Darks Puzzle - Minimum ADU lower than offset

[Tommohawk]

Just been reviewing my darks and notice that the minimum ADU (shows in FITs liberator) is 16, with offset set at 64. How is that possible?? Maybe I'm misunderstanding the units or something.

Grateful for any thoughts!

[david_taurus83]

What camera? Dead pixel? Does the offset necessary translate into 16bit ADU?

[Tommohawk]

Hi and thanks for your reply. ASI1600 cool.

But I think I may be misunderstanding the FITS data. The image below suggests "Input minimum 16" But when I hover over pixels the lowest value is about 592....  so I think its OK and I'm just mis-reading the data.

Does that make sense?

[teoria_del_big_bang]

Can you attach a dark, in Pixelmath I should be able to turn any pixel under the offset red so you can see how many they are ?

Steve

[teoria_del_big_bang]

I take it you are looking at the individual dark frames and not the calibrated master dark ?

Steve

[Tommohawk]

Thanks Steve, thats good of you.  - see attachment. Yes, individual frames not master.

 2020-03-28_23-26-49_SII_-20.00_60.00s_139.00_64.00_64_.fits

Edited November 8, 2021 by Tommohawk

[teoria_del_big_bang]

Hi Tom,

I think I am right in saying all pixels are above the 64 offset.

Problem I always have with these cameras is that they are 14 bit cameras in terms of data they can store but I think everything is still stored as 16 bit because you cannot store a 14 bit number.
I think I am right there, but I cannot say 100%.
So if the pixel value is actually a 16 bit number then there are no values under 64, I tested it with Pixelmath.

Steve

 

[Tommohawk]

Hi Steve - thanks for checking. 

I'm sure its all OK ... what got me started was just checking that I had done the darks with the same offset (64) which Ive adopted for lights and flats. And when I saw that "minimum ADU value" as 16, I thought that cant be right. But as you say no pixels are that low. So not sure what that represents! Anyhow the darks seem to work OK, so hopefully all good.

Some of this stuff is still a bit of a mystery to me ....!

[teoria_del_big_bang]

I tell a lie I have found 1, but it really is an odd cold pixel that is all.

Steve

 

[Tommohawk]

11 minutes ago, teoria_del_big_bang said:

I tell a lie I have found 1, but it really is an odd cold pixel that is all.

Good spot! TBH if there's only 1 dead that's probably pretty good. 

[teoria_del_big_bang]

19 minutes ago, Tommohawk said:

Hi Steve - thanks for checking. 

I'm sure its all OK ... what got me started was just checking that I had done the darks with the same offset (64) which Ive adopted for lights and flats. And when I saw that "minimum ADU value" as 16, I thought that cant be right. But as you say no pixels are that low. So not sure what that represents! Anyhow the darks seem to work OK, so hopefully all good.

Some of this stuff is still a bit of a mystery to me ....!

I think it is correct, the lowest ADU reading I can find is at 1403, 942 and gives a value of 0.0002 on the readout of Pixinsight.

If I use a really good script then it finds the minimum ADU pixel and its value (as well as loads of other things if asked) and it gives the exact value of 0.00024414 which is the Pixelmath normalised range (so 1= 65535) then this is 16 in a 16 bit camera which is what you are seeing I think but as I say it really is an odd pixel that are these sort of values and so will be just a cold pixel which is normal for any camera to have a few of these.

I think I am right in what I am saying here, still pretty new to AP so still learning but thats how I read it anyway 🙂 

Steve

Edited November 8, 2021 by teoria_del_big_bang

[teoria_del_big_bang]

18 minutes ago, Tommohawk said:

Some of this stuff is still a bit of a mystery to me ....!

Well I am with you there 🙂