need some infusion set-up help

General Composites Discussions Composites Talk
#1

ok, so my school team and I tried our first infusion part today and we had quite a few issues. I know threads like this are kind of useless without pictures, but i’ll try to explain the best i can and I’ll have pictures later, they are on my team mates camera.

So we’re just trying to make a flat plate. We had a piece of (waxed)glass for our base with two plies of uni-directional carbon on top of that (0-90 orientation). On top of that we had our peal ply, followed by the breather fabric, and all topped with some painters plastic for the bagging material. We are getting some actual bagging material from our sponsor, but for our test we figured it would work for now…we may have been mistaken. The bagging material was held to the glass with some gray putty tape. All was hooked up and pulling 26"-27" of Hg. My pump does have a minor leak so it cannot be shut off and have the system hold vacuum. I’m buying a check valve for the exhaust so that should fix it, but I don’t think that was our problem.

Any ways, we started the infusion using this resin (pt2712) with the part B hardener. the resin was measured by weight using a precision scale, but it did not cure. It’s suppose to have a 70 min pot life, but after about 90 min, it was the same consistency as when we first mixed it. The resin is left over and has been sitting on a shelf for about a year. I don’t know a whole lot more than that. So can epoxy “go bad” from sitting around or are there any other reasons that could cause resin to not cure? FYI, It was also being done at room temperature.

Back to the infusion itself. it started well but after about 2" into the infusion it just slowed to a crawl, and took maybe 30 min to travel another 6". So being impatient I decided to try and “help” the resin along which resulted in a hole in the plastic. I tapped up the bag and was still pulling 26" Hg, but resin started pooling around the inlet tube and was not infusing anymore. Does anyone have any idea how I could be pulling full vacuum but not getting infusion anymore and why the infusion slowed down so much?

I’ll try and get pics of the system soon, but if you need anymore specs on our set-up i’ll be happy to describe anything. Thanks for the help guys!

#2

  1. Painters plastic is not made to be free of pinholes, and is not treated kindly during transport and on the shelf at the hardware store. I understand you guys wanted to have a go at it, but wait for the right stuff so as to not waste time and materials.

  2. If there’s any noticable leak, don’t infuse. Air will be pulled in as it flows much easier than even the thinnest resin.

  3. Flow media! Breather is for wet layups to allow air to escape and absorb resin. This is highly detrimental to the infusion process. A flow media allows resin to travel over the laminate and saturate down into it, without taking up a significant amount of resin.

  4. If I recall correctly, that epoxy has a 70 minute pot life. At 90 minutes it just gelling, and in the layup it takes a whole lot longer due to the decrease mass and lower exotherm. Wait 24 hours and it will be cured. 48 hours and it’s significantly stiffer. Cure at 150*F for 4 hours and it achieves max strength.

Pics are critical for these kinds of things. Need to see the whole setup.

#3

thanks for the quick reply. I have a few follow up questions/comments.

  1. yeah, I started to find that out that there were little holes everywhere as the infusion went on. We ultimately ended up covering the whole surface with duct tape. Not ideal I know, but it should have covered most of the holes. I would also agree to wait for the proper materials, but our teacher told us to start anyways. At least we have a better understanding of what NOT to do next time.
    2.the only leak that was there was the leak from the exhaust on the pump. When I shut off the pump, air leaks into the exhaust and into to system. this is the pump we are using. I’ve ordered a check valve for the exhaust to deal with this.
  2. Yup. newb mistake. I’ve done several vacuum wet layups and didn’t even think twice about using it for infusion. I have read that a layer of fiberglass can be used as a flow media, is there any validity to this?
  3. You are correct, so we may have tossed in the hat a little early then. Good to know, thanks. Do you happen to know of any resins that cure a little quicker? Ultimately this will be used in production and a quicker curing time would be ideal. Also, at what point can the vacuum be removed? is there a “rule of thumb” that works with the pot life or anything?

Thanks again. I’m very new at this and have a lot to learn yet.

#4

This sounds about the same as my first infusion.

