[PSUBS-MAILIST] correction

Sat Sep 26 22:05:40 EDT 2015

While steels have an endurance limit (stress limit below which fatigue failure is highly unlikely to occur), the fatigue endurance limit of the base material is less relevant than the presence of welds and geometric stress concentrations, which impose their own limits due to localized stress concentrations and crack initiation sites. Characterizing the fatigue life of welded structures was a core focus of my job for several years, in the context of mining machinery subject to continuously variable cyclic loading.  Invariably, four conclusions always popped up in these analyses:

1) Employ full-penetration welds,
2) Profile grind to eliminate stress concentrations,
3) Fully stress-relieve completed weldments with appropriate heat treatments, and
4) Replace welded construction with smoothly-sculpted castings where possible.

Where submarines are concerned, design guides and common practice already address point 1, at least where the pressure hull itself is concerned.  I would additionally suggest to be wary of any highly loaded appendages welded to the pressure hull where a crack initiation could possibly propagate into the pressure hull.  Point 2, profile grinding, is technically prohibited on submarines, but the military does it, and the weld procedures in the commercial guidelines specify bead geometry limits that accomplish much the same thing.  The intent of the prohibition is to eliminate the possibility of grinding below minimum design thickness, but if you're not certifying, I see no reason not to smooth your joints, as long as you are aware of the consequence of error.  Point 3 is a great idea if you have access to equipment or a facility with these capabilities.  Particularly with PSub-sized hulls, I'd look into getting quotes.  Has to be done before any coatings though.  Point 4 is about
eliminating abrupt stress concentrations. Obviously this is complicated for a pressure hull.  If I had a sub with lots of identical windows, I might look at casting their seats instead of fabricating, as smooth transitions which seamlessly integrate into the base hull are easier than 4:1 weld tapers, and create a smoother load path. As with your welds though, any such castings would have to be 100% NDT'd.

All that said, the number of large stress cycles on a submarine hull over its lifetime is so low that fatigue failure is not really an issue, with the possible exception of things like skid attachments and lifting lugs which can see high stresses in the field.  Even if you dive every day, that's 365 cycles a year, or maybe 10k over its lifetime. High cycle fatigue generally starts to show up one order of magnitude higher.

One thing you should be aware of is that fatigue damage varies linearly with number of cycles, but with the cube of the applied stress.  A single cycle at 3 MPa does the same amount of fatigue damage  as twenty-seven cycles at 1 MPa!  Thus, the fatigue endurance of your boat is dictated predominately by only its deepest dives.  This is a counterintuitive result which, interestingly, governs a lot of aircraft design.  The wings of large airliners experience their largest stress reversals not due to the forces on the wings in-flight, but rather to the less frequent but greater magnitude switch between flying, where the aircraft is hanging on the wings, and being on the ground, where the wings are hanging from the aircraft.

