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

How Good Should Guiding Be for 10 Min. + Exposures?

  • Please log in to reply
25 replies to this topic

#1 Bigdan

Bigdan

    Viking 1

  • *****
  • topic starter
  • Posts: 677
  • Joined: 17 Dec 2014
  • Loc: Panacea, FL

Posted 19 June 2017 - 12:47 PM

I am what I kind of call an "advanced newbie." I probably should be farther along than I am. I consistently have done 5 x 5 min. sub.'s for each of LRGB. I have been using my Losmandy Titan with Meade 8 in. SCT, Orion ST-80 guide scope with the SBIG ST-I guide camera. My mount has been permanently installed for several months. I've done several polar alignments, and uploaded a PEC curve to the mount. There was an issue with Gemini as far as turning PEC on, but I uploaded a newer version of Gemini, and that seems to have been fixed.

So.... I have never gone to loner than 5 min. sub.'s because my guiding has been pretty good, but not fantastic. I've been guiding with PHD2 telling me on the guiding graph that I was within 2 arc minutes.

Last night, I polar aligned again..... several 20 min. sessions of drift aligning. I like doing the classic drift aligning with a program called StarTarget (Andy's Shotglass), which puts a reticle on the screen. I feel as though I have higher confidence in using this method than the Drift Align Tool with PHD2..... seeing conditions can affect it too much to suit me.

After drift aligning, I seem to have significantly improved my guiding..... the graph was showing error within 1 arc min. instead of 2 arc min. That makes me wonder if I can go ahead and realistically start doing 10 min. sub.'s.

The next step would be to get it within 0.25 arc min. error on the PHD2 graph.

I think I should go for the 10 min. sub.'s, don't you?

Edited by Bigdan, 19 June 2017 - 12:48 PM.


#2 Jeff Struve

Jeff Struve

    Vanguard

  • *****
  • Moderators
  • Posts: 2053
  • Joined: 05 Nov 2010
  • Loc: Iowa, USA

Posted 19 June 2017 - 12:53 PM

For me... and to answer your question... why not give it a shot? Next... if you have a permanent install on your gear, I'm surprised that you have to continually do a PA... I kinda check the one we have 1 or 2 times a year...


  • Bigdan likes this

#3 xiando

xiando

    Aurora

  • *****
  • Posts: 4919
  • Joined: 27 May 2015
  • Loc: Cloudy NEOhio

Posted 19 June 2017 - 01:32 PM

I presume you have skies that'll support 10 min subs? If so, I agree with Jeff. Give it a shot. I shot some subs of the Crescent using ten minute subs just as a giddy "lark" test a couple weeks ago when I started guiding (the longest sub I was able to consistently shoot prior to guiding was 1 min). My results were pretty "meh", since I hadn't dialed in my guiding parameters yet (at the time) and the sky was poor, but the stars were pretty round, although bloated due to overexposure and jitter due to the >2 arcsec guiding fluctuations.



#4 Jon Rista

Jon Rista

    Hubble

  • *****
  • Posts: 18747
  • Joined: 10 Jan 2014
  • Loc: Colorado

Posted 19 June 2017 - 01:41 PM

Bigdan, when you say "2 arcmin error"...do you mean polar alignment error? Or are you referring to the peak error of your actual guiding? If the former, then 2 arcminutes is pretty good. To get down to 1 arcminute, you would need really stable skies, and you would need very accurate measurements of the test star you are using to determine your polar alignment error. Atmospheric refraction, depending on where you are on the planet, will also affect the accuracy of your alignment. So, for PA error, 1-2 arcminutes is probably about as good as you can realistically expect (even though you might be able to get an actual reading down to 0.5 arcminutes or even less, due to the atmospheric refraction issue, such a reading is likely to be wrong). 

 

Now, if you are actually referring to 2 arcminutes PEAK ERROR in your actual guiding, then that is horrifically bad. :p I am going to assume, if this is what you are referring to, that you mean 2 arcseconds? A peak error of 2 arcseconds may not be that bad, if you have poor seeing. Exactly how that might affect your images would depend on your image scale. A 2 arcsecond peak error with 4 arcsecond/pixel image scale is nothing...you won't even see the guide error. However a 2 arcsecond peak error with 1 arcsecond/pixel or smaller image scale could be reducing your resolution.

