@mmcdara 

I did not know Tadashi Tokieda. Indeed, there are two experiments that I would like to reproduce and one phenomenon that I did not know yet.

I was thinking about industrial use of the contact elements. Without forces, wear and damage of contacts are impossible to evaluate. Your decision to use Maple for an industrial mechanism was probably right. A ball mechanism will likely have more than one contact. This means that the technique I have applied here to determine contact forces will not work.

Maplesoft, as I see them today, has a strong focus on industrial applications. The latest Web Handling library for specialists shows it. I am not sure if training students with toy experiments (to become potential MapleSim users in industry) is enough motivation to enhance contact elements even further. However, students look for fun and cool stuff, only a fraction is excited about tools that increase profitability of others.

The visualization of spinning spheres is non-trivial. I am looking for a while for suitable patterns. MapleSim can visualize CAD models, but these are static and need to be segmented in colored parts. MapleSoft has provided an example of a Tippe Top

Again, a laborious step for a tool (MapleSim) that produces on the other hand quite fast high-fidelity physical results. I am reluctant to waste my spare time on CAD tools since I only do these simulation experiments for fun whenever I need distraction from boring duties.

@janhardo 

Thank you for trying. Do I understand correctly that the cpu load was the same before and after hibernation? IOn may setting it is only after wake-up from hibernation

@mmcdara

The post became unintendedly long. For clarity, and to be able to resume work just in case some new ideas emerge, I had to go into details.

The contact elements are quite sophisticated (see for example the sphere cylinder contact I used) and are a recent development. Some parameters are so advanced that non-experts have to “contact” support for clarification. Taking a component out of the library, slipping occurs quite naturally. Preventing slippage on the other hand requires some effort. In the example of a ball on a turntable I had to use super sticky (not physically realistic) friction coefficients.

The contact elements should be able to determine slippage since there is a parameter “Smoothness coefficient for sliding friction”. Unfortunately, this information, as the contact forces are not exported.

Visualizing an equator and meridians: That is a missing feature. In general, there are no visualization patterns available for that. I hope Maplesoft will do something about it one day.

@janhardo 

Before switching, which I have to do anyway, could I ask you (or anyone else who can spend time one it) a favor?

Could you try the following:

• Enable hibernation
• Close all Maple sessions
• Start Maple
• Open a new worksheet (close the startup page if this is open as well)
• Enter a name
• Open the task manager and note the cpu load and power usage
• Let the system hibernate
• Restart the system and look for any changes (cpu load and power usage) in the task manager
• Optionally: Check if the context panel is still working.

@janhardo 

Unlikely, I checked for network trafic. No trafic at all.

@mmcdara 
That's a good point. I did not check energy this time because I have done such checks on several occasions and gained confidence in MapleSim for simulations of this type (here is, for example, a cross-check with a simulation from Ruben Rostamian).  

A quick check from a side view (from behind) shows a tiny loss of potential energy. The apex of the incomming branch is a bit higher. I should have stopped the amination at the same height of the initial condition to avoid the impression of contact to ground.

Are there any assumptions under which the condition "must be zero"? Your pde still contains allot of parameters that can be complex and Maple cannot simplify to zero. Perhaps of missing assumptions or the simple reason that the pde is too complex.
At least Maple confirms that under certain conditions the pde is true

coulditbe(subs(G1,pde2)=0)
                              true

@mmcdara 

I tought about generating a list of all user defined names. For that a worksheet could be saved and then scanned statement by statement for names in any expression and arguments of procedures/functions. All scanns are merged in a set using union.
This set could be reduced by anames ('user').

This seems to work in principle. However, the indets command I used 

line := " ":
while line <> 0 do
   nm:=indets(parse(readline("input.mpl")),name);
   user_names:=user_names union nm;
end do:
user_names;

is not specific ennough. It captures names but also all other sort of things:

{a, aaa, b, c, false, ln, name, restart, string, user, x}
 

@mmcdara 

Thanks for confirming that my attempt was hopeless. The same seems to be true for anames ()

restart;
anames():
inital_names:={%}:
{anames()} minus inital_names;
anames(`user`)
     {Unit, Units, sprintf, value, _UIUtils, inital_names, 

       debugger/known_lineinfo, debugger/no_output}


                          inital_names

The kernel seems to establish a separate list for assigned user-defined names which is accessible via anames('user').

@nm 
assigned is of course usefull for a particular name that is known. However, often not all names that are defined in a worksheet are known.

I am looking for a kind of unames('user') command that lists all unassigned names that a user has defined. Not all unassigned names the kernel knows. The opposite can be done in with anames( 'user' ): Listing all assigned names a user has defined.

By the way: I did not know the assigned command. The anames and unames help pages, that I new, do not link to the assigned page. On the other hand, the assinged help page links to unames and anames.

@Gabriel Barcellos 

Yes there is.

In the attachment I simplified the "variant" which contained alot of code elements only required to get a formating similar to Rouben Rostamian. The simplified lists needed to be reordered for plotting. This back an forth is a bit unnecessary to generate plot lists.

If you are only interested in point plots, I would put an fsolve statement (to solve for m) and eqG (to compute G) into a time loop and generate all plot points of interest at each time step.

plot_lists.mw

The problem is the Heavyside function which is not contineous. This is a challenge for solvers using the first derivative.
In this case you either use a solver that works with the heavyside function (I don't know one) or you solve for subdomains
 

evalf(subs(Heaviside(_Z - 1433/20)=1,yB_(10)))
                          64.95141449

The above does not work for
 

evalf(subs(Heaviside(_Z - 1433/20)=0,yB_(10)))

It's probably because of the arctan function that has infinite roots (which can be complex) and which is not contineous either. The domain must be restricted again.

 

@Carl Love 

I did not know that a Matrix is also an expression (parameter x of seq).

The evalapply application is something I missed completely. ?evalapply would have been helpfull when I tried to understand the ()() construct. It seems to be impossible to find this help page with a search term. "{" and "[" direct in the right direction but there is nothing for "(" from where a "See Also" could link to pages dealing with function application or round brakets/parentheses. These brackets play an important role in Maple Language but somehow do not seem to fit under the help topics.   

Thank you

@Carl Love 

Interesting way. Is that explained on ?seq or elsewhere?

@acer

On the search for easier ways, I was wondering if a recursive expand could make sense. Something like

expand['recurse'](expr)

Similar to eval['recurse'].

(I would prefer expand because it is more specific than simplify.)

op, with all its drawbacks, is at least easy to get something done and easy to remember.