coremat or not?

[catenahalf]

This is an idea I have for a tiller beam ~5ft long to connect a rudder head and swing up tiller handle for a 27ft boat.

The 2 pieces will be vacuum bagged. My layup plan is 4x 6oz biaxial carbon satin weave on both sides of 3mm coremat (don't know what else to do with this coremat I was given) with 2 layers of 12oz uni carbon extending 6 and 8" from both ends in. There will also be a lightweight fiberglass skin on the outermost layer of the top piece for abrasion resistance and a better look for a clear carbon finish.

Any thoughts would be welcome.

[Stumble]

1) I would be concerned that when driving the boat without the tiller extension the shape of the tiller would be pretty uncomfortable. Even on race boats this is a pretty common occurence, and your shape seems unfriendly.

2) The sharpened edges of the underside of the form don't seem to add much to the product, but could be difficult to manufacture consistently, as well as forcing the CF into a very tight bend.

3) I am not sure that there is any need for a FG skin to protect the carbon. Most tillers don't see that much abrasion.

[catenahalf]

This actually replaces an "I" beam oriented on it's side. It's only a few inches off the surface of an afterdeck. This part of the tiller is rarely touched; there is a comfortable wooden handle about 18" long that swings down from the forward end of this beam. But, I get your concern. That would hurt.
The other issue is my desire to keep the molds and assembly of the parts simple.
Is there a safe radius for carbon?

Guess I could just as easily invert the bottom piece but worried about the negative draft of the mold.

[catenahalf]

Slept on it and came up with a single mold idea with two identical halves to make the whole beam more friendly and hopefully stronger. Inner carbon layers not shown (I'm not that good at Sketchup.)

Don't know if the coremat is going to help me save weight and add strength, or am I better off just adding more carbon?

[rxcomposite]

The coremat and carbon fiber does not seem to be a good idea since coremat function is to separate the inner skin and outer skin and to keep it as far away from the neutral axis as much as possible to make the part stiff.

Looking at your cross section. yours is equivalent to a thin walled section tube, so during bending, one face is in compression, the other face is in tension, with the sides in shear. So I would lay it up in BiAx, Uni's on top and bottom face with a thicker uni laminate in areas (faces) under compression. Carbon is not so great in compression so you need more layers to compensate. Of course if the tube is subjected to two way bending, you design it with equally thick faces but using loads under compression.The -45 x +45 biax will take up the shear load.

In fabrication, it is much simpler to add joggle to the joint or overlap just like you did, glue it with industrial strength epoxy, then finish off with a layer of biax.

I think you are on the right track. You need to make only one mold with a joggle and add a removable caul plate for the upper part so it will not form a joggle.

[rxcomposite]

I am attaching a copy of spar design and analysis from Composite Aircraft Design by Martin Hollmann. Basically, it is a cantilevered beam similar to what you are doing. Hope you will get something out of the page.

Also an excell program to help you design. It is a simple uniformly loaded cantlevered beam formula. Not exactly what you need because you have a concentrated load on the end of the beam pushing and pulling sideways. Just feed in the particulars and the material properties of the fiber you will be using. You should come close.

[catenahalf]

Thanks RX.

Appreciate the time spent on this!

[rxcomposite]

No problem. I just pulled the files from my previous work. Sorry did not have the time to modify the uniformly loaded beam formula into a concentrated load approach.

[catenahalf]

Here's my final design (so far). I ditched the coremat and will be using 7 layers of 6oz biax. I have a bunch of beautiful spar quality spruce that I can mill down to 3/32" and will wrap with the same biax. While each half is in the mold I'll bond in the spruce to act as a shear web. I'll most likely have internal control lines this might help organize.

The design was somewhat dictated by how easy it would be to come up with a plug. The 3/4" radius roundover router bit should make quick work of it. On one side I'll put in the bead by driving the bit in deeper. A fairing fillet will smooth in the hard edge.

This will be my first serious plug and mold... it sounds easy in my head.

[rxcomposite]

catnhalf,

I hope you are not orienting it the wrong way. The reinforcement (uni's) are supposed to be on the side where the tiller arm is pushed and pulled. See the construcuction method attached.

[Phil Locker]

I-Beam on its side.... that sounds like one of those bloody awful tillers that the F27s came with.

Attached are a couple pics of what we came up with when replacing the whole F27 rudder system with a cassette-type style.

The tiller is simple shaped H80 foam, carbon uni's over that, carbon twill over that, fair and paint.

Phil

[catenahalf]

@rx, the image shows the tiller in the proper orientation where the spruce shear web lay athwartships. There will be 18oz uni added to the layup at each end, ...not sure how long to make that layer. Are you saying the web should be vertical? Because of the 1" bonded overlap the sides will be thicker/stronger as well. I'm looking for adequate strength and as light as possible. I don't mind the extra effort and time as this is a learning experience for me.

@phil, yep, you nailed it! The whole rudder system looks like it belongs on an aircraft carrier! Weighs nearly as much too! I'm redoing the the entire rudder system based on Farrier's F82R daggger rudder design. He doesn't support its use on an F-27 like mine so I'm having to improvise quite a bit. The end result will be a huge reduction in induced drag and a much more efficient high aspect foil not to mention a weight savings of at least 50lbs off the transom. I'm nearly done with the foil and will pull an epoxy mold from it soon. The tiller exercise is to familiarize myself with the plug preparation and mold creation and all the materials involved. I'd rather make my mistakes on the tiller than on the 6ft rudder blade.

Do you recall the weights of the three components you came up with?

I think I can get the final molded blade to come in at maybe 10lbs, the tiller under 2lbs and the sleeve maybe 5lbs(?) I'm going to attempt vacuum infusing the blade and tiller, starting with the tiller.

Just picked up a temperature control unit off ebay for post curing... I've got plenty of challenges ahead, but I'm thoroughly enjoying the process.

BTW, beautiful kit you came up with!

[rxcomposite]

Catenahalf,

Because the tiller arm will be pushed from side to side, the bending and compression faces are on the side with the shear web facing top and bettom, hence it is oriented to "the side" as Phil Locker. That is the longer side of the rectangle (web) is top and bottom, the shorter sides (face) on port and starboard.

When a force is applied on the arm, the faces are in compression or compression (taken by the unis in the longitudinal direction) and the web are in shear (taken by the biax at 45 degree angle).

There are many ways of doing it. Sometimes the web side will be stepped upon, or worse, sit upon, so it is reinforced to make it stiff. Honeycomb, foam, or end grain balsa would reinforce the web. (you are correct in reinforcing the web)

For the unis, longitidinal reinforcement can be substituted as long as it is strong in the longitudinal direction (spruce, bamboo, uni's, whatever). Bulky yes, but what is important is that the fibers run into longitudinal direction.

In the spar illustration, it is the spar of an airplane. It supports the airplane so the major force are at the bottom (positive g's) and top (negative g's). In a tiller arm, the main forces are on the side, pushing and pulling so the spar is rotated 180 degrees.

[Herman]

I know you ditched the Core-mat, but am I the only one that seems to think Core-Mat and vacuum don't match?

For vacuum you need either Core-mat Xm or Soric.