Radar Arch Fail - Cautionary Tale

[zboss]

Originally Posted by Cadence

Supporting arch, davits, solar panels, radar, and dinghy from the mast head sounds like a accident looking for a place to happen. JMHO

That was not the only support mechanism. It was the usual arch config with the addition of the wires supports.

[Cadence]

Originally Posted by zboss

That was not the only support mechanism. It was the usual arch config with the addition of the wires supports.

I understand that. It looks like a stop gap for poor design.

[ka4wja]

Bill,
Sorry to read of your troubles!!!


But, I think these words are telling....in that I have sixteen 1/4" machine screws (bolts) holding just my small bow pulpit...they are tapped and screwed into glassed-in plates in the deck, along with fender washers and nylock nuts on the underside....that's four 1/4" bolts per each of the four legs...(maybe overkill for some, but works well for me offshore..)

Have a look here...











I'm thinking 1/4" is pretty small hardware for an arch...yes??


EDIT:
Just saw some other posts here saying this same thing....opps...

Originally Posted by montenido

It appears that all four 1/4" bolts holding one leg to the hull sheared off, causing the others to fail. The reason I am posting all of this is that I have a suspicion that one or more of the bolts might have already been sheared, as today's weather was no big deal. In fact, I would venture a guess that some of the other bolts on the other legs might be sheared.

The forces placed on those fairly tiny bolts holding your arch, even if there were 16 or more of them, would be pretty high....

Again, sorry for this unfortunate problem, but if you don't mind some unsolicited (and friendly) advice....I'd try bigger bolts next time....and maybe even look for some grade 8 hardware...



Fair winds...

John

[zboss]

Originally Posted by Cadence

I understand that. It looks like a stop gap for poor design.

How is this different from running back stays?

[sk Barnett]

Could it be that some of the nuts backed off the screws due to vibration in the hull? It appears from your photo of one of the mounting locations that one screw was in service and was twisted to failure, but the remaining three machine screw holes show no remains of the fastener. Have you checked in board for nuts and washers having fallen from the interior of the backing plate? Do you know it lock washers and/or loctite (blue) were applied at installation? Looking for what is left inside the boat (if you can get to it) might be helpful in determining the nature of the failure.

[techmans]

I know a little about steel; I studied it at university and practiced building nuclear reactor vessels and internals.From the picture of the broken bolts it would appear to me that only one bolt sheared off judging by the ragged shiny metallic surface .The other three had already given way as a result of crevice corrosion. Please note the dark fail surface on those and the blackened immediate surroundings of those bolts suggesting corrosion.
Although it looks that way by the way the arch is mounted those bolts are not really loaded in shear.They are primarily, by their tightening torque, loaded in tension.This tension load ,along with the mounting compound, creates the friction between mounting pad and hull which takes up the shear load.
This shear load is subject to fluctuation as a result of wind load and inertia forces.These forces barely have an effect on the bolt loads.
From an engineering point of view the mounting seems sound enough as long as the bolts can maintain the friction between mounting pad and hull.

[Boatguy30]

Originally Posted by Bill Seal

Not without rewriting the laws of physics. Ask Archimedes.

1/4" screws are too wimpy for anything requiring strength, the minor diameter of a 1/4-20 thread is only 0.188", 5/16 -18 is 0.240". I'd go with 3/8" hardware and heavy back-up plates.

care to share your engineering data? The sheer strength if 1/4" machine screws is more than adequate. Most winches up to around 40:1 use 1/4". 2000 pound of Genoa sheet load is far more than an arch leg SHOULD take.

[Cadence]

Originally Posted by zboss

How is this different from running back stays?

Do you use back stays to support something from flexing. I may be wrong but thought it was for the mast from flexing. I'm sure the architect didn't figure that in? I'd rather see the arch come down not it coming down taking the mast with it. Again. JMHO

[TacomaSailor]

I've had a Atlantic Towers arch on our boat for 16-years and a lot of very hard sea miles including two hurricanes and many 55-knot/12' breaking sea events. We have four big solar panels and we tow our dinghy from brackets mounted on the arch. I once towed that dinghy for an hour in 25-knots of wind while it was full of water due to a wave that pooped it.

But, my mounts are on flush on the deck so all bolts are in compression.

