Radar Arch Fail - Cautionary Tale

[TacomaSailor]

From McMaster-Carr data sheet for 1/4x20 stainless screws

"17-4 PH Stainless Steel—Screws are as strong as Grade 8 steel and as corrosion resistant as 18-8 stainless steel. They have a minimum tensile strength of 135,000 psi and a minimum Rockwell hardness of C28. Material meets ASTM F593 Grade 630 and is magnetic."

Even if we reduce the strength based on the minor diameter of 0.1758" for 1/4x20 that only decreases the strength by 27% (shear strength is directly proportional to bearing area) we see shear strengths over 25,000 psi.

If we use the 5:1 safety factor we see a SWL of 5,000 psi for ONE of four bolts.

How would a STATIC 100 pound arch with 300 pounds of gear on it develop loads approaching 5,000 pounds? If the boat were subjected to motions that would impose those loads on the radar arch - then more than just the arch would come down. After all, the hull of the boat where the arch is attached is moving with the same loads on it as felt by the arch. What can accelerate the arch independent of the hull where the arch is attached?

The one dynamic load might be the dinghy swinging in two axis in a disturbed sea. But, I still do not see much of a dynamic load. My rough calculations is that the 6' lever arm of the arch 4' out to the dinghy hanger block, at an angle of 30 degrees, would multiply the effective weight of the dinghy by a factor of almost 3. So that 300 pound dinghy would put about a 900 pound load on the base of the arch. Still - not a huge load.

Crevice corrosion is the hidden killer - I was shocked at the crevice corrosion damage to my prop shaft inside the stuffing box.

I will be inspecting my Atlantic Towers mounting bolts!

[Cadence]

Originally Posted by TacomaSailor

From McMaster-Carr data sheet for 1/4x20 stainless screws

"17-4 PH Stainless Steel—Screws are as strong as Grade 8 steel and as corrosion resistant as 18-8 stainless steel. They have a minimum tensile strength of 135,000 psi and a minimum Rockwell hardness of C28. Material meets ASTM F593 Grade 630 and is magnetic."

Even if we reduce the strength based on the minor diameter of 0.1758" for 1/4x20 that only decreases the strength by 27% (shear strength is directly proportional to bearing area) we see shear strengths over 25,000 psi.

If we use the 5:1 safety factor we see a SWL of 5,000 psi for ONE of four bolts.

How would a STATIC 100 pound arch with 300 pounds of gear on it develop loads approaching 5,000 pounds? If the boat were subjected to motions that would impose those loads on the radar arch - then more than just the arch would come down. After all, the hull of the boat where the arch is attached is moving with the same loads on it as felt by the arch. What can accelerate the arch independent of the hull where the arch is attached?

The one dynamic load might be the dinghy swinging in two axis in a disturbed sea. But, I still do not see much of a dynamic load.

Crevice corrosion is the hidden killer - I was shocked at the crevice corrosion damage to my prop shaft inside the stuffing box.

I will be inspecting my Atlantic Towers mounting bolts!

Rockwell hasn't got a thing to do with tensile strength.
Given a magnetic stainless has more. I doubt the bolts in question where more than over the counter.





g

[UNCIVILIZED]

Cadence: As to the bolt material, my point exactly.

I did my calc's using 304/306 Stainless, as it's one of the most common alloys that one runs across in marine fasteners. Some of that is due to it's lower cost as compared to a lot of other alloys of stainless. But it also has a lot more flexibility, ductility, & is less prone to work hardening as compared to other stainless alloys, such as 17-4 PH. The latter being far more brittle, & also having some other, less than appetizing, limitations/properties. There's no such thing as a free lunch.

We can run the numbers all day, & hypothesize about what the bolts may or may not be made of. I'm just showing how the calc's are done, to the best of my knowledge.

If we want to play the tensile strength game, there are alloys of 17-4 PH which hover in the 250,000psi range. Although as to their suitability for this application I can't say.
I do know however, that alloys of 17-4 PH are a lot more prone to crevice corrosion (and subsequent catestrophic failure) than are some of the less exotic alloys of stainless. Which is part of why they're not the most common material to reach for when bolting things together, in & around the water.

TacomaSailor: BTW, going down in size from 0.25 to 0.1758, reduces cross sectional area by 50%. It's a squared function.

Also, guys, think about this. Who hasn't been onboard when for periods of time, especially when going to weather, that objects (& people) "levitate" & then crash back down with each wave cycle. Which in lay terms means cyclical loading of in excess of 1g, say 10 - 12 times/minute, given a 5 or 6 second wave period.
That kind of adds up over time if someone goofed on their engineering or materials.

Especially so, if something is on the end of a lever... like the dinghy in question. Just think on what motion feels like when you're at the top of the mast, vs. even half way down it. Let alone at deck level. HUGE differences.

