|
-
POSTING RULES

-
Donate yearly (please).
-
Advertise in here!
-
Today's Posts
|
Insert Pics
|

04-27-2017, 09:38 AM
|
|
|
|
Join Date: Dec 2009
Location: Boulder, CO
Posts: 4,436
|
|
Tom, it's not my job to convince anyone to do anything. Especially on someone else's schedule, and across multiple time zones. You're free to do as you think best, and I'm glad that you're documenting what you're doing and your reasons. I hope that in a few years you'll return to this thread and let us all know how it worked out.
In general, when someone uses an elastomer to manage a difference in thermal expansion, adding thickness to the elastomer is the way to improve the joint. To a degree, the thicker the better - thickness reduces the shear stresses as the other materials expand and contract relative to each other. Generally speaking there's no structural merit to making the elastomer thinner for this sort of joint, although as you suggested, there might be a cosmetic benefit. With canopies, since they're a PITA, I'd put the emphasis on having a good structure and deal with the cosmetics elsewhere.
Dave
|

04-27-2017, 05:29 PM
|
|
|
|
Join Date: Nov 2012
Location: Launceston, Tasmania, Australia
Posts: 774
|
|
So Dave, I think I have worked out why you are right and I am wrong. When a tensile test to failure is performed, the material is allowed to "neck" in the middle. In our case the structure is purely under shear, so the material can not neck to the same degree (essentially not at all). This why what we really need to do is test a sample in shear, increasing the bond thickness until we achieve our desired 1.5mm elongation before failure. This still doesn't tell us much about the stresses within the Perspex at this point, but it will at least tell us what bond thickness is required to sustain the temperature range. That said, I've spent the best part of the day prepping this canopy ready for the layup, and I have only allowed for 0.75mm, so it will be a major PITA to go back and change everything now, thus I'll be the sacrificial lamb in the testing for this one. I can only hope that what I haven't allowed for will be taken up by the complex curve nature of the canopy. At the end of the day 0.75mm spacing on the top and 1/8" spacing beneath is a load better that directly bolting to a steel frame and bonding the skirt with straight epoxy!
Tom.
|

04-27-2017, 07:19 PM
|
|
|
|
Join Date: Dec 2009
Location: Boulder, CO
Posts: 4,436
|
|
The acrylic has some compliance and can absorb some of the differential expansion, as can the fiberglass, if you didn't use carbon. Carbon, being stiffer and having that near-zero CTE, is both the cause of the issue and not a contributor to the solution.
Who knows- maybe you'll get away with it. I hope so, but will take a different, more conservative, approach on mine, when I get that far.
Dave
RV-3B, now skinning the fuselage
|

04-28-2017, 12:12 AM
|
|
|
|
Join Date: Nov 2012
Location: Launceston, Tasmania, Australia
Posts: 774
|
|
I just finished the layups, and I'm having my regrets already.
I used 22oz triax glass on the bottom, 14oz carbon uni in the middle, and 22oz traix on the outside. I suspect this is thicker than what I need.
It was recommended to me to use the triax as it is stronger than the weave for the same weight. I had a dog of a time getting it around the aft bow on the canopy, as the weave was just too stiff, so I had to cut it and double it back over itself in sections. I also had an appalling time getting this heavy weight triax to comply to the profile of the doghouse. The carbon too was also so wide as it didn't conform either around the aft canopy bow, so I had to make some cuts through it and double it over itself. I at least made sure these cuts didn't occur in the same location, so my resulting bulges should be minimized. Total weight was about 950g for material, and 950g for resin, which after edge trimming should come to 1475g combined. I suspect there will be another 100g of filler as the traix was pretty coarse and will require filling. There will be 90g of sikaflex required to fill the 0.75mm gap, giving a grand total of 1.66kg. This compares to approximately 1.2kg for the aluminium fairing, so there's almost exactly a pound difference there.
If I was to do this again (which I hope I never will), I think I would forgo the stiffness of the triax glass, or at least lay down some more conformant woven cloth of lighter weight first and do the layup in thinner layers. This should help conform around the rear bow curves and the doghouse. I may sandwich in the middle some uni glass, but thinner widths so that it conforms better around the rear bow. This super thick triax was tough to handle and wet out. For all the reasons Dave has pointed out, I think I'd also skip the carbon. Time will tell, and I will be sure to report back of this doesn't work out. Great excuse I guess to get out of Tasmania and head to some hotter places, all in the name of science.
Tom.
|