 

Further, what matters a bit more than peak error is the RMS error. You want your RMS guide error to be about 1/3 your image scale for ideal results, and really no more than half your image scale for good results. So if you have a 1 arcsecond/pixel image scale, you would ideally want your guide RMS to be 0.33 arcseconds, and no more than 0.5 arcseconds. If you have a 2 arcsecond/pixel image scale, you would ideally want your guide RMS to be 0.67 arcseconds, and no more than 1 arcsecond. With an SCT, your image scale could be 0.5 arcseconds/pixel, in which case you would indeed want your guide RMS to be no more than 0.25 arcseconds...less if possible. With OAG and a reliable mount that should be possible...but it can be tough. 


  • torsinadoc likes this

#5 einarin

einarin

    Viking 1

  • -----
  • Posts: 564
  • Joined: 23 Dec 2016

Posted 19 June 2017 - 01:43 PM

Maybe you should if your skies is not so light polluted you can take so long subs.

Are you using your scope @ f/10 or @ f/6.3 ? If @f/10 then definitely go for 10 mins.

Mind you could take 10 min L but less RGB, say 5 min.

And you mean arc sec error not arc min.

If it's 1" then I think you can go 10 mins.



#6 xiando

xiando

    Aurora

  • *****
  • Posts: 4919
  • Joined: 27 May 2015
  • Loc: Cloudy NEOhio

Posted 19 June 2017 - 01:50 PM

 

So if you have a 1 arcsecond/pixel image scale, you would ideally want your guide RMS to be 0.33 arcseconds, and no more than 0.5 arcseconds.

So a paper napkin rule for guiding target is 1/3 the image scale? Or is the value ratio you note specific, based on a specific setup?


Edited by xiando, 19 June 2017 - 02:00 PM.


#7 Alex McConahay

Alex McConahay

    Fly Me to the Moon

  • *****
  • Posts: 5597
  • Joined: 11 Aug 2008
  • Loc: Moreno Valley, CA

Posted 19 June 2017 - 02:03 PM

If you can guide for a minute at whatever arc seconds error, you can probably guide for hours at the same rate.

 

I may make a exceptions to that general statement to say the following:

 

1. Periodic error is based on the period of the mechanics of your mount. You probably have a four minute period, and a few other shorter periods. I doubt any error is severe enough to really disrupt the results of your guiding, but if there is something, you would know it after the four minutes.

 

2. While you may be perfectly guided, if you are not properly polar aligned, you may suffer from field rotation. This is not something guiding can correct. How long will it take to show in your images? It depends on the declination of the target, the focal length, and the severity of the misalignment (and your definition of "show in the images.")

 

 

Alex


  • RedLionNJ likes this

#8 Madratter

Madratter

    Voyager 1

  • *****
  • Posts: 10745
  • Joined: 14 Jan 2013

Posted 19 June 2017 - 02:04 PM

Polar alignment of 2' is more than good enough for 10' subs unless you are shooting something very close to the pole.



#9 Jon Rista

Jon Rista

    Hubble

  • *****
  • Posts: 18747
  • Joined: 10 Jan 2014
  • Loc: Colorado

Posted 19 June 2017 - 02:11 PM

 

 

So if you have a 1 arcsecond/pixel image scale, you would ideally want your guide RMS to be 0.33 arcseconds, and no more than 0.5 arcseconds.

So a paper napkin rule for guiding target is 1/3 the image scale? Or is the value you note specific, based on a specific setup?

