Jump to content

  •  

CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.

Photo

Which one filter to use for focusing?

  • Please log in to reply
18 replies to this topic

#1 kisstek

kisstek

    Viking 1

  • *****
  • topic starter
  • Posts: 596
  • Joined: 25 Jul 2018

Posted 04 July 2020 - 04:50 PM

I finally got around to shooting my new flats last night. Afterwards I slewed over to Vega to sync my mount and check my focus. As soon as I take the first image, I could tell the focus was off. All of the little stars were donuts! I don't recall ever seeing this big of a shift, one night to the next. So I throw the Bahtinov mask on and focus. Then I slew over to my nemesis, the Crescent Nebula, to begin imaging for the remainder of the night. (Or until APT couldn't recenter after the Meridian flip, as it happens!)

 

Anyway, as I'm about to start the sequence, I notice that the blue filter is selected. Ah! That's from doing the LRG and B flats! So when I was checking the focus earlier, I was doing so using the blue filter. Normally I use L. That got me to wondering if the big focus shift I detected was actually from the filter change and not some other gremlin.

 

So, if you can use only one filter (LRGBHSO) for focusing, which one is "best", ie the stars will be the sharpest when using the other filters? I guess it coudl be which two filters: one from LRGB and one from LHSO? I think I read somewhere on CN that blue is best for LRGB. Does that mean O3 would be best for narrow band?



#2 endlessky

endlessky

    Mariner 2

  • -----
  • Posts: 235
  • Joined: 24 May 2020
  • Loc: Padova, Italy

Posted 04 July 2020 - 05:14 PM

Yes, I think the focus change was mostly due to the filters being different in the two situations.

 

I haven't had the chance of playing with filters (I only have been shooting with an astromodified DSLR), but it is my understanding that, unless your filters are parfocal, you need to refocus everytime you change filter.

 

If you only focus for one filter and then change filter in the same session, without refocusing, the next set of images taken with the second filter will be out of focus (unless the filters are parfocal).



#3 kathyastro

kathyastro

    Surveyor 1

  • *****
  • Posts: 1,999
  • Joined: 23 Dec 2016
  • Loc: Nova Scotia

Posted 04 July 2020 - 05:48 PM

If your filters are parfocal, then it shouldn't matter which filter you use.  You'd want to use the brightest one, which is Luminance.  If you bought the filters as a good quality parfocal set, there is a good chance that they are indeed parfocal to any reasonable degree of accuracy.  If you picked up different filters at different times or from different vendors, then you should not assume they are parfocal without doing some testing. 

 

Of course, regardless of their origin, if you discover, as you have, that they are not parfocal, then you should focus through each filter separately.  Depending on your software, you may be able to set up filter offsets so that you can use the brightest filter for focusing and have it calculate the focus position for the other filters.


  • psandelle likes this

#4 kisstek

kisstek

    Viking 1

  • *****
  • topic starter
  • Posts: 596
  • Joined: 25 Jul 2018

Posted 04 July 2020 - 07:54 PM

From what I could understand, the thinking that using the blue filter was better for focusing had to do with the wave lengths and that critical focus was more important in the blue band than in the red one. If that were the case, I would have thought using the green filter to split the difference would be the best filter to use.

 

The claim was that focusing with blue and shooting in red would be sharper than focusing with red and shooting in blue. I would have thought it would be equivalent either way. Filter offsets (which I determined at one time if I can find my blue notebook) are based upon that assumption. From a practical point of view, filter offsets will work good enough.

 

I don't know if it had to do with different atmospheric effects on different wavelengths or human visual response to different colors being perceived as different sharpness, or ... The more I muck around in AP, the more things I assumed to be symmetrical aren't.

 

My auto focus "V curve" for instance. (When NINA doesn't lose reference to its objects and sets my focus off by 15000 steps!) Trying to curve fit a parabola to the HFR assumes it's symmetrical about the critical focus point. But in my experience and from what I've seen of other people's V curves, that's not the case. The slope of the curve on the inward focus side is shallower than the slope on the outward focus side. From a practical point of view, as long as the minimum of the fitted curve is within the "critical focus distance" of the true focus, it's good enough.

 

This is kind of a meandering post as I don't understand all of the intricacies of light thru mediums, human visual perception, theoretical vs actual device performance, ... well enough to articulate my thinking.



#5 Der_Pit

Der_Pit

    Surveyor 1

  • *****
  • Posts: 1,527
  • Joined: 07 Jul 2018
  • Loc: La Palma

Posted 05 July 2020 - 07:35 AM

Following the bows of your Meander, some thoughts:

So I understand you do use filter offsets, but wonder which filter to use as reference?

