Glass or carbon filled epoxy mechanical properties

General Composites Discussions Composites Talk
#1

Hi, I’ve been searching all over the web for the answer to my question with no results, so I’m posting here in case someone can help.
My idea is to have an alternative to typical 30% glass filled nylon, which is tough to get in my country and besides needs a molding machine. To be used in prototypes which must be functional.
I would like to use glass filled or carbon filled epoxy, about 1/4 to 1/2 inch strand size for the filler and mixed with epoxy at any suggested weigh ratio. Ideally 1:1 by volume, if possible or more.
I could even mix as much filler as possible, fill an open mold and apply vacuum to remove excess epoxy. And finish the top of the part by machining.

What minimum mechanical properties can I expect? Would they at least match those of common nylon without glass reinforcement?

Thanks for any help and alternate ideas.

#2

I don’t have any experience I can share with you… but just thinking out loud here, it seems like you’re suggesting something that hasn’t been widely attempted before? So why not just go ahead and make up a small batch and do a few test pieces? That’s the only way to really know for sure, and you’ll probably find out very quickly if it will be suitable or not.

#3

I appreciate your comments, however, as far as I know, glass bubbles will modify the rheology of the liquid resin and some mechanical properties like density and hardness, but I don’t see how they could contribute significantly to improving tensile strength which is what I need. I have used glass bubbles in the past but mainly for fillers and bonding putty.

The final density (weight) will depend on how much fibers I can add before reaching the point where the epoxy/fiber mix loses fluidity.

If you have any data on how glass bubbles would improve tensile strength, I would appreciate it.

Have a great day.

Alfredo.

#4

chopped fibers will be hard to get good readings on, unless you have a method of manufacture, and are able to change parameters. 1/4-1/2" are very long, and depending on your part, if you mix the resin/fiber together, and cram it into a mold, the fibers will never be the same direction twice, and then your part will always have different properties. Injection molding uses CAD models, and will include a flow anaylsis.
Try to get the shortest fibers to start with. Also, when vacuuming out the resin, you will achieve VERY high fiber volume, not 30% like a normal filled plastic.
The choice of resin will change everything as well. Even comparing it filled nylon, you still have tons of nylon grades, which change the properties. The best bet for you is to get process down, and start testing parts. Maybe have samples of various filled nylons as well, to compare too.

#5

Just opened up my book titled “Principles of Polymer Engineering” by McCrum, Buckley, and Bucknall. I skimmed the sections regarding filled plastics and I believe you could use the following equations as approximate properties for your short fiber composite properties…

Emax = Phif*Ef + (1-Phif)*Em

Where Emax would indicate the maximum modulus, Phif the volume fraction of filler (or in this case fiber), Ef the modulus of the filler, and Em the modulus of the matrix (or in this case epoxy).

Then if the fibers are random in three dimensions, as I understand is most correct for your situation, then the isotropic modulus would be…

E = (1/5)*Emax

where E is the modulus of the composite.

It sounds as though you are shooting for rather high fiber volume fractions, so you could probably use the equation given below

Sigmamax = E*epsf

The variables are similar to the above modulus equations with the introduction of Sigmamax as the composite failure stress and epsf as the failure strain of the filler (you should be able to get this from your carbon supplier).

All of that said, these equations also should not be trusted as gospel. I think they will be good to give you an idea of where you are, but you will need to test the actual manufactured parts, as others have said.

EDIT: I changed up the equation for stress, as this simple stress calculation should be more appropriate for a rough estimate for your material. I wanted to mention though, that this is my opinion on the matter and not a regurgitation of the McCrum book.

#6

In regards to injection molding, fiber orientation is generally pretty predictable. The fibers align with the direction of fill. So if you “gate” or fill a long 1"x 8"x .125" part from the small end the fibers will align with the length of the part. Knowing that you can design and plan tooling around this.

Thermoplastic reinforced parts have generally short fibers, anywhere from 0.06" inch to 0.25" long. There are a few suppliers that do “long fiber” that are 0.5" to 2.0" long. I only personally have experience with the shorter fibers in polycarbonate. Glass or carbon filled resins are hard wearing on both tools and molding machines.

You could possibly pull your resin fiber mix into a cavity using vacuum? As previously mentioned coupon testing seems a good way to go.

#7

This. Problem about injection molding and shoving the fiber/matrix in a mold to test never leads to consistency.

#8

^ Exactly that, the “flow front” always kind of fans out. That along with the fact that cavities are rarely if ever simple test bars. I’d love to see some test results from this.

#9

Thanks for all the info.

So Fleisch, about your equations:

Assuming E-Glass, with the lowest tensile strength at around 500000 psi, assuming a low 30% fiber content on West Systems epoxy resin, around 7300 psi. Then: Emax = Phif*Ef + (1-Phif)*Em

Emax = 0.3x500000 psi + 0.7x7300 psi = 155110 psi

random fiber orientation so: E = (1/5)*Emax = 31022 psi, which would be an acceptable value to justify farther testing.

Right now I’m in the middle of something but beginning next year I will try to put some samples together and have them tested at the local University and hopefully put some results here.

About the last equation: Sigmamax = E*epsf

epsf= Failure strain, should this be a %?? That’s what I get from techical data sheet, and would give a very low value : 31022*0.12 = 3722 psi which can’t be right.

Sorry, not a native english speaker, so maybe I’m missing something there?

#10

Cletero,

You have used values of strength where values of elastic modulus are to be input (also called the modulus of elasticity or Young’s modulus). Try it again and see if your strength number is still low.

#11

Yeah, my mistake using tensile strength instead of elastic modulus.

Still, my main doubt is about “epsf= Failure strain”: what would be a typical value for E-glass? The value in my data sheet seems too low, but it’s given as a percentage (12%), is this right?

Also, just to be sure, in the equation “Sigmamax = E*epsf”, does E come from the equation “E = (1/5)*Emax”?

Thanks again for your help.