Auto focusser or Bhatinov mask?

[philherbert]

Hi,

I've been imaging for a while now, and only ever used a Bhatinov mask to get fairly good focus. Recently I have been wondering about getting an auto focusser. I have two questions:

1. Is an auto focusser worth it - the added complexity and things to go wrong?

2. How accurate is it? I use APT with the Bhatinov mask and focussing aid, but the reported 'accuracy' varies substantially. And yes, this is after waiting for the scope to cool down. I fear that an auto focusser could be badly confused by this (presumably caused by the seeing) and end up getting it wrong?

Thanks,

Phil

[spillage]

What scope do you have? I have a cheaper skywatcher focus motor and love it. You can remotely focus and make very fine adjustments easily. You can run the auto focus then you either change filters or your subs start to look worse. I will not constantly check and adjust the focus for you over the entire session.

[ollypenrice]

I stick to manual focus because I hate things going wrong - and autofocus systems do go wrong. I host several of them in our robotic sheds. They do mostly work but when they don't they don't. And they're expensive to set up as well. 

However, for me the B-mask is only the first step. Once on target I check the FWHM values of a star in the field and adjust accordingly.

I think the big advantage of robotic is that you can go to bed while it runs the sequence. If you don't need that advantage then manual has a lot going for it.

Olly

[philherbert]

13 minutes ago, spillage said:

What scope do you have? I have a cheaper skywatcher focus motor and love it. You can remotely focus and make very fine adjustments easily. You can run the auto focus then you either change filters or your subs start to look worse. I will not constantly check and adjust the focus for you over the entire session.

I have a SkyWatcher ED100 and currently use a ZWO ASI1600MC-Pro so no filters to worry about.

 

14 minutes ago, ollypenrice said:

I stick to manual focus because I hate things going wrong - and autofocus systems do go wrong. I host several of them in our robotic sheds. They do mostly work but when they don't they don't. And they're expensive to set up as well. 

However, for me the B-mask is only the first step. Once on target I check the FWHM values of a star in the field and adjust accordingly.

I think the big advantage of robotic is that you can go to bed while it runs the sequence. If you don't need that advantage then manual has a lot going for it.

Olly

It sounds like a fairly slow process Olly - either taking subs or using some kind of live view while manually adjusting the focus to minimise the FWHM? Do you find this more accurate than a B-mask?

[Spikey]

I've used both and ended up going back to using a bog standard Bhatinov mask. Getting the focus to shift slightly (even with par focal filters) and adjust for temperature was too much of a pain for what I gained. I ended up using the motor focuser with constant sub preview downloads to adjust the focuser with the mask on from indoors. When happy I used to pop outside, remove the mask (most of the time) and set the sequence off.

[Tiny Small]

47 minutes ago, ollypenrice said:

 

I think the big advantage of robotic is that you can go to bed while it runs the sequence.

Olly

This is by far the biggest selling point for robotic focusers. A few years ago I got an old Robofocus as part of a used bulk buy and it has made life so much easier. If I had to buy one brand new though, I'm not sure that I would as they are very expensive to set up, as Olly said.

[tooth_dr]

1 hour ago, spillage said:

What scope do you have? I have a cheaper skywatcher focus motor and love it. You can remotely focus and make very fine adjustments easily. You can run the auto focus then you either change filters or your subs start to look worse. I will not constantly check and adjust the focus for you over the entire session.

I may have missed your thread on this, but what are you using to remotely control it?

[Xiga]

I like to keep it simple myself, although that may be more down to the fact I have to setup (from scratch) and tear down after each session.

I’ve made s small mark on my OTA drawtube which I know gets me close to focus, and certainly good enough for plate-solving. After that, I pop on the Bahtinov mask, fire up Bahtinov Grabber, and then focus manually using the basic Sky-Watcher handheld auto-focuser. Takes about 5 mins usually. For my setup, the Critical Focus Zone is about 0.8 pixels, but I usually like to get it within about 0.3 pixels, just to allow for some room for movement throughout the night. Although I rarely seem to suffer from rapid changes in temperature. I think I’ve only ever gone out to re-focus once ever!

[jjosefsen]

For OSC i didn't really think autofocus was worth it, with a refractor the focus shouldn't shift as much over the course of a night, it doesn't for me at least.

Now that I am doing mono and filters, things are a little different.. either my filters are not at all parfocal, or my little skywatcher ed80 just isn't as well corrected as I would like.. With it being SW the latter is certainly possible. But this means I will be refocusing between filters, and if at the end of a session I want a complete data set, then I need to focus a few times. It would be nice to not have to go off target to focus with a mask.

I'm thinking about getting a Focus Cube from Pegasus Astro. My 10 cents.. ?

[tooth_dr]

34 minutes ago, jjosefsen said:

For OSC i didn't really think autofocus was worth it, with a refractor the focus shouldn't shift as much over the course of a night, it doesn't for me at least.

