XANSYS Message: 166 [Go back to message list]
[bookmark on del.icio.us]
No rating yet
Rate item:

Subject: Re: How to define units properly? --- please help
Author: Jean-Pierre Dube
Date: 1999-01-11 10:56:00

Dear ANSYS users,

Concerning the use of consistent units someone previously
suggested the true solution is understanding units of
measure and consistent units.

I also feel this is the way to go so here is how I understand the
role of units in case it may help someone else.

Go back to the basic model of the physics of what you
are trying to model.

For the common case of a material deforming under loads
generating a stress field you have

Stress = Young's Modulus x strain (1)

You can make this equation look much more complicated
but for our purposes here this is fine.

This equation will be true no matter what numbers you use
because this is what is built into the program (via the
element behavior)

For example: If in your model length was measured in mm
and you entered Young's modulus in Pa (N/m^2) you can
see the stress result will be a million times too small. It is
as though you modeled your material 1000 times larger
than it actually is (in reality).

You can deal with this by stating your results in micro Pa
(which will confuse everyone) or scale your results by
1E6 (can probably do this) or what is probably best
is enter Young's modulus in N/mm^2.

You could scale your model to use meters as a length unit
but this violates why the geometry was in millimeters to start
with. (2.5 millimeter becomes - 0.0025 meters and that could
also be confusing)

For a problem that involves accelerations an additional
reality applies:

Force = Mass x Acceleration

Check your units against this equation.

When energy is involved (heat, magnetic or electric fields)
units for these quantities will appear in the model of
the problem ex.

Q = kA(T2 - T1)

Some models of reality are empirically based. Unless they
are done in dimensionless terms (other headaches) they
will inherently contain factors only to be used with a
particular set of units.

Is this why ANSYS magnetics has restricted the unit
choice for magnetics?

Jean-Pierre Dube
0Q64 MCAD Tools and Methods
Nortel Networks
(613) 763-8019
dubejp@n...

> -----Original Message-----
> From: Tham [SMTP:cstham@f...]
> Sent: Sunday, January 10, 1999 7:32 PM
> To: xansys@o...
> Subject: [xansys] How to define units properly? --- please help

> From: Tham

> Hello Everybody,

> Regarding the units, Ansys documentation says that "The ANSYS program does
> not assume a system of units for your analysis. Except in magnetic field
> analyses, you can use any system of units so long as you make sure that
> system for all the data you enter (Units must be consistent for all input
> data)

> So, (for static analysis) can I assume that if Im using SI, I should enter
> ALL the units in SI, for example: All dimensions (during creating a model,
> perhaps radius and length of a cylinder) in meter, Applied forces in N,
> Pressure in Pa (or N/m) , material properties such as Young Modulus in Pa,
> density in kg/m3 and so on?

> Any help would be appreciated.

> Rgds,
> CS Tham
> Researcher
> Mechanical Engineering Dept,
> University of Malaya
> Malaysia.

Posts possibly associated with message #166AuthorDateScore
163How to define units properly? --- please helpClement Tham1999/01/10 
164Re: How to define units properly? --- please helpFrank Barcikowski1999/01/11 
165Re: How to define units properly? --- please helpChris Rogers1999/01/11 
166Re: How to define units properly? --- please helpJean-Pierre Dube1999/01/11