[Cheechako]

I seriously doubt many marine bolts out there are 17-4ph. Most are 304 or 316. The shear strength is often about 60% of tensile strength. The cross sectional area of an unthreaded 1/4 bolt shank is about .05 sq inches. substantially less on the threaded portion.
.05 x ~65000 tensile strength x 60%= 1950# in shear. less in the threads.


If the arch had 350# on it up about 5 ft average (?) that's 1750 ft lbs... without movement. Movement could easily triple that.

[model 10]

One of the problems with these common bolts is the un-threaded portion is not the correct length for what you are doing. Good practice says; the un-threaded portion should extend to just beyond what is being bolted together then with the addition of up to 3 washers the nut. This way a shear load is on the solid portion of the bolt, not the threaded part. The threaded part is not made for a shear load.

[TacomaSailor]

It is hard to understand how folks here argue with engineering formulas and concepts.

The arch is static, as is every thing (radar, solar panels) attached to it. That weight is attached to the boat and does not generate any dynamic load just as a winch bolted to the deck does not generate a dynamic load.

The dynamic load is generated by the dinghy swinging. Using scaling and trig functions it is easy to determine the lever arm load of the dinghy hanging in a static position. The swing load is side to side - NOT up and down. A swinging side load on the arch would be a compression load on bolts as mounted on the boat in question.

I know the minor axis and can calculate the bearing surface of the bolt using the formula for the surface area of a cylinder. That is what I did. The minor axis of a 1/4x20 bolt is 0.1758". With 1" length of bearing surface (no spacer pad) - you get a bearing surface of 0.60 square inches - not 0.05" as mentioned above.

The tensile strength of 316 stainless is 81,400 psi. Plug that into the conversion of tensile to shear strength and you get
.60*81400*.3=14,652 psi for one bolt.

There are four (4) bolts on each foot giving a single foot strength of over 57,000 pounds.

How did an arch firmly mounted to a boat generate enough force to shear four bolts at 57,000 pounds? The arch does not move when the boat bounces, rolls, or yaws - at least mine hasn't in the last 10,000 miles.

Or, worse case how did that arch generate over 14,000 pounds to shear a single bolt? What could generate that dynamic load?

And, if the recommended 2" spacer was included in the formula the single bolt shear strength would increase to about 42,000 pounds with 316 stainless.

I have taken 35' free falls (generating about 4,000 pounds of force with a rope that stretches 10%) onto 1/8" bolts anchored in a cliff and not sheared the bolt - and we rock climbers really do calculate these things.

Engineering is a funny business - there are easily measured and calculated numbers.

I am not disputing the failure of the bolt - I am just pointing out that a 1/4" x 20 threaded, common/cheap, stainless bolt has way more shear strength than needed for this application.

As I said, I am not an engineer and maybe someone who is can critique my reasoning.

[model 10]

I thought he said they were mounted on the side of the hull? How could that not be a shear load?

[Djarraluda]

A lot of the stainless steel available for bolts is 304 (sorry, old rating but better known than the new numbers). Lots of folk replace this with 316 because it stays shiny longer, but they have not checked the strength.
Strength to handle sheer force is different, so it could be that the grade of SS has been changed. Must admit, the entire arrangement of plates and bolts looks a bit small for the loads involved anyway, so it may be worth investing in welding some plates on the fittings provided to double or more the area that spreads the load - it will also cover the old holes. That gives the metal to increase bolt sizes to say 6 x 3/8inch per pad.

[montenido]

Wow,
This thing really got going today! To further answer a few questions: Yes, there are backing plates on all four mounting pads. No, none of the fasteners worked loose, as they were sealed with 5200 and used Nylox nuts. I used thickened epoxy under the backing plates to fair the surfaces and make them even. I do not recall anything in the instructions suggesting not to mount the pads in a vertical position.

Do I think that the assembly would be stronger with two or more pads mounted on the deck? Yes, absolutely, I might go that way when I replace the arch. I had not considered corrosion, but I think it might be a possibility and I will check both the broken bolts, as well as those still attached. I am also interested to see if there are any broken bolts in the three remaining pads, and if there is any corrosion in those bolts. I will post any findings.

Thanks for the sympathetic comments, but I feel very lucky that the failure happened close to a marina, and that no real damage was done to the boat or myself. There are some really interesting ideas and suggestions here, and I thank you all for taking the time to post them.

I will be going with hardened anti-corrosion bolts when I replace this thing for sure. I also think that there is a fair amount of side to side flexion going on in a confused seaway, with or without the dinghy, due to the weight of the solar panels. Maybe I should tow the dinghy everywhere and only use the davits at anchor or in the marina .

