Help Planning Mold/Part - Sunroof delete panel

General Composites Discussions Composites Talk
#101

I’m not sure that the material choice is all that important. The main issue will the the thickness increase.

The downside to the UD is that it can be a little more difficult to infuse properly.

If your test laminates were getting a thickness of .048" and the areal weight of the carbon is actually 5.9 oz/yard then you were getting a Vf of 46%. That would be 57% fiber by weight (assuming PAN based carbon). At that same ratio the thickness increase of the added UD carbon would be .064". I think that you will be able to get a Vf of 55%, or better, with the UD which would yield a thickness increase of .054…maybe a little less. 55% Vf would be 66% by weight.

THE ABOVE CALCULATIONS ARE WRONG DUE TO A MATH ERROR ON MY PART. See post #108 for my corrections.

#102

Thanks for the help. I was just assuming thicknesses by what my supplier specifies for the material (Advanced Vehicle Technology). I haven’t actually measured the part thickness. I suppose that would be a good thing to do.

I thought unidirectional was supposed to be easier to infuse, based upon reading on here and otherwise. What issues might arise if I attempt to infuse the uni?

#103

First off, I’m far from an infusion master. I found that the fibers end up being more compact making it more difficult to get full saturation. It looks pretty on the outside but has voids and dry fibers within. This is just my experience.

If you mic the actual part thickness I can recalculate your fiber volume fractions.

#104

Bond the doubler after infusion if u use uni. They also make twill tapes.

#105

I measured my cured laminate. I’m getting 0.038" with 4 layers of the 2x2 twill 5.9 oz/yd fabric. Can you walk me through the calculation of how you determined the Vf just from those parameters.

#106

.038 is very thin… Especially with 4 layers.

#107

I’ll start a new thread specifically on mathematically calculating THEORECTICAL fiber:resin ratios by weight, volume, and thickness.

See post #108 for corrections. I made some math errors.

#108

Hey, I just realized that I made a bunch of math errors in my previous posts. I’ve been doing a bunch of work with Aramid lately and just stupidly used the density of kevlar in my calculations.

.038" for 4 layers of 3K carbon would be a Vf of around 45% (56% by weight). I infused a part yesterday with the same layup and also got .038".

The added UD carbon at Vf 45% would get you a extra thickness of around .068". You can usually do better with UD so might end up being closer to .060"

Sorry about the mislead…

#109

Thanks for the help everyone, this forum has been a great source of information for me.

I suppose I’ll scratch the idea of using uni, and just use additional layers of 0/90 in that area. Will I encounter any issues attempting to infuse up to 8-10 layers of 5.9 oz 2x2 twill? The most I’ve done to date is 4 layers. I’m thinking of doing a small test piece, but what are your thoughts?

#110

I did a test piece to check out infusing additional layers of fabric. I did 14 layers. I also tested using 3M Super77 to help hold the surface layer where I wanted it. I applied a very light mist to the mold surface (a waxed mirror). The fabric adhered nicely. But after the infusion was complete, the mold surface of the part was very sticky, and still had the residue on it. The mirror surface also had the residue on it. There were tiny dry spots from where the 3M was in contact. I was using Adtech 820 epoxy resin, which may have something to do with it. I will not use it again, at least not with this resin system.

#111

I’ve seen the same problem with many other epoxy resins. I’ve never seen an epoxy that will breakdown 3M Super 77. The only solution that I know of is to clear-coat the mold.

#112

3M super 77 does that with any resin that I use. That layer of residue is really annoying because you have to get it all off of your part before you can scuff and clear.

I switched to a adhesive that binds with epoxy and no longer have that problem. The stuff I use is made by Westech Aerosol and called EpokZ. When I first got the stuff I contacted them directly and they sent me a sample spray can for free. Their supplier that I use is Northern Composites. It actually ends up costing the same as a can of Super 77 but you have to by a case of a dozen.

As for the laminate, since this is a relatively flat panel, why not just use a heavier cloth? I use 3K twill for cosmetic reasons, but then I usually use a 12K twill (20oz or so) since laying down one layer will be the same thickness as 3 to 4 layers of 3K. It actually ends up being cheaper than using multiple layers of 3K as well.

#113

If I do use spray adhesive, I go VERY lightly and press more.

Doe…nice Tig welds by the way ! Let me know when you want to knock out a few for me ( De-gass and pot chambers), :wink:

#114

Well I made the real thing finally, but unfortunately I had some issues with the finished part.

Anyhow, here is the process.

