cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
Showing results for 
Search instead for 
Did you mean: 

Community Tip - Have a PTC product question you need answered fast? Chances are someone has asked it before. Learn about the community search. X

MathCAD 15 does not want to solve

pchristensen-2
4-Participant

MathCAD 15 does not want to solve

It should be fairly simple, y = 36 cm

However, MathCAD cannot solve it, how come?

1 ACCEPTED SOLUTION

Accepted Solutions

It should be fairly simple, y = 36 cm

Its not that simple and 36cm is not the exact solution! But Mathcad symbolics is here to give exact solutions and the exact solution in your example is the solution of a fifth order polynomial. There is no general mathematical algebraic solution to this problem and thats the reason Mathcads symbolic gives you this correct, but rather useless answer.

Furthermore the symbolics does not know anything about units, it does not know if m may be negative ore even imaginary and it does not know that 1L equals 10^-3 m^3.

For your task its better to use a numeric approach to get the answer you expect.

This can be a solve block

or you may use the "root" function with a guess value

or you let the "root" function search for a solution within a given interval without providing a guess

Werner

View solution in original post

10 REPLIES 10

It should be fairly simple, y = 36 cm

Its not that simple and 36cm is not the exact solution! But Mathcad symbolics is here to give exact solutions and the exact solution in your example is the solution of a fifth order polynomial. There is no general mathematical algebraic solution to this problem and thats the reason Mathcads symbolic gives you this correct, but rather useless answer.

Furthermore the symbolics does not know anything about units, it does not know if m may be negative ore even imaginary and it does not know that 1L equals 10^-3 m^3.

For your task its better to use a numeric approach to get the answer you expect.

This can be a solve block

or you may use the "root" function with a guess value

or you let the "root" function search for a solution within a given interval without providing a guess

Werner

First off, the equation where you try to solve for y does not contain y. Even though you define  A and P.eri in y, because y is undefined, A end p.eri are undefined and the symbolic processor then treats those (A and p.eri) as (un)known variables.

To get a sound equation to solve for y would be e.g.:

Secondly, the symbolic processor does not deal with units. The L/s appended to 200 are treated as another set of (un)known variables.

Thirdly, the equation is probably too complicated to solve symbolically. Try numeric:

Success!
Luc

LucMeekes wrote:

First off, the equation where you try to solve for y does not contain y. Even though you define  A and P.eri in y, because y is undefined, A end p.eri are undefined and the symbolic processor then treats those (A and p.eri) as (un)known variables.

No, it doesn't! At least not Mupad in Mathcad15. The result of the symbolics repsects the definitions of A, P.eri, etc. and doesn't treat em as unknown variables.

The main problem as already stated is that Mathcad/Mupad is forced to solve a polynomial equation of fifth order which it can't find a solution for.

Here I tell muPad a little bit about s, m and L and also rewrote I.0 as fraction to avoid muPad to autoswitch in float mode (a very nasty habit of mupad):

Making the problem unitless by defining m:=1 and s:=1 we can even get a numeric solution from the symbolics, but of course using of of the numeric ways shown is preferable

WE

LucMeekes
23-Emerald III
(To:Werner_E)

Ah, that's a difference between Mupad and Maple then.

No, I think more between Mathcad 15 and 11, because what is important is the equation that is provided to the symbolic processor.

Thanks!

Luc

Ah, that's a difference between Mupad and Maple then.

No, I think more between Mathcad 15 and 11, because what is important is the equation that is provided to the symbolic processor.

Sound plausible - I am not sure, though. Don't know if the handover of expressions to the symbolics was changed and when.

LucMeekes
23-Emerald III
(To:LucMeekes)

No, there is no difference between Mathcad (11) and Mathcad(15). Also in Mathcad 11 this works:

So the solve problem is really with the order of y in the equation. The maximum oder that is symbolically solvable is 4.

Luc

Ah, I see!

I wasn't sure and could not remember as its quite a time ago when I used Mathcad 11 😉

Was a very fine version and I still miss Maple, of course.

LucMeekes
23-Emerald III
(To:Werner_E)

I still have it running...

and use it for most of my work.

Yes, I know. Sure a good decision.

I lost mine with the last hardware upgrade and there is no way to license MC11 now.

I missed to let a copy license for a VM (guess this should have been possible).

mpchrist.jpg

Top Tags