New Mast Tangs

[RichH]

Roger
That your present tangs have lasted this long only means you havent had 1M load cycles over 30ksi ... and therefore have remained 'under' the fatigue endurance limit (~30ksi).

Safety Factor of 3 is common for ocean boats; 90ksi UTS for normal 300 series stainless divided by FS 3 'usually' insures that one doesnt exceed the endurance limit of ~30ksi.

.... But, but, but, we all know that common rigging does eventually fail via fatigue; so, intuitively that FS should be increased - simply based on the quite common eventual failure of rigging. If the widespread rigging never or rarely had to be replaced due to fatigue then we wouldnt need to have such discussions.

As simple as I can put this: The fact remains that current FS=3 rigging does fail; therefore, more careful design a change of rig materials ... or increased FS is in order. If 'aviation'
had the relative high failure rates as 'sailboat rigging' ... there would be massive correction imposed.

BTW - Ive gradually increased all non-wire components where possible on my deep water rig to FS --->~4: plates, tangs, t-bolts, etc. etc. as I have had failures when I 'proof-loaded' them, thankfully at the dock and not well offshore. I proof-load my rig to 70% every ~two years.

 

[Ross]

Some stainless alloys and some bronze alloys can be cast into any shape you like. They can be heat treated for stress relief They can also be forged successfully. They can be designed to meet the strength requirements for the job. If everybody does their job well the product will perform as designed for as long as needed.

 

[Bill Roosa]

Has anyone calculated the loads?

Design by consensus in not a valid method. It looks like it will work is not a valid design method. It's bigger than the old one so it will work is not a valid design method.

Figure out the max load the part will see,. Include partial failure of the rest of the rig and what that will do to this part. Also consider designed failure. i.e. the chain plates should fail before they rip off the deck. I'd rather lose the rig and have a watertight boat than loose a big piece of deck (and the rig) in a bad blow. An over design in this particular example would be a very bad thing. Calculate the area needed to resist that load (all three loads BTW) in all the various failure modes. Here is where you get to include cyclic loading. Factor in the appropriate safety factor. Draw up the part.

Ross indicated that laying her on her beam ends was the max load you could subject her to. I disagree, the boat has inertia and in a massive wind gust the keel will need to be accelerated AND lifted. Not a static situation. There is also the consideration of another part failing and the load being transferred to this piece. What does that do to the rig (and boat) overall. Perhaps you want this part to fail first to preserve a part of the rig so you can have 'something left' after everything has gone pear shaped!!! This is an excellent example of how building it bigger and stronger is not always a good thing.

I would submit that the original design was thought about by someone who did just what I described above and is "flexy' for a reason.

Hey Roger, you have a line drawing of the rig and the boat specs (ballast, center of buoyancy location etc). For a small fee I'd be willing to "put an engineer’s eye" on it. I work for beer.

 

[Bill Roosa]

Backward engineering

An alternate method (and less preferred) is to figure out the load from looking at the old part and duplicating it (no increase or decrease in capability). If you feel the designers where "silly" in how they put the part together you could address that. Most likely they had cost and ease of construction (labor cost) as a critical controlling element.
I would note that you are accepting risk here. If parts of the rig were designed to fail after 145,456,873 cycles so that the failure is "gentle" and gives the owner a clue that the rig needs some attention then a design that will take 2,000,000,000 cycles is going to fail in a less pleasurable way.

Example, there are three stays coming out of the chainplate attached to the deck. I'd design the lower stay tang hole to tear out from fatigue before the uppers so that the rig will still be standing after the break and the owner can unload it and motor home to fix. Assumed here is a fatigued part will break at a lower than ultimate rig loading. If the owner went and "beefed up" the part because it "looked weak" the whole rig is now in jeopardy of failing in an uncontrolled fashion with possible loss of life.

With that said, a draconian maintenance regime can mitigate some of this. But you would still have to have an understanding of how the designer intended the rig to fail. Alternately you could develop a failure pattern yourself and rebuild to that. But then you are just doing what I indicated in the previous post so you might as well start there.

I cannot stress strongly enough that “making it stronger” is not a safe way to design.