  • Scribe trim line into mold

    IMG_5610.jpg by Kevin_Doe, on Flickr

  • Wax the mold, 4 coats of Meguiars Max Mold release.

  • Lay in 1.2 oz/yd fiberglass, plain weave. The idea for this was to provide a finer layer on the surface. The glass can also be sanded easier if necessary.

    IMG_5615.jpg by Kevin_Doe, on Flickr

  • Lay in two layers of 5.9 oz/yd 2x2 twill weave carbon fiber, first at 0/90 then at -45/45.

    IMG_5616.jpg by Kevin_Doe, on Flickr

  • Place 1/4" divinycell foam core mat. I drilled 3/32" holes on a 2" grid. This is a slightly tighter grid than I used on the all glass test panel.

  • Lay in 4 additional layers at 0/90 for additional stiffness on the mounting flanges.

    IMG_5617.jpg by Kevin_Doe, on Flickr

  • Lay in two more layers of carbon, at -45/45, then 0/90.

    IMG_5620.jpg by Kevin_Doe, on Flickr

  • Peel ply

    IMG_5621.jpg by Kevin_Doe, on Flickr

  • Infusion media

    IMG_5622.jpg by Kevin_Doe, on Flickr

  • Spiral Tubing and Vac bag

    IMG_5623.jpg by Kevin_Doe, on Flickr


IMG_5624.jpg by Kevin_Doe, on Flickr


IMG_5625.jpg by Kevin_Doe, on Flickr

The infusion seemed to go great. No real issues. Yesterday I pulled off the vac bag and consumables. The part looked great. Then when I demolded today the surface has a lot of porosity and dry spots. Looks like the dry spots are only in the glass surface layer but its hard to tell for sure. The dry spots are 2" apart, meaning it certainly had something to do with my hole spacing in my foam core.

I cannot just toss this piece and start over, since I’ve gone way over my anticipated budget already. What is the best course to fix the spots? I have a few ideas.

  1. Sand through the fiberglass down to the carbon and hope the carbon was fully wetted out. If that is the case, I will just clear coat part and all will be well.
  2. Paint on some epoxy and use a heat guy to help it flow into the voids and pull out the air bubbles. I’m not confident this method will work. There is wax probably transferred to the surface, plus I don’t know how well the resin would wet out the cloth.
  3. Fill in with glazing putty and paint. I’d rather showcase my hard work, so this will be a last resort.

Any ideas, I’m open!

#115


IMG_5631.jpg by Kevin_Doe, on Flickr


IMG_5632.jpg by Kevin_Doe, on Flickr

#116

Or another option. Sell it as is, and make another.

Most importantly, what went wrong?

#117

Looks like your infusion was a bit quick, how long did it take to infuse? There a several ways to do this, you can either crimp the line slightly (meaning don’t fold it over) there are line breaks/pinchers you can get, I wish I had a picture…I am at home and they are at the shop, they basically pinch the line to reduce the amount of flow. I picked them up from Princess Auto in Canada (where I am).

Edit: http://www.princessauto.com/pal/product/2990414/Engine/2-pc-Hose-Pincher-Set

^ I use those to slow the flow down when needed. The other method is to create resin breaks in the flow media, this is where you would leave a section intentionally without flow media, I never go bigger than 1.5" when I do it, on a part like yours, I might put 3 in evenly spaced. Thats another option but I have found it can cause small bubbles where it traps.

Last option is to reduce your resin infeed line, are you running 3/8" or 1/2" lines? Others will post come recommendations as well…Mould looks nice by the way! :slight_smile:

#118

I used 3/8" resin inlet, pinched down significantly to slow the resin down. The total time for infusion to complete was about 40 minutes. The part with flow media over it took approximately 10 minutes. Went pretty slow. I did have a small hole in the bag that took a few minutes to find. It was right near the resin inlet line and was distributing air bubbles throughout, which made it tough to find the source. I ended up finding it and plugging it, then all the air was removed and all seemed well. I suppose this could be the issue, but not sure.

#119

The dry spots look focused to one end of the part based on the pic, is that case?

#120

The dry spots are concentrated in one area.

I fed with a manifold of spiral hose down the middle. I placed an additional strip of flow media along the center, about 4" wide to help distribute. The spiral manifold was 3/8" OD, a bit smaller than I’ve used before. I noticed that the resin did not flow across the manifold/media all that quickly. I expected the resin front to flow from across the manifold quickly, then spread evenly from the center outward. What I saw was a teardrop shaped resin front. Hard to explain, but you can see it in one of my pics if you look closely.