Ivar KJELBERG
COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
25.04.2011, 17:35 GMT-4
Hi
COMSOL works in absolute coordinates, for your case it "t" the driver, but be aware that the solver in default Automatic mode will try to increase the steps by doubling them and then interpolate the results. This is OK for asymptomatic responses, but for harmonic or cyclic as for your case you should set the solver stepping to intermediate or strict
You need to define two functions that operate in different time regions if you want to dissociate them (first X then Y etc)
A few weeks ago someone there was a few threads about moving laser beams, try a search
--
Good luck
Ivar
Hi
COMSOL works in absolute coordinates, for your case it "t" the driver, but be aware that the solver in default Automatic mode will try to increase the steps by doubling them and then interpolate the results. This is OK for asymptomatic responses, but for harmonic or cyclic as for your case you should set the solver stepping to intermediate or strict
You need to define two functions that operate in different time regions if you want to dissociate them (first X then Y etc)
A few weeks ago someone there was a few threads about moving laser beams, try a search
--
Good luck
Ivar
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
27.04.2011, 10:24 GMT-4
Hi,
Here is a few hint:
You can use Parametric Sweep to parametrize your Spot position. For exemple you can parametrize a rectangular curve lying on a surface.
Time-Dependant solver will solve for each parameter value determined in Parametric Sweep solver.
You can also discretize your spot path and sets functions for each delimited boundary to emulate your spot movement.
You can also use ALE to move mesh (spot).
Please look at COMSOL user manual for function/solver explanation.
Here are some thread that could help you. Some contain .mph files:
www.comsol.com/community/forums/general/thread/14090/
www.comsol.com/community/forums/general/thread/13688/
www.comsol.com/community/forums/general/thread/15944/
www.comsol.com/community/forums/general/thread/16147/
Hi,
Here is a few hint:
You can use Parametric Sweep to parametrize your Spot position. For exemple you can parametrize a rectangular curve lying on a surface.
Time-Dependant solver will solve for each parameter value determined in Parametric Sweep solver.
You can also discretize your spot path and sets functions for each delimited boundary to emulate your spot movement.
You can also use ALE to move mesh (spot).
Please look at COMSOL user manual for function/solver explanation.
Here are some thread that could help you. Some contain .mph files:
http://www.comsol.com/community/forums/general/thread/14090/
http://www.comsol.com/community/forums/general/thread/13688/
http://www.comsol.com/community/forums/general/thread/15944/
http://www.comsol.com/community/forums/general/thread/16147/
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
30.04.2011, 15:26 GMT-4
Hi Dear Ivar and Gabriel
Dear Ivar Please tell me how can i define two functions that operate in different time regions for dissociate them (first X then Y etc)?
Dear Gabriel
I read COMSOL user manual for function/solver explanation but i don't understand.
please see below pictures and tell me what is the mean of 0 5 2 3 4 8 that is in parametric value?
i813.photobucket.com/albums/zz57/mastercamsml/1-4.jpg
i813.photobucket.com/albums/zz57/mastercamsml/2.png
Thank you
Hi Dear Ivar and Gabriel
Dear Ivar Please tell me how can i define two functions that operate in different time regions for dissociate them (first X then Y etc)?
Dear Gabriel
I read COMSOL user manual for function/solver explanation but i don't understand.
please see below pictures and tell me what is the mean of 0 5 2 3 4 8 that is in parametric value?
http://i813.photobucket.com/albums/zz57/mastercamsml/1-4.jpg
http://i813.photobucket.com/albums/zz57/mastercamsml/2.png
Thank you
Ivar KJELBERG
COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
30.04.2011, 16:02 GMT-4
Hi
for the parameter list, as you have two parameters the set must be read by pairs Param1(1) & Param2(1), Param1(2), Param2(2) etc
I usually write the pairs with commas such as
0 5, 2 3, 4 8,
then it's easier to read, and commas "," or spaces " " are equivalent for sets
I'm not sure what you mean by dissociation for your function, if its that they should be alternatively constant in certain regions, you write them as periodic or repeating functions with slopes and cte regions.