1- if this is for a class, it’s probably good to get started early, failures and all. If budget/material is an issue, I’d practice with fiberglass.

2- If you had a leak in the pump, you couldn’t check for any leaks elsewhere. It is nearly impossible to tell with the pump running. The check valve should help so you can do a leak-down test.

3- Fiberglass doesn’t really work as a flow media. It flows faster than peel ply, but not nearly the same effectiveness as flow media. Plus, it might be harder to de-mold.

4- It looks like there is a hardener for that resin with a gel time of 40 min (part B3?). Would you mind sharing where you bought the resin? I’ve been wanting to try that for infusion.

I’ve tried to help the resin along, too. It has never worked. The flow media should prevent the resin just stopping, though the resin may have “racetracked” around the side and sealed off the vacuum port from the infusion front. This is where pictures would help troubleshoot what’s going on. Good luck on your second attempt with proper materials.

#5

yeah, and it was a good learning experience if nothing else.

we checked for leaks by manually plugging the exhaust and watching for a loss in vacuum. there was no noticeable loss in vacuum over a span of 3-4 minuets until I let off the exhaust port at which time the vacuum was slowly lost.

thanks for the heads up. I’ll see about getting some actual flow media from our sponsor.

I wish I could tell you. Our sponsor bought it directly from one of their material suppliers. I couldn’t tell you any more than that.

I don’t think the resin went around the edges, otherwise we should have seen epoxy being sucked into the vacuum line leading to the catch pot. I’ll work on those pictures asap.

#6

for leak check, just clamp off the vacuum line somewhere, with a gauge between the bag and the pump.

resin bucket----clamp----PART----gauge----clamp-----vacuum pump

#7

Having once half filled a digital gauge with epoxy, I have found it best to set up:

gauge/(resin bucket)----(clamp)----PART----clamp-----vacuum pump

Where you attach the gauge to the inlet line in place of the bucket, this way you can’t leave it attached and risk filling the gauge as you must remove it to attach the bucket.

#8

How would one monitor vacuum during the infusion and cure with that setup? Seems difficult.
We put the gauge between the bag and the vacuum source. Unless the catchpot fills up, no way resin gets to the gauge.

#9

Why do you need to? We use the gauge to see if the bag is good and that the air has been removed, once it hits 99% or better start the infusion. I don’t see the point of monitoring pressure during the infusion as it will change the same every time and once you shut off the inlet pressure it will return. And if you have any leaks you will see a stream of bubbles and they can be easily tracked back to the source.

But then I do only use epoxy which you keep vacuum on during cure, so if you are using an ester and want to reduce vacuum then TET and Riff42’s setup will be the go.

#10

well i’m still working on getting pictures. If my team mate has not emailed them to me by the morning I will just go and take pics for myself. currently we have out set-up:

resin bucket----clamp----PART----clamp----gauge/resin trap-----vacuum pump

I have the clamp before the resin trap because I have come to the determination that the pump is not the only thing leaking. I’m beginning to suspect my throttle valve or the fittings I used. They are all vacuum rated but something is still leaking that is not the bag. I know this because
we did another test today using “food saver” bagging as our bagging material. It actually worked quite well, and should be good for testing until we get our actual bagging supplies in (which are on order). Any ways, we pulled full vacuum (26"Hg) before infusing and clamped off the inlet and outlets. We let it sit for probably 15-20 min without the pump on and there was little to no visual loss of vacuum. So this weekend I’ll be hunting down the leak piece by piece.

Back to the test we did today. We did two test strips in the same infusion. One was two layers of unidirectional with a layer of fiberglass in between, and the other was two layers of plain weave with a layer of fiberglass in between. The fiberglass did speed up the infusion a bit, but it also could have been the more robust bagging material that didn’t have holes everywhere. We are still getting large pools of resin collecting around the resin inlet. Since I’ve never done this before I don’t know if this is normal or not, but I’m guessing it’s probably not. We also had a weird issue were the unidirectional infused much quicker and was very resin rich. I can see it infusing faster because of the difference in materials, but why would just that one sample be rich while the other was not?

