Radar Arch Fail - Cautionary Tale

[Matt Johnson]

Originally Posted by Sandero

If the bolts sheared then the forces on them were too much. This means the engineering was wrong. In addition to the static load (dead weight) you have dynamic loads created from the boat's movement and these sorts of loads are much greater than static loads. That is to say, this rig sitting at the dock would likely not fail. But take it out to sea in all manner of waves and you introduce forces which exceed capacity of the bolts. There are also different grades of bolts and you might have had the lowest grade which didn't help.

The design looks poor as well. The rear post is in compression and the forward one is in tension. Once the starboard side failed and the structure detached and fell... the remaining apparently pinned port connection was no longer in compression and the reduced weight was carried by the floating dinghy. So it seems.

Is there a way to make davits/arch cantilevered over the stern without the rear post in compression and front in tension? We're thinking of having an arch made and cant I see how you could do this.

Matt

[Steadman Uhlich]

Originally Posted by Sandero

If the bolts sheared then the forces on them were too much. This means the engineering was wrong. In addition to the static load (dead weight) you have dynamic loads created from the boat's movement and these sorts of loads are much greater than static loads.

That is to say, this rig sitting at the dock would likely not fail. But take it out to sea in all manner of waves and you introduce forces which exceed capacity of the bolts. There are also different grades of bolts and you might have had the lowest grade which didn't help.

The design looks poor as well. The rear post is in compression and the forward one is in tension. Once the starboard side failed and the structure detached and fell... the remaining apparently pinned port connection was no longer in compression and the reduced weight was carried by the floating dinghy. So it seems.

To Bill (the OP): Thanks for answering my questions.

Once again, what I am asking and thinking about is NOT any kind of criticism of you as a sailor.

I am looking at this simply as an example of a failure of a rig (of sorts) on a boat and wondering how to prevent that on my own boat (in the future). The discussion of the issues and facts and perhaps some supposition by members here may really help others avoid a similar failure.

_______________

I am no engineer, so take my comments below with a big splash of saltwater.
_____________

Over the last year I have been looking at arches used on boats as I search for a boat for myself. I have seen many different kinds and often they are heavily loaded. So the loading issue is a common one.

I have also considered buying an Atlantic Towers "arch kit" because they appear to be a good deal (less than custom) and from what I have read here on CF, they seem to have good reputation for working and fitting easily to the boats.

________________

But, looking at the side view of your boat, with the arch installed and seeing that angle (lever) and looking at the attachment points AND considering the loads when everything is on it AND considering the motion of a boat on voyages, it does look to me like that was not as robustly engineered as I would want at the attachments, or put another way, it looks (to my layman's eyes) overloaded for the structure as it was attached.

I quoted Sandero above, because I like (agree with) his points.

My assumption is that the manufacturer of a piece of boat gear may have marketed it as having a certain "load capacity" that is based on solely static tests or materials tests, but perhaps not on the finished product during any real extensive stress tests.

"Real world" conditions (such as a boat at sea, moving and twisting, shock loads, salt environment, wind, waves etc.) are the "real" test of anything used on boats.

_________________

If this arch was from Atlantic Towers (or another popular arch provider) I would hope they would send you a new one and modify their instructions for loads and attachments and consider modifying their bolt size or supplier etc.

But, if this is a very rare case (of failure) then perhaps it is an anomaly.

In any case, it is good that you shared the failure with us. Thanks!

[onestepcsy37]

From what I read here it seems that the weak point was the side mounting of the forward attachment points. This simply invited a shear problem. Had they been mounted flat on deck as the rear attachments were I don't think this would have happened. Assuming of course the bolts and backing plates were of sufficient size.

So arch buyers might want to insist on deck mounted only arrangements. For those of you not familiar with the CSY, all of their decks are solid - no core.

[UNCIVILIZED]

I'm not entirely sure how you want to handle dealing with the unit's maker on this. But one way or another, at a minimum, the bolts should be analyzed by a lab, to see what they're really made of & if they were up to spec.
I'm thinking that ideally, if it were me, I'd want the testing done by an independent 3rd party. Probably on my dime.

Has your insurer given you any course of action to follow? And when you talk to them, it'd perhaps be wise to hold your cards close/play dumb at first. Including leaving out anything about bolt testing for now.

