• Untitled Document

    Join us on March 29rd, 7pm EST

    for the CBEC Virtual Meeting

    All EYO members and followers are welcome to join the fun and get to know the guest speaker!

    See the link below for login credentials and join us!

    March Meeting Info

    (dismiss this notice by hitting 'X', upper right)

More on Carbon Fiber Whisker Pole Failure

Christian Williams

E381 - Los Angeles
Senior Moderator
Blogs Author
One minute Before pole Break Hawaii 3.JPG...pole break Hawaii 3.JPG

Regarding the failure of my carbon fiber whisker pole under benign conditions, 15 knots, no fouls:

I heard recently from a senior director of engineering at a well known aerospace company, who took an interest in the question. Here are his remarks (He doesn't want his company associated with his comments on composites)


"Typically, failing at one end or the other is usually a sign that the truss beam isn't getting to act like a truss beam. For a long skinny tube to resist compression, it has to be free of loading at either end, which leaves it in "pure compression" or "pure tension"...
pure tension.jpg
"If in fact one end is clamped, then any out of plane motion at the far end will also produce bending in the beam, which will be highest at the clamped end:

after clamped end.jpg

"So it could be that your pole temporarily bound up at one end, and then any "jostling" at the other end would have produced failure in bending near the bound end.

"It's also very possible that the pole wasn't bound up at either end but just went into buckling, and failed wherever it failed. I've attached an Excel spreadsheet with an educated guess at your whisper pole using three materials- Al, steel, and CF laminate. (The CF laminate data by the way is from https://agate.niar.wichita.edu/ which will tell you more than you will ever want to know about the behavior of any composite- make sure to look at "laminate" properties since that's what you'd find in actual products.) Note that this is an idealized model of a pure continuous tube, not a nested set of tubes.

"Looking at your video it seems pretty clear to me that the CF laminate pole was under a serious axial load and then took a good jolt in the off-axis direction that turned your straight pole into a bow:

after into a bow.jpg

"...and the bow of course has much less ability to resist the continuous compression load, which made it bow more, which made it weaker, etc, etc, etc. A few hundred milliseconds later, disaster. That low Euler margin would mean the pole was ready to buckle if it had either a sufficient jolt or any manufacturing errors of any kind."

Materials comparison:

Excell Capture.JPG
 
Last edited:

Christian Williams

E381 - Los Angeles
Senior Moderator
Blogs Author
Control lines:

The outer end of the pole is held in position by the sheet (aft pull), the harness lift (vertical support) and the downhaul (which prevents the pole from rising). The inside end of the pole is attached to the mast with a loose fitting which allows rotation and never jams.

Such a rig is supposed to hold the pole in place such that the forces are only compression, not bending. The rig was set up that way, hard, when it broke, apparently from crippling buckling, the "dimple fail."

Correction invited.
hwhisker pole attachments.jpg
 
Last edited:

goldenstate

Sustaining Member
Blogs Author
I found myself happy with my aluminum Forespar spinnaker pole used to wing out the jib.

Once the topping lift is attached, the weight of the pole can be managed with one hand pulling down on the lift line, and the other to open the pole jaw and attach to the ring.

My jib is 90/100% and the spin pole is J length +/- so if I was trying to wing out a big 140 genoa it would not work as well. Given the potential for failure, however, an aluminum whisker or spinnaker pole seem like good options for an 80's cruising yacht in non-racing circumstances.
 

Kenneth K

1985 32-3, Puget Sound
Blogs Author
So, if the "checkerboarded" force (last diagram) in the center of the pole that "[caused] a good jolt in the off-axis direction that turned your straight pole into a bow," was nothing more than the mass of the pole itself, could that have caused the failure?

Yeah, prolly...

Wiggle one end of a 10ft x 3/4" PVC tube up and down a few inches and you'll cause a good wave/bend/deflection in the middle of the pipe. Now do the same to a rigid tube--the same forces (and tendency to bend/deflect) are present, but the pipe is rigid enough to resist the deflection so you can't see the internal bending forces at work. Now put the rigid pipe under a good deal of compression while wiggling it. A wiggled, compressed pipe will buckle before a stationary compressed pipe will.

Even if the jib sheet, sail, topping lift, and downhaul are pulled tight, there's likely still a good many inches of play at the end of the pole. And that's just movement of the pole end with respect to the boat. The boat itself is rolling side-to-side while rising/falling on the waves. A two degree change in heel angle would try to move the end of a 15 ft pole more than 6 inches through space (no, the ring fitting between the mast and pole won't transmit this rotation to the pole, but the tug of the topping lift/downhaul will. And, since the lift & downhaul pull inwards toward the mast (versus 100% vertically), a tug on either further compresses the pole as it pulls the pole end through space).

