This is not welding related but definitely fabrication related. I thought I'd post it here because it shows how composite component strength differs using different construction methods and materials. I would say this test displays the most advanced vs. the most common materials and processes. In between you have vacuum molding and infusion fabrication methods as well. These methods are availble to the home builder. Well here's the test setup, it may be a little dry for some people but it may give other people some questions to ask next time they buy a carbon fiber part for their street or race car.
All testing was carried out by me. This test was copied from my website


The aim of this post is to use a simple test to display the how materials and processes can dramatically change the stiffness of a given part. In this example I will compare the stiffness of two carbon fiber plate samples.

Plate 1 (.054" thickness): The material used to build this plate is 8 plies of standard modulus prepreg carbon fiber. This plate was cured inside an autoclave at over 300 deg F., 100+ PSI outer pressure, and at least 20" of mercury inside a vacuum bag. This plate was cured using a specific pressure/temperature/time ramp rate per the material manufacturers recommendations. The specific layup stack goes like this:

3K 2x2 Twill Fabric
45 degree Unidirectional (Non-Woven) Carbon
90 "
0 "
0 "
90 "
45 "
3K 2x2 Twill Fabric

This layup is considered quasi-isotropic. In basic terms the layup is evenly stacked. With unidirectional carbon fiber you really have the option of building in stiffness where you need it most without increasing weight and thickness when compared to fabrics. If all plies were laid up at "0" degrees it would only be stiff in one direction, but it would be "really" stiff. As an aside, metals are usually isotropic materials as they offer the same stiffness along every axis. The layup above closely replicates the isotropic nature of a metal plate. The fabric layers are just there for looks basically. Fabric is called "The Money Layer" in the industry because it doesn't lend much to the stiffness of a part but is much more expensive than uni carbon. Fabrics are inherently weak because each fiber has a kink as it travels over and under each other. The fibers are already starting to bend even before they get loaded. Uni-directional carbon is much more efficient when used in a prepreg state as opposed to the dry fiber form. This plate is about as good as it gets from a properties point of view but not everyone has an autoclave right!

Plate 2 (.060" thickness): The material used to build this plate is 5 plies of standard modulus (dry) plain weave fabric as sold by many DIY composite material suppliers. This plate was laid up onto a piece of glass on my kitchen counter using wet layup techniques. This is about the cheapest and easiest carbon fiber plate you can build. It replicates the process used by many aftermarket automotive carbon fiber parts manufacturers to build hoods, engine covers, wings etc. This is what makes many parts relatively inexpensive. I used 5 plies of 3K plain weave carbon fiber stacked on top of each other. I did post cure the plate to give it some extra stiffness. For those that do not know a post-cure is when you let the plate cure at room temperature and then you heat it up for a given amount of time at a given heat depending on the resin used. This gives a part a higher use temperature and/or increased stiffness. Now that you've met the players lets shake hands and come out swinging!

**Please note this plate is slightly thicker than Plate 1 but only has 5 plies of carbon fiber.

First I cut out a section of each plate. Each plate is almost identicle in size. THen I drilled a single hole in the same spot in each plate and tied a piece of fishing line to the end of each plate. I then placed the other end of each plate under my bookcase desk combo. Each plate has the same amount of overhang. See picture:



Once I had each plate positioned I tied a 5 lb. exercise weight to the end of each plate and placed a ruler on end right next to each plate one at a time. I lifted each weight and registered the spot on the ruler where the top of the plate set. Then I would slowly let go of the weight and then gave it an extra push and measured the spot on the ruler where the plate returned to. This gave me a reading of exactly how much the plate flexed.



CAN YOU SEE THE DIFFERENCE?



HERE'S A MORE DRAMATIC VIEW!



AND ANOTHER



AND THE WINNER IS...



Yes you probably guessed it already. The plate on the right is plate #2 (Wet Layup). Plate number one wins by a mile.

The readings I took are as follows:
Plate number 2 flexed a total of .5"
Plate number 1 flexed a total of .25"

The readings were taken several times to ensure the were repeatable. The results for plate #2 could have been improved slightly by using a different weave or resin. The resin used in this test is an epoxy. Using a polyester resin would have decreased the stiffness somewhat.

Some people may say what is the point of this test if I can't afford an autoclave or the tooling that can withstand an autoclave environment? Bear in mind there are many different levels of fabrication between these two extremes. Some process upgrades would include vacuum bagging during cure or even vacuum infusion. Either of the processes will yield improvements. There are many variables to keep in mind when choosing how to fabricate your parts. Hopefully this will give you more information to speak on even if you have no intention of building parts yourself. When it comes to buying products knowledge is power.

