Derivative of two functions

[Marcin]

Hello

Could anybody help me with derivates?

I am trying to replicate some calculations created in different software in Mathcad prime

The biggest problem I have got is Derivative of two functions

I have attached two files:

- derivates_2.png is the calc I am trying to replicate

- Matcad.mcdx

[Fred_Kohlhepp]

Try renaming your derivative function; the apostrophy isn't working the way you think it should.

[Marcin]

Hello

I have changed the way, but this does not look right for me

Is left side correct comparing to right side

[Fred_Kohlhepp]

Your original function t(x,q)' was the first derivative of t(q) with respect to q as a finite difference. Using the derivative functions in mathcad will not give you the correct form of that equation, especially the derivative with respect to x, which you don't want.

just rename t(x,q)' to something else dT(x,q) := (t(x+q) - t(x-q))/(2q) and get the evaluation that way.

[Fred_Kohlhepp]

I don't have Prime available right now, so the attached file is xmcd (regular Mathcad.) I've added a pdf for the mathcad impaired.

I'm worried about your equation for stress area--step changes in area are rare in real life. I've plotted area (assuming I wrote it right) and a simplified area (only the first term. Both blow up at pi/2-beta.

Given area is right, you can then compute tau, shear stress. And I've given you two ways to get the derivative--a numeric computation and a symbolic deerivative of the "simple" stress (available in mathcad and Prime 2.0, not in Prime 1.0.)

Helps?

[Marcin]

That was helpful - many thanks

This calculation has been created to find shear stresses of dog stops area for screw shaft and nut

I am trying to check how accurate all this calc's above are.

[VladimirN]

You work in Pro/ENGINEER?

[Marcin]

Yes I do

[VladimirN]

Why not use in this case, Pro/MECHANICA (for strength calculations by using FEM)?

[Marcin]

I have done FEA using Pro/mechanica

Somebody have done similar calculation previously where dog stop was calculated manually and I would like to use this example for my design to confirm that FEA is correct