(New?) Composite Panel Construction Technique

[StrongFeather]

Yep! Just another example of me not being good at 'splainin' - and trying a little too hard to be brief. The windows will be commercial units with screens. One in each door and one larger one across the top of the rear (above the rear-access storage area). The wood frames that I mentioned (briefly) will be made to fit the windows and then they'll get epoxied into the panel in place of the XPS foam board. The panel skin will then get routed out with a flush-cutting bit so that the windows will drop right in.


Steve

[tony.latham]

Yep! Just another example of me not being good at 'splainin' - and trying a little too hard to be brief. The windows will be commercial units with screens. One in each door and one larger one across the top of the rear (above the rear-access storage area). The wood frames that I mentioned (briefly) will be made to fit the windows and then they'll get epoxied into the panel in place of the XPS foam board. The panel skin will then get routed out with a flush-cutting bit so that the windows will drop right in.


Steve

T

[Atomic77]

I didn't want any fasteners at all and most of the Astro is epoxied together. Really the first time we used any fasteners was when we connected the walls to the frame, but even so we still used a nice bed of Hysol and that's one of the things I'll never regret. I flaked out the interior panels and I absolutely didn't want any fasteners so we used VHB tape to hold them all in place. It's been a couple years now with no failure. One trick is to warm each side of the tape with a heat gun prior to installing it. I would heat it until I see the glue start to just slightly change then I knew it was enough. You only get one shot at application though.

What are you going to do about the seams between the panels?

Looking good so far, keep up the good work

[StrongFeather]

What are you going to do about the seams between the panels?
Looking good so far, keep up the good work

Thanks, Michael. I'm pleased with how things are going, just not the speed at which they're going.

I ran a test with white Elmer's School Glue to hold the panels together, but much to my surprise, it failed. Seriously, though, I'll use some form of epoxy between the panels, whether it's the "thick" laminating epoxy I used on the floor, something like Six-10 out of a tube, or Hysol based on your recommendation - or something else. I haven't spent much time researching that yet, because I know there will be too many viable options to choose from.

The execution will be something like this: Apply painter's tape very close to the edges on the outsides of the panels. Slather the epoxy on the frames of both panels, lightly clamp them together and align them so their flush. The plan in my head is that I'll use just enough epoxy that a little will ooze out of the seam, and whatever does come out will be wiped off with vinegar. The tape should prevent it from migrating onto the surface, so only a very thin line will be exposed, and as long as the epoxy completely covers the seam from top to bottom, it will never leak.

Panel prep will be minimal. The peel ply leaves a nice surface for mechanical bonding, and I'm 98.3% sure that the Adtech820 epoxy is non-blushing so I shouldn't have to wash them first. If not, I'll either leave the peel ply on until I'm ready to put the panels together, or give them a good scrub before hand.


Steve

[Atomic77]

Hysol is a very high quality adhesive. I really like it. But you have to be precise with the mix. Another route I would go, and maybe the better option since you are marrying the same materials rather than dissimilar materials, would be epoxy with Aerosil mixed in. It can be mixed to a less than peanut butter consistency and still sands nice if you get squish out somewhere you don't want it. Hysol doesn't sand well.

Sent from my SM-N950U using Tapatalk

[John61CT]

3M 4000 is great sealant and quite strong adhesive, but not quite as "never undo" as 5200. No mixing.

lots of flexibility/elongation vs 5200, and comes in a UV resistant version

So a damaged panel could be swapped out if needed

[Atomic77]

Depends on the strength you need. 5200 and 4200 are quality marine sealants but don't light a candle to thickened epoxy or hysol

Sent from my SM-N950U using Tapatalk

[StrongFeather]

Making the mold in two easy steps...

