Some questions about ISO 12215-5: 2019

It is, in 12215-7 now, for multihulls. We are just working on that.

 

Thank you, I will take a look.

 

Well, this would be my thinking too. And we did it, say, for HDPE boats or solid fiberglass.
But for composite sandwich structures, the Clients don't want that; they need laminate at the limit of compliance.

 

Composites are a different story than isotropic materials.
However, even in metal hulls, Classification Societies ask to study what they call Main Sections, not just a main section, but at least three sections, one in the center of the ship (0.4 L) and two others in the areas bow and stern (0.1 L). I have never applied the results of the mid section to the rest of the boat because it can lead to very large errors. It is not a matter of the designer's preferences but of the rigor that one must adopt in his calculations.

 

When I have clients like that I tell them:

1...do it yourself
or
2 ..use someone else.

What is the point of using a naval architect if they think they know better!

 

The key is: if we touch the project, we do it completely - from first sketch to cutting files and equipment selection. We don't do partial works.
This means we design the laminates ourselves, with all structural calcs. I would never send the client 'find someone else' for this work - this 'someone else' will screw up the whole project with overweight and multiple changes.

 

One of the improvements, important, of the new 12215-5 is that it asks to calculate the tensions of each fiber at the point furthest from the neutral axis of the laminate. The previous version calculated it at the center of gravity of each fiber.
One thing I'd like to expose to you is, if Method 1 Simplified is correct, no one is going to want to use Method 2 Enhanced. You will tell me about the possible saving of material. The difference in thickness can be of the order of 9% and in weight, probably much less. If that saving is worth it, then why would anyone use the Simplified method?

 

'Enhanced' method is what we have been using for years, under GL HSC 2012 Rules, for composites. We have developed those CLT spreadsheets for panels and different types of stiffeners. I think we can use these spreadsheets with loads from the ISO12215-5:2019. However, who and how in the notified body can check these calcs - big question?

Personally I don't think the 2019 version of the standard will have any use in the industry. The notified bodies are not packed with composite experts, and they can't evaluate all theories while reviewing the designs. It is just very easy to make mistake in those lengthy tabular calcs, and impossible to find it.

PS What really surprises me: for material properties the ISO12215-5:2019 is using CLT approach, even for 'simplified' method. This is very questionable approach.

 

I don't really know what the scope of this new version is going to be, but I suppose that most governments, at least the European ones, are going to adopt it, if they haven't already.
Regarding the 2 Enhanced method, I understand that it is basically the method based on the CLT but studying the forces only in one direction, the longest of the panel or of the reinforcement. The full CLT method, which could be Method 3 Developed, studies stresses in both directions of a panel. But this the ISO does not cover.
Regarding using the old tables, there are several problems. One of them is that the formulas to obtain the properties of each mat or each fabric have changed, they no longer depend on the level of evaluation but on other factors, the old values are no longer valid. As I said before, in the old tables the critical stress in each layer was not calculated but an average stress, which is not correct. The allowable stresses are not the same either. The analysis of the layers of a reinforcement was not totally correct either, now they are studied in more detail, distinguishing if it is the crown, the legs of the omega, the bonding area reinforcement / associated plate. Minimum dimensions of the overlaps are given that were not previously stipulated.
My impression is that the new standard is much better, more technical than the old one, although more complicated. We would do well to use it and discard the previous one. But the ISO Organization, or the author of the standard should correct it, make it clearer for the normal user and, above all, correct the errors it has.

 

Can you expound on the CLT approach? What does it mean or how does it work?

 

Instead of regression formulas, now they use CLT to find elastic constants of the laminate. This means that thy split every fabric (say, BX400) into layers, 0,90,-45,+45 direction and mat, and find compound properties. Then, they use these properties for further calculations of laminate stack.

However, this approach gives equal elastic constants for tension and compression, that is not workable with some fabrics. The old empirical formulas (from -5:2008, also used in LR SSC) are giving different values for ultimate tension and compression stresses.

 

One more observation for 2019 standard, as I am adapting my spreadsheets for use with it.
In the 2019 standard design stresses are calculated 'through the ***' (sorry!). Instead of giving regression formulas for sigma and tau, as it was in previous 2008 version, they find full set of elastic constants as required for CLT and FEA. And then, from that set of advanced parameters and 'breaking strains' they find design stresses for simplified and enhanced methods. If design stresses are required for method 'Enhanced', why the hell we not calculate them directly?? Extra operations not only waste engineering time, but also increase probability of mistakes!

I have no idea why this is done, but to me makes no sense at all. I would approximate data in Table C.9, and use them directly for calculations. This way, my staff can't miss the numbers when playing with strains, E and G.

 

Yes, that's how this sad story is. And, if you use the formulas in C.4, C5, etc, you may not get exactly the same values that appear in Table C.9. If you want to validate your calculations against H.2 tables etc, you will not succeed. Even using the values in table C.9 without thinking further, you will not be able to reproduce the results of H.2, H.6 and H7. There is something else in the standard that does not seem correct.

 

Yes, I am using my CLT spreadsheets with ISO values, and I am not getting these numbers.

Formally, if there are errors in the standard, we can refuse to use it. Here are some mistakes mentioned here. If there is no formal correction of the standard published, then we wait for corrections before using it
https://publications.aston.ac.uk/id/eprint/41209/1/Souppez_JB_ISO_12215_5_IBEX_Symposium_2019.pdf

 

I am using the design pressures and stresses from the new standard, which are correct, and I have eliminated the errors (just 4 or 5) that I have detected in the formulas for forces and bending moments. I do the analysis of layers as in the 2008 version, although with a little more detail in the calculation of the reinforcements. I have also corrected the point where the maximum stress is calculated in each layer, which in ISO 2008, was not correct.
I think that this way, the 2019 rule, which in my opinion is better than the previous one, can be used. Because the general path described is the correct one but in step by step, it is lost.

Another question, how to proceed with aramids in the new standard ?. To calculate its mechanical properties I apply the 2008 standard. Is there, in your opinion, anything wrong with it that I am not able to see?

P.S. How to get ISO to correct its standards? It seems that even Mr. Souppez is not being listened to.