S

szicree

Well Known Member
Two questions:

1. I have a gap of about 1/16 inch between my lower cowl and the side skin. The story as to how this happened is long and boring, but at this point all is well except for this gap. All the hinges are drilled and things are quite secure. I'd like some suggestions on the best way to fill this gap.

2. The upper half of my cowl seems to have a slightly tighter radius than the upper portion of my firewall and this means I have to force the cowl down onto the hinge in order to cleco things together. Obviously, I'd rather not have this preload on things. I notice that there is some sort of reinforcement beam that runs side to side across the top of the cowl just forward of the firewall. I'm thinking that I could saw through this reinforcement from the inside, maybe every inch or so, cleco the cowl to the plane and then layup glass patches over the cuts. I'd like opinions on this or other possible fixes.

Thanks.
 
Steve,
Both issues require splicing glass to glass. Filler alone won't last on the edge of a panel (your cowl gap), and the reinforcement in the upper cowl half is structural.

Take a look at post 14 in this thread for a sketch of a scarf joint:

http://www.vansairforce.com/community/showthread.php?t=22931&highlight=fabrication&page=2

The gap can be closed by sanding a scarf on the edge of the existing cowl and laying perhaps 5 or 6 plies of fabric to extend the edge. After cure trim the new edge as necessary for the fit you want.

The reinforcement can be cut as you propose. Flight loads place the inner cowl surface in compression as cooling plenum pressure tries to blow the cowl outward. You can scarf or just add surface plies to close the cuts, but it is structural, so you must use enough plies either way and make good layups without bubbles or voids.

A quality structural scarf is done by sanding the plies back on each edge of each cut so as to expose about 1" of width of each ply (rule-o-thumb, not an absolute rule). For example, if you sand into it and find 4 plies, your sanded area will be a very flat V six inches wide, three on each side of the cut. You'll lay in one two inch wide strip, one four inch, one six inch, and one eight inch. The advantage to this method is that the scarf can be sanded near flush after cure. It is how you would repair an outside skin for example.

Obviously the scarf method precludes cuts on one inch centers. I doubt you need spacing that close, but if so you can simply sand the surface with 80 grit and stack patch plies without scarfing. You need as many patch plies as plies in the glass layer that you cut. On the inside of a cowl the additional overall thickness doesn't matter. Assuming the reinforcement is in the shape of a rib across the cowl, this would be a good application for glass tape.
 
Last edited:
Dan,

Real nice info. I hadn't even thought of a scarf joint for the gap, but knew that filler alone would never survive. I'm pretty sure I can take care of most of it tomorrow and will report back. Thanks a million.
 
Epoxy Bonding

Dan's description of scarfing is excellent. Just a couple of reminders (don't know what your experience level is...so if this is too much Epoxy 101 disregard; having said that, it may help someone else out there).

You're shooting for a rock solid secondary bond with the old surface. To achieve this you need to ensure that the old surfaces are cleaned before sanding (acetone is good), dried (pretty quick with acetone), and then sanded rough. This will ensure that the epoxy can get a good bite/mechanical link with the old surface by flowing into the newly opened pores and scratches. Don't be afraid to rough up the old glass. Dan's photos and descriptions of sanding boards, blocks, etc. are an excellent resource if you don't already have a collection.

When scarfing, I tend to err on the wide side (Dan's 1" per layer rule is fine as a start, but don't be afraid to go wider) since wider scarfs provide more bonding surface for the new layers.

You also want to achieve good primary bonds between the new layers in the scarf joint. Primary bonding is the chemical linking of the wet layup as it cures-all the layers cure together in a single fused layer. In other words, the layers of the scarf joint should be laid up in a single session. It's okay if the first layer begins to set up, but you don't want it to cure-it should still be sticky as you add the next layer.

I lay out all the layers of the patch/joint, wet down the first layer, smooth it with the brush and get any bubbles out, etc. When I'm satisfied with the first layer, I'll get the next layer ready, and so on, continuing to add layers until I'm done done. Depending on air temp, the hardener you're using etc. you should have plenty of time.

I also do a small test layup to get an idea of how much working time I have between layers, the pot life of the epoxy, etc. This is particularly helpful if it's been awhile since I was working with composites (I sometimes go several weeks or months between projects; YMMV).

