Focussing with sharpcap (bahtinov mask)

[Demonperformer]

I am starting to get to grips with some of the features on Sharpcap and last time out I tried to use the focussing assistant using a bahtinov mask.

The first thing I noticed is how sensitive it is. The attached screenprint involved the tiniest movements of the fine focusser knob on my Star71, which caused it to sweep from +1 to -1 without stopping in the middle. I decided I wasn't going to do any better than that (but could probably waste an awful lot of clear-sky-time trying!), so decided to accept that. After all, from what I have read, an FWHM of about 1 is reasonable.

But now comes my confusion. I was assuming that the scale was FWHM units. But, as can be seen from the screenprint of the log, the FWHM of each individual sub was just under 4. I realise it would not necessarily be the exact same figure as when I was focussing, but 4x that figure? This would seem to suggest that either the scale on the focussing assistant is not FWHM units (so what is it?), or my focussing went way out during my imaging session. The resultant image can be seen here.

So, are these the sort of results I should be expecting? Or where am I going wrong?

Thanks.

[LightBucket]

Is it because the top image shows a single exposure average, and the bottom is the stack average.... ?

[SamAndrew]

When using a Bahtinov mask it is giving you a score based on how well the 3 lines intersect. 0 would be perfect. It's not the fwhm value.

I've not had my star71 for long, but a good FWHM values seems to be under 3.0; I haven't seen it below 2.5

I've found during the first hour after the scope has come out from indoors I need to refocus every 10-15 minutes. I lost a load of subs on one of my first outings as I didn't expect to have a particular long window of clear sky, so didn't recheck focus and the FWHM values drifted from 3.0 to 5.5 over about 2 hours ? 

I believe it's better to be slightly outside of focus (i.e. drawtube over extended) as it will contract as it cools.

[Demonperformer]

19 hours ago, LightBucket said:

Is it because the top image shows a single exposure average, and the bottom is the stack average.... ?

The log gives you both. For example, the 06:15:10 INFO gives the FWHM of the stacked image as 4.295013, and the FWHM of the frame image as 3.782281.

Thanks.

[Demonperformer]

19 hours ago, SamAndrew said:

When using a Bahtinov mask it is giving you a score based on how well the 3 lines intersect. 0 would be perfect. It's not the fwhm value.

I've not had my star71 for long, but a good FWHM values seems to be under 3.0; I haven't seen it below 2.5

I've found during the first hour after the scope has come out from indoors I need to refocus every 10-15 minutes. I lost a load of subs on one of my first outings as I didn't expect to have a particular long window of clear sky, so didn't recheck focus and the FWHM values drifted from 3.0 to 5.5 over about 2 hours ? 

I believe it's better to be slightly outside of focus (i.e. drawtube over extended) as it will contract as it cools.

So does the bahtinov score relate to FWHM linearly? I was getting a score that was either +1 or -1 and that equated to a FWHM of about 4, so if I wanted a FWHM of 3 I would need a score of 0.75? Or is it not that simple?

Thanks.

[Demonperformer]

Just had another "bright" thought ... will the longer exposure time (15s vs 5s) cause the stars to be more "bloated" and so produce a worse FWHM figure? Not totally sure how the FWHM thing works, so I may be talking complete anagramed-fish there (so nothing new!).

[SamAndrew]

3 hours ago, Demonperformer said:

So does the bahtinov score relate to FWHM linearly? I was getting a score that was either +1 or -1 and that equated to a FWHM of about 4, so if I wanted a FWHM of 3 I would need a score of 0.75? Or is it not that simple?

Thanks.

It wouldn't be linear, 0 would be perfect focus, but that might equate to FWHM of 2.5 or 3.0 etc depending on the seeing.

[SamAndrew]

Not sure on how FWHM compares vs exposure length, I would think you could get lucky with a short exposure and get a better value, but over a long exposure if your tracking is sufficient then it won't increase with time (I'm sure someone else can correct me)

[Merlin66]

FWHM values should not be different until you start to saturate and over expose the star image.

All unsaturated star images have the same FWHM.

 

[Stub Mandrel]

Bahtinov mask score does not relate directly to FWHM. The point where the Bmask lines cross is unaffected by seeing, while FWHM is affected by focus AND seeing.

[Demonperformer]

So although I have an ideal figure of zero (which in practice is not going to happen), I don't really know what figure I should be aiming for as an "acceptable" result (ie when I should stop fiddling with it and get on with doing some actual imaging)?

[Demonperformer]

Now i'm not on my imaging laptop atm, so I will need to check later, but I think there is a FWHM option for focussing on Sharpcap. That may give me a figure I can use ...

[Stub Mandrel]

The problem with these assistants is that the number you get depends on your system  (on in the case of Bahtinov mask the rate it changes away from zero).

I've found that it also depends on things like the conditions, setup, size of the target, bit depth and the exposure time.

In short, they all need to be seen as 'relative' systems.

What I have found (in spades) is that the contrast detection in Sharpcap is VERY good for planetary, so good I can spend ten minutes of precious time on focusing because it give such an improvement.

For guiding I use FWHM in Sharpcap for focusing. With my 50mm 138mm f3.5 guidescope and ASI120MC 4.5 is about as good as I can get, and allows me to get decent guiding. Clearly you would probably want a much lower FWHM for imaging, but you would have a much larger aperture/higher quality scope.