Some questions about ISO 12215-5: 2019

My idea is to use design pressures from 2019, and run either FEA or old laminate stack spreadsheets. Also this will work for catamarans (most of our designs are catamarans).

1) So now it comes to FEA, see the attached file. I can't understand - what kind of alternative physics is that??
Stress multiplied by pressure is <= to stress?? Have the authors ever heard about units, such as N/mm2??

2) And the line about 0.6Pbottom and 0.25Pbottom is not understood, too. Do they mean I can use these pressures for FEA?? If so, why do we need all previous very detailed calcs of pressures?

 

I don't have the 2019 version but the older version so I am going ask, what is the difference between the simplified version and the enhanced version? Is the tabular method such as the H2,,,,H4 the enhanced version?

 

'Simplified' is quasi-isotropic single skin - formulas for strength/stiffness. 'Enhanced' is laminate stack analysis with some amendments.

 

Aahh..nuances of the language.
What I mean by... find someone else. Is - find someone else for the whole job, I walk away from it.

I don't have the 2019, but the citation you posted for ref #2, that appears to be correct. It is suggesting that you can use 0.6P for the sides and 0.25P for the deck, as your pressure.
How you decide to analyse the structure with said P, is then of course up to you... unless it explicitly states which method must be used - in those new 2019 rules.

 

Tansl, I think that is a typo or translation error. We do not calculate the stress from the “center of gravity” of each fiber nor from the center of the laminate (mine) of the individual layer. It is calculated from the neutral axis to the outermost distance of the fiber being investigated.

I was not able to fully investigate the tables you posted as it does not contain the formulas for each cell. It will take sometime to plug in in the formulas. But at a glance, I can see that you got it right against what you are saying. The inner laminate is at a distance 3.47 mm away from the Zna the neutral axis. In the illustration however, the Zcrit is the distance of the laminate before the last layer. Zcrit is the critical distance away from the Neutral axis as this the outermost laminate which has the greatest stress. The one farthest away from the neutral axis.

In the tophat analysis, The illustration of the distance of the Neutral Axis points to the inner laminate of the plate. It is not so as the neutral axis determination includes the plate. It is the sum of the plate, the tab, the web, the crown, and the cap reinforcement that is rationalized as one simple I beam.

As for the CLT, I will explain it in another post as Alik has explained what is “simplified” and “enhanced” calculation is.

 

There many forms of CLT or Classic Lamination Theory. The ROM (Rule of Mixture), Hart Smith otherwise known as the 10% rule, and the use of Krenchel efficiency factor are among the simplest method in use. The more sophisticated ones for CLA (Classic Lamination Analysis) use the matrix method of calculations, commonly known as the ABD matrix and the use of Engineering Constant. Both borrows some formula for predicting laminate properties for Shear modulus and Poisson’s ratio as it requires input of at least 4 mechanical properties of the individual laminate to be analyzed. Attached are the summary of results of the use of the methods.

The use of Engineering Constant has been for awhile and has been in use for quite sometime since 1989. It is mentioned in ISO, Used by BV, and a similar formula by LR. The formula has its limitation as it is limited to Unidirectional fibers. Shoefield, a known Naval Architech realized this and came up with fiber directionality factors for different types of fabric weaves from CSM to pure unidirectional fibers held together by glue. You might say this is an expansion of Krenchel factor.

So each method produces different results. I find the ABD matrix to be the most accurate but is difficult to program. The most practical is the tabulated method or the stack analysis we have been using for years. ISO uses it, LR uses it, and is exactly the same sans the assignment of “allowable” or “designed stress”.

Simplification may be good if you just too lazy to find the material property of each ply. It can be a unification of a 0/90/+45-45 ply, the property of +45-45+CSM. If it comes out too thick say 1+mm, then errors come into play. You are 2 to 3 plies away from the neutral axis distance. Using the stack analysis method, you can easily see at a glance which ply is failing no matter where it is placed. A simple rearrangement of the stack will solve this as nearer the NA, the stress is less or nearly zero.

