Joined: Sep 2010
Location: Tacoma, WA
Project #7: Getting Started With Resin Infusion
I have been thinking of trying infusion for a long time, and now I have a reason to give it a shot. I just laminated the first test piece in my Hand Guard Deflector project, and there are bubbles and imperfections galore. I want the parts to be perfect, and infusion is a way to get there.
I actually don't have all the proper materials needed to infuse a part, but I figured if I just do a small scale test, I may see some success with the stuff I had laying around.
To start, I grabbed a piece of 1/8th inch ply off the shelf that I had used for some past paint tests. The ply had been painted and cleared, so all I did was add a couple coats of wax and called it good. Ideally you would want to use a pain of glass to lay up a flat sheet of fiberglass or carbon. I only had a board, but will procure a nice piece of glass soon to use with future testing.
I added bagging tape to the board, and layed my front facing piece of 8oz glass onto the waxed surface. The fiberglass will act as a sandable surface so that I don't sand into the Kevlar that will be below it.
Next I cut two strips each of carbon fiber and Kevlar/carbon highbrid fabric. I arranged these into a creative pattern on top of the first piece of fiber glass.
I followed up the strips with three layers of 8oz fiberglass, added the resin inlet tube (right), vacuum tube (left), and sealed it all up with the nylon vacuum bag. Ideally, you would want to use infusion mesh, and spiral tubbing or MTI hose for your setup. More on this in posts to come.
I pulled 25"ish inches of vacuum on the composite stack, letting the vacuum run for a little while to remove moisture from the pile of composite materials. If you look closely, you'll notice the imprint of the pattern below the three layers of 8oz fabric. Notice that the resin inlet hose is clamped off.
Next, I mixed up a little batch of resin and let it degas under the heat lamps. Ideally, you would want to degas the resin in a vacuum chamber. I haven't made a vacuum champer yet, but will soon. Also, this resin isn't designed for infusion. A proper infusion resin would be thinner and have a longer open time.
After the vacuum had been tested and I was sure that there were no air leaks, I placed the end of the resin inlet hose into the resin cup, and slowly unclamped the resin hose. The resin flowed onto the composite stack and started it's saturation process.
To keep track of how quickly and efficiantly the resin front is progressing, simply trace the front with a sharpi marker, then come back and look at it a few minutes later. In this photo, you'll notice a couple of black lines with the resin front progressing beyond them. Most of the part infused nicely, suprisingly.
Twards the end of the infusion, the resin front really started to slow down. This would make me very nervous if I were working on a realy part, but not so much in this case. For this test, the resin didn't completely infuse the part. No worries because I know why. More on that soon.
In this pic, you'll see the resin trying to make it to the vacuum hose. You can do it!!
The stack mostly infused.
So thats where I'm at now. I hope to use infusion on the hand guard deflectors, but first I need to spiffy up my process. I'll need the follwoing:
Spiral tubbing or MTI hose. MTI hose is awesome BTW
Pressure pot for resin degassing and to act as a resin catch pot
Probably infusion resin, but I'll test the 635 thin resin some more
I think that's it.
When infusing composite material, you need a vacuum pump that pulls a deep vacuum. I have been reading about vacuum pumps for a long time now and I noticed a trend. Lots of folks chose cheap two stage oil rotary vain pumps from HF and the like. These pumps work OK because they do pull a lot of vacuum, but they also smoke a lot of oil in the process. These type of vacuum pumps are primarily designed to be used to pull moisture from AC lines due to the very low pressures that they are able to achieve. Thankfully, there are some good ones on the market as well. The RobinAir 15600 pump seems to be a very popular choice of infusers out there with some shops running ten or more. I have been looking for one for a while, but everything was $260 or more, so I held out. Just yesterday I found a new one on CraigsList for $200. The price seemed right to me.
With the new pump in hand, I needed to figure out how I was going to splice it into my existing vacuum system. I need to make some refinements, but I have come up with a temporary solution.
Here is the new pump next to the existing vacuum system.
I ran a hose from the new vacuum port into a barbed T fitting attached to the existing vacuum feed hose. Now both pumps can pull vacuum from the pressure tanks which pull vacuum from the vacuum bags.
Here you can see how I T'd in the new vacuum feed line. The hose on the leg of the T leads to the new pump.
With the pumps set up this way, I can pull down to 25" mercury with the piston pump...
Then switch to the oil vane pump to pull all the way down to 29.5".
These two vacuum pressures will come in very handy for a number of different operations when degasing resin and infusing.
I made a run up to Fiberglass Supply just north of me in Burlington, WA. They carry a fair selection of infusion materials, so I stocked up.
Here a few pics of a couple of their stock shelves.
Following are some of the materials that I picked up...
Spiral tubbing. This tubbing is used under the vacuum bag to act as a pipeline for vacuum pressure or resin inflow.