Now that I am doing mono and filters, things are a little different.. either my filters are not at all parfocal, or my little skywatcher ed80 just isn't as well corrected as I would like.. With it being SW the latter is certainly possible. But this means I will be refocusing between filters, and if at the end of a session I want a complete data set, then I need to focus a few times. It would be nice to not have to go off target to focus with a mask.

I'm thinking about getting a Focus Cube from Pegasus Astro. My 10 cents.. ?

This is my experience too. With my mono camera I have to refocus between Ha and oiii. You make a good point that it could be the scope ED80 rather than the filters being entirely responsible themselves. 

[jjosefsen]

55 minutes ago, tooth_dr said:

This is my experience too. With my mono camera I have to refocus between Ha and oiii. You make a good point that it could be the scope ED80 rather than the filters being entirely responsible themselves. 

Yeah maybe, it's just because the red is really far out compared to all the other filters (LRGB). My oiii and ha filters are also quite far apart. But again it is the red end of the spectrum vs. the blue/green end.

[ollypenrice]

6 hours ago, philherbert said:

I have a SkyWatcher ED100 and currently use a ZWO ASI1600MC-Pro so no filters to worry about.

 

It sounds like a fairly slow process Olly - either taking subs or using some kind of live view while manually adjusting the focus to minimise the FWHM? Do you find this more accurate than a B-mask?

Taking a load of out of focus subs for five hours is a seriously slow process. Five hours equals zero hours. In AP, capture is essentially a simple, mechanical process with only three things to get right, exposure time, tracking and focus. Exposure time is non-critical since, within reason, more shorter subs will not be a million miles away from fewer, longer subs. So that leaves focus and tracking. They have to be right. How long does it take? Well, the B mask approximation is very quick, likely three minutes, max. Fine tuning using FWHM might, on a night of unstable seeing, take 5 minutes. On the other hand it might only take two. 

B-mask or FWHM? Well, when we were using a 14 inch reflector at 0.66"PP we had no choice since the FWHM readings bounced around so much as to be useless. It was B-mask or nothing. Using refractors down to 0.9"PP I think that the FWHM is more precise but I'll believe the B-mask for RGB if I'm pressed. I do think FWHM is more accurate and use that without exception for luminance.

Olly

[ollypenrice]

2 hours ago, jjosefsen said:

For OSC i didn't really think autofocus was worth it, with a refractor the focus shouldn't shift as much over the course of a night, it doesn't for me at least.

Now that I am doing mono and filters, things are a little different.. either my filters are not at all parfocal, or my little skywatcher ed80 just isn't as well corrected as I would like.. With it being SW the latter is certainly possible. But this means I will be refocusing between filters, and if at the end of a session I want a complete data set, then I need to focus a few times. It would be nice to not have to go off target to focus with a mask.

I'm thinking about getting a Focus Cube from Pegasus Astro. My 10 cents.. ?

Indeed I think all that you're finding is that you can now see the shortcomings of your optics when in OSC you couldn't. Obviously, in OSC you're taking an average value from all wavelengths whereas in mono you're not. At high resolution it's the same, even with high end optics. We use a TEC140 and at 0.9"PP we are not effectively parfocal. At 1.9"PP we are. I think that Baader and Astrodon filters are close to, if not perfectly, parfocal.

One advantage of FWHM is that, unlike B-mask focus, you don't need to leave your image field to find a suitable star. (The exception is provided by the Astrodon 3nm Ha filter which holds down stars so well that focus is always a nightmare!!!)

Olly

[tooth_dr]

I didn’t realise the need for an adjustment in focus between filters existed before I switched to mono. I had it in my head I would shoot the three NB filters per night and have data for a lovely Hubble palette style image the next day.  This is impossible unless I stay up or buy an auto focuser!

[ollypenrice]

21 minutes ago, tooth_dr said:

I didn’t realise the need for an adjustment in focus between filters existed before I switched to mono. I had it in my head I would shoot the three NB filters per night and have data for a lovely Hubble palette style image the next day.  This is impossible unless I stay up or buy an auto focuser!

But surely NB imaging is such a slow process that three nights would be an absolute minimum anyway? No need to to scroll between filters during one night. If the seeing looks like improving, hang fire with the Ha till it does.

Olly

[tooth_dr]

18 minutes ago, ollypenrice said:

But surely NB imaging is such a slow process that three nights would be an absolute minimum anyway? No need to to scroll between filters during one night. If the seeing looks like improving, hang fire with the Ha till it does.

Olly

There is no doubt you are right ? It took me 5 nights to get data for Pacman.  I just assumed parfocal meant just that. I had no idea the scope can influence it! Every day is a learning day. 

[RayD]

What about temperature change? With my FSQ106 it seems to make a huge difference and I definitely need to refocus even on the same filter. Not tried it with the FT yet but assume it isn't actual focuser related. 

[don4l]

My 106 is very difficult to keep in focus. 

 

The lower the F ratio, the smaller the "sweet spot".