epoxy resin degassing 1

[CarbonMike]

Hi,

I have a problem with vacuum degassing of epoxy resin (300-500mPas viscosity) before resin infusion. After mixing I put cup into vacuum chamber and make 28,8" of vacuum (I'm in 1000feed above sea). I can see that the bubbles are forming and rising. Lot of them will pop, but I still have lot of them in epoxy after 10-15min degassing. Is the vacuum level low for perfect degassing? Microbubbles in epoxy cause my problems with dry spots in the carbon weave.
I don't have any leaks during infusion, holding 28,8" vacuum for a day with pump shut down.

Thanks for your help.

Mike

 

[Compositepro]

Resin degassing problems may not be the main issue causing your dry spots. If you are only drawing 28.8" you still have residual air in the mold during infusion. Where it ends-up will be determined by your preform and flow pattern and by your venting arrangement.

As to your question - a few drops of solvent on the surface of your resin can act as a degassing aid. Air bubbles can be very stable and not pop. A solvent like acetone or heptane will boil under vacuum and will expand air bubbles so that they do pop.

Without a gage that reads absolute pressure rather than gage pressure it is difficult to know what is really happening when you try to degas resins. I use a mercury manometer but these are not allowed at most companies these days. There are alternative gages which are more expensive, however.

 

[CarbonMike]

Nice to ,,read,, you again CP

My infusion setup is very simple. I stopped trying to infuse usable parts, instead I came back to small rectangle (150mm x 80mm) test pieces on piece of glass so I can control and change some parameters and compare results. I use Frekote 770 NC as release, few layers of 160g/m2 carbon twill fabric, 80g/m2 peelply, then Enkafusion flow media (thinnest one), simple inlet outlet tubes. All bagged with sealant tape and high quality vac. bag. I have no leaks, I can hold same vacuum for a day without any drop on gauge. I use only epoxy. After the epoxy is sucked into the inlet tube I can see that bubbles are expanding and are visible. I made few test with just epoxy sucked into the tube and clamped. As a result, epoxy in tube is not clear but there are lot of bubbles in it. Maybe this is not the only problem, but I think its biggest.

BTW its a long time you haven't look at werksberg forum

Mike

 

[Compositepro]

I hope you are clamping-off the vacuum side of the tube and not the feed side. The epoxy must be under pressure (albeit only ambient pressure) and not under vacuum when it cures.

I don't understand your reference to werksberg.

 

[CarbonMike]

I'm clamping feed side of the tube while resin is in the tube. Clamping off vacuum side will not have any effect on air bubble expanding. Resin will be under normal outer atmospheric pressure and there will be no visible air bubbles. Like it is in resin cup. But I get your point. Thanks.
But if Im right, after vacuum infusion of the composite part, when the feed line is clamped, so no more resin can flow in, the resin will be back under vacuum right?


Sorry I thought that you were someone else... you have same nickname. CompositePro was one composite engineer. I was discussing with him about carbon bicycle frames long ago. Sorry.

 

[Compositepro]

The standard procedure for resin in fusion is to pull vacuum until the vent line fills with bubble-free resin and the clamp the vent line.

If you want to maintain some compaction pressure on a laminate after it is infused, you you can clamp the feed line and regulate the vacuum to something like 10". You never cure resins under full vacuum. It can only cause void problems. That is not to say that it always will cause voids, but it does not help.

 

[CarbonMike]

Sorry but I have never ever heard/read about clamping vent line and not feed line. Its not a standard procedure and it will never be. You will NEVER have a good fiber/resin ratio with your ,,standard procedure,,. After resin flows in the layup it is in equilibrium with outer atmosphere so there is virtually no compaction pressure on the layup in areas filled with resin. After about 70% of the part is infused with resin, feed line is clamped and layup will be drained so whole part is infused and also there will be rising of compaction pressure from outer atmosphere back to value of vacuum level on gauge (if resin is still able to flow). Thats how it is working.

 

[Compositepro]

In RTM when a match mold is used, it is always the vent line that is clamped and often additional pressure is is applied to the feed port. When using a vacuum bag, the feed can be clamped or a partial vacuum can be pulled on the resin reservoir to remove excess resin from the laminate around the feed port. But it is very ill-advised to pull a high vacuum on resin while it is curing. Almost anything will cause voids in the laminate under those conditions. The factors that will allow you to get away with pulling full vacuum is that the resin is close to gelling so it cannot bleed out the vacuum port and the vacuum is irrelevant or the vacuum pump is not capable of pulling a high vacuum so it does not need to be regulated to a lower level.