So yeah, likely the end (or ends) of the pole were jolted (accelerated) in one direction while the mass in the center of the pole was still being jolted (or at least displaced) in the other direction (due to a previous force), all to an extent great enough to cause the pole to buckle. This explains why you might have sailed for hours under similar conditions without failure until an unlucky combination of opposing and accumulating forces just happened, in one split second, to exceed the limits of the pole.
 
Last edited:

Loren Beach

O34 - Portland, OR
Senior Moderator
Blogs Author
Having had a pole end deform on our previous boat, I kinda wonder if there could be a slight mismatch with the jaw and the eye. I recall that when a couple we know was outfitting their Valiant 40, they were advised to use a fitting that self-aligned the forces. This is a socket fitting on the pole end that locks onto a mast toggle pin (the rigger had somewhat rude anatomical name for it, as I recall). :)

When/if the force on the present ring fitting gets slightly off center from the direction of the pole, the jaw can jam rather than maintain a straight alignment, and then load one side of the pole. If that happens the fitting can bend or break, or try to force a curve into the pole, which would bend or break the pole.
I certainly do not have enough experience to have real value to my opinions, but this type of force and vector problem does make sense to me.
 
Last edited:

Christian Williams

E381 - Los Angeles
Senior Moderator
Blogs Author
--As Loren suggests, a socket/pin fitting is recommended for poles like mine, rather than my UXP (piston) fitting. But I've never had an issue, and in this case the ring fitting showed no damage. I could not reproduce any jam factor, and the pole did not sky.

--The most persuasive explanation for me is a temporary/instantaneous flexing with complicated dynamics that, as Ken mentioned, "exceeded the limits" of the carbon fiber tube.

New theory/recognition:

The break occurred about where the collar on the small tube was located inside the larger tube. The sliding collar exists to equalize the two diameters, such that bending the small tube exerts pressure on the large tube at the collar. I now think that is why the pole broke there, rather than in the middle, or at the point where the small tube enters.

Did I have the lines set up too hard (no slack)? Perhaps a little sloppiness in the rig would have tolerated sudden forces better. In this setup the only control line that gets winched is the sheet, but with a downhaul led forward to a bow cleat, and a pole lift secured, such winching really locks the outer pole in place and creates a lot of compression.
whisker pole collar.jpg
 
Last edited:

Joliba

1988 E38-200 Contributing Member
To me it seems that the failure was most likely due to one of or a combination of a couple possible problems:
1. The UXP fitting bound on its ring. This can happen when the shoulder of the hook flips under the ring when the pole bounces and rotates in just the wrong way. (I get this often with the piston snap hook and ring on my dog's leash fitting.) It only needs to be momentary and would not create any visible damage to the ring.
2. There is definitely a stress riser at the point where the the break occurred due to the collar of the small tube section located there. Either a small manufacturing defect at that spot or a fatigue crack that developed over time could predispose to the failure ultimately caused by the normal micro-motion flexing of the pole. (We see this type of prosthetic hip stem failure in a spot where there is a stress riser and no space for anything more than repetitive cycles of micro-motion.) The internal collar itself may also have caused the initial scratch that propagated unseen over time.
Mike
 

Kenneth K

1985 32-3, Puget Sound
Blogs Author
The problem with my "buckled under its own weight due to motion" argument (post #4) is that it requires fault in the engineering design. Rather than implying so, I was staying consistent with aerospace guy's assumption--that no bending occurred, and his conclusion--that the pole had a very low safety margin against buckling (154%). If those are true (and we know the pole's failure was true), then buckling under its own weight due to motion seems the only remaining cause. However, since you weren't out sailing in a hurricane, the forces you encountered (and much higher) should have been design requirements considered by, and accounted for, in Forespar's design. The pole's design should prevent it from buckling under its own weight due to normal (expected) motions.

But if you forego the assumption that no bending occurred, because, say, of the common occurrence of binding at the ring-and-piston fitting, you now have a likely explanation of the collapsing "jolt" that broke the pole, without implying an engineering flaw. It's seems the most likely explanation.

This doesn't discount what the aerospace guy said. It's likely that the pole still failed due to buckling (at a force below its pure buckling limit) when a bending jolt shook the pole out of column.

I bet the Forespar guys would like this explanation better as well. And, for another $1K, they'll happily sell you a socket & pin fitting to ensure it doesn't happen again.
 
Top