Sikaflex sliding canopy

I've read a bunch of posts about Sikaflexing the canopy, and have chosen to do that on my 7 slider. I'm trying to think through the details. The drawing shows my proposal, but it needs at least one more point of attachment for the canopy skirt. I'd like to keep at least 1/8" Sika thickness for the differential expansion of metal and plexi, and I don't want to add more holes. Thoughts?

How about Sika between the skirt and the plexi? That stuff is pretty tenacious and I don't think you would have any problems with separation. If I have read correctly thats what a number of slider people do and is my current plan. Thats a great drawing by the way...

Don't get excited about thickness.

We have a tip up 7 and we had a similar assembly for the side skirts.

What you are trying to achieve is a bond between the inner and outer elements of the canopy structure.

Scuff the canopy inside and outside - mask and prime both sides. Prime the inside of the frame and the outer strip and assemble.

When the Sika ha just gone off - rub it off and you will have a perfect line.

Finished

I'm sure I'm overthinking it, and I appreciate the advice to put it together and go flying...

but the Sika brochure recommends 6mm bond line thickness to allow for differential thermal expansion of a 1.5m window. So, I will use 1/4" spacers on the inside of the window, and then see how wide the structure is.

With luck I will be able to use a 1/4" thickness between the window and skirt, and still be able to rivet a tight fit against the fuselage. If I'm not so lucky I may have to joggle the top of the skirt to make room for the Sika.

Thanks,

A 1/4" sika thickness between the plexi and skirt is too thick IMO. 1/8" is lots, again in my opinion. My skirts are bonded with sika (on a tipper) @ 1/8" and they do not move, period. The Sika brochure's specs are based on a relatively large sheet of plexi in a marine environment.
regards

I did some practice joint pieces with with the various materials involved and what I learned is that the prep is VERY VERY important. If you do it right the material will be destroyed before the Sika comes loose. If you do not do it right your 3 year old can pull it apart. Keep in mind that you are not gluing the plexi to the metal canopy frame. You are gluing the plexi to the paint that is on the frame and that seemed like kind of an unknown to me. What I did for what it is worth was to build the canopy as the instructions call for with the rivets and such. Then when I was through I put Sika on in fillet shape along the tubes with the hope that the Sika will prevent movement of the plexi that would create pressure against the rivets. Kind of a "belt and suspenders" way of doing it.

No problems here

This is exactly what i did on mine, 10 hours flying and she is looking good!

some info on my blog if needed.

http://rv6aj.wordpress.com/2012/04/09/canopy-side-skirts/

recommend not adding spacing as drawn

I am finishing up a sika'd sliding canopy on my -7A with zero rivets through the plexi. I also struggled with the question of thickness and at one point had the exact same idea for how to bond the canopy. I think you definitely don't want to have the setup in your drawing for two reasons.

The first is clearance between the canopy frame and turtledeck. You can't really have any space between the canopy and frame at the bottom most portion of the canopy. This is because the canopy frame must clear the turtledeck when you slide the canopy aft. If you have space between the canopy and frame, and want the side skins to be flush against the fuselage, the canopy will be too narrow to clear the turtledeck. You could end up with a canopy that you can't slide back.

I ended up with 1/8" spaces between the canopy and frame along the canopy spine and forward and aft bows, but tapered this to zero space at the very bottom. I should stress again that this was required to allow the frame to clear the turtledeck. I wasted a bunch of time having to re-bend the canopy when I figured this out - hopefully you can benefit from my experience.

The second reason is it doesn't seem necessary from a strength perspective. I created a test panel with plexi trimmed from my canopy and bonded a representative piece of skirt material to one side. The result is that the bond is so strong, the aluminum bends before the joint fails. In fact, I can apply 30-ish pounds of force along the joint line of a 2-inch piece of skirt and not get the joint to do anything. This is with a bond that is only 1/2" wide. With the actual canopy, the situation is even better because the inner and outer skirts are jointed together and to the canopy frame - basically, there should be no bending load on the joint. The total bonded area of the actual canopy is much, much greater than the test piece as well. Based on this testing I decided the gap is not necessary for the skirt application. I ended up installing the parts as shown on the drawing with sika bonding the inner and outer skirts to the canopy.

I also did a test piece with the steel tube from the canopy frame with 1/8" spacers. I have only a 1" piece of tube sika's to the plexi, but I physically can't pull it apart. This stuff is really, really, really strong. I think as long as you follow the instructions regarding prep you will be fine.

Thanks bikes_and_planes. I am worried about having the canopy and skirt wider than the fuselage, which is why I am obsessing about the thickness of the Sikaflex.

For those who are comfortable with TLAR engineering, you are probably fine with a 1/8" thickness of Sika. Please click on to another page...

The problem we are trying to solve is cracking of canopies. Many mechanically fastened canopies have failed either because of poor fabrication techniques or failure to leave room for thermal expansion around the fasteners. I also seem to remember at least one Sika-bonded canopy that cracked, but I'm not sure of the cause.