1. Buy a rickety old sliding glass door for $20.

DSC00028.JPG (175.7 KiB) Viewed 2393 times

2. Snap your fingers and you end up with this...

DSC00033.JPG (152.65 KiB) Viewed 2393 times

Okay, it was a little more than a snap of my fingers, but it ended up looking great. After ripping the wood frame off (actually, it nearly fell off) I cleaned the glass and trimmed the rubber perimeter (double pane gasket) before adding the aluminum frame, which is 1 x 1.5 x .125 T-6063. It's not real easy to see, but I decided to seal the frame to the glass by using gum tape. The bottom and right side frame pieces are flush with the edge of the glass, so I just ran the tape where they come together. The top piece was done in a similar manner, but it is not flush with the edge of the glass, so the gum tape is pushed into the corner where they come together. The bottom, right and top pieces will remain fixed, while the fourth piece (left side in the picture) will move to adjust to the width of the panels. The mold will remain in this configuration for all of the 5' tall panels, which is the majority of the panels for this trailer. I also spent a good amount of time waxing and polishing the glass one side of the fourth (adjustable) frame piece. The frame piece needs to be polished because some of the panels will need cosmetic edges, e.g. the door openings - the doors will need 4 cosmetic edges, which should be a little tricky.

Here's another picture. It's kind of hard to see, but the white strip on the bottom of the frame is the paper backing of the gum tape. You can't really see the edge of the glass.

DSC00032.JPG (153.06 KiB) Viewed 2393 times

Steve

[StrongFeather]

Big day today. I infused the first full-size panel today. I learned a few more things and made some rookie mistakes, so hopefully the panel turns out okay. Here's some pictures...

This picture shows the carbon fiber, fiberglass and wood frame in the mold. The clamps aren't really clamping as much as holding the frame and fabric in place. If you look at the corner closest to the camera, you can see that the corner of the carbon fiber fabric is cut out. This prevents bunching and makes layup easier. I cut this piece a little differently than the prototypes by cutting almost into the bottom of the frame, whereas before I was pushing the frame and fabric into the corner and then cutting to the top of the frame.

DSC00034.JPG (179.83 KiB) Viewed 2354 times

I wanted to show that the flow media (the green mesh) is under and outside of the frame to allow the resin to flow under and behind the wood frame. This is kind of unconventional because the flow media will stay in the part, and it's a significant change from the prototype panels. In the prototypes, I used flow media and peel ply against the mold frame, which was a PITA because 1" strips of both (flow media and peel ply) had to be taped to the mold frame before laying the fabric. The new method is much easier and there's no chance of them getting pushed down into the visible surface, which did happen in testing. The only downside is that the panel edges will not have the advantage of the rough surface left by the peel ply, which is great for bonding. I'm sorry, I know this is wordy and probably difficult to visualize.

DSC00037.JPG (207.04 KiB) Viewed 2354 times

And a picture of the fabric folded up and over the wood frame. I used 3M composite spray tack adhesive to hold the fabric in place, which works awesome! Again, the clamps aren't really clamping as much as just holding everything in place.

DSC00038.JPG (167.25 KiB) Viewed 2354 times

More to follow...

[StrongFeather]

In this picture the peel ply (white material) and flow media (green mesh) were laid on top of the carbon fiber and fiberglass (the wood pieces are just holding the flow media in place). This is also where I made my first rooking mistake, and it may have ruined the panel. You see the extra flow media channels going to the four corners, right? Welp, that was me thinking that would be enough to give the resin a faster path to the corners. Nope! From now on I'll add some spiral tubing to give the resin an easier path. I thought this infusion was going to stall, ruining the part. More on that later.

DSC00041.JPG (149 KiB) Viewed 2351 times

The bagging file (with pleats), resin inlet and MTI vacuum tubing are all done. Ready to pull a vacuum, check for air leaks and infuse!

DSC00042.JPG (153 KiB) Viewed 2351 times

Here's 880 grams of resin in the degassing chamber to remove entrained air, which could end up as tiny bubbles in the part. That's a lot of air!

DSC00044.JPG (107.52 KiB) Viewed 2351 times

More to follow...

