3D RTM

[earlgreyhot]

We are using RTM on a square preform woven in three dimensions. I know most RTM is done on flat two dimensional weaves.
How do I minimize voids? All of the literature I'm finding only deals with 2D preforms.
1. Should vacuum on the mold be maximized while injecting the resin into the mold?
2. After the resin starts coming out of the exit line, it's closed and the mold gets pressurized by injecting more resin in. Should this hydrostatic pressure be maximized?
3. Degassing - Is pulling vacuum and mixing enough to degas the resin and the hardener or do I need to use some nucleator?

 

[Compositepro]

A 3D preform should be more permeable than a 2D, so infusion should not be difficult. The main challenge is locating the injection and vent ports so that the vent is at the last point to get infused. The most common cause of voids is entrapping air in a portion of the preform that has not infused yet when the resin flow front blocks the vent port. When this happens, the residual air will be compressed into one void, as there is no way to remove it. Even with a high vacuum is applied, it is surprising how much air volume is left.

Vacuum should be maximized if possible (29.9"Hg or better). Some use no vacuum and use the resin flow front to push air out. This requires careful control of the resin flow rate so that tow bundles can impregnate by capillary action before the flow front encapsulates air in the fiber bundle.

Generally, the higher the pressure, the better. The mold will be more expensive as pressure increases. Some mold expansion due to pressure can be very beneficial, however. Resin shrinkage during cure is a much bigger problem in closed cavity molding than in vacuum-bagging. If the mold gets slightly inflated, then it can shrink with the resin and prevent voids due to cavitation.

Be aware that there is a difference between degassing and devolatilization. A volatile is a liquid that can evaporate. Vacuum is sufficient to remove gas bubbles, and volatiles actually help this process. Volatiles will not cause voids if there is sufficient pressure during cure.

 

[earlgreyhot]

The problem is not one big void but lots of "microvoids" between the fibers but still visible to the eye.
The epoxy isn't styrene based and it doesn't really have volatiles. My thinking regarding the degassing is that there are gas bubbles, but there is also dissolved gas that I don't think would come out with just a vacuum. The less gas load in the resin, the better it will absorb the gas bubbles caught in the infusion and thus there will be less voids.

The maximum that the mold can be pressurized to is 50 psi. Is that enough?

 

[Compositepro]

As long as you aren't injecting visible bubbles, I don't think degassing is your problem. As I mentioned, a 3-D preform is usually fairly permeable, and if you inject too rapidly you will trap air in the tow bundles. Later in the cure cycle this air will end-up as bubbles in the spaces between tows due to capillary action impregnating the fibers and pushing the air out.

 

[earlgreyhot]

We are only injecting a few visible bubbles at the beginning of the infusion.

Is it true that the higher the hydrostatic pressure of the resin, the smaller the bubbles will be? So if we pressurize at 100 psi the bubbles will be half the volume of if we pressurized at 50 psi, due to boyle's law?
I still wonder if 50 psi is enough to collapse bubbles or if we are off by an order of magnitude even...

And cavitation during curing... does this occur primarily on the sides of the mold? Because we are getting a lot of porocity towards the center as well.

 

[earlgreyhot]

also the resin we use has very low shrinkage.

 

[Compositepro]

Air bubbles get smaller with pressure but will generally not completely go away. 1% shrinkage will result in 1% voids in closed cavity molding. Hopefully the void shows as a separation from of the part from the mold surface. Often this shows as "worm-holing"; tracks on the part surface that look like worm holes under the bark of a tree. This is caused by cavitation in one spot on the mold surface which grows linearly as more shrinkage occurs. An air bubble will be a nucleus for cavitation inside the part. Fast cures cause temperature gradients, which results in areas curing at different rates. This causes most of the cavitation to show-up in the last area to gel, which is usually the center of the part.

 

[berkshire]

Earlgrey,
item 3 It helps to keep the air bubbles down, if the resin is mixed under vacuum, prior to resin transfer. If you have reactive dilutants such as DGE they may become volatile.
B.E.

You are judged not by what you know, but by what you can do.

 

[earlgreyhot]

I'm stumped.
The micron gauge came in today and our vacuum is fine. I pulled vacuum from one end of the part and the other reached 500 micron (29.98") in half an hour.
Shrinkage in our resin system is only 0.35%. Meanwhile we cut the latest part in half and the estimated porosity was almost 2%, with clusters of voids.

So if I'm understanding correctly, the only two ways to lower porosity would be to tweak the injection speed or increase hydrostatic pressure at the end of the infusion?
And if I figure out a way to increase pressure to say 500 psi, would that effectly make voids dissapear, since void volume would be 10x smaller than what it is at 50 psi?

 

[Compositepro]

RTM is very sensitive to air leaks in the mold or tubing.