I can imagine my plane experiencing as low as -20 in the hangar, and as high as 140 degrees on a sunny ramp. Let's call that a 150 degree range.

3 feet of steel expands about 0.034" over this temperature range. Plexi expands about 0.221". Aluminum is intermediate at 0.070. We are most concerned about the differential expansion of plexi on steel, which amounts to 0.188 or 3/16".

I accept the argument that the curved bows of our canopy will probably change shape, reducing the differential. I am most concerned about the differential expansion along the 3' of the square tube on the canopy sides. I don't think it can change shape as easily.

If I glue the canopy at an intermediate temperature (70 degrees), I only have to worry about half of the expansion. Also, the difference will be split between the two ends. So I guess we are talking about an expansion of 1/16 or 1/8" on each end. This means an 1/8" bead of Sika will be stretched 100%. The Sika literature indicates it stretches 500% before failure, so I guess we are OK.

The Sika folks have a chart that recommends 4mm thickness for bonding windows 3 feet long. That's how I got started with the 3/16" number.

Seems to me that tapering the Sika thickness to zero at the edges of the canopy is exactly the wrong approach, since the stresses will be largest there?

I worried about that as well, but even when there is zero clearance between the canopy and frame, there is still plenty of room to allow for movement due to thermal changes. The plexi is flat at the bond location, but the canopy tube is round. There is really only zero clearance at one line of contact between the two, but the clearance increases moving outward from the center of the canopy frame.

Basically what you end up with are fillets of sika along both sides of the round canopy frame. I ended up making fairly generous fillets that i felt were wide enough to ensure that along most of the joint there is actually a 1/8" separation between the two surfaces. Even if the joint line where the canopy and frame have zero clearance breaks down over time due to the sika exceeding the 500% strain, the fillets are still there and should have plenty of margin. If you wanted to completely follow the guidance, you could even omit the sika in the lower-clearance center area of the frame and only have the fillets.

I sika'd at what I felt was an intermediate temp for my climate to split the difference as far as thermal expansion. Additionally, most of the canopy does have a 1/8" gap, which seems to be flexible enough to allow for some deformation of the canopy even if the edges are more or less constrained by the thinner sika.

One other thing to think about - on the forward side of the canopy, larger fillets will reduce visibility since you will have to add a wider band of primer to the canopy. I have smaller fillets on the aft side than the forward side for this reason.

Thanks for all the suggestions and helping me figure this out. This is what I am going to do:



Let's me get my 1/8" Sika thickness, and keep the width outside the canopy frame to a minimum.

Great discussion

Very timely discussion. I'm not too far out from working on the the canopy.

I have several sites bookmarked regarding sliders and sika, 4 or 5 maybe. Anyone have more sites logged? Maybe we could compile a good list and have Doug add it to the references.

It seems like those lists of www sites go stale pretty quickly.

A Google search for 'sikaflex vans canopy' brings up a pretty nice list of current www sites.

I bonded my slider canopy with Sika in 2010. It went through some pretty severe thermal cycles just sitting in the hangar while I finished the plane, due to my location in the Sacramento Valley of northern California. Finally started flying in June, 2012. About 180 hours on the hobbs now, and have exposed the canopy to temperatures ranging from sub zero to 140+ F since then. So far, so good.

Being an engineer with some experience with urethane adhesives, I obsessed about some of the details. Prep is crucial follow Sika's procedures to the letter, using all the materials- cleaner, primer, etc. I made several test coupons, then tore them (or attempted to; I was not always successful) apart. Sika is good to about 300% strain, which is to say that a 1/16" thick bond line can be sheared up to 3/16". On an RV-8 with its long canopy that's conceivable, but a side by side RV isn't likely to see that even at a temperature extreme. The fact that the canopy is curved and far from rigid means that the entire thing will distort a bit, reducing the actual amount of shear due to thermal expansion/contraction. This is apparent in my canopy, as the fit is somewhat different on a very hot vs. very cold day.

To create the side joint, I copied the procedure described in this thread. Note that the link to a picassa photo album in the first post is dead, but the one on a later page is good. On the side bow, I did what the OP of that thread did, as shown on this photo from his picassa album. Rather than trimming the canopy to lie about 3/16" above the side bow per plans, I cut it closer- more like 1/16". The big fillet of Sika as shown on the photo link makes the joint very forgiving.

Additionally, rather than making the skirt out of aluminum per plans, I laid a one-piece skirt up with carbon/epoxy. The skirt is also bonded to the canopy and frame with 295UV. For "belt and suspenders" I also riveted the skirt to the side bows of the canopy frame per plans. There are no holes through the canopy however, other than the hole for the latch, on top.

So far, so good. No regrets. Feel free to PM me for pix if you think they might be helpful.