It is just marketing gimmick when they say woven 0/90/+45-45 has better strength. A uni oriented at +45/-45 is a biax and has only 10% of the strength remaining. Placed that in the outermost layer and it is a disaster.

To prove, insert the ISO formula for off axis load into the laminate stack analysis and see where the failure is. It is most obvious when you are analyzing tophat stiffened panels. The crown or cap with a biax fail first.

 

I know that you do it correctly because we have already mentioned it on a previous occasion, a long time ago. But the ISO 2008 did not calculate the Zcrit but the Zg.

The figure that accompanies the table is correct. The Zcrit that it indicates is not that of the outer layer but that of the previous layer. Of course, the maximum stress will occur in the extreme layers, but the stress in one of the inner layers, which is what the figure refers to, could be more critical than that of the outer edge of the outer layer. Sorry, tongue twister.

Of course it includes the plate, there is nothing in that table that says otherwise, but the reference point that I take, because it is more comfortable for me, is the internal surface of the panel. The barycenter of the piece (reinforcement + panel) will be what it is but its coordinates will depend on the coordinate system that I have chosen. This has no influence on the actual distance between the critical surface of a layer and the NA of the whole.

 

I think it means, although the choice of symbols is not very adequate, that Sigmai is the stress due to the application of a unitary pressure (value = 1). Therefore, the total stress in element "i" will undoubtedly be the design pressure, Pd, multiplied by the unit stress.

I think that here it means that instead of using a pressure of value 1, as indicated in point 1), a pressure of value 0.6 is used for the bottom and of value 0.25 for the deck.

 

Hmm, the FEA software we use does not give the 'stresses', but minimum RF which is reserve factor - for composites. Because layers inside the laminate stack have different stresses.
We will try to see how it works next week, maybe try to find some procedure to do it. But seems clear as mud, with the standard written in 'you guess!' way.

 

I totally agree. So I started this thread, to share the "assumptions" that each one has made when trying to apply the rule.

 

Ok, I just got the info that HullSchan has done the new version to 2019 standard. Probably they have been in touch with ISO commette for interpretations and bugs.
I don't think anyone can use this standard legally without getting formal interpretations.

 

It should be so, but I'm afraid it isn't. In the preamble to the standard for Europe (I don't know why only for Europe), it says:
"This European Standard shall be given the status of a national standard, either by publication of an identical text or by endorsement, at the latest by December 2019, and conflicting national standards shall be withdran at the latest by Jun e 2021"
and then gives the list of the European countries involved.
There will be no choice, I suppose, but to ask each Administration what we should do.
In any case, once the errors that we know are in the 2019 standard have been corrected (the 2008 standard also had errors, which have never been corrected), in my opinion, it is logical that the new standard is applied instead of the old one.

 

We have been reviewing the 12215-10 standard for the last few days, as we need rig loads for a sailing catamaran (those are required to calculate global strength).
Outcome: Table E.2 is incorrect, about 50% of cells do not give the numbers provided in the text of the standard as sample.

Some cells give 5-10% difference, some cells give 40-50%. See attached document. Looks like low quality standard that can't be used.

 

This issue of ISO standards is frankly embarrassing, especially the latest editions. Everyone can be wrong, of course, but the most outrageous thing is the refusal to correct mistakes, treating users, who pay for the rule, with total disregard.
I have standard 12215-10 but I have not yet been able to study it in depth. I am in these moments with the 12215-9, which is not easy either.

 

We should probalby set a ISO Small Craft corrections group in Facebook... And discuss it there.

I was reading the text of ISO12215-10 last night, and definitely the writers have lack of academic education. Quick modern ediucation, 'bachellor in 3 years', hehe...

Say on p.11 of the standard they talk about righting moment and heeling moment, 'the righting moment upwind is greater than the heeling moment'. Inf fact, for a vessel sailing steadily the righting
moment is equal to the heeling moment. They should talk about 'maximum righting moment'! There should be two design cases:
- governed by max righting moment from stab curve - this is case of monohulls.
- governed by righting moment at sailing angle (demihull take-off, etc.) - case of multis and 'stable monos'.
What they say 'governed by righting moment' and 'governed by heeling moment' is nonsense!