The question is what makes a focus curve reliable.  IMO it is mostly scatter from seeing.  The better the seeing, the better the curve looks in my experience.  Now seeing is caused by refraction in the atmosphere, and refraction is stronger for blue wavelengths.  This is why seeing usually is better in the red.  So from this, you would like to use the red filter as 'reference'.

I'm not sure how strong the asymmetry of your V-curve is.  IMO it is caused by your intra- and extra-focal star images not being identical.  No idea if that can be improved by better collimation.  If it is so strong that it really shifts the detected focus point you could either try to reduce the number of points on one side of the curve, or switch to using a Bahtinov mask.  That of course has the drawback that you can't do automated refocus in-sequence :(



#6 schmeah

schmeah

    Fly Me to the Moon

  • *****
  • Posts: 5,864
  • Joined: 26 Jul 2005
  • Loc: Morristown, NJ

Posted 05 July 2020 - 09:27 AM

Some may  disagree with this, but .... if you are doing LRGB imaging, it really doesn’t matter if your RGB filters are out of focus. The detail is carried in the luminance. You can blur your RGB stacks into oblivion in post processing, and when combined with the luminance you will not likely notice any drop off in detail. So get perfect focus with your luminance filter and don’t worry if your color filters are not parfocal. Now if your color filters are so out of focus that the stars are swollen blobs, then you may have to deal with color halos in post processing, but this is manageable. When doing narrowband, it does matter because some detail is typically carried in each band. Though for many bicolor  targets, the detail is primarily in the Ha and so the OIII being a bit out of focus may not detract from the image. 
 

Derek


  • Miguelo likes this

#7 Peregrinatum

Peregrinatum

    Apollo

  • *****
  • Posts: 1,292
  • Joined: 27 Dec 2018
  • Loc: South Central Valley, Ca

Posted 05 July 2020 - 12:37 PM

if all your filters are parafocal then you would want to focus with the Lum



#8 kisstek

kisstek

    Viking 1

  • *****
  • topic starter
  • Posts: 596
  • Joined: 25 Jul 2018

Posted 05 July 2020 - 01:44 PM

I don't think my ZWO filters are parafocal. I spent some time last night focusing each filter with the Bahtinov mask. I went through the exercise twice and came up with this table:

 

Screen Shot 2020-07-04 at 9.47.20 PM.png

 

I have no experience to know whether these values are typical or unusually abnormal.

 

I normally focus using L (easier to see and faster exposures) and then start my sequence and let it run all night. As I said, when I switched to the blue filter after focusing using L, I noticed that all of the small stars were donuts. If I had processed them, I'm pretty sure all of my small stars would have had a blue ring around them.

 

You mention that blue has the largest amount of dispersion in the atmosphere and therefore you should focus using the least dispersed color: red. I believe the argument I read was that since blue is the worst, you should focus using that and make it the best it can be. Then let everything else take care of itself.

 

Until APT adds auto focus, I'm pretty much limited to one focus per night at the beginning of the session. So either I focus using one filter and stick with that band for the night or I have to come up with some other method to compensate. Filter offsets are one tool for that. If my offsets can be believed.



#9 schmeah

schmeah

    Fly Me to the Moon

  • *****
  • Posts: 5,864
  • Joined: 26 Jul 2005
  • Loc: Morristown, NJ

Posted 05 July 2020 - 02:43 PM

Again, if you are doing LRGB imaging, you absolutely want your luminance filter to be the most sharply focused. But something else must be going on there. With your setup, the critical focus zone (say for the blue filter) is a generous 232 microns. It is hard to imagine that your filters are so non parfocal (not parafocal) that just switching between filters is going to turn focused stars into donuts. How many microns does the number in “offsets” represent, ie: how many microns/step for your focuser?

 

I suspect this is simply related to temperature related focus change from one night to the next.

 

Derek


Edited by schmeah, 05 July 2020 - 02:44 PM.


#10 kisstek

kisstek

    Viking 1

  • *****
  • topic starter
  • Posts: 596
  • Joined: 25 Jul 2018

Posted 05 July 2020 - 03:47 PM

I have no idea how many microns per step as that's not a specification I've ever needed to use. I'm not sure the manufacturer gives enough info to derive it. Heck, the manual for my scope doesn't even mention my scope! I can usually notice a visual change of focus with 20-30 steps. That's eyeballing it. The Bahtinov mask is more sensitive than my aged eyes.

 

The two nights were within a degree or two F of each other. Similar time of night. Same focus target. Weather around here is pretty consistent this time of year. And if the day to day variation was enough to cause a visible problem, then the overnight variation should really be causing me trouble. And I've only seen that one time last fall at the All Arizona Star Party out in the desert. Hot during the day, cold at night. I even had dew one year out there! But in the heat island of the Phoenix Metropolitan area, I just haven't seen an overnight temperature induced focus issue.

