Davits, And Why They Fail

[GoneAstray]

Hi all,

Instead of replying to an old string on davits and having it lost, I thought I'd start a new thread and talk a little about davit welds.

This is just a little engineering perspective to the topic of davit construction.

First is material selection. For anything structural on a boat, I'd definitely use either 316 or 317. These two types or stainless steel are best suited for a corrosive environment. 317 is a rarer and used in manufacturing processes with highly corrosive materials, so 316 is the best approach with common materials.

Now comes a little engineering. Steel has a unique property in materials in that there is a stress range where the material will return to its original shape after being relieved of its stress. (i.e. it will "spring" back) Not only that, if the stress is kept at a point significantly lower than its maximum stress, it can spring back indefinitely (unlike aluminum). For any application where there is going to be millions of cycles of stress (bouncing around on a boat) you will need to design the system such that the steel never sees stress above the point where it can infinitely spring back. If it is above this point, after some number of cycles, a crack will develop (fatigue crack) which will continue to propagate until failure.

For most steels, a general rule of thumb is that the most stress that you put into the material and still have it return to it's original shape is around 60,000 lbs / sq. in. This is just a guideline; each type of steel has a unique property according to additives and heat treatment.

For sake of argument, most steels have a point around 15,000 lbs / sq. in. where they can have infinite cycle loading without failure.

The 15,000 lbs / sq. in. is a perfect specimen, in the real world we have different "stress risers" or dimensional changes that reduce the strength in a local area of the material. Stress rise factors vary from 0 (no dimensional change) to 3 (a notch). Since welded joints have microscopic inclusions, the safest stress factor to use is 3. That means, for infinite life, you should try to keep the maximum stress in a welded steel structure to less than 5,000 lbs / sq. in. (this notch factor is also the reason why stays usually fail close to crimps)

5,000 lbs / sq. in. sounds like a lot, but let's look at the application. First, I'll assume the entire load is carried by the bottom weld. (In actuality there is an upper support clamp, but I have seen many times davits with upper supports attached to less than rigid supports, so for sake of this analysis I'll assume the full load is carried by the bottom weld) The arm on the davit can hold 300 lbs. Let's just assume that in a heaving ocean, you will get about 2 g acceleration / deceleration as the boat goes up and down. That's now a 600 lb alternating force. Now we look at how far the load is carried from the base weld. In the davit originally talked about, this is 42", or 3 1/2 ft. The moment on the base weld is now 600 lbs X 3 1/2 ft or 2100 lbs.

That 2100 lb moment has to be held by the weld bead on the forward part of the tube (the rear of the tube is in compression). The tube is 1 1/4 diameter with a 32 mm wall, so the forward part of the tube has a weld bead approximately 2 inches long by about an 1/8 of an inch or about 0.25 sq. inches. That adds up to around 8400 lbs / sq. in., above our 5,000 lbs / sq. in. design criteria.

So what does this mean? It means that your upper brace is critical in protecting the full cyclic load from the base weld. Don't use an upper attachment point that can "give", i.e. not allow for a secondary load path. The load should ideally be shared between all points lowering the stress in any one weld.

What does it mean if the upper attachment is "springy" and the load is carried by the lower weld? The davit will definitely hold the loading of 300 lbs comfortably, but will not last indefinitely with a heavy dingy bouncing around on it.

To make your davits last indefinitely is really quite easy, don't overload it. That means keeping the hoisting weights below 300 lb limit and it means storing the dingy elsewhere if there are going to be prolonged periods in less than tranquil seas.

So for some last thoughts...

1.) Keep your davits loaded lightly and avoid high cyclic loads.

2.) Make sure that the welds at the base (where the stresses are highest) are BEEFY and fair nicely with big radiuses and no discontinuities (slag bumps, etc.)

3.) Make sure the weld is free of anything that will prolong exposure it a corrosive environment (no towels or rags that will remain wet wrapped around the base, that goes for stanchions as well). Once a crack starts, a corrosive environment will greatly speed its propagation and eventual failure.