 

Yeah, paper napkin rule. I always aim for 1/3 image scale nowadays, regardless of the setup. I did a bunch of testing in the past, and with my 5D III @ 600mm (2.1"/px image scale), I was a bit surprised to find that I kept seeing better and better stars until my guide RMS hit ~0.65". Below that, there wasn't really any observable improvement, but above that I definitely started seeing my stars go softer and softer, and with an RMS above 1.1" my stars started to look a bit out of focus. These days with a 1.3"/px image scale, I continue to see improvements down to about 0.45"...but below that I'm not seeing much further improvement (although I've only had a couple short opportunities to check, so I cannot say for certain yet that there is no further improvement). Above a 0.67" RMS, my stars start to soften up. A lot of the time recently, I've been unable to get below 0.8" guide RMS, seeing just hasn't been good, and my stars are just not as good as they were last year during fall and winter, which had better seeing. 

 

With the RC, my stars were always soft, but my image scale was 0.48"/px...and my guide RMS was rarely below 0.6". With that scope diffraction was certainly playing a role as well, though...so, I guess once you get up into the higher magnifications, it's a bit more complex. Your guide RMS probably doesn't need to be much smaller than half your airy disc size. With my 8" RC, I guess the diffraction spot was 1.26", so I honestly don't know if there would have been any real value in keeping the guide RMS below 0.6" anyway. You would need to get the geometries normalized to really know for sure...the formula 2.44ᵞ# gives you the linear size of the airy disc, which isn't entirely compatible with a guide RMS. I forget the formula for airy disc as an RMS off the top of my head...but it's definitely smaller than the linear size. My guess is any improvement with better guiding would have been marginal at best. So, something to take into account here for sure is the size of your diffraction spot. 


Edited by Jon Rista, 19 June 2017 - 02:12 PM.

  • xiando likes this

#10 xiando

xiando

    Aurora

  • *****
  • Posts: 4919
  • Joined: 27 May 2015
  • Loc: Cloudy NEOhio

Posted 19 June 2017 - 02:24 PM

No offense intended re: "napkin". Not generally a fan of paper napkin, at least when it comes to more technical matters (engineering design for instance....suggesting "conservative padding" as we might do when establishing a safety factor is one thing, but in this case it sounds a reasonable generalization. Keep one's guiding error to 30% of their image scale to minimize guiding bloat within the presumed resolution of the optical system...
 

Thanks for  the verification. It's actually a question I'd been meaning to ask now that I'm guiding, but I'm never quite sure how to pose them without the answers going way over what I asked wink.gif

 

Sorry for the momentary thread hijack...back to your normal programming....


Edited by xiando, 19 June 2017 - 02:24 PM.


#11 Bigdan

Bigdan

    Viking 1

  • *****
  • topic starter
  • Posts: 677
  • Joined: 17 Dec 2014
  • Loc: Panacea, FL

Posted 19 June 2017 - 02:34 PM

I forgot to mention I am using a 6.3 F/R. So my focal length is 1260, guide scope F/L is 400mm. I also think my guiding improved when I switched from a 100mm F/L guide scope to 400mm.

My impression was that if I reduce the focal length for the imaging OTA, that will lessen the chance of getting poor results (streaky stars) if my guiding is not quite up to snuff.

Prior to this last polar alignment, I was using a scale of +/- 4 arc sec.'s and getting error on the graph of usually no more than +/- 2 arc sec.'s. Now I've switched to a +/- 2 arc sec scale, and it looks like I'm normally within 1 arc sec. on the graph.

I will continue to polar align, and probably use PEMPro again, and try to get the error on the graph down to +/- 0.25 arc sec..... go for dimmer targets with longer exposures.

Edited by Bigdan, 19 June 2017 - 02:42 PM.


#12 Jon Rista

Jon Rista

    Hubble

  • *****
  • Posts: 18747
  • Joined: 10 Jan 2014
  • Loc: Colorado

Posted 19 June 2017 - 03:28 PM

No offense intended re: "napkin". Not generally a fan of paper napkin, at least when it comes to more technical matters (engineering design for instance....suggesting "conservative padding" as we might do when establishing a safety factor is one thing, but in this case it sounds a reasonable generalization. Keep one's guiding error to 30% of their image scale to minimize guiding bloat within the presumed resolution of the optical system...
 

Thanks for  the verification. It's actually a question I'd been meaning to ask now that I'm guiding, but I'm never quite sure how to pose them without the answers going way over what I asked wink.gif

 

Sorry for the momentary thread hijack...back to your normal programming....

