I have said often enough that I am often wrong. I'll try and see where I am wrong better, and perhaps you and Rick can help me see the light.

| Not sure what to make of your message. To me it seems completely wrong. You are comparing the core of a sandwich panel to a web in an I-beam as an argument that the core is in compression. But the web is in shear!

In your link https://core.ac.uk/download/pdf/1332963.pdf

under 2.1 there is the following

"The three terms on the right hand side (2.1) correspond to bending of the skins about their centroidal axes bending of the skins about the centroid of the whole beam and bending of the core respectively. We can simplify this equation by assuming that bending of the skins about the centroid of the beam is the dominant term."

Would you agree that generally the shear of the web is ignored as it is a lower order term? Or is it just my mistake?

| And there is certainly no shear force in the skins, as you state.

Would not the bending of the skins about the centroid have a shear force in said skin?
Is not the stiffness of a thick panel a function of the shear in the skin?

Looking at Rick's link
http://www.oneoceankayaks.com/Sandcore.htm
Under the section "The Skin" there is this graphic.
http://www.oneoceankayaks.com/Images/coreE.gif

Bjorn, What is the horizontal force depicted on the skin in the above graphic? Is it crazy to call it shear?

| And there is no need to paste wiki-links with information about basic mechanical calculations, because I know then by heart.

I added that link to make sure I knew WTF I was talking about, and perhaps the odd reader who may not be fully familiar with the domain. I did not mean to imply you lacked anything.

I am not attacking you. If you feel I am, I apologize, completely, now and forever. That is not, and never will be, my intent. If that is not enough then let me know. This also goes generally for everyone on the list. Except H.W.

I kid, Same goes for H.W.

| I studied and graduated as a mechanical engineer already as a teenager. And as you may know, things learnt at a young age remains for life.

Perhaps you should not have said this, as, now, I have to bug you a lot regarding these matters. As I know nothing about these matters, perhaps this will frustrate you. Sorry!

Given the skin MUST move, as I understand it, in order to convey any stress to the core, the core can only be effected by the movement of the skin.

Thus, and correct me here, the core only needs to take a load strong enough to deflect/stretch the skin. which is why the core strain is a third order/ much smaller term, right?

In the book you link, under 2.8, the author argues that the difference between ignoring the core strain term and including it can theoretically approach 26% in the case of the specific example of 2.3.1. But I do not understand the 26% number. Can you explain how it practically effects design choices?
This is for honeycomb cores, which the author called 'low density'. How we may assume a more robust core like a foam would effect this value?

As I understand it, there is a given schedule of skin, that has a practical limitation of available weaves. thus, it seems to me, you can choose from a discrete list of skin thicknesses. Each of witch would seem to only require specific needs in the core in order for the skin to fail/yield before the core. Given the skin carries a much higher load, why is there value in a stronger than minimum core?

Thus it seems if one specifies a given layup, say 400gsm bd, than one can say this layup's core needs a min of X compression strength and Y shear value, right? So cannot most all of this core discussion be eliminated by just making a chart of needed core values for a given layup? This chart should be relatively small right?

Finally I would like to understand how maximal loading is relevant to this discussion.
Given we have no idea what is a reasonable maximal load value is for large portions/majority of the surface of a HP design. We only know things like perhaps local expected values and some safety factor.
Knowing that whatever layup schedule is chosen via domain experience and for other practical observed requirements, then what relevance is core failure modes and core optimization? If Rob says some layup is needed, then core needs should follow from that, right?
Or is there some sort of well understood engineering that says one can put a tougher than needed core under a laminate and that when the skin yields(fails), the core will be fine, thus the composite panel will be fine?

Thank you for your time.