PE tubbing. This solid tubbing is used to pull vacuum from the bag out, or to inject resin from the outside into the vacuum bag. The stuff is rigid, and cheap because you throw a lot of it away when it's exposed to resin.
T barbs and vacuum/resin ports. T barbs are used to T off vacuum lines. The vacuum/resin ports are used to connect vacuum or resin lines with spiral vacuum or resin lines in the bag. You simply place the feed port tunnel over the spiral tube, lay the vacuum bag over this setup, and when ready, simply make a little cut over the hole on top of the feed port, slip your vacuum or resin line in, and seal with vacuum tape.
Demolding wedges. I have been meaning to pick up some demolding wedges for a long time, and finally I remembered when I saw them up at the shop. They are used to gently wedge/pry parts from molds withoug causing damage.
Enka-Flow. Enka-Flow is made to be used as a resin fast track at the primary resin delivery point. The product also allows you to place your resin feed line directly on the part without the feed hardware imprinting the cloth beneath it. More on this later...
Flat glass mold surface. I scavanged a little piece of glass from my grands house to use as a flat mold surface when performing coupon tests as they are called. Coupons are stacks of composites that are layed up flat. Usually they are created simply for testing pourposes, when you want to learn how a stack of composite materials will flow resin and infuse. The glass will come in very handy, but I'll probably need to upgrade to a larger piece in a little while. Note, this piece of glass already has vacuum tacky tape applied just inside of the blue masking tape. I am preparing for the first test infusion.
I'll probably get into setting up a "real" infusion test this weekend, so stay tuned...
When you are infusing, you want to protect your vacuum pumps from sucking in resin. I don't have the funds to buy a propper resin trap, so I made one out of a mason jar. I will use this as a resin trap, as well as a vacuum chamber to degas resin in.
To begin the setup for my test infusion, I applied the vacuum tape as a continous strip around the perimiter of the glass mold. Applying the tape this way gives less of a chance that a leak will occur.
Next, I cut all fabric from a 7"x10" template that I made from card stock. I made the template so that all test are done with pieces of fabric of the same size. After the fabric is cut out, it is layed onto the glass FG, CF, CF, FG and taped in place with masking tape.
When the fabric was in place, I layed down the peel ply fabric, green infusion mesh, resin inlet line (bottom of pic) and vacuum line (top of pic). The vacuum bag goes on after all these items are applied to the mold.
After making sure that the vacuum bag setup was holding perfect vacuum, I mixed up the resin, degassed it in the little vacuum chamber, and then injected it into the mold.
It just happens that the large mixing cups that I have sit perfectly on the lip of the mason jar, so I melted a few holes in the top of the cup in order to equalize pressure. I just pour the mixed epoxy into the cup, set the cup into the chamber, screw the lid back on and apply vacuum pressure and watch all the little bubbles come out.
With the resin degassed, the fabric seemed to infuse perfectly. Of course I won't know until at least tomorrow, but I have a good feeling about it.
No bubbles. Yay!!
Here is a little video of the resin infusing with the fabric.
When I pull the sheet tomorrow, I'll post the updates.
More fun to come soon. I think that I may have the confidence to infuse the hand guard deflectors now.
I pulled the composite stack from the glass mold and the results are fantastic. I'm super excited. No voids, no bubbles. Just shiney carbon fiber.
Now I feel that I am really getting the hang of this infusion stuff. I want to infuse Zoomzu's hand guard deflectors, but I need to figure out how I'm going to set up the mold, and I'm not really sure how that is going to work.
What I would like to do is pull vacuum along the right and left flanges, with the resin inlet running down the middle feeding resin to the parts on both sides. Remember, the mold looks like this.
I had requested a couple of infusion resin samples to test, and they came just in time, on the same day as a matter of fact.
Through a lot of research I choose the Resin Tech infusion resin, and the PTM&W in fusion resin systems. To be honest, I'm kindof rooting for PTM&W because their resin is used a lot in aerospace apps, and is supposed to be a good product, not to mention they sent me a gallon of this stuff. Thanks PTM&W guys!!
The Resin Tech is supposed to be good stuff too, with a higher Tg of around 300 degrees F, which is the temp that the resin will start to distort and become dynamic. The Tg for the PTM&W stuff is aroung 150 degrees F, which is kind of low, but we'll see how some tests pan out.
OK, back to figuring out how to infuse our deflector mold. We need to inject resin from the middle, and pull vacuum from the edges, so why not mock up an infusion test with our new resin.
To begin, I cut out two stacks of 7"x10" 2x2 8oz glass, and set them as far apart on the glass as I thought the fabric would be in the deflector mold.
Then I cut, and set the peelply in place leavin room for a resin brake to slow the resin down twards the end of the infusion.
I wanted to make sure that the resin really soaked into the fiberglass, so I cut the infusion mesh leaving about 2.5" open to the vacuum lines. The resin inlet will sit in the middle, so I placed a second strip of mesh there to make sure that the resin was able to travel out eaisly to the stacks of fiberglass.
Next I placed the inlet hose in the middle, with the vacuum hoses on the outside edges. I decided to use the 1/4" PE hose for all connections because it is so much eaiser to work with than the .5" stuff.
To connect the 1/4" vacuum lines coming out of the mold to the .5" hose coming out of the resin trap, I just stuffed them in there and sealed the deal with some vacuum tape. That vacuum tape comes in handy for all kinds of stuff.
The time had come to fire up the heat lamps and the vacuum pumps. I pulled the bag down and did a leak check. There was a BIG leak somewhere. I could even hear it, but I couldn't figure out where it was, then it dawned on me that I hadn't clamped off the resin inlet hose. After that, the leak check tested out just fine.
While I ran the leak check, I started to measure out the new resin. I used 200g resin with 44g part B hardner. The resin is nice and clear.
But the hardner is quite amber, leaving the batch looking like this after it is mixed together. One nice thing about infusion resin is that it generally has a very long open time, so you can take your time mixing, degassing, and infusing.
Speaking of degassing, this is what a the batch looks like degassing in the little vacuum chamber. You can see it bubbling and boiling in there.
This resin degassed like a dream. It didn't foam up crazy like the 635 resin did. Rather, the gas came out of it in a much more controlled way. This one aspect of the resin is very nice. I'm excited!!
Then I infused. The resin didn't travel as fast as I though it would have, but it infused beautifully. Here is the whole package just after infusion.
I got lucky with the ammount of resin I mixed. Here is what I had left at the end.
Here is a video of the infusion sped up 400%. In real time, the infusion took 14 minutes.
Here is a shot of one of the fiberglass panels that were infused in the video above. There are a few voids in the surface, but the piece is generally in very good shape, and would do just fine as a structural part. If it were carbon, one could make quick repairs to it to make it look grade A. Notice the color of the sheet. It doesn't have the deep amber color of the PT2712 resin when it is mixed up. The resin actually looses most of that color as it cures. I don't know how this works, but it's pretty cool.
With the relative success of the split stack, middle feed experiment above, I decided that it was time to do a test shot on the real deal deflector mold. I needed to make a template to cut the fabric and consumable items. I messaged a piece of aluminum foil into one side of the mold, then filled it with rice to keep it down and in place. With a sharpie marker, I drew a line around where the part edges would be.
I cut the foil template out. With the help of a couple of weight bags, I used the template to cut all the pieces of carbon fiber fabric and peelply. I cut ten pieces in all; 5 plys for each side of the mold.
The mold had already been waxed (I chose to ONLY use wax with this layup, so pleas cross your fingers for me and say a little prayer tonight), so the time had come to place the fabric. When setting up the composite stack for infusion, sometimes it's necessary to use a spray adhesive to hold things in place as you build the package. I used 3M Super 77, but there are sprays that are actually made just for this. You only want to use a very light spray.
Ready to rock and roll.
Here I have finished applying all 5 layers to both sides of the mold. This is the first time I have used a spray adhesive to hold fabric this way and I'll tell you I learned a lot. I'm pretty sure that one part out of this mold is probably going to turn out a lot better than the other.
Next I used the foil template and weight bags to trace the shape onto the peelply fabric. I cut the fabric with scissors (this fabric is really tough to cut with the rotary cutter for some reason), and place it into the mold using tape to hold it into place. You could also use a little Super 77 to hold it in place if you wanted.
Then I added the infusion mesh and vacuum and resin plumbing. Notice I stopped the infusion mesh about 2" before the vacuum lines. These areas will act as our resin breaks to slow that resin front down to a nice crawal. This is done to ensure that all the fabric gets fully soaked in resin, however it doesn't always work. Thats why you experiment.
The flanges on this mold are fairly small, with room only for the vacuum lines. Because of this, I had to make a vacuum bag, and tuck the mold package in it nice a tidy.
I sealed the vacuum and resin inlet lines to the bag, and pulled down the vacuum. The pressure test wasn't perfect, loosing 1.5" after 15 minutes.
Based on my previous test, I decided to mix 366g or resin for this infusion, but it ended up being too much. This is good or bad. Either I'll be able to save resin on future layups, or the composite stack didn't soak up enough resin. I hope it soaked up enough resin.
Anyhow, here's the resin and hardner in the cup before mixing.
I mixed it up, degassed, and infused.
Here is a quick clip of the resin degassing, and there I am folks, the man with no plan.
The infused package.
And of course, a video of the infusion.
Setting up an infusion in a real mold is much more difficult then on a piece of glass, but I learned a lot and I'll put that knowledge to use the next time around.
therivermonster screwed with this post 02-07-2013 at 07:03 PM