Pics will be here tomorrow…one way or another…
thanks for the help so far guys. I appreciate it.

#11

“I tried our first infusion part today and we had quite a few issues.”

LOL… yep infusion will either drive you nuts or bring out the perfectionist in you.

#12

If epoxy isn’t oven cured ( not all epoxies are meant for the oven) it takes about 7 days to reach total cure, even though it may seem cured after 24 hours.

#13

If epoxy isn’t oven cured ( not all epoxies are meant for the oven) it takes about 7 days to reach total cure, even though it may seem cured after 24 hours. As for how old your epoxy is… 1 year it’s probably not very good for infusions now.

You guys need flow medium, such as Enkafusion or Airtech Greenflow or Redflow netting. This gives the resin a path to flow thru. Cover at least 95% of the lay up, on top of the peel ply with this netting.

Also if you can get some, buy the yellow AT200y sealant tape. Not critical but it works much better than the grey or black tape does.

Also if your part is critical, double tape the vacuum bag to the mold… two lines of tape all the way around.

Peel ply… if you want a rough surface to be able to post bond something to the part or the part to something else… you will need to have peel ply. I tried peel ply B from Airtech and it worked awesome for infusing parts, but other brands should work ok… i think, now this is my opinion… that the more space there is between the threads of the peel ply the better your infusion will work. That’s why i say plyB type.

Make sure the mixed viscosity of your resin is 500 cps or less. Thicker resins won’t flow well and can end up waisting time and materials. Also degass your epoxy resin before infusing… mix epoxy for 3 minutes scraping the sides of the mixing cup and the bottom too. Microwave a container of water for 1 1/2 minutes on high setting… drop 4 small objects into the water to support the resin cup off the bottom of the water container but make sure there is enough water to meet the level of the epoxy in the mixing container. So you are setting one container ( plastic ) into another container of hot water. Leave it in the water until all the bubbles in the epoxy are gone or at least 85% gone. While the epoxy is degassing drop in your resin inlet line and tape it or fasten it to the side of the epoxy container … drop the line all the way to the bottom then lift it up 1/8" so as not to block resin flow. I usually cut my resin line on an angle so it won’t block the flow inside the resin cup. Mix at least 3 ounces more epoxy than your laminate requires. You do not want to suck air into the infusion. Use a spring clamp or vise grip style flange clamp onto your resin inlet line until you are ready to let the resin flow. Once your laminate is 95 to 99% complete - clamp off the resin inlet line. Let the vacuum pump run for 12 to 24 hours. Use a resin trap if you don’t want your vacuum pump to suck up resin. I also wrap a small amount of breather cloth and tape it onto the end of the vacuum line that is inside the mold or cover the spiral wrap with one layer of peel ply if using spiral wrap. That will help keep resin out of the vacuum line going to your pump.

Also make sure your resin feed line is sitting on top the laminate so there is no space between the hose line or the laminate. Extend your laminate onto the flange area of the mold to obtain that. Or there are resin flow products for distrubuting the resin onto the laminate without harming the laminate IF you need to have the resin dump onto the part surface. This is so you don’t end up with an impression on your final part and sometimes it just helps the resin flow distrubute evenly.

Also with infusion debulking is necessary. Otherwise there will be voids/air pockets in the laminate. Take your time setting it all up, make sure you have every detail covered.

#14

I don’t get why people go to the effort of degassing and debulking and then put a feed into the resin from above. Unless you pre-fill the line there will be a big slug of air trapped in the pipe between the resin and the clamp and as soon as you remove the clamp it will shoot into the part.

However if you use a funnel or funket (a funnel bucket with a hose fitting on the bottom in a cnr) and clamp it just below the outlet of the funnel the air in the line will float up and there will be no air introduced when you open the lines. As an added bonus if you don’t pay attention and the bucket runs out of resin, quickly clamp the line near the laminate and add more resin, lift the lines up to let the air float up again and restart with no harm done. As an extra added bonus there is less waste as you don’t need to leave 1/4" of resin to cover the inlet.