One final comment/question for other builders: I've noticed a number of threads that mention abrasion issues in the airbox, cowl, etc. I'm new to RV building, but in other projects I've used a layer of Kevlar (tm) in areas prone to abrasion. West Systems 420 aluminum powder also increases the hardness and abrasion resistance of the coated surface and improve its moisture resistance. Cures to a metallic gray color. Has anyone tried these or other layers/fillers to improve abrasion resistance on RVs, or does this tend to generate other issues (i.e. abraid the aluminum)? Suggestions?

Regards,
Mike
 
Last edited:
1/16" Gap

A 1/16" gap is not much of a gap. I have been advised in the past not to close up the gaps too far as they will close themselves up considerably in the painting process. Something to think about...
 
Hey Mike, do you have any reliable reference material you can share regarding scarf slope? I don't have a lot on the subject in my own library. Couldn't remember where I got the 1" per ply rule of thumb for structural scarf repair (probably some old Rutan stuff about repairing EZE's), so I got digging around with my coffee this AM. A favorite design reference (Niu) suggests 20 to 1 for production scarf joints, but those would be aerospace-quality, bagged and autoclaved. Niu lists 30 to 1 for field repairs, but again we're talking aerospace; prepreg, vacuum bag, and heat. An FAA circular says 50 to 1 for "lightly loaded composites", ordinary wet layup, but they don't define lightly loaded. If a typical mediumweight cloth is 0.010" thick, 50 to 1 is 1/2 inch per ply, less than you and I have been thinking.

Niu suggests a reason for less slope, at least for highly loaded stuff. The idea is to balance adherent and fiber strengths. The adherent is in shear and the fiber in tension/compression. As the scarf gets more flat (it would be 100 to 1 with .010" plies and 1" per ply) the shear area goes up but the feather edge of each ply gets weaker (the fibers part where they have been cut thin), and optimum balance is lost.

If the 1" per ply source was in fact Rutan, he may have been thinking about the scatter factor in hombuilder adherent strengths.

Geek stuff, mostly irrevelant for RV glasswork. Inquiring minds and all that rot <g>
 
An old manual...

DanH said:
Hey Mike, do you have any reliable reference material you can share regarding scarf slope? I don't have a lot on the subject in my own library. Couldn't remember where I got the 1" per ply rule of thumb for structural scarf repair (probably some old Rutan stuff about repairing EZE's), so I got digging around with my coffee this AM.
Click to expand...

Dan... I have an old (mid 70's?) factory repair manual for Glasflugel sailplanes... one of the earliest fibreglas gliders...

They talk about a 20 mm scarf on the outer surface of the wing skins (wings have a 5 mm balsa or foam core), which are two layers of 92145 cloth (6.5 oz uni)...

This is much less than modern standards, but many Libelle sailplanes are still flying that are repaired to this standard...

Fuselage holes require a 50 mm scarf, but the fuselage has more glass layers and is not a sandwich construction.

Resins specified (probably now banned as carcinogens...:)...) are Shell Epikote 162 with a Laromin C 260 hardener.

I'll see if I have any other old factory manuals around...

gil A

Remember, Rutan copied much of the technology from this time period... his original Mojave operation was next to Fred Jiran - one of the first composite sailplane repair guys...
 
Thanks Gil. The Glasflugel example works out about 40 to 1 at .010" per ply.
The specs in the motorglider .pdf seem to suggest ratios in high 40's with the same ply thickness assumption. Good data points!
 
IMHO a full scarf joint for the 1/16? cowl to fuse gap would be an overkill. The 1? per ply taper will give a full structural repair. Filling this 1/16? gap is not a structural issue. I had a similar gap on one side of the lower cowl. I sanded about a ?? overall taper and filled the gap with flox. I was going to leave it at this but decided to lay one ply on the inside lapping on to the cowl and the flox. Sixty hours so far with the cowls on and off many, many times with no problems. If you do this, make sure everything is clean, sand/rough up the surface with very coarse paper and don?t make the flox too dry or too wet. Possibly, also a ply on the outside, like the one on the inside would be a good idea.

Fin
9A
Vari-Eze
 
A Little Research

This is a timely topic since it looks like industry standards are being reviewed/revised. In the FWIW category:

The Commercial Aircraft Composite Repair Committee (CACRC) (http://www.sae.org/servlets/aerostd/committeeHome.do?comtID=TEAAMSCACRC) is working to standardize repair techniques since repairs on Airbus composite parts don't use the same procedures as those on a Boeing, etc. even for the same materials. Drives repair stations nuts trying to keep all the different pre-preg etc. on hand in case a bird is damaged....

