Discussion Closed This discussion was created more than 6 months ago and has been closed. To start a new discussion with a link back to this one, click here.

shock acceleomter

Please login with a confirmed email address before reporting spam



hai all,

Iam doing poject on shock accelerometer.

i want to apply input as semisinusoidal acceleation(1e5g (with in 50micoseconds ).

if you know any thing to define a function(i.e semi sinu accelaation),

please help me, it could be very helpful my poject.





sampath kumar

4 Replies Last Post 13.04.2010, 07:08 GMT-4

Please login with a confirmed email address before reporting spam

Posted: 10 years ago 11.04.2010, 13:50 GMT-4
Hi

I would try something like :

Fmax = 1E5*9.81[m/s^2]

F=Fmax*sin(pi*t[1/s]/(50[ms]))*(t<50[ms])

, or just

F=Fmax*sin(pi*t/(0.050))]))*(t<0.050)

But the turn "on" and "off" is rather rough so you might have some convergence issues, you should step carefully around t=0 and t=0.050

Some sort of smoothing for the turn on/off could also help

Have fun Comsoling
Ivar

Hi I would try something like : Fmax = 1E5*9.81[m/s^2] F=Fmax*sin(pi*t[1/s]/(50[ms]))*(t

Please login with a confirmed email address before reporting spam

Posted: 10 years ago 13.04.2010, 02:11 GMT-4
thank you sir
but this is expression for acceleration,i have a choice of applying only force at particular boundary. i want to convert this acceleration as a force,i have mass of 1.216*10e3.
could you tell me how to apply that acceleration as a force per area at particular boundary.
thank you sir but this is expression for acceleration,i have a choice of applying only force at particular boundary. i want to convert this acceleration as a force,i have mass of 1.216*10e3. could you tell me how to apply that acceleration as a force per area at particular boundary.

Please login with a confirmed email address before reporting spam

Posted: 10 years ago 13.04.2010, 02:25 GMT-4
thankyou sir
i want to apply this acceleration as a force per area at particular boundary.
i have a mass of 1.216*10e3.
could you tell me how to apply this acceleration as a force per area.
it will help me a lot.
thank you
thankyou sir i want to apply this acceleration as a force per area at particular boundary. i have a mass of 1.216*10e3. could you tell me how to apply this acceleration as a force per area. it will help me a lot. thank you

Please login with a confirmed email address before reporting spam

Posted: 10 years ago 13.04.2010, 07:08 GMT-4
Hi

I'm not 100% I fully understanding you, but one thing that confused me in the beginnng was that COMSOL wants all forces "F" as pressures [Pa=N/m]. In between I understand why and have got used to this, so if this is the issue I can explain a little:

Forces in Newton [N] are typically spread out onto edges or better surfaces (or even volumes/bodies such as with gravity) to act in a more homogenious way, all depends on youre model and desires.

Therefore using
F[N] = M[kg] * g[m/s^2]

force is mass times acceleration or Netons law, you easily transform this to

P[Pa=N/m^2] = M[kg] * g[m/s^2] / (Area[m^2])

for forces applied to surfaces or to

Flin[N/m] = M[kg] * g[m/s^2] / (Length[m])

For forces per length for forces applied over edges.

I ususally calcualte my volume "V", areas "A" or lengths "L" on my COMSOL geometry by using the "Options - Integration Coupling Variables - Subdomain variables" respectively Boundary variables or Edges variables, by integrating the value "1"

When you write in the Boundary Integration Variable GUI, under name = expression something like

Area = 1

This means in fact Area = double finite integration over the boundaries selected of the expression "1" or
Area = int_A 1 dA

The programmers could really have added this in the GUI title as a reminder, it took me quite some time to get catch on and really learn all these special implicit elements in COMSOL. But once its in your head it becomes so usual that you do not even think over it, probably that is what is happening for the COMSOL programmers, they forget that we new users need to understand and learn too ;)

Back to your case if I get it right now:

you are applying an acceleration field, similar to a gravity, you should apply body (volumic) forces and not surface pressure, which means that you do not apply an external force as such on your structure surface/boundary, but you apply an external body load on the full volume.

In this case you should add it as a "Physics - Subdomain settings - Load tab" select the bodie(s) and apply a force of

rho_xxx*Acc

where rho_xxx is the density of the material(s) with the xxx to be replaced by the name of the application mode used (for 3D structural "rho_smsld", for 2D perhaps "rho_smpn" = standard naming convention of COMSOL)
and Acc is a constant you have defined as the acceleration desired in your case
Acc = 1E5*9.81[m/s^2].

If you check the units you will see that they match with the density (except if you are in 2D and you have selected "Body load is defined as force/area" then you must use the default thickness and divide by "thickness_smsld" (replace smsld by whatever your active module name is).

hope this is closer to your need
Have fun Comsoling
Ivar
Hi I'm not 100% I fully understanding you, but one thing that confused me in the beginnng was that COMSOL wants all forces "F" as pressures [Pa=N/m]. In between I understand why and have got used to this, so if this is the issue I can explain a little: Forces in Newton [N] are typically spread out onto edges or better surfaces (or even volumes/bodies such as with gravity) to act in a more homogenious way, all depends on youre model and desires. Therefore using F[N] = M[kg] * g[m/s^2] force is mass times acceleration or Netons law, you easily transform this to P[Pa=N/m^2] = M[kg] * g[m/s^2] / (Area[m^2]) for forces applied to surfaces or to Flin[N/m] = M[kg] * g[m/s^2] / (Length[m]) For forces per length for forces applied over edges. I ususally calcualte my volume "V", areas "A" or lengths "L" on my COMSOL geometry by using the "Options - Integration Coupling Variables - Subdomain variables" respectively Boundary variables or Edges variables, by integrating the value "1" When you write in the Boundary Integration Variable GUI, under name = expression something like Area = 1 This means in fact Area = double finite integration over the boundaries selected of the expression "1" or Area = int_A 1 dA The programmers could really have added this in the GUI title as a reminder, it took me quite some time to get catch on and really learn all these special implicit elements in COMSOL. But once its in your head it becomes so usual that you do not even think over it, probably that is what is happening for the COMSOL programmers, they forget that we new users need to understand and learn too ;) Back to your case if I get it right now: you are applying an acceleration field, similar to a gravity, you should apply body (volumic) forces and not surface pressure, which means that you do not apply an external force as such on your structure surface/boundary, but you apply an external body load on the full volume. In this case you should add it as a "Physics - Subdomain settings - Load tab" select the bodie(s) and apply a force of rho_xxx*Acc where rho_xxx is the density of the material(s) with the xxx to be replaced by the name of the application mode used (for 3D structural "rho_smsld", for 2D perhaps "rho_smpn" = standard naming convention of COMSOL) and Acc is a constant you have defined as the acceleration desired in your case Acc = 1E5*9.81[m/s^2]. If you check the units you will see that they match with the density (except if you are in 2D and you have selected "Body load is defined as force/area" then you must use the default thickness and divide by "thickness_smsld" (replace smsld by whatever your active module name is). hope this is closer to your need Have fun Comsoling Ivar

Note that while COMSOL employees may participate in the discussion forum, COMSOL® software users who are on-subscription should submit their questions via the Support Center for a more comprehensive response from the Technical Support team.