IW STi Forum banner

Carbon and Aluminum Drive Shaft Overview

30K views 30 replies 19 participants last post by  Scargo  
#1 · (Edited)
With the popularity increasing the last few years of using aftermarket drive shafts , we wanted to go over some information in regards to the subject and cover some common questions that we see our customers ask us about.

Why:
The reason to change out the drive shaft to a Aluminum or Carbon Fiber unit is quite simple, it is to reduce rotational mass and increase response.

By replacing the heavy stock steel drive shaft with a lighter 1 piece unit you save about 1/2 the weight, and allow more energy to go into rotating the tires instead of rotating dead weight. Similar concept as doing a lighter flywheel and pulley, however other additional benefits are also to be had.

Many new modern cars are now coming equipped with light weight shafts for the reasons of performance, such as the GT-R and other high end exotics.



Aluminum Vs Carbon:
2 of the most popular materials to use when replacing the steel shaft will be either aluminum or carbon fiber. Below I will cover some of these basics.

1) CF shaft is smaller in OD to handle the same tq capacity as the aluminum counterpart unit. Means you have more clearance and requires less modification under the car to run a carbon shaft vs aluminum. Some applications require modification to heat shields or transmission tunnels to allow adequate clearance for the larger OD Aluminum shaft.

2) The CF shaft has a higher critical speed. For example a aluminum STi application shaft start to flex and vibrate above 155MPH do to the aluminum hitting its natural frequency. The CF shaft is closer to 195MPH. It might not sound like 40MPH is a big difference, but a Stg2 STi can get very close to 155MPH...to hit 195MPH you got to have a crazy fast car (well in access of 500WHP to hit that, and only a small hand full of tracks in the world have straights to accommodate that).

3) The CF shaft can twist about 2X as much compared to the aluminum counter part before it fails from sheer stress. This will directly apply to shock load capacity.

Carbon shafts will be able to absorb good amount of shock. They have very good elastic memory, meaning that they will return to the original shape if deformed from twisting.

Directly from ACPT
"All driveshafts twist to some degree when torque is applied. The resistance to this twist is measured as torsional spring rate. Standard carbon driveshafts have a torsional spring rate a little less than aluminum and about half that of steel. The advantage of a lower spring rate is less driveline shock and a reduction of stress on other drivetrain components, as well as increased traction."

If you are having issues breaking your rear axles, or ring and pinion this might be a good way to prolong the life of those parts. Reason is that the carbon drive shaft will be like a rubber band, when launched hard it will twist slightly to absorb some of the shock and then spring back to its normal shape. The twist would be minimal (only a few degrees 5-10) on VERY hard launches, however that twist is absorption of energy. Thing that breaks many parts such as gears is shock to the part from abrupt and high application of power. F=M*A, so if you slightly reduce the acceleration of the power application you prolong life of components. *simple way of thinking this, you can put a 10lb hammer on your finger slowly with no damage, however if you slam the hammer you will do allot of damage. Similar concept with the carbon shaft twisting slightly, as it slows down the application of the energy on the component.


4) When the CF shaft fails, the fibers will come apart and un-wind. When a Aluminum shaft fails...well you have a 14lb sledge hammer spinning at over 1000RPM going to town under the side of your car and you just hope it does not penetrate.

The big concern many people have with the CF units is the CF shaft separating from the metal yokes. The adhesive that actually holds the 2 together, how its assembled, and how its cured is a trade secret of these manufactures as this will separate many of the manufactures and really make or break a manufacture.

That was a big concern for us as well when getting into the CF drive shaft market, however after many conversations with PST about they have yet to have a failure of the carbon or the adhesive itself. They state that the failure point is....the metal yokes actually. Not something that they have seen on import applications much in the past, but more or less on 1/4 mile domestic applications. And at that point they have a option to custom make a billet yoke if your car really needs it

5) A equivalent CF shaft will be about 5-10% heavier then a Aluminum counter part.

6) Aluminum shafts are only about 30-50% of the price of the carbon unit.

So what one to get?
This strictly will depend on your application and budget to be honest.
Overall both aluminum and CF are significant improvements over the stock unit. But the CF will be better then Aluminum as its safter, smaller diameter, has a higher critical speed, a better elasticity, and higher degree of torsion before fracture/separation. However that is not saying that aluminum is a bad, as aluminum shafts have been used for decades before the composite technology got to the point where it was better and more affordable for this type of application.

