HKS Kansai carbon driveshaft coming - lighter than current carbon drive shafts

[chrisl]

Originally Posted by Turdinator

Interesting info

Could the circumferentially wound carbon be a safty measure, to hold the shaft together better at high rpms?

Only if their manufacturing process is awful. The centrifugal loads on a driveshaft are pretty small, and the 45 degree winding is still fairly strong in that direction. Even with the driveshaft spinning a bit over 12,000 rpm (which is the equivalent of redline in 6th in an AT car, even though they actually couldn't come close to pulling that off in reality), assuming a 3 inch diameter driveshaft, the load is only about 6000G at the outer diameter. This sounds like a lot, but it's only about the equivalent (assuming all the weight is concentrated around the outside) of the load that would be caused by pressurizing the driveshaft to around 100psi (based on some rough, back-of-the-envelope calculations). That's a pretty easy load for a CF tube (even one wound at 45 degrees) to handle. Far and away the biggest load on the driveshaft will be from the torque of the engine (especially clutch dumps/launches/jerky shifts), not from rotation.

You could maybe make a case that you would want some fibers oriented to maximize the driveshaft's bending stiffness, so you wouldn't set up any transverse vibration modes due to any slight eccentricities or asymmetries, but in that case, you'd actually want the fibers running more parallel to the driveshaft, not circumferentially.


(isn't engineering fun?)

[OICU812]

Well I'd like to try another cf shaft hope there's more info on this one soon!!

[retoocs]

What's being done about the carbon fiber / aluminum interface?

[SubieNate]

Originally Posted by retoocs

What's being done about the carbon fiber / aluminum interface?

Hopefully something. CF/Aluminum makes a pretty gnarly galvanic couple. No bueno.

@chrisl - are you sure about the wrap on other driveshafts? I haven't seen any up close but in pics the most apparent thing to me is the imprint left by what looks like a shrink tape (or just plain tensioned plastic wrap) final overwrap done before cure to consolidate the wind. That would be nearly perpendicular to the shaft. I can't see the other driveshaft manufacturers being so dumb as to waste material (and therefore $$$) on an inefficient wrap pattern but maybe I'm missing something.


EDIT-

If you look at the above pic you can pretty clearly see the carbon is running more closely to a 45* angle than the lines left in the surface by the manufacturing process would indicate. This is the PST. Arguably a less pretty wrap job, but probably nearly as functional.

Cheers
Nathan

[chrisl]

Originally Posted by SubieNate

Hopefully something. CF/Aluminum makes a pretty gnarly galvanic couple. No bueno.

@chrisl - are you sure about the wrap on other driveshafts? I haven't seen any up close but in pics the most apparent thing to me is the imprint left by what looks like a shrink tape (or just plain tensioned plastic wrap) final overwrap done before cure to consolidate the wind. That would be nearly perpendicular to the shaft. I can't see the other driveshaft manufacturers being so dumb as to waste material (and therefore $$$) on an inefficient wrap pattern but maybe I'm missing something.


EDIT-

If you look at the above pic you can pretty clearly see the carbon is running more closely to a 45* angle than the lines left in the surface by the manufacturing process would indicate. This is the PST. Arguably a less pretty wrap job, but probably nearly as functional.

Cheers
Nathan

I've only looked at 2 aftermarket CF driveshafts in person, and they definitely had the incorrect wrap. I can't say for sure if that's common or not though...

As for that one in your picture, it's kind of hard to tell. I don't really like the small voids visible in the surface layer, but they aren't a problem so long as there aren't any deeper within the composite. The fiber direction is also difficult to see, though it does look like it's somewhere between diagonal and longitudinal, which isn't a bad choice. It does look like the surface layer is entirely wrapped in one direction, instead of alternately in both directions, but again, that isn't too much of a problem so long as a layer just underneath is wrapped the other way.

[z3ro]

Originally Posted by ciro

Finally a company uses the correct weave.

correct weave?

[SubieNate]

Originally Posted by chrisl

I've only looked at 2 aftermarket CF driveshafts in person, and they definitely had the incorrect wrap. I can't say for sure if that's common or not though...

As for that one in your picture, it's kind of hard to tell. I don't really like the small vids visible in the surface layer, but they aren't a problem so long as the aren't any deeper within the composite. The fiber direction is also difficult to see, thigh it does look like it's somewhere between diagonal and longitudinal, which isn't a bad choice. It does look like the surface layer is entirely wrapped in one direction, instead of alternately in both directions, but again, that isn't too much of a problem so long as a layer just underneath is wrapped the other way.

Thanks for the info. It looked to me like there was some fiber running in the opposite direction in the bottom left of the shaft in that pic, but it is hard to tell. Would be pretty easy to sort out in person.

The voids look typical of a wet wound shaft done without an exterior mold/shaped vacuum bag. Not likely to have a huge strength impact but as you say, they are less than ideal.

Nathan

[RallySport Direct]

Just to start off, this is NOT to start an argument and more to help people understand about carbon fiber and how it is used.

As chrisl stated carbon fiber strands are strong in tension, and weak in every other possible aspect(think of a rope). So to get the most strength from the fibers you want the fibers to be in tension. In any other direction all the loads are being taken up by the Resins that are used to hold the carbon together.

The interaction of the resin, the type of resin, and amount of resin has more to do with how a carbon fiber acts when loaded then just the carbon itself. Resin makes up between 35-50% of the material, and can have not only your basic components but also things like nano-carbon tunes and fillers to make it stronger or flexible. The better the resin for the application the more load the entire matrix(term used to describer carbon and resin sample) can hold in all directions but ultimately it will be the strongest in the direction that puts load on carbon in tension.

