Differentiating $f(F,\theta)$: Why the Answers Differ?

  • Thread starter Thread starter agnimusayoti
  • Start date Start date
  • Tags Tags
    Differentiating
Click For Summary
The discussion centers on the differentiation of the function f(F, θ) = F cos θ, leading to two different results for df based on the methods used. The first method yields df = 5.28 N, while the logarithmic approach results in df = 4.28 N. Participants highlight the importance of checking derivatives and the implications of error correlation in calculations. It is noted that the first method can indeed be used to derive the relative error found in the second method. The conversation concludes with a request for clarification on the concept of correlated errors in this context.
agnimusayoti
Messages
239
Reaction score
23
Homework Statement
A force of 500 N is measured with a possible error of 1 N. Its component in a
direction 60◦ away from its line of action is required, where the angle is subject to an
error of 0.5◦. What is (approximately) the largest possible error in the component?
Relevant Equations
For ##f(F,\theta) \rightarrow dF= \frac{\partial f}{\partial F} dF +\frac {\partial f}{\partial \theta} d\theta ##
From the question,
$$f(F,\theta)=F \cos \theta$$

1. If I use:
$$df=dF \cos{\theta} -F \sin {\theta} d\theta$$
and using radian,
$$df=dF \cos{\theta} -F \sin {\theta} d\theta \frac {\pi}{180^\circ}=5.28 N$$

2. But, if I take logarithm to both side:
$$ln f=ln F+ln \cos{\theta}$$
differentiate both sides:
$$\frac{df}{f}=\frac{dF}{F} + \frac{\sin\theta}{\cos \theta} d\theta$$
Using radian, it gives ##df=4.28 N##My question is, why the answers are different? Thanks.
 
Last edited by a moderator:
Physics news on Phys.org
Signs matter.
 
  • Like
Likes Delta2, fresh_42 and DaveE
Recheck your derivatives
 
I think it should also be added that there is no reason to assume that the errors are correlated.
 
Yeah the second should be negative ##-\tan \theta d\theta##. But then for the worst case or should be positive, because we can take d\theta ia negative then ##-\tan \theta d\theta## becomes positive.

With this condition (worst case), so the first one becomes positive and brings ##df=5.28 N##What do you mean by the errors are correlated? Could you please explain itu further? Because the book show to get relative error by the second method. But i just wonder why can't be calculated by 1st method.
 
agnimusayoti said:
Because the book show to get relative error by the second method. But i just wonder why can't be calculated by 1st method.
You can. If you divide your expression for ##df## you got using the first method by ##f##, you get the expression you found for ##df/f## using the second method.
 
  • Like
Likes agnimusayoti, joshyewa and Delta2
Uh youre right. Thanks!
 
But again what is the meaning by the errors are correlated?
 

Similar threads

  • · Replies 14 ·
Replies
14
Views
2K
Replies
3
Views
2K
  • · Replies 4 ·
Replies
4
Views
1K
  • · Replies 6 ·
Replies
6
Views
2K
  • · Replies 1 ·
Replies
1
Views
2K
  • · Replies 3 ·
Replies
3
Views
2K
Replies
28
Views
2K
  • · Replies 11 ·
Replies
11
Views
2K
  • · Replies 8 ·
Replies
8
Views
2K
  • · Replies 3 ·
Replies
3
Views
2K