Scarf repair will be outlined in AIR 5367 "Machining of Epoxy and Polyester Matrix Thermosetting Composite Structures" and they discussed this topic in Nov 2007 at the conference hosted by the National Institute for Aviation Research at Wichita State University. As I understand the status, AIR 5367 is being sent out for balloting. Don't know when this will be available.

I also took a fresh look at AR03-7 "Bonded Repair of Aircraft Composite Sandwich Structures-2004"; this is the DOT/FAA study regarding sandwich repairs using both OEM and CACRC recommended procedures to repair cored composites (see www.tc.faa.gov/its/worldpac/techrpt/ar03-74.pdf).

For those into math/engineers out there, this should give you some light reading :D. The study outlines a variety of repair and testing methods used to compare the strength of composite repairs. OEM method used a .25" overlap per layer; CACRC method used a .5" overlap per layer, plus one additional layer on top of the repair. Interesting results: in shear testing, the wider overlap produced a stronger joint as you might expect.

However, in overall testing of the repair samples, the .5" overlap didn't show much advantage over the .25". This makes sense since more good material is being removed around the damage to achieve the wider scarf...as a result, the report concludes that a .25" overlap per layer may be enough for cored composite repairs if other repair procedures are being followed (no contamination, sufficient resin, etc. etc.).

I also found a reference to the 1" overlap/layer in Chap 3 of AC 43.13-1B regarding step joint repair of composites. "Extend the cleaned-out area for a distance equal to the number of plies to be removed, less 1 inch. For example, if you must remove three plies, extend the repair for 2 inches beyond the cleaned-out area. Each layer should be 1 inch beyond the layer below." The scarf joint section in the same chapter recommends 50 times the thickness of the face ply (going back to the approx .5" standard for a .01" thick layer).

I think the real answer depends on the repair required. Cored part? Structural or not? In tension, compression, or both?

Wider scarfs provide better shear resistance, so if the part is in tension, you may want to go with a wider joint. If the part is subject to compression loads/is a structural part, I'd go with a narrower joint/follow the sailplane repair recommendations discussed in this thread.

Fin's comment regarding "...possibly a ply on the outside, like the one on the inside would be a good idea" is worth mentioning again; if the part being repaired is in tension, you don't want to beef up one side and not the other, since the stress loads will not be evenly spread. If you add a layer on the bottom, add a layer on top. Keeps balance in the universe...

Whether or not a 1/16" gap is acceptable/whether a scarf joint is overkill is up to the builder...personally, I've had mixed results (mostly poor) trying to build up an edge with flox and not have it chip later. Could work if the area is small enough, but for a longer edge I'd at least wrap the flox-built edge with a layer of cloth-biaxial wraps real well. YMMV.

Regards,
Mike
 
I hope I didn't leave Steve with the impression that he needed a 1" per ply scarf to add 1/16" of glass to the edge of his cowl. I probably wouldn't sand back more than an inch total just to add glass to an edge.

The scarf discussion was targeted to the structural issues of cutting the inner cowl reinforcement so he could spread his cowl for better fit.

Found an old Rutan reference! The 1" overlap per ply instruction is in "Moldless Composite Homebuilt Sandwich Aircraft Construction". The illustration shows a lap repair rather than a scarf.

For those who would like learn fiberglass work with a great "how to" reference in hand, the above is a classic. Spruce still sells it I think.
 
Dan,

Actually I did take your advice to mean I should do a full-blown structural scarf joint. However, upon looking at my piece today, I made an executive decision to bevel only about 1.5 inches back instead. I already laid everything up and am waiting for it to cure before trimming it back. I'm very confident that it will hold up great and really appreciate the advice, even if I did ignore some of it :D. Hooray for experimental aviation!
 
Well durn, Steve, sorry about that....I wasn't clear enough. Let us know how it turns out.

Mike, I've only quick-scanned the ARO3-7 paper, but it looks like great stuff! Thanks.
 
A 1/16" gap? Man, I wouldn't even mess with it. Get it too tight and the paint will chip. A nice even 1/16" gap will look fine. I just completed an extensive "fix" of the one on mine. The fit was so bad the builder added an overlap extension. I pulled that off and had to deal with up to a 1/4" gap in places. In hindsight I should have just ordered new parts. That was painful ordeal.
 
You must log in or register to reply here.