[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [PSUBS-MAILIST] Design depth
Frank, there's a potentially serious problem with this - despite software
packages making FEA tools available to almost anyone (i.e. COSMOS is a pretty
cheap add-on to SolidWorks, for example), and the associated interfaces being
quite easy to use, getting useful results out of a FE analysis requires some
critical background knowledge in order to properly construct a simulation. In
my professional experience, I have come across a number of analyses which were
originally contracted to another party for FEA, but which failed in service
despite a positive FEA result - typically, these were because of improper FEA
load cases applied to a detailed model, and this is exactly the sort of stuff
that the software can't tell you. Knowing how to properly constrain a model
to avoid stress singularities, knowing the differences between plate/shell,
solid and gap elements, and where to appropriately use them, how to construct
a FE model to make best use of available processor resources instead of
modeling everything as solids, knowing where and to what extent to conduct
mesh refinement in order to obtain useful results, and identifying critical
locations which you can subsequently strain gauge to verify the FE model and
calibrate it against the actual physical part - these are all critical aspects
of proper FE analysis which software does not address, no matter how user-
friendly it is. For my own part, I have about 240 hours of formal FE
education, plus years of experience with both FE modeling and strain gauging
in the field, and I consider myself a novice. It doesn't matter how much
computing power you have available. Garbage in equals garbage out, and as
your system gets more complicated (i.e. multiple parts, multiple combined
forces and accelerations, dynamic vs quasi-static behaviour, etc.), the
problems get less intuitive to solve, which makes it easier to make mistakes
which are not obvious in the results. As far as subs are concerned, it is not
only hydrostatic pressure loading we need to be concerned about, but rather
that hydrostatic pressure considered in combination with maximum component
misalignments considered in combination with a collision impact to the
structure which could occur at any location or angle - typically assumed as
the worst possible case for FEA, but identifying that worst case is where the
experience comes in. Software is just a tool, and on several occasions as a
technologist I used to program FEA simulations based on load cases defined by
my then-boss (Ph.D., P.Eng.) on paper. There is a reason professional FEA
analyses run into the thousands of dollars, and that reason had very little to
do with either the cost of my time or the processor hours.
-Sean
On Monday 13 December 2010 05:46:48 you wrote:
> Would it be possible for PSUBS to gather donations from the members and
> purchase outright a good FEA program ? It seems like such a useful tool
> with pretty much universal need by anyone in the design phase. It
> apparently takes a "WHOPPER" of a computer but if a computer savvy member
> was willing to take on the task of running the calcs ( for a fee ? ) it
> would sure be helpful. I for one would be willing to help pay for the
> original program purchase and a reasonable fee could be taken by the
> member doing the work so it wasn't a process where the member got abused
> by hair brained ideas. Naturally it would have to be completely liability
> free for the guy doing the FEA's, but for running basic hull shapes,
> thicknesses, sizes and keeping it down to simple shapes and more or less
> "standard" hulls etc. it would be a great resource available to "members
> only" and maybe even make a little dough for the guy runnin' the numbers.
> Frank D.
>
>
>
>
> -----Original Message-----
> From: vbra676539@aol.com
> To: personal_submersibles@psubs.org
> Sent: Sat, Dec 11, 2010 7:04 am
> Subject: Re: [PSUBS-MAILIST] Design depth
>
>
> It would be interesting to see a work-up and FE-calculations on Gamma, as
> it has some odd stress points in the battery box and, of course, the
> drilled through viewports. It works, and has done so for many years, but
> I'd love to see modern computer technology applied to see in realtime what
> Doug Privitt actually did. Vance
>
>
>
>
>
> -----Original Message-----
> From: MerlinSub <MerlinSub@t-online.de>
> To: personal_submersibles <personal_submersibles@psubs.org>
> Sent: Fri, Dec 10, 2010 11:24 pm
> Subject: Re: [PSUBS-MAILIST] Design depth
>
>
> Hi Vance,
> we finalise just the FE-calculations for the Javasub (3+1seater) and the
> orthsub (3seats).
> oth boats have the same design - Java is just a little longer.
> Resulst are crush at 900m (3000 feet), work depth 400 (1300 feet) test at
> 500 1600 feet).
> hell thickness is about a half inch, the first calculation by hand shows
> crush t 500m
> he difference between the first 500 crush and than later increase crush up
> to 00 m was a
> ptimize process of the stiffteners by the FE method and the fact that the
> hand ethod
> s far away from the optimum to find local high force spots.
> Here are the Table fore classification to Germanischer Lloyd:
> d x factor = td and dd x factor2 = cd
> design deep - test deep - crush deep (all in feet)
> 164 x 1,7 279 x 3,2 525
> 328 x 1,4 459 x 2,4 787
> 656 x 1,25 820 x 2,0 1312
> 984 x 1,2 1180 x 1,87(!) 1840 (1968)
> 1312 x 1,2 1574 x 1,80(!) 2361 (2624)
> 1640 x 1,2 1968 x 1,76(!) 2886 (3280)
> 1968 x 1,2 >2362 x 1,73(!) 3404 (3936)
> (!= factor shall be 2,0 if water deep is more than design deep)
>
> br Carsten
>
> vbra676539@aol.com> schrieb:
> You guys have fallen head over heels in love with a logic trap, heavily
> basted ith semantics. Give it up. Design depth is your operating depth
> PLUS the ercentage required for testing. It means the depth to which the
> pressure vessel ay be taken repeatedly without incurring irrecoverable
> damage.