 

[Roger Long]

For a small fee I'd be willing to "put an engineer’s eye" on it. I work for beer.

No line drawings but stand by for an upcoming post.

Actually, I've designed a number of rigs including this one in complete detail right down to the shipyard drawings for all of the spars, line leads, and fittings:

and this one:

and a bunch of yachts.

I also did the indeterminate computer structural analysis (with a program I wrote) for the Coast Guard SSV approval of the Frigate Rose a.k.a. Surprise in "Master and Commander" and worked on SSV rigging standards with USCG MTH as chairman of the ASTA Technical Committee.

I'm not designing by consensus here, just having fun and stimulating some discussion I hope will be useful.

Regarding weak points and failure modes, masts are usually designed to be close enough in column compression that failure of a rig not compromised by in-service fitting weakening, say during a wave induced roll over, is probable enough in that portion of the structure that I don't see any point in making part of the standing rigging intentionally weaker than I could. You have to introduce significant weak points to insure failure at a particular point.

Thanks for the offer though. If you are ever in Portland, I'll still buy you that beer.

 

[Roger Long]

Whenever I have designed rigs, I have had stability data for the boat. It’s not always understood but hull stability is the design basis for rigging; not wind speed (I’m sure Rick and Bill know this). I don’t have stability data on my boat but I can make some reasonable estimates based on observed performance and other vessels.

Strider heels about 20 degrees in about a 15 knot wind. She is on the tender side and likes to heel but has a hull form which performs better at these angles than modern, flat floored hulls. Typical design practice back in slide rule days was to assume 16 knots of wind prior to applying safety factors.

From the calculations in my reefing post

http://forums.sbo.sailboatowners.com/showthread.php?t=123499&#post757681

this would indicate a heeling moment of about 7,200 pounds. Displacement is about 12,500 pounds so the righting arm is 7,200 / 12,500 = .58 feet. GM at 20 degrees would thus be 1.59 feet. This is pretty low by modern standards but she is a heavily constructed boat with a high volume keel. Modern hulls would have much higher GM but lower displacement for similar righting moment.

If the reported range of maximum stability of 114 degrees is accurate (it seems about right for the hull form) I can draw a rough righting arm curve. This shows a righting arm of .73 feet, equal to a righting moment 9,125 pounds at the limit of normal service (deckedge angle, 30 degrees heel) where cyclical loading is experienced and about 1.12 feet at the peak, a knockdown condition which would require the rig to resist 14,000 foot pounds of force.

The chainplates are 4.4 feet off centerline so chainplate load would be 2074 pounds at the limit of the range of cyclical loading and 3,182 maximum. It’s been a long time so I’ve forgotten the typical load sharing ratio between uppers and lowers on a single spreader rig like mine but I would like to think that the uppers could handle the entire load.

Industry standard breaking strength for 1/4” 316 wire is 6,900 pounds so I am currently at a F.S. of 3.33 in normal service if I neglect the contribution of the lowers. F.S. for knockdown is about 2.17. I’m going to go to 9/32 wire which riggingonly.com batch tests at 10,056 so my safety factors will be 4.85 and 3.16.

If I want to keep my chainplates below 30,000 psi in the heeling range of normal service, I need > .07 square inches of area. My chainplates are currently .25 thick with 7/16” of metal on each side of the pin hole so I have .219 square inches. To stay below 30,000 psi for the knockdown condition, I would need > .11 square inches so I’m still good. Even adding 20% for rig pre-tensioning leaves me with good margins.

Accepting Rick’s figure of quarter strength for a bent beam, I would need > .28 square inches in the bent part of the tang assuming every cycle is to the righting moment of 30 degrees heel. My 3/8” tang would therefore need to be > .74 inches wide. I’m planning on 1.5 inches so I think I’m good.