Aluminum Shaft
Image

Image




Carbon Shaft
Image

Image




Our Review of a shaft installed on our 2011 STi here at the shop:


Stock Driveshaft: 27.6lbs
PST: 14.2 lbs!​

Weight Savings of 13.4lbs means that your engine spends less energy rotating the driveshaft and more energy on spinning the tires. This in turn decreases drivetrain power loss and increases power at the wheels. So even though your car is not making more power, its able to transfer more of it to the wheels increasing performance, drive-ability, and response.

After installing it on our 2011 STi shop car, we where nothing short of amazed. The fitment was spot on and was a direct factory replacement. Install time was about 10 min after the mid pipe was removed off the catback.

Test Drive Results where nothing short of AMAZING! The vehicle was much more pleasant to drive in very low RPM with no more bucking or shacking, and it was much more eager to accelerate even at partial throttle. Throttle input response was also drastically increased with less "dead" time between when the throttle was applied to when the vehicle started to accelerate. As for all out acceleration, well the thing is MUCH faster then compared to the way the car was with a stock driveshaft. The vehicle is much more eager to go through the RPM range at a much faster rate, meaning that its accelerating faster. The shifts where also crisper and faster as up shifts and down shift delays where cut to a minimal in terms of hard downshifting and fast upshifts.

We where curious to see the results the addition of the driveshaft made, so we did a quick dyno comparison of before and after.

Performance Mod List:
PST Carbon Drive Shaft
Cobb Tuning Accessport Running the 91 octane AEM Stage 2 map
Cobb Tuning Catted Down pipe
Cosworth Intake
Invidia Q300 Catback

Temps: High 30's
Altitude: 4300ft
Fuel: 91 octane winter blend (talk about bad fuel)

As you can see, consistently through out the entire powerband both Tq and Power increased!
Image
*sorry about the starting RPM range, it was getting later in the day and we didnt think twice about the starting RPM of the initial pre-driveshaft starting point. But you can see that after 3000RPM the gains are nice and constant.


Kirill
RallySportDirect.com
 
#7 ·
Very informative thread. Thanks for the test/ write up

I have an ACPT carbon drive shaft and I'm very happy with it
 
#8 ·
Are there any issues with the Aluminum driveshaft option and the Invidia Q300 exhaust on a 2011 STi Sedan? I see where you've mention that there is only 1/4"-1/2" gap to the resonator with the PST in another post. I'm guessing the PST CF driveshaft may be my only option potentially.

Great write up!

Thx
 
#9 ·
Given how much larger OD aluminum must be to handle this type of torsion, I will guess there would be a issue.
Interesting thing here is this, if they just dropped the OD on aluminum shafts and slightly increased the wall thickness the shafts would be easier to balance and smaller in size eliminating almost all the common issues found with them :-/

Kirill
RallySportDirect.com
 
#12 ·
It is a balancing act. The torsional stiffness of a round hollow section is driven primarily by its outer diameter. A solid section will experience zero torsional shear at its center, so most of the load is being carried by the material at outermost diameter. Basically, shear stress is directly proportional to the radius and is a maximum on the surface

Shrinking the diameter would require a much greater wall thickness, not just a slightly increased one. You will end up with a heavier shaft even while your diameter is shrinking.

If you know your maximum torque and you comfortable factor of safety then you can optimize the drive shaft size through an iterative process.

One of the key things you need to pay attention to is polar moment of inertia:

The polar moment for a tube is j=pi*(D^4-d^4)/32

What can be seen here is that the polar moment of inertia goes up very fast as it is proportional to the diameter to the 4th power!

THe max shear stress is Tau(max)=TR/J where T is the applied torque (in our case vehicle crank torque multiplied by a factor of safety) R is the radius and J is the polar moment of inertia from above.

If you know your max shear stress for the material (which thanks to books we do) you can iterate and get the diameter that meets your strength requirements with minimum polar moment of inertia.

Im sorry to say, that I think the current aluminum drive shafts on the market are already optimized.

Sorry if that was a little dry.
 
