Chain Plates 316 or 304

[rleslie]

I am replacing 18 chain plates and have received a quote for either 316 S.S. or 304/304L S.S. The price difference is NOT a consideration.

I realize that the 316 will hold-up to salt water better but the 304 is stronger. Do you have any thought on which should be used? (Increasing the size of the 316 is not an option)

Thanks
Roger

[GordMay]

What makes you think type 304 S/S is stronger than type 316?
S/S Properties Table: http://www.sppusa.com/reference/stai.../physical.html

316 SS is generally preferred to 304, and duplex stainless (2205) is even better.

[rleslie]

Gord,

I was always under the impression (mistaken) that 304 was 10% stronger than 316. Thanks for the info!!

Roger

[Alan Wheeler]

Strength is not quite the correct term. 304L and 316 are very similar. The main difference being a slightly different percentage of the hard metals that are alloyed into the base metals to achieve their corrosion resistance, such as Chromium. The 316 becomes a little more brittle, the 304, just a little more malleable. The end result, apart from machineing differences, is 316 work hardens and can fracture due to fatique earlier than 304. In the situation of a chain plate, this would be of no concern, as the loads are longitudeinal. If you where trying to see tha differences, the 304 would be ever so slightly more elastic than the 316. However, In the length of a chain plate, the elongational stretch diferences between the two would be too minimal to be of concern. If the plates are heavy enough, they should be way in excess of the failure point of the stays.

[delmarrey]

316 material will cost more per pound. But the 304 will cost more to machine.

Personally I'd go with the 316 mostly because of the corrosion factor. 304 will turn red on ya!

Make sure the bolt holes are as tight as you can get them. Any slop will elongate the holes. Once in motion, always in motion!!!

[By Invitation]

Go with the 316.

[rleslie]

Thanks for the input. For $115 more, the 316 sounds like the way to go.

Thanks
Roger

[Richhh]

304 or 316 .... really doesnt matter for chain plates. 304 is stronger for tensile values; but, you DONT (or shouldnt) design chain plates based on 'Tensile' values. 316 has better corrosion resistance; but, still will ultimately fail due to fatigue, if the boat is 'really' sailed. I'd seriously consder one of the newer duplex stainless steels ..... but my jury is still out and only TIME will tell if these newer grades will actually live up to the 'promotion'. 316 wasnt the 'wonder drug' as it was originally marketed.

300 series stainless is prone to fatigue or endurance failure when cyclical loads are applied. Boat designers have wrongly designed these metals for years, designing the plates to a tensile value of about 90ksi. In fact (as with most other metals/materials) designed for *cyclical stress loading* (like chainplates) should be designed at BELOW 30ksi so that whats called the 'fatigue endurance limit' is never approached. Actual testing in laboratories under the duplicated stress application is the REAL way to evaluate; but, if you dont know any better you should only design stainless to 30ksi when there are fluctating loads applied. Further, fatigue is additive and due to the inherent metal 'errors' in metallic grain structure. Fatigue starts from the very first load you put on it. Every time you stress (above the endurance limit) micro/macroscopic cracks begin to form and keep adding up every time you overstress the part. Usually such a metal will only bear about a million load cycles, then it gets - 'tired'. Surface finish is also critical in retarding the crack formation ---- mirror polishing and then electopolishing is probably getting close to 'good'. So what a poor boat owner to do: Answer - reinstall with thicker and wider plates, polish the hell out of them so they look like mirrors, avoid 'bends' and welds, etc.

Chainplates dont fail by 'ductile failure' therefore tensile values simply dont apply. They fail by fatigue cracking that opens the matrix by teeny cracks that allows water to enter .... and you wind up with two simultaneous failure modes: brittle failure/fatigue AND crevice corrosion. Design a proper plate to less than 30ksi (with 90ksi material), polish the hell out of it .... and this constant plate failure you hear about all the time will diminish.

[delmarrey]

Richhh,

Actually your statments are incorrect about tensile strengths. According to my references from the Curcible Steel Co., 304 has a rating of 85 at 1000 psi and 316 has a rating of 90 at 1000 psi. And even the Rockwell hardness B, 316 is 85 vs 304 is 80 which makes 316 a stronger material. Which isn't really a whole lot of difference. But the corrosion factor for 316 is better. And for the 30ksi, the holes will start to elongate and then one will get back lash on every tack. I'd rather have the micro cracks.

As for the 2205 SS (Ferrillium) there are 4 different grades. But I must admit it is superior to the 300 series. I just spent this past week machining (3) 6" ball valve seats for steam valves at the Boise Cascade plant. The stuff is tuff but the price is is exxxstream and hard to get in normal sizes. It took a month to get some raw material plazma cut into ring blanks................._/)

[Richhh]

Only a daft idiot would design to the UYS or UTS of a material ... unless he/she WANTED the part to break at a certain value.