Sean

On September 26, 2015 10:29:16 AM MDT, hank pronk via Personal_Submersibles <personal_submersibles at psubs.org> wrote:
>Tim,Yes, I want a  safety margin.   As far as I know there is no limit
>to number of cycles on a steel pressure vessel.  That would be a Sean
>question.  Hank 
>
>
>On Saturday, September 26, 2015 9:46 AM, T Novak via
>Personal_Submersibles <personal_submersibles at psubs.org> wrote:
>   
>
>#yiv5503147854 #yiv5503147854 -- _filtered #yiv5503147854
>{font-family:Helvetica;panose-1:2 11 6 4 2 2 2 2 2 4;} _filtered
>#yiv5503147854 {panose-1:2 4 5 3 5 4 6 3 2 4;} _filtered #yiv5503147854
>{font-family:Calibri;panose-1:2 15 5 2 2 2 4 3 2 4;}#yiv5503147854
>#yiv5503147854 p.yiv5503147854MsoNormal, #yiv5503147854
>li.yiv5503147854MsoNormal, #yiv5503147854 div.yiv5503147854MsoNormal
>{margin:0cm;margin-bottom:.0001pt;font-size:12.0pt;}#yiv5503147854
>a:link, #yiv5503147854 span.yiv5503147854MsoHyperlink
>{color:blue;text-decoration:underline;}#yiv5503147854 a:visited,
>#yiv5503147854 span.yiv5503147854MsoHyperlinkFollowed
>{color:purple;text-decoration:underline;}#yiv5503147854
>p.yiv5503147854msonormal, #yiv5503147854 li.yiv5503147854msonormal,
>#yiv5503147854 div.yiv5503147854msonormal
>{margin-right:0cm;margin-left:0cm;font-size:12.0pt;}#yiv5503147854
>p.yiv5503147854msochpdefault, #yiv5503147854
>li.yiv5503147854msochpdefault, #yiv5503147854
>div.yiv5503147854msochpdefault
>{margin-right:0cm;margin-left:0cm;font-size:12.0pt;}#yiv5503147854
>span.yiv5503147854msohyperlink {}#yiv5503147854
>span.yiv5503147854msohyperlinkfollowed {}#yiv5503147854
>span.yiv5503147854emailstyle17 {}#yiv5503147854
>p.yiv5503147854msonormal1, #yiv5503147854 li.yiv5503147854msonormal1,
>#yiv5503147854 div.yiv5503147854msonormal1
>{margin:0cm;margin-bottom:.0001pt;font-size:12.0pt;}#yiv5503147854
>span.yiv5503147854msohyperlink1
>{color:#0563C1;text-decoration:underline;}#yiv5503147854
>span.yiv5503147854msohyperlinkfollowed1
>{color:#954F72;text-decoration:underline;}#yiv5503147854
>span.yiv5503147854emailstyle171 {color:#1F497D;}#yiv5503147854
>p.yiv5503147854msochpdefault1, #yiv5503147854
>li.yiv5503147854msochpdefault1, #yiv5503147854
>div.yiv5503147854msochpdefault1
>{margin-right:0cm;margin-left:0cm;font-size:10.0pt;}#yiv5503147854
>span.yiv5503147854EmailStyle27 {color:#1F497D;}#yiv5503147854
>.yiv5503147854MsoChpDefault {font-size:10.0pt;} _filtered
>#yiv5503147854 {margin:72.0pt 72.0pt 72.0pt 72.0pt;}#yiv5503147854
>div.yiv5503147854WordSection1 {}#yiv5503147854 Hi Hank,Sounds like a
>good way to go.  Is your plan to restrict manned dives to 66% of the
>maximum unmanned test dive?  This would seem to be a good policy in
>general.  Is there a design fatigue life on the hull (i.e. so many
>cycles before overhaul)?Tim  From: Personal_Submersibles
>[mailto:personal_submersibles-bounces at psubs.org] On Behalf Of hank
>pronk via Personal_Submersibles
>Sent: September-26-15 7:59 AM
>To: Personal Submersibles General Discussion
><personal_submersibles at psubs.org>
>Subject: Re: [PSUBS-MAILIST] correction  Hi Tim,I have only been to 105
>or 110 with you.  I will deep test Gamma with the pod as soon as it is
>back together.Hank    On Saturday, September 26, 2015 8:46 AM, T Novak
>via Personal_Submersibles <personal_submersibles at psubs.org> wrote:  Hi
>Hank, What has been the max test depth of Gamma itself with the dome so
>far? Tim From: Personal_Submersibles
>[mailto:personal_submersibles-bounces at psubs.org] On Behalf Of hank
>pronk via Personal_Submersibles
>Sent: September-26-15 5:44 AM
>To: Personal Submersibles General Discussion
><personal_submersibles at psubs.org>
>Subject: [PSUBS-MAILIST] correction Hi All,Correction, I estimated that
>the pod sank 800 feet based on the amount of rope left on the spool.
> It actually sank 638 feet,  my target depth was 600 feet.  I will be
>diving at 400 feet for the next season.    Hank
> _______________________________________________
>Personal_Submersibles mailing list
>Personal_Submersibles at psubs.org
>http://www.psubs.org/mailman/listinfo.cgi/personal_submersibles  
>_______________________________________________
>Personal_Submersibles mailing list
>Personal_Submersibles at psubs.org
>http://www.psubs.org/mailman/listinfo.cgi/personal_submersibles
>
>
>
>
>------------------------------------------------------------------------
>
>_______________________________________________
>Personal_Submersibles mailing list
>Personal_Submersibles at psubs.org
>http://www.psubs.org/mailman/listinfo.cgi/personal_submersibles
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.whoweb.com/pipermail/personal_submersibles/attachments/20150926/106bfa84/attachment.html>