Atlantic Towers was quite specific - the support hardware has to be mounted on a flat, smooth DECK so the bolts will be in compression. Their instruction manual in 1999 did not permit mounting where the bolts would experience a shear load.

Did something change in the Atlantic Towers technical instructions?

[Steadman Uhlich]

Originally Posted by TacomaSailor

I've had a Atlantic Towers arch on our boat for 16-years and a lot of very hard sea miles including two hurricanes and many 55-knot/12' breaking sea events. We have four big solar panels and we tow our dinghy from brackets mounted on the arch. I once towed that dinghy for an hour in 25-knots of wind while it was full of water due to a wave that pooped it.

But, my mounts are on flush on the deck so all bolts are in compression.

Atlantic Towers was quite specific - the support hardware has to be mounted on a flat, smooth DECK so the bolts will be in compression. Their instruction manual in 1999 did not permit mounting where the bolts would experience a shear load.

Did something change in the Atlantic Towers technical instructions?

Thanks for posting the above comment. Your experience with the product would give me a lot of confidence in the product and brand as a good choice, if mounted as you say.

[TacomaSailor]

The shear strength of a bolt is calculated approximately as follows:
bearing surface area x tensile strength x constant

The constant for a fluctuating load is 0.3

Assume 1/2" of horizontal bearing surface and 1/2" diameter bolt which would give a little more than 1.1 square inches of bearing surface

The tensile strength of stainless 17-4 PH or A286 is 130,000 psi.

1.1 x 130,000 x 0.3 = 42,900 pounds

That is a lot of strength in a single bolt - and there were four of them

Feel free to correct the above calculations or measures - I am a statistician and sailor but not really an engineer - but I did a lot of calculations and measurements before I installed my Atlantic Tower arch. An engineer friend helped with all the calculations.

[Cadence]

Originally Posted by TacomaSailor

The shear strength of a bolt is calculated approximately as follows:
bearing surface area x tensile strength x constant

The constant for a fluctuating load is 0.3

Assume 1/2" of horizontal bearing surface and 1/2" diameter bolt which would give a little more than 1.1 square inches of bearing surface

The tensile strength of stainless 17-4 PH or A286 is 130,000 psi.

1.1 x 130,000 x 0.3 = 42,900 pounds

That is a lot of strength in a single bolt - and there were four of them

Feel free to correct the above calculations or measures - I am a statistician and sailor but not really and engineer

I thought it was 1/4 inch bolts. Practically, what difference does it make. They broke!

[TacomaSailor]

I need to clarify or correct my earlier statements. I dug out my old calculations and the installation manual that I used in 1999 to mount my Atlantic Tower arch.

The mounting bolts were 1/4" x 20 x 3.5" Grade 8 stainless.

The manual does show the mount on the external vertical surface of the hull - but with a serious spacer. The mounting plate is 1/4" thick and I'll assume the hull is 3/4" thick so assume the bearing surface with no spacer is one inch.

The instructions are very clear that a 2" thick spacer block is required. That thickness increases the shear strength by a factor of 2.78. This is due to the increase in bearing surface area from 0.88 square inches with no spacer to 2.45 square inches with the spacer.

Using my formula from above for the shear strength - the shear strength WITH the spacer block would be:
2.45 x 135,000 x 0.3 = ~99,000 psi

IF NO spacer were used the shear strength would be:
.88 x 135,000 x 0.3 = ~36,000 psi

Even with no spacer and four bolts that seems like a very strong connection for each of four legs. I doubt that weakness or lack of strength in the bolts is the cause of the failure.

[UNCIVILIZED]

Posted by Boatguy30
"care to share your engineering data? The sheer strength if 1/4" machine screws is more than adequate. Most winches up to around 40:1 use 1/4". 2000 pound of Genoa sheet load is far more than an arch leg SHOULD take."

There aren't a lot of 40:1 winches out there with 2,000lb leads on them, especially not from genoas. When you get into sheet loads like that, you want winches with close to double that power ratio. Especially as with the shock loads which sheet winches see, they'd rip winches that size out of the deck. Unless, that is, you have some monster over-sized backing plates, & a Herculean deck laminate.

So I'm curious as to where you're getting your numbers. Because, generally, to get those kind of sheet loads, you need a racing boat in the mid 40'ish size range, carrying it's #2 Genoa (say 750sqft) hard to weather, in winds north of 25kts. Which is NOT a chore that a 40:1 winch can even begin to touch.