4.) Put those base welds on all welded joints (not only your davits) on your monthly inspection list and look closely to make sure a crack has not started. Fatigue cracks close when not loaded and are hard to see.

My two cents worth and hope I didn't bore anyone...

All for now,

Jeff

[MMNETSEA]

Originally Posted by GoneAstray

1.) Keep your davits loaded lightly and avoid high cyclic loads.

2.) Make sure that the welds at the base (where the stresses are highest) are BEEFY and fair nicely with big radiuses and no discontinuities (slag bumps, etc.)

3.) Make sure the weld is free of anything that will prolong exposure it a corrosive environment (no towels or rags that will remain wet wrapped around the base, that goes for stanchions as well). Once a crack starts, a corrosive environment will greatly speed its propagation and eventual failure.

4.) Put those base welds on all welded joints (not only your davits) on your monthly inspection list and look closely to make sure a crack has not started. Fatigue cracks close when not loaded and are hard to see.

My two cents worth and hope I didn't bore anyone...

All for now,

Jeff

Jeff - Great - 2 cents ? That was 200 bucks worth of really good info

[Lighthouse]

Thank you for the very informative engineering "lesson" Jeff. Most appreciated.

[Trim50]

Jeff,

Since you're going to be welding...wouldn't 316L be a better material selection?

Ken

[GoneAstray]

Originally Posted by Trim50

Since you're going to be welding...wouldn't 316L be a better material selection?

Ken, great question. Here is my opinion... (I'm not a materials "expert" but just an old engineer with some practical application knowledge)

Bear with me because I'll delve a little deeper than my original post and it will get a little technical.

Let's look at the chemical composition of the two.

Type---C Max---Mn Max---P Max---S Max---Si Max------Cr-------------Ni------------Mo--------Cu Max

316------.08------2.00-----.040-----.030----1.00----17.00/18.00---12.00/14.00--2.00/3.00----.50

316L-----.03------2.00-----.040-----.030----1.00----17.00/18.00---12.00/14.00--2.00/3.00----- -

The "L" version only differs by the lower Carbon content. Since the corrosion resistance comes mainly from the Chromium and Nickel content, both types are basically equal in their corrosion resistance.

The lower Carbon content comes into play when the steel is heated within the Temperature Range of 800 - 1500 degrees.

At these temperatures, the steel matrix loosens and certain materials can propagate and some elements, such as carbon, tend to bind together to make little microscopic "pockets" of one element within the matrix. This "pocket" tends to be more brittle then the surrounding matrix which is more ductile. Under stress, this discontinuity between the two areas can cause the Carbon to "break away" from the surrounding matrix. This microscopic inclusion can cause high local stresses in the matrix and become the origin of a stress crack. Once a crack has started, even microscopically, the tip of the crack is always under much higher stress than the surrounding matrix, causing the tip to crack further and move through the material.

Three major factors come into play when it comes to Carbon precipitation in the surrounding matrix, the first two being temperature and time. The third factor is "work hardening" which is when the material is put under extreme localized stress which opens the matrix and allows precipitation at lower temperatures. Many stainless steels are considered non "work hardening" because their low Carbon content does not allow a change in material properties after "work hardening" such as occurs in mild steel. None the less, it could allow for carbon precipitation in the matrix.

Now lets look at the application. There are only two times after the steel leaves the furnace that it is subjected to high temperatures and stress, when it is formed into it's final state for sale as a product, and when it is welded. The forming process generally does not alter the matrix appreciably because it is done in a fashion (usually with annealing, i.e. heating and slowly cooling the material to return the material to it's original matrix) so that the material properties are maintained. When welded, localized areas of the base material can have elevated temperatures, and in a perfect world, the weldment would then be annealed by heating it and slowly lowering the temperature to preserve the original matrix structure. In a welding application, the material is under high temperature for only a short period of time, and this steel has a fairly low carbon content to start out with, so the benefits of annealing are not large, but would be done in a perfect world.