Today I was able to remove the three solar panels from the arch, which is still attached to my dinghy in the slip next to mine. They are 175w BPs. Pretty hefty, but the arch is designed for loads like this aloft.

Thanks again for all the great input. Now go have a beer or cocktail, it's Friday night!

Cheers, Bill

[UNCIVILIZED]

I pretty much thought that this stuff had been covered, but for some reason my translation of engineering/math into lay-speak must be flawed.

Originally Posted by TacomaSailor

The arch is static, as is every thing (radar, solar panels) attached to it. That weight is attached to the boat and does not generate any dynamic load just as a winch bolted to the deck does not generate a dynamic load.
EVERYTHING on a boat is both subject to & contributes to, dynamic load. For if the boat is in motion, so is everything on it.

Take for instance a coffee cup.
If the boat's underway, & you set it down, does it stay in one place as you sail through 5' waves? Nope, as it's subject to a dynamic load - the ones imparted by; the boat's motion, the forces of the sea acting upon the boat & everything in it, & the law of conservation of energy, which says that everything at rest or in motion will stay that way unless & until acted upon by another force.

So... it requires force & or energy to stop said java mug from sliding onto the floor (deck). And both that energy being used to restrain it, as well as that which is trying to dump it on the deck are the same kind of dynamic loads acting on a dinghy hung in davits. Period.

For this last Volvo Race, when they were laying down the rules for sails, & sail handling gear, one of the more significant changes was the heavy shift towards the use of a lot more furling gear.
The reason being was that when working on the bows of such high performance boats, they were seeing vertical accelerations of up to 5g's at times (positive & negative). - Which obviously can make it both difficult & dangerous to do sail changes. Ergo, the rise of furlers relative to previous generationns of such boats.

The same things transpire on all boats, everywhere on the boat, to varying degrees. If you choose to believe that such loads aren't dynamic, then Sir Issac Newton wont be happy to meet you in the next life.
The dynamic load is generated by the dinghy swinging. Using scaling and trig functions it is easy to determine the lever arm load of the dinghy hanging in a static position. The swing load is side to side - NOT up and down. A swinging side load on the arch would be a compression load on bolts as mounted on the boat in question.
The dinghy doesn't have to move, relative to the mother vessel in order to be subject to movement. As the whole vessel, & everything on it is moving. And the mass of the dinghy contributes to the overall energy of things.
Ah, & the side to side motion would impart different force vectors on each bolt, & most of it/the force vectors wouldn't be compression.

There are four (4) bolts on each foot giving a single foot strength of over 57,000 pounds.
Okay, here, let's stop & use common sense for a minute. The total cross sectional area of FOUR 1/4" bolts is unarguably 0.19625 square inches.
So, given this, for them to have a combined strength of 57,000lbs, then the steel out of which they're made, would have to have a tensile strength of 290,445psi

Does that truly sound realistic? Or fall within the realm of common sense?

I am not disputing the failure of the bolt - I am just pointing out that a 1/4" x 20 threaded, common/cheap, stainless bolt has way more shear strength than needed for this application.
Have you ever seen what a cold chisel can do? Even to a fairly large bolt or piece of metal.
Now think of the edges of the plate through which the bolts pass as a big chisel, & the G-forces from the combined weight of the arch & everything on it as the hammer which strikes the chisel.

As I said, I am not an engineer and maybe someone who is can critique my reasoning.

One other thing which is easy to forget, but very important. With 4 mounting bolts per plate, it's a rare moment indeed when all 4 bolts are under the same amount of load, acting upon them in the same direction. Let alone a direction which is ideal to get the maximum theoretical load bearing strength out of them.

Rather, in the real world, it's likely that a couple of them are taking 75% of the load on the plate which they're supporting. And some of them are in sheer, while orthers are more in tension. Just like the loads on the different legs of the arch.

And here's one final common sense question regarding dynamic vs. static loads.
If there weren't going to be any dynamic loads of significance on the arch & it's payload, then why use 4 strong bolts Per Plate? For if such loads weren't present, then theoretically, a few #10 screws would be strong enough to do the job?


PS: Bolts don't get stronger just because you make them longer. Come to think of it little else does either. Quite the opposite actually. But I won't get into explaining that now.


montenido,
If you plan on putting a wrench (or any other tools) onto what's left from the arch, which is attached to your boat, without the insurance guys present at the time, might I suggest that you both document your actions on video, & still photography, as well as witnesses. Perhaps including a surveyor of whom the insurance company approves.
And then, from there, maintain a chain of evidence, much as you see them do in criminal cases. As in a situation like this, some forensic engineering & testing may be called for.

[hoppy]

Looking at the photos I can see that on one side, two poles broke 1 ft from the mount and on the other side one broke at the mount and on the other the bolts sheared off.

