4 mm carbon parts infusion problem

General Composites Discussions Composites Talk
#1

hi
I recently tried to infuse some 4 mm thick all carbon parts. I did a small flat panel as a test about 150x 150 mm. Everything went well and the part came out flawless. Next I had to do a 300x400 mm panel. I started the infusion and it seemed all the part reached wet out but when it came out of the vacuum the epoxy did not reach some of the bottom layers from about half way of the panel. I also did some parts in a mold with similar results:mad:. After this epic fail I hit the forums again and found 3 possible problems with my process:

  1. My epoxy is not good for the infusion. although I got a low viscosity resin it’s work time is only 20 minutes. The infusion for the panel took probably 30 min or more.
  2. The thickness of my composite has to be 4 mm. Someone suggested in a thread that carbon can be infused only up to 2-3 mm. This can’t be true…I’ve seen thicker carbon panels that were infused.
  3. Also read the mesh could be a problem…there are faster an slower mashes. I went cheap and used mosquito mash but bought peel ply. However when I let the resin in, the infusion line seemed to go quite fast for the first quarter of the panel.
    I added some pictures to showcase the process and the results. Hope I get a good lesson from you guys and find out what i have done wrong. Also if someone from Europe can suggest where to get some good resin for the job it will be welcomed. Thanks!
#2

My opinion and I think it’s obvious is you have to change your resin, and with a higher pot life time. 2nd; the first picture you posted is something… Beautiful

Sent from my GT-I9100 using Tapatalk 2

#3

Resin temperatures can change dramatically over large parts as well so keep that in mind. My work room is usually 68-70F. The resin in my pot is usually 75 F, and for the first few inches the laminate stack is that temperature as well. The slow down you’re experiencing could be because your wide open part is significantly cooling down your resin. by 8 inches in my resin has dropped to ambient RT (68-70F) and it thickens up quite a bit, causing it to slow down significantly (I can get 5cm/s at first and I’ve seen my front slow down to literally .5cm/s by the end of the part, even with a mesh.

Do you know the CPS of your resin at varying temperatures?

#4

Arriverderre,

Speed is directly and lineair dependant on viscosity. Viscosity twice as high, speed reduced in half. However, when doing an infusion the speed is not only dependant on viscosity, but also permeability, and geometry. The further you go, the higher the resistance from the laminate stack. Hence the slowdown.

Question:
In the first picture, did the resin come from left or right? (as you are telling the bottom layers did not infuse)

Small hints for the next test:
-lower viscosity epoxy (for instance Sicomin SR1710, or Resoltech 1800)
-perhaps a change in laminate. Do you need all carbon, or can you change from 3K woven to 6K or 12K, or multiaxials
-high vacuum (5 mbar or less)
-degassed resin
-MTI hose to reduce porosity

This definately should improve quality.

#5

looks like it race tracked to the outlet… you can see a nice even front, then its dry, and then its wet again at the outlet… It looks like its race tracked around the areas with tight curvature in your mold. This is predictable given tight radii molds and its difficult to remove all the bridging in these areas.

Consider changing your strategy so the areas of tight curvature are used as resin feeds. In other words, feed around the perimeter and pull resin to a central vacuum port in the center of the flat surface.

This doesnt explain why you had problems in the flat panel test tho. Most likely fibre lockout which happens to thick carbon layups. Need to infuse very slowly, long open time resin, very high vacuum, very low viscosity resin. You could also consider intra-laminate (stays in the laminate permanently) flow media such as CFM - continuous filament mat - in the center of the layup to provide an increased permeable layer help with the fibre lockout.

#6

thank you craj1031tx. I did not know temperature can affect the viscosity of the resin so much. In fact now I remember opening the window because the resin container sayes not to breath in the vapours. So with the winter outside the temperature drop was significant indeed.
Don’t know what CPS stands for…

#7

herman thanks for the hints.
The flat panel in the first picture was infused on glass so the mirror finish shown was on the bottom. Resin came from right to left in the picture. The mesh, peel ply and some of the layers got wet i could clearly see that in the bag (see last picture).
Yes I need it to be all carbon, thinking of buying multiaxials for my next try. Will this improve the flow?
The vacuum was under 5 mbar. I’ve read an article about double bagging infusion that said: “The inner bag pressure is set for the optimum resin flow, given the materials and infusion setup. Vacuum pressure in the outer bag is defined to achieve desired fiber volume and is applied only after the laminate is completely infused.” So lower vacuum for infusion and higher for compaction.

groper your answer is much appreciated. Feeding around the perimeter and pulling in the center was on my mind ever since i finished bagging (pic 3). But I continued like that and failed my parts. I know my mold has problems and will work on that too!
fiber lockout is new concept for me…will look into that and CFM media also.
How will I infuse slower with even lower viscosity resin?

#8

For me case is simple - you use fast hardener and want work longer than it last liquid…that doesnt make sens
Bear in mind that provided pot life of 20min is for spilled resin and resin in vacum cures much more faster.