An additional treatment after welding is passivation. To passivate the steel means to apply a chemical to the surface of the material which lifts off any pure iron from the surface, therefore removing microscopic areas of the surface matrix that is not uniform stainless steel. This is actually what you are doing to some degree when you apply stainless steel cleaners to remove rust spots. In all but a very few specific examples, the highest stress seen by a material is on the surface.

But all of this is a little digression, because in most applications, the weld bead itself is what is undergoing the maximum stress, not the base material. Since the material in the weld bead comes from weld rod/wire used in the welding process, that is the material that we must examine.

There are in general use three types of TIG wire used when welding stainless steel. Lets take a quick look at those.

Type-------------C-----Mn-----Si------Cr---------Ni---------Mo------Cu

18/8-308L-------.01----1.8-----0.4-----20-----10.00----- - ------0.10

18/8/3Mo-316L---.02----1.6-----0.5-----19-----12.00----2.8------- -

23/12-309-------.02----1.6-----0.4-----24-----13.00----- - -------- -

As you can see, all three of these have very low Carbon content specifically to avoid Carbon precipitation as high temperatures.

Since the weld material should match the base material as closely as possible for uniform material properties, the 316 weldment should be welded with the 316L wire. Though the 308L is not a bad choice, it has slightly less corrosive protection to salt.

So for a practical application where your davit is manufactured by a professional fabricator, your weld bead is in actuality probably 316L and you didn't realize it.

On another note, 316L is usually not made into pipes and tubes and to find tubular 316L would probably require a special order. It is generally only available in plates and billets. 316 is readily available in the market in most forms and dimensions.

So when is 316L used in the real world? It is generally used for weldments that will see extended periods of elevated heat and stress such as occurs in manufacturing processes for chemicals.

So in a nutshell,

1.) Your weld bead is most likely 316L to begin with.

2.) In a perfect world, your weldment would be annealed and passivated, but probably isn't. (and in this application would be a topic of debate whether it would be worth the extra effort for a marginal gain)

3.) Have another sundowner and take this one off your worry list.

Hope this explains the material selection a little better and didn't make your head hurt too much.

Jeff

[Trim50]

Jeff,

It was something of a leading question and your answer is superb. I've often had this arguement with engineers that demand that 316L be used for reasons mostly associated with hydrogen embrittlement. The same type of arguement goes for 304 and 316, but I'd alway pick 316.

Well written!

Cheers,

Ken

PS, My next project is Davits.

[GoneAstray]

Originally Posted by Trim50

....engineers that demand that 316L be used for reasons mostly associated with hydrogen embrittlement.

I don't like to bash on my fellow engineers, but I have seen a lack of understanding of steel and its properties by many new mechanical engineers. Us older engineers (I can't believe I'm saying that and I'm only 50) spent more time in material sciences and as new techniques in computer modeling and finite element analysis were developed the curriculum has changed. There is never enough time in a 4-5 year degree to cram as much in as one needs to know so I can't blame the universities or the students, but as an engineer you have to be very diligent to know what you DON'T know. Just ask me to fix the computer and you'll quickly find out.

I personally have only seen two types of cases of hydrogen embrittlement in my career with parts I was involved with, but each industry has its own Achilles Heel in this problem area.

(Hydrogen Embrittlement is where hydrogen goes into the steel, then combines together or with some other material making a "bubble" or inclusion that is high in pressure making the surrounding steel crack. A fatigue crack then eminates from this original microscopic crack)

The first is high strength fasteners that have been electroplated.

The second are axle spindles that have been repaired by electroplating a new surface and remachined and heat treated to the tolerances and hardness required for the bearing.

