Signal to noise ratio

  • #1
Summary:: An image was taken with a ##60## second exposure time of a 6th magnitude star and the signal to noise ratio was detected to be ##S/N = 20##.

a. What should the exposure time be if you wanted a ##S/N = 100##?
b. Now calculate the ##S/N## if it were a 2nd magnitude star for a ##10## second exposure.

For part a I got the following:

Let ##S = \mu t##, where ##\mu## is the count and ##t## is time, therefore we have $$S/N = 20$$ $$\frac{\mu t}{\sqrt(\mu t)} = 20$$ $$(\frac{\mu t}{\sqrt(\mu t)})^2= 20^2$$ $$\frac{\mu^2 t^2}{\mu t} = 400$$ therefore the count ##\mu## is ##400##. Therefore in order to get ##S/N = 100## we have $$S/N = 100$$ $$\frac{\mu t}{\sqrt(\mu t)} = 100$$ $$(\frac{400t}{\sqrt(400t)})^2= 100^2$$ $$\frac{400^2 t^2}{400t} = 10000$$ now solving for ##t## I got ##1500## seconds.

But for part b of the question, do I apply the same logic even though the magnitude of the star is different? Or will the logic be different?
 
  • Like
Likes Delta2

Answers and Replies

  • #2
34,947
11,128
The count rate ##\mu## will be different and you can calculate how much it will differ based on the magnitude.

You only consider Poisson noise here - which might be fine (and might be expected within the scope of the problem), but in general there can be other noise sources, too.
 
  • #3
The count rate ##\mu## will be different and you can calculate how much it will differ based on the magnitude.

You only consider Poisson noise here - which might be fine (and might be expected within the scope of the problem), but in general there can be other noise sources, too.
Yes, the scope is only Poisson. So the count rate for part b will be different due to the change of magnitude? So since the magnitude difference is ##4## will I get the following: $$2.5^4 * \sqrt\frac{400}{\frac{60sec}{10sec}}?$$
 
  • #4
34,947
11,128
Why did you put the magnitude difference outside the brackets?
 
  • #5
Why did you put the magnitude difference outside the brackets?
Hmm which brackets?
 
  • #7
Eh, I meant the square root.
I thought the equation was magnitude times the count? Or does the magnitude need to be inside the square root?
 
  • #8
34,947
11,128
2.54 is the ratio of photons per time. Just like all the other terms that scale with the signal it should be in the square root if you calculate signal to noise ratios.
 
  • Like
Likes learningastronomy

Related Threads on Signal to noise ratio

  • Last Post
Replies
0
Views
2K
Replies
4
Views
1K
  • Last Post
Replies
2
Views
611
Replies
2
Views
504
Replies
0
Views
805
Replies
1
Views
809
  • Last Post
Replies
3
Views
4K
Replies
2
Views
2K
  • Last Post
Replies
1
Views
1K
  • Last Post
Replies
3
Views
2K
Top