 

And I understand and agree with your luminance analysis. So for now I'll keep using L for focus, enable the filter offsets, and keep and eye on things. My Ha subs from last night look pretty sharp, just not enough Ha.



#11 kisstek

kisstek

    Viking 1

  • *****
  • topic starter
  • Posts: 596
  • Joined: 25 Jul 2018

Posted 06 July 2020 - 11:24 AM

And I spoke too soon. I had noticeable focus shift last night, which I'm attributing to the temperature change during the night. I don't have a temperature sensor on my rig so I don't think I can do temperature compensation. And APT doesn't support auto focus so I can just refocus every once in a while.

 

Ideas?



#12 Der_Pit

Der_Pit

    Surveyor 1

  • *****
  • Posts: 1,527
  • Joined: 07 Jul 2018
  • Loc: La Palma

Posted 06 July 2020 - 01:09 PM

Well, the obvious one is 'Change Software'.....  tongue2.gif

 

Or at lest get some computer-readable temp sensor at the telescope so you get an idea when to (manually) refocus.


  • OldManSky likes this

#13 kisstek

kisstek

    Viking 1

  • *****
  • topic starter
  • Posts: 596
  • Joined: 25 Jul 2018

Posted 06 July 2020 - 07:46 PM

What's not obvious to to which software!

 

APT really wants the temperature to be reported by the focuser. My Celestron Focus Motor for SCT doesn't support a temperature sensor. APT will adjust the focus for temperature changes if I could get the temp in somehow.

 

NINA does do auto focus when the bugs don't interfere. NINA also supports the ASCOM ObservableWeather drivers, so I could potentially get the temp in that way.

 

I wasted my SGP trial period while trying to get my then new mount to work. The interface was confusing to me coming from other apps so I never got it set up.



#14 ekallgren

ekallgren

    Viking 1

  • *****
  • Posts: 634
  • Joined: 23 Dec 2011
  • Loc: Northern Nevada, Reno

Posted 07 July 2020 - 07:50 PM

I would ask SGP to extend the trial period. The worst that can happen is they say no.



#15 OldManSky

OldManSky

    Skylab

  • *****
  • Posts: 4,331
  • Joined: 03 Jan 2019
  • Loc: Valley Center, CA USA

Posted 07 July 2020 - 09:57 PM

What's not obvious to to which software!

 

APT really wants the temperature to be reported by the focuser. My Celestron Focus Motor for SCT doesn't support a temperature sensor. APT will adjust the focus for temperature changes if I could get the temp in somehow.

 

NINA does do auto focus when the bugs don't interfere. NINA also supports the ASCOM ObservableWeather drivers, so I could potentially get the temp in that way.

 

I wasted my SGP trial period while trying to get my then new mount to work. The interface was confusing to me coming from other apps so I never got it set up.

I haven’t experienced a “bug” in NINA that affected an unattended night of imaging (including AF dozens of times a night, multiple targets, automated meridian flips, etc.) in over 9 months. And that one 9 months ago was partially my fault.

What “bugs” were a problem?


  • nimitz69 likes this

#16 kisstek

kisstek

    Viking 1

  • *****
  • topic starter
  • Posts: 596
  • Joined: 25 Jul 2018

Posted 07 July 2020 - 11:14 PM

I keep getting little red error boxes complaining about losing references to objects. Usually during the AF process. That cause AF to kick out and leave the focus set to who knows what. All of the subsequent subs look like flats! The focus is off that far. This was with RC200X. I have RC3002 that I'm going to try tonight, once my 5 minutes darks are done.

 

NINA-error.jpg

 

 



#17 kisstek

kisstek

    Viking 1

  • *****
  • topic starter
  • Posts: 596
  • Joined: 25 Jul 2018

Posted 08 July 2020 - 10:32 AM

Got the error again last night with the latest release candidate:

 

Screen Shot 2020-07-08 at 1.07.02 AM.png

 

 

But NINA was able to work through it and finish the night. I'll try it a few more nights to see if it will continue working. If so, great, I have my app. If not, I'll have to keep looking.



#18 OldManSky

OldManSky

    Skylab

  • *****
  • Posts: 4,331
  • Joined: 03 Jan 2019
  • Loc: Valley Center, CA USA

Posted 08 July 2020 - 08:22 PM

Mike, what focuser are you using?



#19 kisstek

kisstek

    Viking 1

  • *****
  • topic starter
  • Posts: 596
  • Joined: 25 Jul 2018

Posted 08 July 2020 - 08:41 PM

Celestron Focus Motor for SCTs, so no temperature probe.


Edited by kisstek, 08 July 2020 - 08:42 PM.



CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.


Recent Topics






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics