symbolic derivation inconsistency

[bmulyanto]

I am trying to write a sheet for grad, div, and curl.

I am finding that one symbol (rho) is considered as a constant instead of a variable.

what should I do to rectify this?

[Fred_Kohlhepp · Accepted solution]

Prime has an annoying habit of labeling your variables and constants.  You (apparently) have rho as a constant:

[bmulyanto]

Yes...I think prime has this weird behaviour. I changed rho with other symbol and changed it back to rho...and it showed a correct result.

The thing is I dont always have the opportunity to debug this kind of error..although I have sort of an expected result beforehand.

Do you have any tips for me?

[Fred_Kohlhepp]

As Luc says, the way Prime labels is a mystery to us all.  Prime does format the variable differently from label type to label type, although there are very minute differences.

The original Mathcad didn't really use labels a lot.  So if you defined a variable m := 24 gm (call the mass the letter m, and assign it the value of 24 grams), Then a little further on defined length := 6 m (define the length to be six meters), what you really told Mathcad was "define the variable "length" and assign it the value of 144 grams." (6 x 24 = 144)

So when they redid Prime, they used labels; you can have m (as a variable or constant) equal to 24 grams, AND m (labeled as a unit) to represent one meter.

The result:  They solved one issue and created another.

Tips?  About this?  When you see something that doesn't seem to make sense, stop and figure it out.  It's usually time well spent.

[LucMeekes]

Tip:

Make sure that you have well distinguished looks of the same 'name' when it is labelled differently. E.g. Have your 'variables' be italic, your 'units' be upright and maybe in a different colour, etc. That will show more easily when two items are different(ly labelled).

Success!
Luc

[LucMeekes]

Why would rho be (labelled as) a constant? Prime doesn't 'know' the constant rho.

When I open the sheet in Prime 4:

- rho as the argument to the differentiation is labelled as a variable.

- rho as the differentiatioon variable (rho after 'd') is labelled as an 'automatic'.

Apparently that suffices to see them as different beasts, hence the differential results in 0 instead of 1.

Anyway, the issue occurs as a result of Prime's automatic labelling.

Success!
Luc