>
>
> 125% doesn't make some people happy. I'll just stick in my 2 cents and
> tell ou that I have done a whole bunch of dives in vehicles built and
> tested to that tandard. Crush depth is an arbitrary derivative of the
> overall calculations, uggesting that all things being perfect, if you go
> this deep (whatever THIS appens to be) then the hull MAY suffer
> irreparable damage, and if it does, then HIS is the site of most probable
> damage. The THIS is typically a measure of the eakest link in the design,
> which is specific to the design, not general iscussion. It is NOT a weak
> link, but simply the first failure point.
>
>
> A C-class Perry for instance would be rated for 1200 feet maximum
> operating epth. It's design depth (and unmanned test depth) would be 1500
> feet and it's rush test.....hmmm, I forget. About 1800 and change, I
> think. Maybe 19 and hange. That kind of thing is in the initial ABS
> calculation package, if you appen to have it. However, it is not normally
> in the general ABS certs. I do ot have a crush depth number for Gamma, for
> instance, but do have her ertifications.
>
>
> For general interest, Nekton Alpha, Beta and Gamma are all rated at 1000
> feet nd were all tank tested to 1500. Crush depth was ascertained by model
> testing, hich verified the calculations handily (just over 2000 feet).
> Perry was doing epetitive construction on two proven primary designs
> (tubular hull, internal or xternal ring stiffeners). Once proven, in other
> words, all they had to do was eplicate and prove that they had done so.
>
>
> As to accepting one safety factor over another, I'd say 125% should be the
> ccepted norm, based on experience and history. Anything over that is a
> personal hoice. I feel one way, somebody else feels another. Potato,
> potahto.
>
>
> Shallow boats are easy to double up on. The old Submaray, for instance,
> was /8" plate and they held it to 300 feet for safety. The PC-8, on the
> other hand, as also 3/8" plate and I can't even tell you how many times
> I've been to her perating maximum of 800 feet. Different construction
> techniques. Different esign. Different results.
>
>
> Vance
>
>
>
>
>
> -----Original Message-----
> From: Jon Wallace <jonw@psubs.org>
> To: personal_submersibles <personal_submersibles@psubs.org>
> Sent: Fri, Dec 10, 2010 4:53 am
> Subject: Re: [PSUBS-MAILIST] Design depth
>
>
>
> Jim,
>
> Design depth is derived from the parameters (material type, hull
> hickness, number of ribs, spacing of ribs, etc) used in ABS/ASME
> formulas to etermine the maximum external pressure that the submarine
> can operate in. I on't think ABS or anyone else can reliably calculate
> "fail depth" or "crush epth" because those depths are highly dependent
> upon quality of material, abrication, vessel use/maintenance, etc. The
> crush depth could be some ercentage above or below design depth, however
> you could only find out efinitively by testing the vessel to
> destruction. For that reason, you on't want to perform a manned dive
> to 375 feet. Make sure that is an nmanned test. You'll note that ABS
> Underwater Vessels Section 3.5 limits he manned test dive to the design
> depth.
>
> Jon
>
>
> On 12/10/2010 9:04 AM, JimToddPsub@aol.com wrote:
>
> Jon,
>
>
>
> I need to read and study the whole ABS publication thoroughly,
> but rom what I've read is seems that the Design Depth I would
> need to laim would be 300 feet. During the survey I would need
> to dive in ncrements to 375 feet (300 x 1.25). The vessel
> would then be rated o 300 feet. Per ABS, Design Depth is the
> depth to which the essel is "designed and approved to operate." It's
> going to be pproved to 300. If I want to have it approved for
> anything more han that, I'll have to have it surveyed to
> 125% of whatever I want ts operating depth to be.
>
>
>
> What is surprising is that [so far] I see nothing in ABS that
> pecifies how the Design Depth must be derived. It seems rather
> rbitrary rather than stating something such as "60% of the
> alculated fail depth."
>
>
>
> Jim
>
>
>
>
> In a message dated 12/10/2010 1:24:29 A.M. Central Standard
> Time, onw@psubs.org writes:
>
> A different approach is to say, I know I don't want to dive
> eeper than 300 feet (my max operating depth) and I want a 2x
> afety margin. That means using material capable of
> withstanding 00 feet depth. Now with this approach you
> certainly may not pass 1.25 overpressure test for a 600 foot
> design depth, however ou've already self imposed a 2x safety
> margin. If you wanted ABS ertification, then you simply
> claim the design depth as 480 feet. ow you meet the ABS
> requirement (480 x 1.25 = 600) and also are till well within your 300 foot
> self-imposed depth limit.
>
>
>
>
>
>
>
>
>
>
> ***********************************************************************
> ***********************************************************************
> ***********************************************************************
> he personal submersibles mailing list complies with the US Federal
> AN-SPAM Act of 2003. Your email address appears in our database
> ecause either you, or someone you know, requested you receive messages
> rom our organization.
> If you want to be removed from this mailing list simply click on the
> ink below or send a blank email message to:
> removeme-personal_submersibles@psubs.org
> Removal of your email address from this mailing list occurs by an
> utomated process and should be complete within five minutes of
> ur server receiving your request.
> PSUBS.ORG
> O Box 53
> eare, NH 03281
> 03-529-1100
> ***********************************************************************
> ***********************************************************************
> ***********************************************************************
************************************************************************
************************************************************************
************************************************************************
The personal submersibles mailing list complies with the US Federal
CAN-SPAM Act of 2003. Your email address appears in our database
because either you, or someone you know, requested you receive messages
from our organization.
If you want to be removed from this mailing list simply click on the
link below or send a blank email message to:
removeme-personal_submersibles@psubs.org
Removal of your email address from this mailing list occurs by an
automated process and should be complete within five minutes of
our server receiving your request.
PSUBS.ORG
PO Box 53
Weare, NH 03281
603-529-1100
************************************************************************
************************************************************************
************************************************************************