Proving the Properties of Curl

  • Thread starter Thread starter SithsNGiggles
  • Start date Start date
  • Tags Tags
    Curl Properties
SithsNGiggles
Messages
183
Reaction score
0

Homework Statement


The curl satisfies

(A) curl(f+g) = curl(f) + curl(g)

(B) if h is real values, then curl(hf) = hcurl(f) + h'·f

(C) if f is C2, then curl(gradf) = 0

Show that (B) holds.

2. The attempt at a solution
I'm not quite sure how to interpret the "h is real valued" part. Does it mean that, as is, h is a scalar quantity and its derivative is a vector?

Any help is appreciated.

- - - -
So far I've been reviewing the definition of curl.
I let f = f(x, y, z) = (f1(x, y, z), f2(x, y, z), f3(x, y, z))

hf = (hf1(x, y, z), hf2(x, y, z), hf3(x, y, z))

curl(hf) = (D2(hf3) - D3(hf2), D3(hf1) - D1(hf3), D1(hf2) - D2(hf1)),

but I don't see where this route is taking me.
 
Last edited:
Physics news on Phys.org
Probably means h is a real valued function. Something like
h(x_1,x_2...)\vec{f}(x_1,x_2...)
so the values of h are scalars, but it is not necessarily constant.
 
yeah I would probably take it as real scalar function of position h = h(x,y,z), multiplying anything else by a vector does not make sense unless it is a vector product
 
also the fact that h' is in a dot product means it is probably the gradient of h
 
I looked at the problems ahead, and there's a similar one involving divergence, where I have to show that

div(hf) = hdiv(f) + gradh·f,

so I'm not sure if h' is the gradient in the curl question.
 
SithsNGiggles said:
I looked at the problems ahead, and there's a similar one involving divergence, where I have to show that

div(hf) = hdiv(f) + gradh·f,

so I'm not sure if h' is the gradient in the curl question.

In my reference the h'.f term is the cross product of grad(h) with f.
 
SithsNGiggles said:
So far I've been reviewing the definition of curl.
I let f = f(x, y, z) = (f1(x, y, z), f2(x, y, z), f3(x, y, z))

hf = (hf1(x, y, z), hf2(x, y, z), hf3(x, y, z))

curl(hf) = (D2(hf3) - D3(hf2), D3(hf1) - D1(hf3), D1(hf2) - D2(hf1)),

but I don't see where this route is taking me.

I ended up continuing with process and was able to simplify the expression to

curl(hf) = h(D2f3 - D3f2, D3f1 - D1f3, D1f2 - D2f1) + ((D2h)f3 - D3h)f2, (D3h)f1 - D1h)f3, (D1h)f2 - D2h)f1)

curl(hf) = h curl(f) + ((D2h)f3 - (D3h)f2, (D3h)f1 - (D1h)f3, (D1h)f2 - (D2h)f1)

Is there any way to simplify the rest of this expression?
 
SithsNGiggles said:
I ended up continuing with process and was able to simplify the expression to

curl(hf) = h(D2f3 - D3f2, D3f1 - D1f3, D1f2 - D2f1) + ((D2h)f3 - D3h)f2, (D3h)f1 - D1h)f3, (D1h)f2 - D2h)f1)

curl(hf) = h curl(f) + ((D2h)f3 - (D3h)f2, (D3h)f1 - (D1h)f3, (D1h)f2 - (D2h)f1)

Is there any way to simplify the rest of this expression?

The rest of it looks like grad(h) x f. Where 'x' is the vector cross product.
 
Dick said:
The rest of it looks like grad(h) x f. Where 'x' is the vector cross product.

I suppose there's no way of expressing "grad(h) x f" as "h' f", is there?
 
  • #10
SithsNGiggles said:
I suppose there's no way of expressing "grad(h) x f" as "h' f", is there?

I have no idea what "h' f" is supposed to mean. It's no standard notation I've ever seen. I wouldn't torture yourself over it. You've got the vector formula right.
 
  • #11
Good enough, I suppose. Thanks everyone for the help.
 
Back
Top