[StrongFeather]

So the infusion... It started off nicely, but began to stall somewhere before the point you see in this picture, which was taken at least 40 minutes into a 60 minute pot life resin (25 minutes of that was degassing the resin). If that wasn't bad enough, I ran out of resin and nearly sucked air into the part, which according to what I've read/seen, that's a sure way to ruin a part. So, I clamped the inlet and frantically mixed another 200 grams. Then another 100 and another 100. Along the way, air did get in, but the bubbles seemed to just work their way to the edges and disappear. In the end, even after at least 90 minutes, the part did infuse entirely. I'm anticipating a bunch of surface bubbles when it comes out of the mold, but we'll have to wait and see.

DSC00047.JPG (158.39 KiB) Viewed 2346 times

One other thing that may have contributed to the slow flow problem may be that I did not run flow media under the MTI tubing. The tubing sits right on top of the wood frame, and I only ran flow media up the sides of the frame. So, with the bagging film essentially choking all but the bottom of the tubing, and no flow media to channel the air into the tubing, I think I essentially choked the tubing off, making it difficult for the vacuum pump to pull the air out and allow the resin to take it's place. I hope that makes sense.

DSC00048.JPG (208.7 KiB) Viewed 2346 times

Sorry for the information overload. Hopefully some of you made it all the way through. I'll post a picture or two of the finished part tonight or tomorrow.


Steve

[Atomic77]

I would have run Enka mat down the center and "T'd" off a couple times with multiple inlets for a part that size. Would have cut the infusion time in half or more

Sent from my SM-N950U using Tapatalk

[StrongFeather]

I would have run Enka mat down the center and "T'd" off a couple times with multiple inlets for a part that size. Would have cut the infusion time in half or more. Tapatalk

EnkaFusion is on order. Thanks for the tip, Michael.


Sent from my iPhone using Tapatalk

[StrongFeather]

I pulled the first full-size panel from the mold yesterday and...

IMG_3152.jpg (812.3 KiB) Viewed 2273 times

...It is B-E-A-Utiful!
The picture doesn’t do it justice. The skin is nice and smooth, pattern is true and the corners/edges are square. I’m really happy, relieved and surprised at how nice it came out.

And this happened as I was demolding...

57411346307__11605F00-FD2B-404D-962D-8F0FCA8207C9.jpg (463.6 KiB) Viewed 2273 times

Another lesson learned. The entire top pane of glass on the mold shattered as I was removing the part. The cause was my shortsightedness when I failed to apply mold release to the mold frame, causing it to be glued to the glass. The part released nicely, but was locked in the corner and just a little pressure on the part and BOOM!

So now I’m going to build a new table (the old one was warped so I had to shim between the table and glass) and rebuild the frame on the pane that didn’t break. If this one breaks, and there’s a chance that it will, I’ll buy a nice thick piece.

Cheers!
Steve


Sent from my iPhone using Tapatalk

[Atomic77]

I pulled the first full-size panel from the mold yesterday and...

IMG_3152.jpg

...It is B-E-A-Utiful!
The picture doesn’t do it justice. The skin is nice and smooth, pattern is true and the corners/edges are square. I’m really happy, relieved and surprised at how nice it came out.

And this happened as I was demolding...

57411346307__11605F00-FD2B-404D-962D-8F0FCA8207C9.jpg

Another lesson learned. The entire top pane of glass on the mold shattered as I was removing the part. The cause was my shortsightedness when I failed to apply mold release to the mold frame, causing it to be glued to the glass. The part released nicely, but was locked in the corner and just a little pressure on the part and BOOM!

So now I’m going to build a new table (the old one was warped so I had to shim between the table and glass) and rebuild the frame on the pane that didn’t break. If this one breaks, and there’s a chance that it will, I’ll buy a nice thick piece.

Cheers!
Steve


Sent from my iPhone using Tapatalk

If one were sensitive to disappointment, this would be tough. But... you now know more than you did before, therefore mark another success. The panel looks great. Next time spray PVA (Polyvinylacetate) in your mold first. Simple and foolproof. Your part will practically fall out on it's own and the thin green film washes off with soapy water making clean up a snap.



Sent from my SM-N950U using Tapatalk