Steps to do:
1.Use spiral tubing 10mm at the inlet and outlet side
2.Use longer lasting low viscosity resin like 200-300mpas
3.You can heat the resin before infusing to lower its viscosity

good luck

ps. i infused 8 milimeter cf

#9

herman is right, it looks like a vacuum and resin problem. Resin cures to fast and your vacuum seems not to be high enough.
And it is not a thickness problem, I have infused laminates up to 30mm.

#10

So far everyone agrees to the resin being the major problem. I found the technical data to the Epoxy I used in italian. Translator said:
Viscosity of the mixture at 25 ° C CPS 1200-1300
Density at 20 ° C gr/cm3 1,06
Ratio base / hardener 100/50
Pot-life mass gr. 200 at 25 ° C min. 20
Apparent hardening at 25 ° C Hours 12
Total hardening at 25 ° C hours 72
Hardening apparent at 50 ° C Hours 3
Total hardening at 80 ° C with post
cooking Hours 5

This clearly tells me: read all the technical data to all the products next time you moron! Anyways this is the only resin I could find in my country that doesn’t sell by the barrel so was tight on choices.
Herman suggested: Sicomin SR1710 or Resoltech 1800. Much longer pot life here!
The pump goes down to 0.05 mbar ultimate vacuum. This should be enough, right?

#11

I am planning an infusion of a structural carbon spar, maximum thickness is 13mm. i know infusing thick UD carbon laminates can be difficult. What are your recommendations for approaching an infusion like this to avoid fiber lockout? Any considerations or advice would be appreciated.

#12

Do you have everywhere the same thickness? Is it all UD? Which resin viscosity do you have and which resin do you use. How is the geometry?
There are a lot of things you have to take care.

#13

The geometry is an oval section tube. The wall thickness reduces from 13mm at one end, down to 4mm at the other end. The laminate is 70% UD at 0degrees, 15% +45 deg and 15% -45 deg. The resin I will use can be any type of epoxy, what would you recommend? Most resins are available here…

Infusion in a 2 piece split mold. Hoping to do it one shot with tube bag inside the center and envelope bag around the mold, but may have to do it in 2 halves then bond together if internal webs are required- don’t know yet…

What about the use of intralaminate flow enhances such as Continuous filament mat, are these used in these types of thick laminates?

#14

I would not use a integrated flow media. Start infusion from the thinner side to the thicker side. Make a brake zone of about 40mm, I mean let the flow media end 40mm before the laminate ends.
Resin I would use Momentive Rim145, other resins I know are not available for smaller companies. Sika also has a very thin resin for UD carbon laminates with about 30mpas.
If you work with MTI hose place it direct to the laminate, even when infusing thick UD laminates it is important to have a good and direkt to the laminate working vacuum.

#15

I do not have temperature controlled mold capabilities, so the rim145 cannot be used. I can heat the preform after infusion completed but I do not have complete or ideal control over the temperature.

So what physical properties of the resin do I need, what mpa.s range of the mixed resin should I be looking for? How fast flow front speed is ideal for 13mm thickness laminate?

#16

If you have about 25 Celsius that is fine for the Rim 145 if you can heat up after infusion. It is not only the viscosity that makes the descission. Some resins with higher viscosity work better on carbon UD than some with lower viscosities. That also depends on the polarity of the Epoxy.
Most resins that are made for UD resin infusion are only sold in big amounts ( for example 50 tons) for the wind industry.
How long is your part? More important than flow speed is Vac level and vacuum integrity. And you should let the flow media end about 4times part thickness before the part ends.

#17

Part is 13m length, and i have 2 of them to build. Total resin consumption is about 50kg per piece, so 100kg total. I have 2 stage vac pump, it will pull down to 25 microns if i could get 100% integrity…

I noticed momentive also have the RIM235 -also for wind turbine and boat building manufacture. They specify room temperature working conditions with post cure. Do you know about this resin suitability or its availability?

Id like to hear more about the polarity of the epoxy - do you have a link to reasearch this information if you do not have time to explain?

Thankyou…

#18

I can not explain you in details what it means. I know from tests of membranes that dome resins that have a lower viscosity got through the membranes and some resins that were like water did not pass the membrane. Thats what I can tell you from the research of membranes when we started to produce a membrane that will perfom best for the MTI hose.
I talked a lot to epoxy people. I have not worked with the Rim 235, just with the Rim 135 wich should also work. We have a lot MTI customers in the wind industry that infuse beams with up to 40mm UD carbon. They use a stitched UD so that the resin can travel easier through the UD.
At a 13m long part you can not infuse aceoss the long side. What are you building? If I know that I can give you a infusion strategie.

#19

Building freestanding rotating wing masts for a catamaran (no stay wires), a pic for you;

I have done many infusions of thin laminates, but never attempted a very thick carbon laminate as i have been told by others in the industry that these types of infusions have given them problems with fiber lockout. I know it must be possible to get good results as the wind industry does it all the time - just need to know the key reasons of what causes the problem and how to avoid it.

#20

Contact me via PN.