I guess one thing I should be clearer about here. The guide RMS is a lagging indicator...for lack of a better term. The RMS helps you gauge how well you are doing, but fundamentally it I guess it's probably best to say, it doesn't do anything in and of itself. It's just a reading of how good (or bad) things are. In my case, at least when I'm more on the undersampled rather than oversampled side of things, I have found, in practice, that I continue to get improvements in my star quality well below what people generally recommended to me when I was first learning. As I said above, I was a bit surprised to find that I continued seeing (visually) improvements in my stars with the 5D III down to ~0.65" guide RMS, as at the time I'd been told I just needed to keep it at or below my image scale. It makes sense that better guide RMS would lead to better stars, as that indicates that either your PE or your seeing is under better control by guiding, and a tighter RMS means the star centroid isn't moving around as much. 

 

What I described in my previous post is more about the relationship between image scale and guide RMS than anything. I've found that when my guide RMS gets to about 1/3 of my image scale, I stop seeing (visually) any improvements. There might be measurable improvements, I actually haven't done those tests (probably should). However since for the most part what I do is visual art more than science, what I see matters more to me in the end. So it seems when the star centroid moves around at a scale 1/3 of my undersampled image scale or less, then I've effectively hit the limit beyond which improve guiding accuracy won't do me any good. With oversampled data, I suspect once your guide RMS 1/2 or 1/3 the diffraction spot, the same thing would happen...further improvement in guide RMS would have minimal impact to resolution, and possibly no visual impact. 

 

There are certainly things you can do to improve your guiding, and the smaller the RMS is, the better. Additionally, if you think about it, once your peak errors become smaller than half your image scale, then technically speaking you wouldn't really have much room for further improvement (or any further improvement would be very, very small, maybe measurable but probably not visually observable). I usually have peak errors larger than my image scale by a but (1.5-1.8" vs. 1.3"/px), but on occasion last year I had peak errors around 0.8". My stars were probably almost as good as they could be with a 1.3"/px image scale in those cases. On the flip side, there are also things you plain and simply cannot control...bad seeing is bad seeing, wind is wind (you might be able to block it, sometimes you can't), you can only guide within the limits imposed by your environment, and that may be a lot worse than the ideal guide RMS that indicates you have things well enough under control that you are getting optimal results. 



#13 xiando

xiando

    Aurora

  • *****
  • Posts: 4919
  • Joined: 27 May 2015
  • Loc: Cloudy NEOhio

Posted 19 June 2017 - 03:43 PM

I'm aware of what you meant, but thanks. Like I said (not that you need it or want it but) I approve of the rule you stated. It makes sense statistically (in a general sense).

 

I'm operating at about 1" RMS under "average transparency, average seeing" (per Clear skies metrics) now that I've dialed things in a bit tighter after four & 1/2 nights out guiding. I expect to improve that number. Your rule gives me a target of about 0.5" arcsec to hit an ideal for my ~1.53" image scale (I'll shoot for as low as I can get it, but 0.5 seems a reasonable goal to shoot for, as I've barely scratched the surface on tweaking PHD guiding and locking down my guidescope itself. (or possibly even re-investigating my OAG now that I'm building confidence in the process, since I may or not be able to hit 0/5 with the external scope) 

 

Dan, yes, reducing your main's F-number should help a bit in reducing the effects of jitter. Best wishes with your goal



#14 Stelios

Stelios

    Skylab

  • *****
  • Posts: 4317
  • Joined: 04 Oct 2003
  • Loc: West Hills, CA

Posted 19 June 2017 - 04:50 PM

What I'm trying to understand is... why go from 5 mins to 10 mins *for LRGB*? 

 

I'm no expert, but based on what I've read in a LOT of posts here, I think that if, instead, you take twice as many subs, what you will lose is a *very small* improvement in SNR, but you gain in some secondary considerations:

1) You can dither more often (since you have more subs) and help FPN that way.

2) If you lose a sub due to a plane, satellite, kid stomping by the mount, etc., it's half the damage.

3) With many targets you would be over-exposed with 10-minute subs (depending on camera) for LRGB.