#15 ·
swedishSTile, thank you for the nerds eye view on the explanation :)
I would have put in a bunch more math based info into the original post, but I think it would simply have gone over many peoples heads and caused them to get bored and never finish the article. As unless you are accustomed to reading engineering/math/science text books the info is quite dry to say the least LOL




Would be no issues at all, we got plenty of customers running a pulley, flywheel and driveshaft with no issues. Not imaging getting some nice 2 piece light weight rotors, and some light weight wheels....that would be one heck of a quick car due to the fact that your not spinning a bunch of dead weight ;)

Kirill
RallySportDirect.com
 
#17 ·
While I can visualize a flywheel and rims and quantifying the results of these lighter components is pretty well accepted, and deemed important, from a performance aspect, I have trouble accepting subjective claims that a CF driveshaft has measurable "feel" or butt dyno impact. Its diameter is so small... Its weight is similar to a flywheel.
I also don't understand how the lightening of the driveshaft impacts torque and horsepower as a claim of improving it. It doesn't improve either of these or make them "more" does it? It just allows what already exists to be seen a few hundreds of RPM sooner? Those few hundred RPMs (how much sooner?), can significantly affect the acceleration of the car? How many feet will I gain on my competitor when I accelerate out of a turn in third gear at four grand?

It is appealing that I could shed 13 pounds, but for $1000?

I am also pondering how a one-piece shaft can be used with no consequences (or more why a two-piece was ever used). I don't have my car to look at right at the moment but my application is road racing and I guess there are no clearance issues or downsides to using them on any GR?
I may sound like a contrarian, but I want to know the minuses as well as the positives and its value be quantified relative to something like a light flywheel, which is relatively cheap to replace and which has a very obvious positive effect for a race application. And no disrespect to a vendor, but they are trying to sell these things and they are expensive.
 
#20 ·
The shaft does not "make" any more power, what its doing is reducing drivetrain losses that are consumed as energy. In this case the reduction in rotational mass through a driveshaft that is lighter then stock allows for more energy to go into spinning tires instead of being consumed by using the heavier steel driveshaft. Not sure if your familiar with bike racing, but pedal bike road racers for instance will use carbon fiber wheels for this exact reason over aluminum wheels for instance.
The power gains are not huge, so do not think that all of a sudden you will go from running a 13 second 1/4 mile to a 12.5, the gains will be in the fact that the car will be able to go through the powerband quicker with out having to waste energy through rotating a heaver shaft. Same principal as doing lighter wheels, crank pulleys, flywheels.

As for fitment, we have not seen any issues at all when using a carbonfiber shaft. Aluminum shafts do require tunnel modifications, but the aluminum ones are a simple bolt on and go.

Hope this helps explain it a little more for you.

Kirill
RallySportDirect.com
 
#18 ·
From a road racing perspective I know that the CF driveshaft made the most improvement to me in drivetrain smoothness. I don't think you can really notice any real gains in free'd up hp or tq though, even on a dyno I don't really see these showing any real difference at all. For me the harshness of the drivetrain on downshifts and such was reduced quite a bit by adding lighter rotors and the driveshaft with the less rotational mass if you can see what I mean.

As for clearance I don't see any issues on my car and have been running one for over a year now even through winter. Plenty of room.

This is a modification you should do when you really have nothing else left to do and you have some money burning a hole in your pocket.
 
#22 ·
The engine produces a fixed amount of crank HP & TQ. Reduce the drive train parasitic loss (weight) and the transfer of the fixed crank power is more efficient. This is seen on the first page dyno chart. So say the typical Subaru loss from crank HP to WHP is 25% - the driveshaft reduces the loss to say 20%.

Flunked math and still want numbers, try this hypothetical scenario: your stage 2 car produced 400hp at the crank, 25% drive train loss puts you at 300 at the wheels. With the c/f driveshaft, loss is reduced to 20% so you've gained 5% power. Your same 400hp at the crank is now 320hp at the wheels.

Clear as mud now?
 
#23 · (Edited)
The time to upgrade your driveshaft is when the central bearing in the stock unit needs to be replaced. Then your just spending only a few hundred extra dollars. Those doing this (like me) can not give you an apples to apples comparison because we replaced the driveshaft due to vibration so the new shaft had to make a difference. In my GD there may have been a small difference performance.