Everyone seems to be assuming that it's the bolts shearing off that caused the failure. Can the OP say for certain that this was the cause? Maybe the weld on the mounting bracket that broke failed and the bolts sheared when they started taking the full load on that side?

The design of having the mounts on the side of the hull seems wrong to me as a non-engineer.


Sent from my iPhone using Cruisers Sailing Forum

[Kenomac]

Went back and looked at the original pictures up close. Three corroded cheap (probably Chinese) crap stainless bolts of insufficient size. And one sheared bolt. Poor monkey bar design. Would you even allow small children to play on it? Let alone load it up with all that weight, dinghy, solar, radar etc. I'm sure the OP meant well and probably relied on the manufacturers recommendations and trusted the contraption would work for its intended use. Sad out come.... Terrible design.

Glad this was posted to warn others of possible design shortcomings and having the same trouble. Thanks for your post. To bad many tend to fault the end user rather than the lousy design.

We need more posts like this one on CF. I also try to post failings whenever possible to warn others.. I recently posted trouble we are having with a piss poor bilge pump set up. Hopefully, others will heed our warnings.

Ken

[Jon Eisberg]

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

Not to mention, such additional 'support' would likely go a bit slack when you might most desire it... Namely, when sailing to weather in a breeze, and you've cranked up the backstay tension to help flatten the main...

Oh, wait, what the hell am I thinking ??? Adjusting backstay tension ??? This is a CRUISING forum, after all...

;-)

[montenido]

I'm back to clarify another point. When the arch failed it failed on the starboard side first, which had the pad that sheared its bolts, and then one broken foot on the other pad. The cables from solar and radar enter the deck on this same side so initially the arch was held close to the boat by the cables. As everything was banging around I reduced speed to maintain steerage into the wind, then set about trying to secure the arch from being lost while also trying to get it loose from the boat. As I mentioned before, everything was floating on top of the dinghy, secured by the davit rigging.

The two port side legs were bent and flexing in the waves and with the motion of the boat, finally breaking like the afore mentioned paper clip that is bent too many times once I cut the cables on the starboard side. I was pretty sure that I was going to lose the whole rig, dinghy included, if the dinghy was holed in this situation. The dinghy never lost air, which is just amazing to me with all the sharp pieces banging into it.

As far as adding a 1" block under each foot, wouldn't that just create more of a shear situation with a longer lever? The boat sides are probably 3/4" or more of FRP, with thickened epoxy and backing plates inside. I think the mounting surface was plenty up to the task without mounting blocks, although I like the suggestion of having larger plates welded to the existing pads to increase the load spreading area. Now would the weld be a source of concern? The dinghy was being carried, but was strapped to the arch securely with very little movement. As I mentioned before, while under way in anything but lake conditions, I would think that most arches have some flexing and movement in all directions going on. Those of you with arches, take a close look next time you are out and see if this is true. It is true for mine.


Cheers, Bill

[Kenomac]

Originally Posted by montenido

I'm back to clarify another point. When the arch failed it failed on the starboard side first, which had the pad that sheared its bolts, and then one broken foot on the other pad. The cables from solar and radar enter the deck on this same side so initially the arch was held close to the boat by the cables. As everything was banging around I reduced speed to maintain steerage into the wind, then set about trying to secure the arch from being lost while also trying to get it loose from the boat. As I mentioned before, everything was floating on top of the dinghy, secured by the davit rigging.

The two port side legs were bent and flexing in the waves and with the motion of the boat, finally breaking like the afore mentioned paper clip that is bent too many times once I cut the cables on the starboard side. I was pretty sure that I was going to lose the whole rig, dinghy included, if the dinghy was holed in this situation. The dinghy never lost air, which is just amazing to me with all the sharp pieces banging into it.

As far as adding a 1" block under each foot, wouldn't that just create more of a shear situation with a longer lever? The boat sides are probably 3/4" or more of FRP, with thickened epoxy and backing plates inside. I think the mounting surface was plenty up to the task without mounting blocks, although I like the suggestion of having larger plates welded to the existing pads to increase the load spreading area. Now would the weld be a source of concern? The dinghy was being carried, but was strapped to the arch securely with very little movement. As I mentioned before, while under way in anything but lake conditions, I would think that most arches have some flexing and movement in all directions going on. Those of you with arches, take a close look next time you are out and see if this is true. It is true for mine.



Cheers, Bill

Bill,

Judging by your last post, it doesn't seem to me that you've learned from your arch experience. The one you purchased was crap, the bolts used were crap... it's easy and obvious to most respondents from your excellent photos, that three of the bolts did not sheer off, they were corroded. Your arch shouldn't be flexing that much under load, if it is more than just a small amount.... IT'S TOO WEAK, for the intended purpose.

If you mount it the same way or similar way a second time... Do you actually expect a different result?

Ken