Carbon fiber is made in layers, a single layer is thin (1/4" thick piece of carbon fiber is anywhere between 75-150 layers) and each layer can be oriented in a different direction to give you the desired results. The worst way to use carbon fiber is to make it quasi-isotropic (making it the same in all directions), so manufactures will mix layers and put material in different directions depending on what properties they want from the material.

For this application the ideal placement of fiber direction is along the direction of load, which is around the shaft in the perpendicular direction. However due to the weak bond between the resin and fibers the shaft would simply sheer if all fibers were in that direction. Then if you put all fibers in the parallel direction there would be no sheer issues but you have load issues were the shaft is not able to take high loads with out bending. So it comes down to a fine balance of each different aspect.

Last but not least, unless you are the manufacture you do not know what the manufacture has done to the carbon, what carbon they use, what resin they use, and in which direction they have placed the carbon elements. Carbon fiber is typically 8-20 layers deep and each layer plays a critical part in how much load the matrix can take and in what direction. The automotive industry is driven by aesthetics, so everyone makes the outside of parts look shiny and pretty. But with carbon fiber its on the INSIDE that counts. And given that people buy majority of parts on how they look on the outside, many manufactures go out of their way to make outsides of parts look more impressive then 98% of the rest of the product. No carbon manufacture will simply come out and tell you what resin and carbon they use, what manufacturing technique, what carbon lay up, and the curing temperatures/pressure as that is the "secret sauce".

So the moral of the story is simple, you simply cant look at a true carbon fiber part and say if its "good" or "bad" with out running tests on it or getting the manufacture to tell you the recipe to their "sauce" on how they made it and so forth. Its like looking at a cut of raw steak and guessing the age of the cow and where the cow came from with out knowing all the details.


Kirill
RallySportDirect.com

[chrisl]

I agree with most of what RallySport says above, with one exception. Namely, this:

Originally Posted by RallySport Direct

For this application the ideal placement of fiber direction is along the direction of load, which is around the shaft in the perpendicular direction. However due to the weak bond between the resin and fibers the shaft would simply sheer if all fibers were in that direction. Then if you put all fibers in the parallel direction there would be no sheer issues but you have load issues were the shaft is not able to take high loads with out bending. So it comes down to a fine balance of each different aspect.

(warning: engineering-speak incoming)

In a torsionally loaded shaft, the direction of load is not around the shaft in the perpendicular direction. Instead, as I mentioned above, the principal stresses are at a 45 degree angle to the shaft. In addition, CF typically has a compressive strength nearly as high as its tensile strength (admittedly with some exceptions, depending on the type and grade of the carbon fiber). As one of the principal stresses will be in tension and the other will be in compression, a shaft with windings at the surface layer that are at ~45 degrees in both directions is ideal, since the fiber direction will align with both principal stresses (which are the two directions in which there is no shear stress). The only time you want the carbon wrapped perpendicular to the tube is if the tube has to withstand a crushing or expanding load, such as a weight set on the side of the tube, or if the tube is part of a highly pressurized system.

In a brittle material (in other words, one which fails due to tensile, rather than shear stress), the failure mode will be a helical fracture surface, as can be seen if you twist a piece of chalk to failure. A carbon driveshaft is admittedly a bit more complicated than this due to the interaction of the resin with the fibers, but this is still the basic failure mode for most fiber-reinforced plastic composites (fiberglass and CF, along with a few other exotic ones). By orienting the fibers at 45 degrees (perpendicular to this failure surface), their usefulness is maximized.

Interestingly, after going back and looking at some of my engineering textbooks, I might have to take back what I said above about the voids in the surface of that driveshaft posted by SubieNate not being a big concern. A torsionally loaded shaft has the maximum stress at the surface, and surface imperfections can cause stress concentrations that significantly weaken the overall shaft. Admittedly, most of these driveshafts are probably manufactured with such a large safety factor that you would never get close to their failure point anyways, but surface imperfections do apparently have a larger influence on the failure of a shaft in torsion than I had remembered. The driveshaft in the OP does appear to be better made in that regard, as it appears to have a uniform, smooth surface finish with an equal number of fibers wrapped in each direction.

[RallySport Direct]

Originally Posted by chrisl

(warning: engineering-speak incoming)

Sounds like we speak the same language

You are correct, its been a couple years from when I had to deal with stresses in a shaft under torsion.

We can go for quite some time going back and forth but I think most of what needed to be said has been cleared up.



If someone wants to learn a little more about composites and fall asleep here is a good way to get your toes wet and possibly get confused of how loads in composites interact.

This is posted by one of the pioneering research Universities that specializes in the composite industry research.

http://www.mech.utah.edu/~rusmeeha/l...omposites.html


Kirill
RallySportDirect.com

[OICU812]

Any updates on this HKS Shaft? Will production model come with front protective cup that will mate in nicely as well to keep out debris?

[kbye]

Originally Posted by OICU812

Any updates on this HKS Shaft? Will production model come with front protective cup that will mate in nicely as well to keep out debris?

+1 would like to know as well.

[OICU812]

Did these guys ever make this shaft for sales/production?

[xwd]

I don't see it listed as an available item anywhere. There are a couple other brands of JDM CF propshafts and they are in the $3500-$4500 range... I can't see it ever being worth it compared to something from the US vendors.