[FONT=&quot]I am going to 3/8 inch thickness and going to beef up my upper chainplates for a different reason. Turnbuckle and toggle jaws are 1/2” for the 1/2” pins I have and the 1 /4” inch thick chainplates thus put a bending strain on the pins. With so much slack in the jaws, there is also an often overlooked failure mode where the pin is held in the hole by the rigging tension and a sideways force applied to the fitting sheers off the cotter pin by sliding the toggle jaw over the clevis pin. I found one of my cotter pins broken this way when I took the rig down. There are some new clevis pins on the market with threaded heads, expensive, but I’m going to take a look at them. [/FONT]

 

[Bill Roosa]

Think you are missing the point Roger

Clearly you understand HOW to calculate the loads. That is usually the hard part to explain. I'd ask you how the rig will fail when it sees its failure load?
Are you left with a deck?
Are you left with a long or short stick, where does the mast buckle?
Do parts (please say individual shroud wire threads) fail without total rig failure? Giving advance warning that you need to unload before total failure.

This got drilled into me in college, design the building so that it fails gracefully and people have time to leave before total collapse.
Then in the Army they beat me up with PACE; primary, alternate, contingency, emergency (plans)

My concern is that just making part x stronger can actually change the failure mechanism. The mast top breaks off verses the mast buckles in the middle. The former leaves you with a longish piece to work with (assuming you can recover it from the water) the later leaves you with two smaller or one small piece to work with. I'm thinking that after I get de-masted I'd rather have the longish piece than the short one so I can jury rig something to get me to safety. The rig can be designed to fail that way and not the other by making parts of it stronger or weaker so they break first. That is called designed failure or failsafe. It gives you a second or even third chance to be stupid and live to tell the story. The rig being carried away during a capsize and the deck remaining intact being an excellent example. If you redesigned the chain-plates to be stronger, then capsize you could rip the deck off, get swamped in short order and end up in the life raft verses the rig is just carried away and you have a watertight vessel to haunker down in.

With that said, I'd just warn folks that real smart guys like Roger and Rick can get away with this stuff because, well, they are real smart. They have an engineering sense from doing engineering and while they don't specifically state that they considered xxx and yyy they did in fact give it some thought and determined it either didn't apply or was insignificant. Perhaps I'm being overly cautious about things or about encouraging folks to "fix their own boat." I think self sufficiency is one of the most important safety items you can get on a boat. I just do not want to see someone "fix their boat" and set themselves up for failure.

 

[Roger Long]

The rig can be designed to fail that way and not the other by making parts of it stronger or weaker so they break first. That is called designed failure or failsafe.

It's hard to do though. One of most instructive lessons I ever got in engineering was while working at Woods Hole Oceanographic Institution where they have a tension testing machine. Someone called me down and said, "Look at this."

He was breaking a bunch of bolts all purchased at the same time and presumably from the same production run. The aim was to design a weak link into an instrument mooring system so they could get most of the gear back if it was overstressed. The breaking strength of the identical bolts, being stressed as consistently as a hydraulic machine can do, was all over the place. This was long before China started flooding the world market with cheap stuff and things like bolts were still made in the good old USA.

Factors of safety are applied largely because of this kind of uncertainty. The problem with trying to design in failure modes in weight critical structures like aircraft and boats is that the weak links have to be the strength you want and everything else has to be oversized by additional amounts which are an appreciable portion of the factor of safety to insure that the break will occur where you want it. Beefing up the weak link often then becomes trivial from the weight standpoint and why not have the whole thing that much stronger?

You're fooling yourself if you think there was any careful design of my boat. I doubt there were many calculations done at all. Like 95% of production cruising boat builders, they just threw together components that were about the same size as used on similar boats. Engineering is expensive and there is very little of it in fiberglass production boat building.

Oh yes, they gave up on trying to put the weak link in the mooring system for the reasons above.

 

[Ross]

When I was in the Air Force we received several hundred bolts for Rockwell"C" hardness testing. I was in the welding shop in those days. You can get a fair idea of tensile strength from the hardness in heat treated alloy bolts. As Roger stated the hardness was all over the place. None of them were too hard but some of those bolts were still in the "as machined" hardness and hadn't been heat treated.

 

[Roger Long]

As I stated in one of the "Titanic" shows, "Any idiot can make something strong enough, engineering is the science of using as little material as needed."