There are several threads on the forum discussing laser moving paths, or load paths
to draw a circle of radius R around X0,Y0 at a speed of w0[1/s] you write
X-X0+R*sin(w0*t[1/s])
Y-Y0+R*cos(w0*t[1/s])
if you want it to look more like a square you replace the sin/cos by smoothed periodic "rect" functions in V4
--
Good luck
Ivar
Hi
for the parameter list, as you have two parameters the set must be read by pairs Param1(1) & Param2(1), Param1(2), Param2(2) etc
I usually write the pairs with commas such as
0 5, 2 3, 4 8,
then it's easier to read, and commas "," or spaces " " are equivalent for sets
I'm not sure what you mean by dissociation for your function, if its that they should be alternatively constant in certain regions, you write them as periodic or repeating functions with slopes and cte regions.
There are several threads on the forum discussing laser moving paths, or load paths
to draw a circle of radius R around X0,Y0 at a speed of w0[1/s] you write
X-X0+R*sin(w0*t[1/s])
Y-Y0+R*cos(w0*t[1/s])
if you want it to look more like a square you replace the sin/cos by smoothed periodic "rect" functions in V4
--
Good luck
Ivar
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
06.02.2013, 00:54 GMT-5
Hi to all
I want have laser beam movement.
In circular movement i used below equation :
X : R * sin(omega * t )
Y : R * cos(omega * t )
Z : X
and now i want have a rectangular movement :
i813.photobucket.com/albums/zz57/mastercamsml/Untitled-1-1.png
I want use below equation for it :
X : V * t --> V : laser velocity
Y : D * t --> D: 0.025 ( Y step movement)
Z : ??
The question:
What method does Comsol software use? (Incremental or absolute coordination)?
In circle movement does the comsol take before value plus with new value (10 omega + 10 omega = 20 omega for new coordination position)? or the coordination system rotation step by step?
what do you offer for rectangular movement? if i use ( X: V * t ) it is ok for first line and when the Y axis step to up level, X movement must be in the opposed direction. what should i do for move in opposed direction in x axis?
What about in Z axis?
Please attention that the Y axis movement must be done when the X axis movement is finished.
Thank you
Hi everybody,
Did you guys solve this issue? I am really confused by such professional term you are talking about in the forum. Could you please just tell me how could I define movement in both direction (x and y)? my case is exactly similar to the Larson's.
Thanks
[QUOTE]
Hi to all
I want have laser beam movement.
In circular movement i used below equation :
X : R * sin(omega * t )
Y : R * cos(omega * t )
Z : X
and now i want have a rectangular movement :
http://i813.photobucket.com/albums/zz57/mastercamsml/Untitled-1-1.png
I want use below equation for it :
X : V * t --> V : laser velocity
Y : D * t --> D: 0.025 ( Y step movement)
Z : ??
The question:
What method does Comsol software use? (Incremental or absolute coordination)?
In circle movement does the comsol take before value plus with new value (10 omega + 10 omega = 20 omega for new coordination position)? or the coordination system rotation step by step?
what do you offer for rectangular movement? if i use ( X: V * t ) it is ok for first line and when the Y axis step to up level, X movement must be in the opposed direction. what should i do for move in opposed direction in x axis?
What about in Z axis?
Please attention that the Y axis movement must be done when the X axis movement is finished.
Thank you
[/QUOTE]
Hi everybody,
Did you guys solve this issue? I am really confused by such professional term you are talking about in the forum. Could you please just tell me how could I define movement in both direction (x and y)? my case is exactly similar to the Larson's.
Thanks
Ivar KJELBERG
COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
06.02.2013, 01:14 GMT-5
Hi
COMSOL uses a basic Cartesian coordinate system, so if you want to move in a circle, set up a cylindrical coordinate and define a radial velocity*time for your position, or write the equations out fully
x=R*sin(2*pi*freq*t)
y= R*cos(2*pi*freq*t)
it's not more complicated than that ;)
--
Good luck
Ivar
Hi
COMSOL uses a basic Cartesian coordinate system, so if you want to move in a circle, set up a cylindrical coordinate and define a radial velocity*time for your position, or write the equations out fully
x=R*sin(2*pi*freq*t)
y= R*cos(2*pi*freq*t)
it's not more complicated than that ;)
--
Good luck
Ivar