Although, some of the testing "doesn't have to happen" right now. Meaning that there are some forms of NDT which don't require anything much being done to the bits of the bolts which you have left.
For example, just taking them to someone who knows metals, & having them take a good look at the grain structure in them at the sheer point, under magnification.
Have a friendly expert look them over on his "lunch hour", for cash.

FYI, there are NDT (Non-Destructive Testing) forums out there, which likely could offer a lot more insight than will be found on here. And push come to shove, I may have the addy for a couple. But they shouldn't be tough to find.
Here's one as a starting point, if you're interested. http://www.ndt.net/forum/ And there's lots of information on the topic(s) in lay-speak, out there on the web.

With regards to the bolt's failure, did they sheer at a point somewhere along their non-machined, cylindrical portion, or was the threaded portion loaded in sheer also, where they broke?
I ask as steel is notorious for being weaker when notched, which is a title that the threaded portion falls under. And while engineers know this, & account for it (assuming proper calculations), it can still be a problem.
One other thought is that stainless is prone to work hardening & becoming brittle over time, especially when subject to cyclical & or vibratory loading. It's molecular grain structure can literally change due to such.

Also, out of curiosity, how big was/is the bolt foot print? Meaning how far apart were/are they. As the closer they are together, the harder they're going to be working.

You might hit up a couple of the surveyors on here, in order to get some more information on the engineering & materials science involved in this. And a couple of other memebers who would know come to mind as well.
So getting some good answers shouldn't be too tough, knock on wood, as there isn't any overly fancy engineering involved.

I'm sorry to hear of your troubles, especially at what's definitely the wrong time. And I'll be curious to hear what you find out, if you've a mind to share it.

Good luck getting things sorted out.


PS: Like Steady Hand, I'm curious to know what the total weight of the items bolted onto the arch was. As well as that of the arch itself.

[montenido]

Sandero,
Thanks for your comment. If I understand correctly, most truss designs, of which this is one, require compression and tension for their strength. That is the design of a truss. I'm still leaning towards the undersized fasteners being the weak point.

Thanks for the kind words and other ideas everybody.

Cheers, Bill

[Valmika]

Was there any kind of galvanic corrosion on the foot where the bolts went through. I have been looking at these arches also and read good reviews but this did give me a lot to think about. My neighbor has one on a B423 and I think his bolts are larger, 1/2 to 5/8. I will look

[montenido]

Hi folks,
I have some weight information. There are 3 solar panels at about 33# each, plus the mounting hardware, say another 10#. The radar antennae is probably about 25#, I'm guessing. The weight of the tower is about 100#. My dinghy is 130+#. So, yes, it is a lot of weight, with about 150# of it up high. The company website says that the tower can support two people while they are installing equipment on top.

If I remember correctly, the mounting bolts are threaded their entire length, which would make them weaker to begin with. The company website shows all kinds of equipment mounted high up like mine, including davits, solar panels, and wind generators. They also show mounting similar to mine, with all 4 legs mounted on the outside vertical surface of vessels.

I have definitely had this set up out in heavy seas over the course of about two years. I did notice some flex side to side on occasion, but I figured that the design took this into account. There was no galvanic corrosion that I can see. I am leaving everything as is until the insurance guys have a look.

You all are bringing up some great things for me to consider, and I do not get offended or insulted easily, so don't worry about that. Thanks again for all the great input everyone.

I have contacted my insurance company, but have had no reply yet. I will keep you guys in the loop as things progress.

Cheers, Bill

[Roy M]

Have you found the underdeck portion of the fasteners? It could provide some more information, specifically, did they still have the nuts attached? Wiggling bolts have the ability to allow nuts to loosen and fall off. Were you using lock washers, nylock nuts or Permatex Red? Take a peek into the bilge to see if those parts are still down there.

[S/V Alchemy]

This happened to me in similarly "no big deal" conditions on my steel boat, which has welded plate with angled supports. One of the cast davits, holding a Zodiac RIB on two falls with no engine or fuel aboard (just oars and some line), snapped at its base. We were motoring at the time and we had to act quickly to keep the falls out of the prop and rudder. It took some time to rig a pole to bring aboard the RIB, which was still attached to the second, unbroken davit.

I went off davits at that stage. I carry my tenders (a nesting dinghy and a Portabote) lashed on the foredeck and cabin top, and the steel plate now has a deck crane on one side and will sport a wind generator pole on the other.

The davits in question were these Danforth types. Had I had davits previously, I might have guessed they were garbage, but they came with the boat.