Even when a structure is not weight critical, more material usually means more cost. The idea failure mode for a structure is for it to vanish completely in a puff of fine powder because every particle experienced failure at the same instant. You then know that you weren't carrying around anything extra and the structure was as strong as it could be for the amount of material used. Obviously, this idea is never achieved in practice and lots of good and practical reasons for making some components stronger than this ideal would indicate.

 

[RichH]

Ross indicated that laying her on her beam ends was the max load you could subject her to. I disagree, the boat has inertia and in a massive wind gust the keel will need to be accelerated AND lifted. Not a static situation. There is also the consideration of another part failing and the load being transferred to this piece. What does that do to the rig (and boat) overall. Perhaps you want this part to fail first to preserve a part of the rig so you can have 'something left' after everything has gone pear shaped!!! This is an excellent example of how building it bigger and stronger is not always a good thing.[/FONT][/COLOR]

I would submit that the original design was thought about by someone who did just what I described above and is "flexy' for a reason.

Hey Roger, you have a line drawing of the rig and the boat specs (ballast, center of buoyancy location etc). For a small fee I'd be willing to "put an engineer’s eye" on it. I work for beer.

Ross is probably correct on this one. When one 'back-calculates' proven designs it appears that the designer mathematically pulls the boat over from the top of the mast to a 45° heel angle (to ascertain the max. trigonometric rig cap shroud load) about the metacenter, then arrives at the calculated force (imparted rig tensile load to do this) THEN multiplies by the nominal (from historical scantlings for the intended service) Safety Factor to cover the 'unforseen and unexpected' loadings (though/ by scantling history).

When I back-calculate Bob Perry, etc., I keep coming up with a FS of close to 3.5-4 for his rigging designs and the 'numbers' always seem to 'stack up'. When I back-calculate 'coastal designs' ... FS less than 3; inshore designs FS @~ 2; 'gonads to the wall' race designs FS = 1.5.

 

[RichH]

It's hard to do though. One of most instructive lessons I ever got in engineering was while working at Woods Hole Oceanographic Institution where they have a tension testing machine. Someone called me down and said, "Look at this."

He was breaking a bunch of bolts all purchased at the same time and presumably from the same production run. The aim was to design a weak link into an instrument mooring system so they could get most of the gear back if it was overstressed. The breaking strength of the identical bolts, being stressed as consistently as a hydraulic machine can do, was all over the place. This was long before China started flooding the world market with cheap stuff and things like bolts were still made in the good old USA.

Factors of safety are applied largely because of this kind of uncertainty. The problem with trying to design in failure modes in weight critical structures like aircraft and boats is that the weak links have to be the strength you want and everything else has to be oversized by additional amounts which are an appreciable portion of the factor of safety to insure that the break will occur where you want it. Beefing up the weak link often then becomes trivial from the weight standpoint and why not have the whole thing that much stronger?

You're fooling yourself if you think there was any careful design of my boat. I doubt there were many calculations done at all. Like 95% of production cruising boat builders, they just threw together components that were about the same size as used on similar boats. Engineering is expensive and there is very little of it in fiberglass production boat building.

Oh yes, they gave up on trying to put the weak link in the mooring system for the reasons above.

In my 'youth day' I did this testing (Tinius Olsen - the then prime mfg. of such testing equip) and will relate that MILL CERTIFIED materials are usually quite uniform in tensile, etc. properties. I will also state that the 'rate of load application' is one the principal reasons for non-uniformity of tensile properties even in MILL CERTIFIED materials.
Roger makes the statement of 'chinese-asian' materials .... indeed they are CRAP but virtually noone makes anything in the USA anymore ... so you really HAVE to increase the Safety Factors when designing/engineering structural components made from CRAP.

 

[Bill Roosa]

The differance between a mechanical and civil engineer

Not being able to recover a piece of test equipment is significantly different than having a building fall down and kill everyone in it.

A word to the wise is sufficient. For they unwise this is an example of natural selection removing your genes from the pool.

See you on the waters

 

[Ross]

Some one could do a complete data page on the physical properties of CRAP. Others would probably look at it and say that it stinks.

 

[Roger Long]

Ross is probably correct on this one.