The applicable principle is that pressure makes bubbles smaller, and vacuum (or lack of pressure) makes bubbles grow larger and creates voids. In VARTM it is important to get the right balance between using vacuum to create a compaction pressure on the fiber but still maintaining pressure on the resin so that it does not boil or cause small bubbles to grow into large voids.

 

[Demon3]

I'm no expert in this area but I did do some work years ago and was always taught that one applies pressure at the gel point of the resin. Apply earlier than that and you squeeze out the resin (obviously not a problem if you're in a sealed bag but it is with loose prepreg). That's why there's so much work trying to detect the gel point (i.e. using capacitance, IR or chemiluminescenc).

Chris DeArmitt

http://www.phantomplastics.com

Consulting to the plastics industry

 

[CarbonMike]

I know that in RTM after injection a pressure is used to reduce voids. But there are constrains from both sides of tooling, so it don't have any influence on fiber/resin ratio. I get your point. You are absolutely right about void size/vacuum level. It will be really better for void size to reduce vacuum level on exit. But I will lost compaction pressure. I will make some experiments with it. I saw lot of parts made via infusion and cured at full vacuum (28"+) and visually void free. Hardly depends on resin of course.
Read the Boeing CAPRI patent. Really beautiful idea.

Thanks for your ideas.

Mike

 

[allcomposites]

Hi Mike, here few comments on this issue.

1) During degassing, low molecular weight components on the resin are expanding due to a phase change (from liquid to gas). If still foam after degassing means that suface tension of the resin is too high (trapping heaviest components). To improve degassing, you can reduce surface tension of the resin either by increasing temperature or by additives, such as BYK 051-054 (according to curing agent).

2) During infusion, voids are trapped in the reinforcement due to improper impregnation. If resin viscosity is too low and vacuum too high, a fast resin flow happens through open pores between fibers. Tows are then impregnated radially leading to trapped voids. To improve impregnation you can reduce vacuum level till observing a clear flow. We actually optimize vacuum level according to resin viscosity and architecture of fibers.

3) During cure, few components on the resin volatize (usually less than 1% in weight). This leads to an important content of voids if not properlly achieved (1% in weight can be as high as 10% in volume of gas). To avoid this, you need some compaction pressure on the resin during cure (ideally higher than vapour pressure of volatiles). In resin infusion, you should avoid applying vacuum during cure (clamp both vent and entry ports). Note that infusion resins are improved to minimize this effect. You can easily identify this kind of void since gas will flow to the surface of the laminate and collapse forming big voids almost the size of the tows (you dont see much voids within the laminate).

 

[allcomposites]

You can improve Vf by cycling compaction before infusion. The laminate has an hysteresis behaviour due to fibers nesting and deformation. Try to run several compaction/decompaction cycles before infusion and you will see the improvement in Vf and thickness.

 

[CarbonMike]

thanks for excellent reply allcomposites.

I made a special test on glass, so Im able to see tool side of composite. There are no visible voids, pockets or anything during infusion. Whole piece is infused beautifully. Problems occur after clamping inlet port. After a while small voids are made in tow intersections in twill weave. Second test, layup was infused at 29" vacuum, when the feed port was clamped, I reduced vacuum to half. Drastic difference in surface void count and size. Not sure about resin/fiber ratio change. Problem is, that after infusion, area around feed port is really resin rich. With full vacuum this resin rich area was drained to normal in few minutes after feed port was closed.

I think I use insufficient vacuum for vacuum degassing. Will make more experiments this way.

I make cyclic compaction before infusion. I read study about this effect in Boeing patent.

Thanks for your help!

 

[allcomposites]

It looks like you are draining the resin out of the part. When you apply vacuum after infusion, the vent port is at vacuum level while the inlet (clamped) is close to atmospheric pressure. Both pressures equilibrate after a while, when resin no longer flows. So, if you still apply vacuum, it will take some time to reduce thickness near the feed area (while you are draining the resin near the vent).

You can either reduce vacuum level to avoid draining the resin or just clamp the vent after infusion and wait for consolidation. In the last case you will need a longer gel time to allow thickness compensations all around the part.

In Boeings patent they suggest that recycling resin in a close loop (from vent to inlet) will reduce thickness variations. We did some experiments but didn't work that way.

Another way to go is to control resin flow rate at the end of infusion by using a 2-way plastic valve. Before resin arrives to vent, you start closing the valve (so reducing resin flow rate). This will increase vacuum level near inlet draining excess resin to the vent. This works nicely but requires some expertise.

 

[CarbonMike]

Thanks for great help...