DeepSkyStacker FWHM values and guiding RMS?

[ChrisEll]

Hi,

I would expect the FWHM values in my images when processed in DeepSkyStacker to come down proportionally with improved guiding performance (i.e. as RMS gets lower) but I'm not seeing a strong correlation... any ideas?

Chris.

[kens]

Are you excluding all substandard subs from the stack?

[Shibby]

I suppose a better RMS doesn't necessarily mean better FWHM. If you're guiding on a decent length exposure, the seeing will be averaged out giving you a decent centroid position for the star, while affecting the FWHM of the captured star.

[vlaiv]

It is very deep topic, I did some "research" on it, and these are my findings:

There are 3 factors that contribute to FWHM:

- airy disk size, depends on aperture

- seeing

- guiding RMS

First factor is straight forward it is always the same, but second and third have correlation that depends on so may factors and is really hard to describe/calculate.

If we were to obtain precise values for second and third parameter, then FWHM can be calculated without problem. Airy disk PSF can be approximated with Gaussian shape, seeing PSF also tends in limit to Gaussian shape. Guiding is not that straight forward. It can also be approximated to Gaussian shape under certain circumstances - when there is enough randomness to mask all non random components like fast periodic error and such. That would be what we call well behaved mount. So most contributions to guide error need to come from random influences like gear imperfections that are not periodic (on a small scale), wind, shake, error in star position measurement, whatever that is random in nature.

When you have these conditions, calculation works, and it is simple calculation, since you convolve 3 gaussian PSF - you end up with gaussian PSF again that has sigma equal to sum of individual sigmas.

But problem is that some of guide errors come from seeing, so in order for calculation to give better value, one needs to minimize impact of seeing on guide RMS - usually by using long guide exposure and trying not to chase the seeing.

So while result FWHM depends on guiding it is not linear, but there is formula to describe dependence if all above conditions are satisfied.

This also has another implication. Seeing FWHM can't be judged from an image, it also depends on guide precision and aperture used to record seeing. So having two or more different setups one would be able to get good estimate of seeing FWHM, or alternatively very reliable mount that tracks to perfection and then adjust for aperture.

[ChrisEll]

Thanks all, I'm not surprised it's more complex than it seems though I did suspect seeing was playing a part.