All comments or additions are welcome.

I forgot to mention both plates weighed about the same but #1 could have been lightened up significantly and still beat plate #2. If you built a one piece front end using the same process as plate #1 you could build it much lighter yet still maintain the stiffness of a wet layup part.

 

intersesting write up, good stuff

 

Yes very good. A little out of some peoples scope but i'll still give it a

 

nice write up, found it a very intersting read.

but the result is something i would have assumed but theres nothing like proof to back up something you believe.

 

I think a slightly more subjective point of this test would be what are the typical costs involved in each assembly, aka - what would consumer expect to pay. Also what applications would be suitable for different types of layup? At what point should we be concerned with using something stronger? I can't see that for a gauge panel or switch panel we would need something stronger than a wet layup, but I could be wrong.

 

Well that's an easy one. The largest parts should be made fabricated using the methods that produce the lightest parts. As with everything you will get the best parts you can afford. Everyones budget is different. Some guys wouldn't pay for that extra 20% weight loss but to some its a big deal. Of course a wet layup gauge panel wouldn't matter really but a hood, yes. When I wrote this post I wasn't thinking of cosmetic interior parts per say. On a hood a meterial; and or process upgrade would matter. I wouldn't even say you need an autoclave to build the hood. Vacuum infusion would be the route I would take for Honda hoods or a one piece front end. Vacuum infusion can net a very low resin ratio compared to regular wet layup. All this with not much added cost. Another huge upgrade would be to build hoods just using carbon fiber. I think using fiberglass under the fabric is the worst thing you can do from a weight point of view. Also using unidirectional carbon could replace glass easily. Nobody that I know of uses uni-directional carbon in their hoods (manufacturers). Fabric may look great but it is inherently weak due to the fibers being bent over and under each other. Why not use full uni for the hood and the just paint it or use a top layer of fabric and the rest uni. At least the you could design strength where you need it. Sure a hood like this couldn't be had for $300 but I would never suggest using cf if cheap parts are the goal.

 

Most people don't understand a good cf hood made with weight in mind (if they have it made by someone) will usually be over $1,000 (I've checked with 4 people and they all charge over a grand).

 

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by Aquafina &raquo;</TD></TR><TR><TD CLASS="quote">Most people don't understand a good cf hood made with weight in mind (if they have it made by someone) will usually be over $1,000 (I've checked with 4 people and they all charge over a grand).</TD></TR></TABLE>

That sounds about right especially if its a one-off part since you have to build two molds for it inner and outer and then build the parts from the molds. In a production situation you could get the price to around $850. But then again people will pay $400 for a regular hood and only drop a 10lbs vs. a stock hood. If the hood was built well you might drop another 8-10 lbs. Its possible to build a streetable hood that weighs 6-7lbs. It would be interesting to plot this hood example on a graph comparing weight lost to dollars spent.

 

Great write-up

I'm going to be trying some DIY carbon fiber frabrication soon and wanted to know if you could compare the carbon fiber pieces to similar metal pieces. (ie. how would a carbon fiber splitter compare to one with the same thickness metal)

for a good write-up

 

Great idea I'll see what I can do to get some samples of different metals to compare it to. One of the benefits of using uni-directional carbon is that you have the ability to design strength in a given direction. Itis possible to skew this test badly toward the prepreg plate. I could have laid all of the uni-directional carbon in the long direction of the plate. This would of course make it super stiff in the direction the plate is bending for the test. Instead the lay up is even with fibers running at 0/90/45 degrees. An even layup would be the most fair way top compare composites to metals since metals are isotropic in strength and stiffness (not directional).

 

dam thats **** is stronger than i thought!

 

I just bought a cf braclet from the site in your sig

 

How much do you think the results would have been effected if you had used unidirectional carbon in the wet layup, Say two at 0deg and one at 90. Not exactly a "general purpose" piece but do you think a purpose built wet layup could come closer to comparing with a quasi-isotropic prepreg piece?

I'm in the process of building some aero pieces for a motorcycle and I was planning on backing them with FG because the woven CF cloth I was looking at was expensive, but unidirectional fabric is less than 1/4 the price.

 

Yes uni is priced much better than fabric. I think the wet layup may have done slightly better with uni instead of fabric but the fabric does offer 0/90's just not the 45's that the prepreg plate has. When the weight is applied as in this test you really don't need the 45's. So the gains would be a result of material and not layup in your sceario. The only tough thing about uni-directional in complex parts is that it doesn't conform very well the the shape of the mold at least without vacuum. Pressure is better