Except initial cost there is no downside to using the CF shafts, only benefit! If produced properly they are stronger and lighter, and have less rotational mass. Eliminating the central bearing couples less noise into the floor and leaves one less source of friction too. Probably all, but at least my Drive Shaft Shop shaft has replaceable U-Joints, so I should not have to replace the shaft if they were to wear out. So few decisions are so easy to make!
 
owns 2020 Subaru STI
#24 ·
I want to apologize for some of my comments. There might be a bit of hyperbole from some sellers and users but "snake oil" was inappropriate. I can see that a light driveshaft has measurable benefits. In a useful power range it seems to be about three percent (for that setup). I didn't flunk all math, just higher math.
I was also de-coupling power from torque, when the are inseparable. I was confusing power with energy.:p
Three questions come to mind: One, is there a diminishing return with increased power? If I produce 400 wheel horsepower won't I see less benefit? In other words, those with closer to stock power are more likely to get the "woopie!" effect...
Secondly, if I have lightened my brake rotors, wheels, flywheel and crank pulley, will I still see the same order of magnitude of change with a lighter driveshaft?
Lastly, the AFR numbers amazed me. How could they be so different? You can see more power and have that much less fuel consumed with the lighter driveshaft?
 
#25 ·
One, is there a diminishing return with increased power? If I produce 400 wheel horsepower won't I see less benefit? In other words, those with closer to stock power are more likely to get the "woopie!" effect...
Secondly, if I have lightened my brake rotors, wheels, flywheel and crank pulley, will I still see the same order of magnitude of change with a lighter driveshaft?
Lastly, the AFR numbers amazed me. How could they be so different? You can see more power and have that much less fuel consumed with the lighter driveshaft?
1) The % change should not change in theory. So the more power your car makes the more difference you will see. But realistically if you have a 600WHP car, the gains will not be as noticeable on the butt-dyno as your car is already fast. On a stage2 car, it will be quite noticeable to the driver, as at that power level even small changes such as getting a catback light weight pulley are noticeable.

2) Going back to the previous explanation. The % difference in theory should not change and there is a noticeable difference. When we installed this shaft the car had already light weight wheels and rotors on all 4 corners.
With that being said, the more energy/power a engine makes the larger the friction losses as a result of friction and efficency. The more energy something makes, the less efficient it becomes. If you went into the field of studying engineering this will be covered in your advanced Thermodynamics class, Heat Transfer Class, and Fluid Dynamics classes, in essence the more power/energy a motor makes the less efficient it becomes as output is increased.

3) There was no AFR reading on the pre-shaft test. We were in a hurry to get the car on and off the dyno that day to get the shaft installed and re-tested before ambient conditions changed drastically. However AFR's will be identical as the car was using the exact same tune and exact same hard parts with only change being the driveshaft. The results shows were the best of 3 runs for each pre and post driveshaft test. Given you dont change your driving style your car will get slightly better efficiency and fuel economy, as less energy is used as a result of reduced drivetrain loss to produce the same amount of energy at the wheels.

Kirill
RallySportDirect.com
 
#26 ·
Thanks for that. Perhaps I should go back and finish engineering. I worked as an engineering tech for almost 12 years and then rubbed shoulders with engineers the rest of my working life. Still dislike the higher math and at 66 I'm not to inclined to finish getting a degree. I have a lot of experiential education.:D
Just curious... What do you think the real-world drivetrain loss would be for a race prepared car with all the light rotating parts, racing fluids and 400 whp? Is it still going to be above 20%, given that racing is almost all hard acceleration (or slowing)? It seems like it would be more than average since you factor out the cruising, low load parts of driving and interject high-friction loading of the gears, bearings, etc. What do dynos tell us for crank VS wheel dyno results (for race cars). Do any race teams worry about that and make a concerted effort to reduce parasitic losses, beyond the usual lighter race parts? The only thing I can think of is playing with lubricants and lowering viscosities.
 
#27 ·
Is it still going to be above 20%, given that racing is almost all hard acceleration (or slowing)? It seems like it would be more than average since you factor out the cruising, low load parts of driving and interject high-friction loading of the gears, bearings, etc. What do dynos tell us for crank VS wheel dyno results (for race cars). Do any race teams worry about that and make a concerted effort to reduce parasitic losses, beyond the usual lighter race parts? The only thing I can think of is playing with lubricants and lowering viscosities.
I think the parasitic loss percentage is measured under power, meaning it ignores "the cruising, low load parts of driving..." etc.

And they're going to run the viscosities/fluids that give them the protection/performance for as long as the parts in question need to last (a race, a season, etc.) I can't imagine they'd pick up anything worthwhile by trying to run extra light viscosities. If they get one failure in one race, they've lost more than they'll ever pick up.