04-28-2017, 06:21 AM
|
 |
|
|
Join Date: Jan 2005
Location: Twin Cities, MN
Posts: 1,565
|
|
Quote:
Originally Posted by tgmillso
I used 22oz triax glass on the bottom, 14oz carbon uni in the middle, and 22oz traix on the outside. I suspect this is thicker than what I need.
|
Holy heavy cloth Batman! I think I used one layer of 6oz CF BIDI sandwiched between two layers of 8oz FG BIDI. Then again I was just making a cosmetic/fairing thing. I think you made a roll bar
The heavier weight + uni weave would make it more difficult to deal with any compound curves I think. I've never worked with triax so I'd be interested in commentary from someone more knowledgeable than me around that.
Looking forward to seeing how it all turns out!
__________________
Brad Benson, Maplewood MN.
RV-6A N164BL, Flying since Nov 2012!
If you're not making mistakes, you're probably not making anything
|

04-28-2017, 12:22 PM
|
|
|
|
Join Date: Dec 2009
Location: Boulder, CO
Posts: 4,436
|
|
Tom, thanks for an informative report on the build-up. Very interesting.
Dave
|

05-30-2017, 06:47 AM
|
|
|
|
Join Date: Nov 2012
Location: Launceston, Tasmania, Australia
Posts: 774
|
|
Complete
Hi Dave/Brad and the 50,000 or so other VAF geeks,
So I finally finished the canopy. Talk about one monster of a project. No wonder it had an entire section of the instruction manual dedicated to it. I started work on March 8th, and finished a day ago, working on it almost every day. Admittedly some days were only a couple of hours of work, because I'd have to wait a full 24hrs before I could sand the West 407 filler back due to the low temperatures (even with 205 fast hardener).
So the canopy skirt (or rear roll bar as Brad refers to it) came in at 1.8kg total after surface filling then priming the interior with Wattyl (Valspar) UC230 high build primer, and top coating with a couple of coats of Colourthane Polyurethane. Heavier than I would have liked, but crikey it is stiff. I had just bridged between the canopy and the fuselage with packing tape, so like others, I experienced a bit of sagging when the layup cured, so I had to use more fill than I would have liked, which I suspect lead to the higher than calculated weight, although I still don't know a better way of bridging this gap. I used the proscribed MK319-BS rivets to attach the fairing to the canopy frame, so I had to countersink pretty deep, then filled over the top with West 407. I went with the monel rivets instead of the longer aluminum CS4-4 due to Dan Horton's concerns about corrosion with the carbon fiber. I suspect given that these will be completely sealed, then the risk of corrosion should be very low. The canopy skirt brace is riveted to the skirt using soft -A instead of -AD rivets, and on the inside of the steel canopy frame using the regular LP4-3 rivets. The top slider dog house was laid up over the supplied aluminium part (it was taped down to the aft fuselage when the packing tape was laid as the release agent) then much later in the fabrication the C-679 slider seal was shaped, then taped with packing tape so a tighter mould could be taken to ensure an airtight seal. The skirt was attached to the perspex using Sikaflex 295UV (and 209 primer of course). 0.75mm spacers were fabricated from layers of black electrical tape and places at 2-3" intervals When the original layup was completed, 0.75mm thick layers of electrical tape were run all the way around to simulate the future sikaflex layer. The big issue faced when bonding the skirt to the canopy was getting the sikaflex to squeeze out of the layer. The problem was the sikaflex was so thick and I didn't have clamps deep enough to clear the aft of the skirt, so I had to squeeze using my hands as hard as I could and progressively work from one side ot the other. It came up fine, but in hindsight I was just lucky I got it down before things cured with an overly large gap. I probably should have fabricated some deep clamps for this purpose.
Once the canopy was complete, I then laid up a coating of West G-Flex epoxy and West 407 filler and placed it around the windscreen. My windscreen is also a bit non conventional. The windscreen was originally bonded to the fuselage using West G-Flex and flox, on both the inside and outside, with Sikaflex 295UV around the roll bar portion. I then I added more flox and G-Flex around the front of the windscreen and sanded this into a smooth transition, before laying up just one layer of 12 oz double bias glass cloth. This glass was laid onto a coat of G-Flex but was wetted out using West 105 resin, as G-Flex isn't much good for wetting out (but bonds well to Aluminium and Perspex). A couple of final layers of West 407 filler was progressively laid down and sanded accordingly. Around the roll bar I laid a transitionary coat of G-Flex mixed with 407 filler so I could have a smooth aerodynamic flow from the windscreen to the canopy. My canopy didn't match the windscreen perfectly, as I had clamped the windscreen down a little hard in places when the sikaflex was setting, so this 407 filler masked my incompetence. The gap between the windscreen and the canopy is very consistent (I spent a day and a half sanding it, so I guess it should be) thus I have decided to do without the 1" lip that normally extends from the windscreen over the canopy, and instead will just add a foam seal, as it the practice on aircraft like the Grumman Tiger. I can always add something in the future should I deem it necessary, however this got around the problem of people leaning on the strip when entering and exiting the aircraft and potentially causing cracks, or making something so stiff (carbon) that it causes thermal coefficient of expansion problems and subsequent perspex cracking. It also buys back some of the weight I added with my overly massive canopy skirt! Anyway, I'm glad that's a chapter of this build completed, and I'm definitely looking forward to the engine installation as a change of pace from sanding fiberglass.
Tom.
PS. Having trouble uploading pics, so will try to add more tomorrow.