He's not correct about maximum load being at 90 degrees knockdown though, at least in most normally proportioned vessels. Maximum rigging loads will be at the peak of the righting arm curve which is typically in the 35 - 50 degree range.

"Traditional" yacht rigging design distributes the sail force along the main sail luff. This doesn't quite makes sense with triangular sails but, because of wind gradient, is probably closer than you would think. Jib loads are added as if the entire jib was pulling on the attachment point at the head of the stay 90 degrees to the mast. This doesn't make sense either but is one of those things that seems to work. Loads are assumed to be 1 pound per square foot about 16 knots of wind (according to a since discredited antenna design source copied by yacht designers until it became gospel) or the wind load that will produce 20 degrees heel if the vessel's actual righting moment is known.

There's a lot of eye of newt and toe of frog in this as you can see.

 

[RichH]

Clearly you understand HOW to calculate the loads. That is usually the hard part to explain. I'd ask you how the rig will fail when it sees its failure load?
Are you left with a deck?
Are you left with a long or short stick, where does the mast buckle?
Do parts (please say individual shroud wire threads) fail without total rig failure? Giving advance warning that you need to unload before total failure.

This got drilled into me in college, design the building so that it fails gracefully and people have time to leave before total collapse.
Then in the Army they beat me up with PACE; primary, alternate, contingency, emergency (plans)

My concern is that just making part x stronger can actually change the failure mechanism. The mast top breaks off verses the mast buckles in the middle. The former leaves you with a longish piece to work with (assuming you can recover it from the water) the later leaves you with two smaller or one small piece to work with. I'm thinking that after I get de-masted I'd rather have the longish piece than the short one so I can jury rig something to get me to safety. The rig can be designed to fail that way and not the other by making parts of it stronger or weaker so they break first. That is called designed failure or failsafe. It gives you a second or even third chance to be stupid and live to tell the story. The rig being carried away during a capsize and the deck remaining intact being an excellent example. If you redesigned the chain-plates to be stronger, then capsize you could rip the deck off, get swamped in short order and end up in the life raft verses the rig is just carried away and you have a watertight vessel to haunker down in.

With that said, I'd just warn folks that real smart guys like Roger and Rick can get away with this stuff because, well, they are real smart. They have an engineering sense from doing engineering and while they don't specifically state that they considered xxx and yyy they did in fact give it some thought and determined it either didn't apply or was insignificant. Perhaps I'm being overly cautious about things or about encouraging folks to "fix their own boat." I think self sufficiency is one of the most important safety items you can get on a boat. I just do not want to see someone "fix their boat" and set themselves up for failure.

To put this another way, you dont the design to be 'so perfect' so that EVERY SINGLE component fails 'simultaneously'.

Engineering, whether formally gained in university or by the school of hard knocks, prevents one from using 'cook-book' solutions. A cookbook solution would be having a 9,000 lb. load hanging from a 1/10 inch square cross section of a 90,000 psi material. An 'engineer' is going to apply a Safety Factor to 'de-rate' the load bearing capacity of the material (90,000 psi / 2) etc. = 45,000 psi. to cover 'unforseen events' for 'safety'. ;-)

 

[Bill Roosa]

Using CRAP

I'd submit that all you have to do is understand "how crappy" the material is and design for the least common denominator. Is it "strong crappy" or "weak crappy?"
In Civil engineering we literally work with materials that we have no way of knowing what their specific properties are. If we dig up the foundation material to measure it that changes its properties when you put it back and you still don't know what is 1" beneath the stuff you dug up. That is why statistical analysis was developed.

There comes a point in every project where you have to kill the engineers and just build it.

 

[Bill Roosa]

Rich is right but...

You can design structures so that they break where you want to give you a fighting chance to live to tell the story.

 

[RichH]

There's a lot of eye of newt and toe of frog in this as you can see.

Thats why someone who commands such expertise and knowledge, gets paid the mega-bucks, .... usually not because most of them work in the same office as "Dilbert".
:-D

 

[tommays]

When its all said and done after all the FEA computer work they still go out and make aircraft parts fail in testing before they turn them lose on the public and even after that they still get it wrong a bit too much lately