How long will interior paint on steel last?

The correct term is endurance limit and the operative words in the definition are " smooth polished samples subjected to millions of cycles...the curve approximates a 50% probability of failure". For most mild steels, the endurance limit is ~ 28 ksi @ 10^7 cycles with a yield of 32-36 Ksi and an ultimate ~50-60 Ksi depending on form (i.e. plates, sheet, strip, as rolled, etc.). From the stand point of design, the endurance limit defines the life, indeed, the original reason for calculation of endurance limit was for locomotive rolling stock....i.e. how often do you need to replace the axles? Theoretically any crystalline metal will fail under minute loads if subject to enough cycles, but I have to ask the wife on that one as that is her field (Metallurgy).

Fatigue is NOT academic for small vessels, as the many failures over the yearts will attest. From a design standpoint, endurance limit is even more critical on smaller vessels than large ones. This is because the are more and proportionally larger openings and bulkheads/houses. Every one of these is a stress riser, potentially increasing the local load 200-500% over the nominal stress. This, coupled with less sectional moment, designing for cost, and weld stresses, make any vessel, especially a small relatively open hull, susceptible to fatigue.

Also note that every weld is a potential crack, and the selection of steel for crack growth propensity is an entirely separate issue.

A welded steel hull is fraught with problems, otherwise the classification societies wouldn’t spend so much verbiage on what you can make them out of and how you shall treat it.

 

Thanks. I love it when I learn something.

I can see how fatigue is an issue for, say, plastic boats. And I can see how fatigue might be an issue for a commercial boat. But I've never heard of a recreational steel boat having a fatigue problem--probably because electrolysis or corrosion does them in first. Correct?? Hence the importance of paint, foam, maintenance, and the like.

Or, should I favor thick steel plating over thinner versions, particularly if the boat is already a veteran of a circumnavigation, because of fatigue?

 

You can design for virtually infinite life but it comes down to a good analysis of the fracture mechanics and stress levels. If you build small boats heavy with due attention to welds you will not suffer any fatigue failure . Mild Steel uniquely does have a limit below which fatigue does not occur, this is indicated on material S-N curves. We use S-N curves for rough surface and even salt corroded steel for critical designs. The scientists use smooth polished S-N curves.

Mr Hardiman is correct in his critique of commercial shiping which is not commercially viable to build bullet proof but in a small steel well built hull the higher levels of stresses should not occur.

I have seen seen major fatigue induced failure in aluminium vessels especially in the larger heavier boats, split bottoms detached floors and fractured keelsons, but so far in steel fishing vessels the fatigue failures have been due only to heavy localised stresses in poorly built components and even then the repair was simply a re-weld and the addition of some extra stiffening.

Some designers have a bad reputation for skegs falling off yachts due to the fatigue of the plating to which the skeg was welded, but this is a case of poor design not a problem with the material.

There is some good reference material available from the scantling societies and welding technology institutes.

I would also put more weight behind observations from vessel in-life problems than from boat-builders (sorry fellas) but in reality boat builders have their rigid prejudices and they are not engineers nor do they often understand stress fatigue .......... particularly with welds.

The classic weld induced fatigue failure was the "Alexander Kieland" Norwegian oil platform collapse in the Eighties which killed over a hundred people after a fatigue failure due to a welded attachment on a main brace.

Cheers

 

I asked the wife and she said that the curve is not truly asymptotic, though it approximates asymtosis, and that theoretically the sigma-n curve goes to zero (eventually). Typically, fatigue life is measured in the hundreds of thousands to millions of cycles, and there is an approximation of the practical (effective) limit for normal design life. Other environmental factors combine with the fatigue endurance limit to affect design load & life. General corrosion (or wear) can reduce the section until the stress exceeds the endurance limit....and temperature affects the curve too.

She did point out however, as I did, that the value given is a statistical one. That the location and distribution of defects (like inclusions or voids) has a significant effect on fatigue life. Then she started about slip planes and microcracking and .. I got lost somewhere around there.

FWIW, she came up and read this over my shoulder then took the keyboard away and made corrections....so the big words are hers.....(asymtosis?)

 

Composite design is undergoing somewhat of a rethinking right now.. the operative idea is "cumulative damage" or "total stress/flexure". This concept is also needed when working with things like spectra fiber ropes. It has to with the physical failure of the hydrogen bonding in ultra long chain polymer matrices. While they are suited to rapid irregular loading, long term low strain causes slipping of the bonds as the chains slide alongside each other. Well beyond my P-chem classes.

True, most (with some notable exceptions) modern (post 1950 or so) hulls rust away (or the rust causes the failure) before they fail from fatigue/loading. But remember that loving it too much is also a bad thing. The military has had to do extensive repairs to hulls because they repainted them too often. That used to mean grit blasting. Blasting off good coatings just to recoat seriously thinned hulls. Recently this has changed, and as Milan pointed out, a good UT tester will tell you coating thickness, adherence, and plate thickness. We have one in the office and we are now painting less often.

With reference to hull thickness, that is a touchy subject. The load-to-flexure energy tradeoff in steel design is important. Beware of boats that the builder just up'ed the hull thickness without the designer re-calculating hull strength. The inter-relationship between primary, secondary, and tertiary loading of the structure is critical to overall hull strength. If one part becomes too strong or too stiff, load flow changes, often to the detriment of structural integrity. Thicker is not necessarily better. What is best is a well designed hull that all members share the load in the best way suited for that member (i.e. the zen of structural design).

 

Sorry getting a bit off the paint topic here...

I have been FEA modeling a 60 foot steel hull design for the past 2 weeks looking particularly at fatigue issues. Computers are so usefull now!

We model the hull with FEA then model the higher stressed areas then even model the welds in those higher stessed areas. The welds are the key of course and the time and energy of supervision goes into making sure those welds are done as well as we can test (within reasonable cost)

When you start to consider slip zones you have gone too far. The problems with structures occur in the weld zones long before we get down to that level in the unadulterated stock material.

Practically.... asymptotic behaviour of S-N curves has been shown experimentaly to exist for all practical design purposes (eg bridge building eg caissons in high vibration zones etc) when using mild steel ......providing you stay below the critical thresholds (which are very low). For all other metals it is not asymptotic at all. Mild steel loses that advantage under certain conditions eg immersed in salt water unpainted.

Fracture mechanics is made easier if you consider that the fracture has started and then model whether it will propogate, this is the favoured approach for critical structures. We can predict fairly well the practical life of our designs. Yacht hulls are not very highly stressed unless built light, small commercial vessels have very low stresses as the plate is usually relatively thicker than required for other reasons.

Your Zen of structural design 101 is just the old boat builders proverb "Avoid stiff spots" small steel hulls are a lot more forgiving than ships in this regard .

Aside
Riveting is good for avoiding the weld issues, very expensive now and little expertise in ship building yards but still usefull at times.

Cheers

 

Them words too big for me.

My wife is a mechanical engineer doing aerospace design. I may have to print and ask her to translate.