Document conventions when defining geometry in contact mechanics

[sannant]

Positive direction for heights ? Out of the surface I guess.

x is first index in the array in the nc file, y is the second index.

We should probably provide a demo python script with the appropriate labels to the axis.

[mcrot]

Yes, or simply a picture which is shown on the web page.

[sannant]

For the signs of heights and displacements, we have one in the contactmechanics repository.

[sannant]

Here is a suggestion, based on the svg we have on surface topography and what is claimed to be outputed .

I also noticed that some people are not aware that they are supposed to upload heights and not the gap in a rigid contact, which changes the sign.

Also, the fact that we always substract the mean.

hraw would be the uploaded data and h the detrended data we display and work with

[sannant]

we also need an illustration of what the composite roughness is.

If you want to do the contact mechanics of the contact of topography h1 against h2, assuming h is height data following our conventions (peaks positive), the composite roughness with which you should do the contact mechanics calculations should be h = h1 + h2

[pastewka]

We should probably have a sketch like this in the ZIP-folder with the contact mechanics results.

[sannant]

Is this picture actually wrong ?

In our code I see

return (disp[self.comp_slice] -  # TODO: Check 1D Compatibility
Systems.py-                    (self.surface.heights(*profile_args, **profile_kwargs) +
Systems.py-                     offset))

so that offset is the mean rigida body penetration and not separation.

I.e. you have large gaps at negative offset .

See also our HardWall simulation example.

[sitangshugk95]

• What are the 2 dotted lines in this image?

• When the code computes an offset (corresponding to a certain mean pressure as input) after solving for the contact mechanics problem which is "negative" in value, is it assuming that the rigid rough body is fixed and the elastic half space is pushed "down" against it, and the sign convention is upwards "positive" pointing out of the rough topography?

I am planning to make a nice, colorful schematic with the positions before and after, to make this clearer.

[sannant]

I see three. The bottom most is h=0. The next one is mean(h), and the top most is the surface of the elastic halfspace before deformatio .

For the second point, you are right, at least according to this picture. I am not sure of the convention. However, I am not sure the figure is right. I say this because in my thesis, I used the opposite sign for the offset (See Figure 3.1 and equation 3.1). Can you check which one is consistent with topobank ? with the code ?

[sannant]

We would indeed appreciate a figure illustrating this more clearly. But we need to think about how to put it into the web-page.

[pastewka]

You mean technically? That is quite straightforward, I suggest we start using Jupyter notebooks for the webpage, Sphinx supports rendering them.

[pastewka]

@sitangshugk95 If you make a nice version of the picture and are willing to provide this for the documentation, this would be super!