|

05-30-2017, 08:13 AM
|
 |
|
|
Join Date: Oct 2005
Location: 08A
Posts: 9,500
|
|
Quote:
Originally Posted by tgmillso
I used 22oz triax glass on the bottom, 14oz carbon uni in the middle, and 22oz traix on the outside. I suspect this is thicker than what I need.
|
It's also bass ackwards. The carbon cannot add much stiffness when placed at the neutral axis of the layup.
__________________
Dan Horton
RV-8 SS
Barrett IO-390
|

05-30-2017, 04:23 PM
|
|
|
|
Join Date: Nov 2012
Location: Launceston, Tasmania, Australia
Posts: 774
|
|
Hi Dan,
I too was a little sceptical about this at the start, as I am in agreement that the moment of inertia and benefit of the carbon would be greatest when it is out at the geometrical extremities of the layup, but was told that the intent wasn't to transfer the bending load at a point directly back to the canopy frame, but to add hoop strength and transfer the subsequent load from the load point around the curved surface of the rear hoop under compression, hence the reason that it didn't matter that it sat in the middle of the layup. The composites engineer that came up with the layup for me said that this would be the most cost effective application of the carbon and the easiest to apply. I didn't argue, as I frankly wasn't keen on having the carbon layup on the outside that I had to sand. I see carbon as simply high strength asbestos. I can give you this engineers number and you two can hash it out, but at the end of the day, it's done and **** is it stiff. I think if I were to do it again however, I would have skipped the carbon layer as the stuff is a-- expensive and a pain to cut, and would have used thinner and narrower layers of triax glass (but probably close to the total thickness I used) as the thick stuff I used was tough to work with. All part of the education I guess.
Tom.
Last edited by tgmillso : 05-30-2017 at 04:25 PM.
|
Posting Rules
|
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts
HTML code is Off
|
|
|
All times are GMT -6. The time now is 09:07 AM.
|