As a baseline, hydrogen embrittlement is generally only an issue in high strength steels that have a hardness greater than 30HRC and have been through some process that introduces hydrogen ions into the steel matrix. Stainless is generally not considered a high strength steel (because of its low carbon content cannot be heat treated) and is seldom used in an electroplating process or other industrial process that would induce hydrogen. Not to say it can't occur, but is generally not a design issue in typical applications.

I will also reiterate that all of these posts are made by an engineer that hasn't practiced in quite some time (I was moved up to management many years ago). So there are probably a few small errors here and there, but the gist of the conversation I feel comfortable as being true and factual.

For those few who actually interested in the subject of different materials and their properties, a few basic reference books I have in my library are:

Machinery's Handbook

This probably up to its 25th edition by now. It's the basic bible of mechanical engineering and is a great reference source for engineers and machinists (and those that are endlessly fascinated by the workings of the mechanical machinery of their boat ). A little expensive, but absolutely packed with information. A must have. (this reference book has cross references on fastners, stock materials, etc. so is great when you're ordering a replacement or "close enough" in a foreign port)

Steel and Aluminum Stock List and Reference Book - Earle M. Jorgensen Co.

This is an engineering properties reference guide from an old American steel manufacturer. It's been long out of print, but you can occasionally find it on EBay. A good buy if you work in metals.

Good luck on your davits,

Jeff

[Trim50]

Well, your PH stainless steels can be heat treated...but we could talk for days about all the various grades. I'm presently working with A286 which is a fun steel super alloy.

[MMNETSEA]

Jeff, Ken,

Have you the name of the Dye used in the Dye Penetrate Test for crevice/crack corrosion in 316 stainless steel? How is it administered in marine applications?

Richard

[GoneAstray]

Originally Posted by MMNETSEA

Have you the name of the Dye used in the Dye Penetrate Test for crevice/crack corrosion in 316 stainless steel? How is it administered in marine applications?

Richard,

There are a lot of choices out there, and a lot depends on the application. If the part is going to be tested in situ, then the choice may be different than in a manufacturing environment.

In a manufacturing environment, you would probably choose a flourescent dye while checking a part on your boat you would most likely choose a visible dye. The flourescent dyes give better indications, especially in small cracks, but you have to have a dark inspection area and a black light.

If you were going to test a part on a boat, then you have two concerns, how well can it clean up and not leave red stains everywhere and how environmentally nasty is the stuff. Luckily, there are many choices. Water soluable dyes may be easier to clean up than non water soluable, but you have to match the developer with the type of dye you are using.

For all practical purposes, you have two choices. Take your part to a shop that is set up for inspection and have them test it for you, or, buy a portable inspection kit and do it yourself. The actual procedure is not complicated, and as long as you are not looking for very small imperfections a do it yourself approach can be fairly easily accomplished. If you have never seen it done, search for some pictures on the internet so you know what you're looking at.

Do it yourself test kits run about $100 and contain small aerosol cans for cleaning, dyeing and then a developer. Cleaning the part is the most critical part of the procedure for good results, so don't leave out or under estimate this step.

On a warning note, if the part is critical, i.e. you can't stand a failure, I would take it to a professional and have it done. If it is a part that will just be an inconvience if it fails, than give it a shot.

Here is a website for a little more info.

http://www.ndt-ed.org/EducationResources/C...cc_pt_index.htm

I know it's not a direct answer, but this should get you going in the right direction.

Jeff

[MMNETSEA]

Hello Jeff,

Thanks for the good website.

The specific application was a PI dye to be used in-situ. Mainly S/S terminal fittings on standing rigging - eg : chain plates - bob stay fittings - swaged terminals. On the mast 'T' terminals and other forged terminals.

Generally, early signs of problems (rust colouring - hole and pin distortion ; wire strand breaks etc) are worth very close examination, it was with this in mind that my question was framed.

For some reason, Standing Rigging is not given the attention it deserves. Maintenance inspections are often cursory affairs.

My internet search was fairly unproductive - the factor of how hard to clean up after testing would probably be secondary to that of corrosion detection.

Richard