The sheet load formula is Wind Speed (in knots), squared x Sail Area (in square feet) x 0.00432 = Sheet Load (in pounds). Harken
750sqft x 25kts (squared) x .00432 = 2,025lbs

Originally Posted by TacomaSailor

The shear strength of a bolt is calculated approximately as follows:
bearing surface area x tensile strength x constant

The constant for a fluctuating load is 0.3

Assume 1/2" of horizontal bearing surface and 1/2" diameter bolt which would give a little more than 1.1 square inches of bearing surface

The tensile strength of stainless 17-4 PH or A286 is 130,000 psi.

1.1 x 130,000 x 0.3 = 42,900 pounds

That is a lot of strength in a single bolt - and there were four of them

Feel free to correct the above calculations or measures - I am a statistician and sailor but not really an engineer - but I did a lot of calculations and measurements before I installed my Atlantic Tower arch. An engineer friend helped with all the calculations.

On this, from how I was schooled, the math is something like this:
(0.5 x diameter) squared x 3.14 x 0.85 x metal tensile strength = Breaking load.

The 0.85 is a constant used for bolts, due to their being notched, & thus not as strong as if they were straight cylinders, or smooth pieces of base metal.

The rule of thumb, strength wise, is that you want a safety factor of 4-5:1

And I'd bet a few thousand $, that those bolts were at best 304/306 Stainless. Which has a ballpark tinsle strength of 85k psi, depending on the grade, how it was machined, etc. As can be seen here
MatWeb - The Online Materials Information Resource

Using semi-exotic alloys for these kinds of apps is fairly uncommon I would think, as it's cheaper for the manufacturer to simply go with a bigger bolt, than it is to use those alloys.
And if such is in doubt, go & look up the price difference between 2 shackles, the same size, made out of the differing alloys of stainless in question.


So, back to the math at hand, AKA figuring out the approximate strength of the bolts.
If we skip the safety factors for a moment, & use TacomaSailor's assumed 1/2" of bearing surface, then.
(0.25" x 0.5") squared x 3.14 x 0.85 x 0.5" x 85,000psi = 1,772lbs/bolt

Where TacomaSailor's 1/2" diameter figure for the bolts in these calc's is coming from, I don't know.


If folks want to use stays in order to reduce some of the cantilvered hanging loading on the arch, then...
- Put a gusset/chainplate on the upper, forward section of the arch frame.
- Put a chainplate or reinforced (backed) padeye on the hull or deck, forward of the foot of the arch, by somewhere in the neighborhood of it's height+/-
- Run a stay, using standard rigging practices between said points, & tension appropriately

***NOTE*** - If you do choose to go this route, the whole load bearing dynamics of each leg of the arch will change. Most notably, that the stay will induce a good bit of compression loading in the forward most leg of the arch & it's attaching hardware.

[Cadence]

Originally Posted by TacomaSailor

I need to clarify or correct my earlier statements. I dug out my old calculations and the installation manual that I used in 1999 to mount my Atlantic Tower arch.

The mounting bolts were 1/4" x 20 x 3.5" Grade 8 stainless.

The manual does show the mount on the external vertical surface of the hull - but with a serious spacer. The mounting plate is 1/4" thick and I'll assume the hull is 3/4" thick so assume the bearing surface with no spacer is one inch.

The instructions are very clear that a 2" thick spacer block is required. That thickness increases the shear strength by a factor of 2.78. This is due to the increase in bearing surface area from 0.88 square inches with no spacer to 2.45 square inches with the spacer.

Using my formula from above for the shear strength - the shear strength WITH the spacer block would be:
2.45 x 135,000 x 0.3 = ~99,000 psi

IF NO spacer were used the shear strength would be:
.88 x 135,000 x 0.3 = ~36,000 psi

Even with no spacer and four bolts that seems like a very strong connection for each of four legs. I doubt that weakness or lack of strength in the bolts is the cause of the failure.

Stainless steel bolts do not have an indicated proof load. Grade 8 bolts have a proof load of 120,000 pounds per square inch. The minimum strength of stainless steel bolts is 40,000 to 90,000 psi. The minimum strength of grade 8 bolts is 130,000 psi. Grade 8 bolts are much stronger than stainless steel.

Read more : http://www.ehow.com/facts_7723407_stainless-vs-grade-8-bolts.html