, without a corrosion issue.

Regards, Ian

[Thrainer]

For me, carbon driveshafts have been a fascination since I started doing FSAE many moons ago. Since they're not critical for a car's performance (at least not to an extent like carbon wheels), they don't seem to be of high priority to teams, which makes them rather rare and most often not very optimized. (If you know of an exception to this, I would love to read about it)

My first concept of carbon driveshafts, for Simplon in 2009, was basically identical to the pullrods on the same car, but three times the diameter and with TRE tripods instead of rod ends.

http://www.amzracing.ch/amz/files/amz1308.jpg

I would suggest it was like most carbon driveshafts you see on FSAE cars: Vastly overdimensioned, thus only saving 200-300 g compared to steel driveshafts, roughly the same inertia, but 2-3 times the strength required (if bonding and tube quality is good).

After a lot of static and dynamic testing with bonded "pullrods", I came to the conclusion that there's a lot of potential for improvement. But to reduce the risk of failure, even more testing needed to be done. Now, after completion of the project, I'm trying to fool myself into thinking the original target was to build a reliable driveshaft weighing <500 g (half the weight of Simplon's carbon driveshaft). The diameter should not be much larger than steel driveshafts and 1" tripods should be used.

So, here's a little How To:

First, you need a great bond. With smaller diameters, bonding becomes exponentially more difficult. Further, inserts must be kept short to save weight.

http://www.amzracing.ch/amz/files/amz1914.jpg

Second, you need a good supplier of quality carbon tubes. High strength and the possibility to choose the angles of the fibres is a requirement to push the envelope. I was very fortunate to be able to work with Peak Technology.

http://www.amzracing.ch/amz/files/amz1915.jpg

Third, think about how to integrate all the functions (CV joints, plunger, boots). If you're using off-the-shelf tripods with a spline, you have more interfaces than necessary and therefore more weight. Make sure your tube, bond and insert are dimensioned to the same (strength) requirements.

http://www.amzracing.ch/amz/files/amz1912.JPG

Fourth, it helps to have a sponsor with good CNC equipment and you should also do a bit of FEM (but don't rely on it too much). Did I mention you should try to keep the weight down? An ideal lightweight design has the stresses spread evenly over the entire volume.

http://www.amzracing.ch/amz/files/amz1911.JPG

Last but not least, testing. And more testing. Then redesigning, building, and more testing. You might want to verify some calculations and adjust your design to achieve the targeted strength and durability. I built a test bench just for driveshaft testing.

http://www.amzracing.ch/amz/files/amz1913.JPG

The result is a driveshaft of 540 mm length and 34 mm diameter that weighs <500 g including tripod rollers and plunging system, while enduring a simulation of 1000 km autocross and >900 Nm of static load. The weight including both tripod housings and boots is <800 g.

http://www.amzracing.ch/amz/files/30.jpg

I hope there was some interesting information for some of you and would love to read about your own carbon driveshaft.

Regards
Thomas

[Thrainer]

I made some more calculations, with interesting result. Simplon's carbon driveshafts are 21% lighter than Furka's steel driveshafts, while their Moment of Inertia (MOI) is 78% larger. On the other hand, Novena's driveshafts offer 61% less weight and 8% less MOI compared to Furka's.

Has anybody made similar experience? I would conclude that most carbon driveshafts don't even offer a performance benefit.

Regards
Thomas

[Electrofreak]

DanVaan: "One of our members did a senior research project to find the best method of joining the aluminum ends to the shaft."

DanVaan, did he write a paper? Any way I can have access to it?

Freak

[JWard]

I will try and find out what publication the article was in, but warwick worked with ctg to produce a carbon halfshaft with integrated tripod ends. The tripod ends were machined in house from a sintered maraging steel. CTG have a proprietary method of mating the ends to the halfshaft, that is covered in this article I believe. They were a great design excercise using much novel / advanced manufacturing techniques and they turned out super light.

[mech5496]

Interested in that as well... Do you have any resources/links/more info?

[vikram.]

for adhesives , i used to use jb weld on engine casings when they get damaged, haven't experienced if they have the strength to handle the amount of torsion and bending moment over there

[Ben K]

Has anyone else seen RCV's CFRP halfshafts systems? USF had them on their car and frankly--they looked awesome.

Ben