When done let the funnel drain and reuse it next time.

#15

ok guys, here are the pictures. I unfortunately didn’t get to the set-up before my team mate tore it all apart, but maybe it’ll be helpful. The tube on the left hand of the catch can goes to the bag. I’ll try and get pictures of the actual bag set-up. tubing is 1/4" nylon, fittings are vacuum rated, as is the gate valve that is teed into the system right underneath the gauge.

#16

little more info:
the catch can is 3" PVC tubing with machined Al end caps that are epoxied in place. Fittings are loctited in using blue loctite. Fittings are all rated to 29"Hg as is the gate valve. The gauge is my most likely culprit for the leak as it’s a cheap gauge. I pulled full vacuum (25.5"Hg now that the check valve is installed on the pump), and crimped both lines into the catch can, but it’s still leaking a tiny amount. I tried putting that gray tacky tape on all the fittings but that didn’t seem to help any so I scraped most of it off.

any constructive criticism is always welcome.

Also… sorry for the large pictures. trying to get as much detail in as I can for you guys.

#17

yeah, we have some Enkafusion on order along with the actual bagging material.

We have no way to test the viscosity of the resin, but we could degas it easy enough. since we have no microwave or water source in the lab we are using could we possibly just use the vacuum from the pump? I mean making a lid for the mixing container and just applying vacuum for a few minutes to draw all the air bubbles out?

Just for my clarity debulking is done while the resin is infusing? And the process is simply to pull full vacuum, vent, pull vacuum again, vent…etc?

Moke, that seems like a good idea. I’ll have to look into that.

#18

Degassing the resin helps draw out the dissolved air that is in the resin from transportation and mixing, along with some chemical reactions. When you pull the resin from ATM to a vacuum, it will expand the bubbles. Depending on your part and infusion setup, those bubbles can stay in the part because it becomes trapped in the fibers, and stay expanded due to the vacuum pressure in the bag.

Also, as for debulking, it is NOT necessary, but when cyclic debulking (pull vac for 60seconds, release to 10" for 30 seconds, pull vacuum again, etc etc) is done on dry fibers, it can nest the preform, and you will get better fiber volume, however, your permeability will do down, due to the lack of voids where resin flows into the part.

As for the resin and air in the lines, it doesn’t matter if there is air in the lines when you open the resin clamp, because the vacuum will suck all that out right away…air flow through the part and distro media will not be slowed down…all the air in the resin lines that are injected into the part will be sucked out MUCH faster than the resin flowfront. Keeping the resin BELOW the part helps because you are trying to pull a vacuum, and having the gravity work with you helps. Else you are letting resin pour into the part, and the vacuum isn’t really doing anything. The point of the vacuum is to let the resin suck into all of the voids in the fibers, not flood the bag. However, technique being what it is in composites, you can start the resin above the part to help it flow in faster from the start. Great if you have a short gel time…but make sure that resin goes below the part before you clamp so you can limit the resin intake!

#19

Yeah, you can use the same pump. Maybe if you want, add the vacuum chamber (bell jar is normal) on another line before the pump, with some clamps you can degas at will.

Debulking really means to pull vacuum to get all the air out. Normally done with prepregs, where you have tacky resin filled fabric. Debulk every few layers, to make sure there are no trapped air bubbles between the layers, that otherwise can’t escape due to a thick layup.
In infusion, some people cyclic debulk (vac, release, vac, release, etc) to nest the fabric into each other. This will lead to more fiber volume in the part. However, because those spaces are gone, the resin will take alot longer to flow in the part…it has to rely on how the resin flows in the fiber tows, not all the little spaces and channels between the tows.

#20

so your saying to lay up 2-3 layers, bedulk, lay up another 2-3 layers, debulk, etc…?

our last effort resulted in fairly decent parts, but we still have lots of air bubles at the surface of the laminate. however, we did not degas the resin before hand. we’re giving it another go this friday, and I’ll hopefully be playing around with it this weekend.

i appreciate the advise guys, keep it coming!