[montenido]

Roy,
All fasteners used washers and nylox nuts, which are all still in place since I used 5200 to seal them. Nothing was loose, and the holes in the fiberglass are perfectly round, not oval, which would indicate wear and/or movement before the fail. If you look at the picture that just has the 4 holes and no pad on it you can see the sheared bolts still inside their round holes.

I am itching to get started on this but I think I need to let the insurance people take a look first. Of course, I have tons of pictures from start to finish of both the original installation and the failure at sea.

Cheers, Bill

[Roy M]

Bill, good followup, as it clearly eliminates a fairly common cause of failure. In your case, it will be clear that it was bolt failure. I'm assuming, based on your description of the holes and the presence of the nuts and washers, that no backing plate was present. Still, the nuts and washers held, therefore, only the bolt failure remains as the cause of the loss. Why they failed is still a mystery, but when you get your insurance settlement, you can have the replacement bolts sized upwards. Thank goodness no one was hurt, and everything was covered by insurance. Still, it was a disturbing failure. Sorry.

[wgoodbye]

Hi Bill,

Sorry to hear about your misfortune. I also have the same arch from the same company as yours. I had 2 x 250watt panels on the arch initially that weighed about 50lbs each and after watching the sway in heavy seas, I sold those and replaced with some thin panels from MiaSole that weigh around 8lbs each. My arch is mounted with all four posts on the stern, so the sheer forces might be a bit different than yours but now going to redrill the mounting holes and go up a size in fasteners just to make sure.

Chris

Here is a picture of my setup:

https://cti.ctera.com/invitations?sh...70d29b359&dl=0

[hamburking]

Originally Posted by Sandero

The rear post is in compression and the forward one is in tension. Once the starboard side failed and the structure detached and fell... the remaining apparently pinned port connection was no longer in compression and the reduced weight was carried by the floating dinghy. So it seems.

Agree. The rear posts would be in compression, the forward posts in tension. In either case, all the bolts would be in shear, since the mounts are on the side of the boat. Had the bolts been on a horizontal surface, in tension rather than shear, they might have sufficed.

The bolts were likely chosen adequately for their strength, but not for fatigue. The constant motion of the boat (IMHO) fatigued the bolts to the point of failure. Think of bending a paperclip back and forth...once for each wave...for years.

Overall though, it looks like a good design, with the one weakness being the bolt size. Yes, bolts come in different grades. My metal cradle uses grade 5 bolts. i was surprised to see that the entire weight of the pads passes through a total of 4 grade 5 bolts (5/8" I think).

I also agree with your conclusion...bigger bolts.
And your warning for others is appropriate as well...check your bolts, your pins, your shackles...check everything. Personally, I start at the bow and work my way aft.

Thanks for sharing your story, and for taking the time to get some excellent photos at a time when you had a lot of other things to do.

[gbgreen59]

I just finished installing a similar arch from the same company. Backing plates were provided with it. The bolts for the pads and backing plates were not provide and this is stated on the web site and in their video. They recommend 1/4" bolts. The inside of the deck/hull on my boat, where backing plates are mounted, is not flat. So a nice surface for the backing plate is not there. My plan is to fiberglass the area where the backing plate is mounted to make a nice fit for the backing plate and distribute the load around the backing plate. I may used G-10 as a part of this when I get to it.

Are there pictures available of the underside of the installation where the backing plates are (or should have been) located? It would be nice to see how flat the surface is where the mounts are located. Were backing plates used or maybe oversized washers? Was there a regular inspection and re-tightening of bolts? Just curious and gathering information for future reference.

[Cheechako]

Looks like a lot of weight on that arch. As mentioned the dynamic forces are much higher than just the weight. My guess is some bolts broke in your trip from Lapaz to SoCal. That trip likely had a lot of pounding I imagine?
Not sure I like the design having all the bolts in shear by attaching to the sides of the boat rather than the deck either. That puts the load directly on the bolts. Essentially the load is trying to twist the bolts off. Stupid design. On a deck mounting.... the deck is supporting a lot of the force and the bolts simply keep it all in place. (to some extent anyway)
Lastly, sometimes we have to use bolts that are threaded nearly to the top of the bolt, so they have no unthreaded shank. These bolts are terrible in shear. If you had those that's another reason.
Backing plates were not the issue unless you tore a chunk of deck out!