You are mistaken on several account. First the values given are for *Cruicible Steel's* own produced stainless steel and only if the metal is accompanied by the usuall mill certifications. These values wouild not apply to metal produced by "Nan Fujngs Art Foundry and Delicatessen". Each mill's values will be different, and different from lot to lot. Without certifications based on actual testing values you MUST 'de-rate' those value to be prudent/safe.
Hardness is only a correlation to the strengths and if you are designing for other than hardness, then you need to apply de-rating 'safety factors' that lower the apparent ultimate values. And, one must KNOW that these values are for slowly applied direct tension/compression; and, thats NOT how a chainplate works !!!!!!!!!!!!!!!!!!!!!!!! The portion of the chainplate (as normally ill-designed by boat 'architects') that attaches to the bost is usally WRONG or EXTREMLY WEAK with respect to 'proper' / normal stress transmission. These are cantilever/shear connections that usually require that the attachment materials (bolts) be used at ONE FOURTH of their tensile values .... and then further 'derated' by the 'normal' safety factors for such service.

As for 'elongation' .... sorry but 30ksi is 1/3 the fiber stress of 90ksi ... and if the bolting holes elongate it means that the metal has yielded (ductile failure) and the designer didnt know what he/she was doing and probably simply 'hung' the bolting in simple shear (w/cantilever) instead of making a 'compression/friction joint' wherein the compressional force of the torqued bolting was used to force the plate and its base together and that 'friction' is what transmits the stress. 99% of most chainplates are improperly designed as flat plates and are unsymmetrically loaded where the plate meets the base attachment .... this puts the bolting into 'cantilever stress' which automatically decreases the load bearing ability of the bolt by 1/4th !!!!!! If boat designers actually knew what they were doing, they would design the plates so that the terminal bottom end would be a CLEVIS connection (and free of 'stress risers' of bolt threads touching chainplates) .... just like how the other rigging components are designed. If properly designed, chainplates would hardly every fail !!!!!!!

Those 'book' values of metals (yield and UTS) are ONLY for straight slowly applied tensile loads. Applying impact, cyclical, cantilever loads, etc. and then use a material that is long term incompatibble with sea water (crevice corrosion) .... you better KNOW what you are doing and KNOW what 'safety factors' are normaly applied (historical scantlings). If you think that one can read a straight tensile value from a book and apply those values to non-straight line tension compression .... then you risk having the mast come down into your cockpit. I state again those book values for metals are for DUCTILE failure mode, not FATIGUE modes and those values are WITHOUT the safety factors that MUST BE additionally applied by the designer. If you dont know what safety factors by which you DE-RATE the material strengths, you risk catastrophic failure when the material eventually 'work hardens' becaues it was a BAD design in the first place. A minimum *safety factor* of 3 (4 or 5 is better) to derate 300 series to usage at less than 30ksi is VERY prudent for usage versus cyclical loads in a halide corrosion atmosphere.

[GordMay]

Gee, RichH - don't be shy, say what you mean ...

Gord

[delmarrey]

Only a daft idiot would design to the UYS or UTS of a material ... unless he/she WANTED the part to break at a certain value.

Richhh,
I think you just insulted every boat builder out there. They're only building boats as economiclly as possible, not space shuttles.

If parts didn't fail, sailors wouldn't have anything to do!

[Alan Wheeler]

Hehe I like that Del, if parts didn't fail part, it would make a good footer at the bottom.
Hey Richhhh, Ooookaeey, so what's the point you are trying to make? Sounds like you are over analyizing and getting just a little to much into the over engineering side of the argument. They are only plates mate and it's pretty rare for a plate to fail before anything else in the rig goes. If you wanted to really get technical, then SSt of any discription is not the best choice. Titanium would be the way to go. But who cares. Most of us aren't in the Americas cup, we just want a plate that doesn't rust.

[Richhh]

Well, if you only have a marina queen that occasionally leaves the dock when the winds are under 15kts, then the 'other' arguments apply.

Fact is that chainplates fail all the time and that even boat designers (the best of them) will agree that they misunderstood the complexities of proper design; and, thats why now they employ those who who are slightly more knowledgeable in such matters, especially in light of the new euiropean standards.

Even ancient 'insurance scantlings' apply to chainplates..... chainplates fail, people get hurt, etc. and the insurance actuarial department advises to 'beef up the damn plates' or we will charge your customers more to cover their risk. Nothing wrong in replacing in exact kind and thicknesses, etc. ... and know that that part will fail again after the same number of load cycles .... take your choice. Its called design evolution.

[delmarrey]

Richhh once whispered in the wind:
Fact is that chainplates fail all the time and that even boat designers (the best of them) will agree that they misunderstood the complexities of proper design; and, thats why now they employ those who who are slightly more knowledgeable in such matters, especially in light of the new euiropean standards.

Even ancient 'insurance scantlings' apply to chainplateschoice.

Can you post some documentation on this?

I have yet to see any chain plates ever FAIL. They are usually replaced before they get that bad. Maybe in the old wooden boat days when the fasteners use to pull thru the wood.