4) Despite guiding perfection, your stars would end up being a bit more bloated, as statistically speaking you will generate more points 'further out.' Mediocre seeing will aggravate this.

 

As I said, I would love to hear why I'm wrong. 


  • hnau, Jon Rista and xiando like this

#15 xiando

xiando

    Aurora

  • *****
  • Posts: 4919
  • Joined: 27 May 2015
  • Loc: Cloudy NEOhio

Posted 19 June 2017 - 05:18 PM

nonetheless, it's a good learning experience and he may get something deeper (dark nebula maybe?) for it.



#16 Bigdan

Bigdan

    Viking 1

  • *****
  • topic starter
  • Posts: 677
  • Joined: 17 Dec 2014
  • Loc: Panacea, FL

Posted 19 June 2017 - 07:01 PM

Why not be accurate with your guiding..... then you have the choice of doing many shorter sub.'s if you want, or fewer, longer exposures?  I think there is a group of people out there that thinks longer exposure is the way to go, and you need a quality mount that is well aligned to do it. 

 

There is an interview with Scott Losmandy on the internet right now that I think was done in March.  In it, he discusses his desire to build a substantial, quality mount that can achieve longer exposures, instead of many shorter exposures.

 

I would think better results can be achieved for dimmer targets using longer exposures, instead of stacking shorties.


Edited by Bigdan, 19 June 2017 - 07:44 PM.

  • xiando likes this

#17 xiando

xiando

    Aurora

  • *****
  • Posts: 4919
  • Joined: 27 May 2015
  • Loc: Cloudy NEOhio

Posted 19 June 2017 - 07:04 PM

The beauty of improving your guiding is that it will help improve virtually any exposure time. (obviously for excessively short exposures, as might be used in planetary, it might not have any palpable effect.)



#18 rgsalinger

rgsalinger

    Mercury-Atlas

  • *****
  • Moderators
  • Posts: 2532
  • Joined: 19 Feb 2007
  • Loc: Carlsbad Ca

Posted 19 June 2017 - 07:07 PM

I always thought that the airy disk wasn't the issue. Leaving aside diffraction spot size (which definitely is an issue) I thought that the key was going to be spatial resolution for most focal lengths. I don't have the numbers to hand but my memory is that a 2 arc second sky is going to give you a 10micron star at 1 meter of focal length. With my smallest pixels that's much larger than the airy disk for all of my instruments. What am I missing? 

(Don't all pile in at once.)

 

Also, I have found that if I can get the guiding to run at a particular RMS level for one worm cycle then it will just continue to run that way forever. Now, it's also true that sooner or later there will be wind, a satellite, a plane, etc to spoil the image. I might just try an experiment this week and see if my experience/intuition is actually correct.

 

Finally, since the guiding error just sums in quadrature, when you get under .5 arc seconds, I think that you are done unless you have a really unusual use case to deal with. I'm certainly happy with my 10 minute subs using a KAF8300 or 16803 chip. The subs aren't better at 5 than ten because both systems that have these cameras end up with pretty consistent guiding.. 

 

Rgrds-Ross


  • psandelle likes this

#19 Jon Rista

Jon Rista

    Hubble

  • *****
  • Posts: 18747
  • Joined: 10 Jan 2014
  • Loc: Colorado

Posted 19 June 2017 - 07:08 PM

What I'm trying to understand is... why go from 5 mins to 10 mins *for LRGB*? 

 

I'm no expert, but based on what I've read in a LOT of posts here, I think that if, instead, you take twice as many subs, what you will lose is a *very small* improvement in SNR, but you gain in some secondary considerations:

1) You can dither more often (since you have more subs) and help FPN that way.

2) If you lose a sub due to a plane, satellite, kid stomping by the mount, etc., it's half the damage.

3) With many targets you would be over-exposed with 10-minute subs (depending on camera) for LRGB.

4) Despite guiding perfection, your stars would end up being a bit more bloated, as statistically speaking you will generate more points 'further out.' Mediocre seeing will aggravate this.

 

As I said, I would love to hear why I'm wrong. 

waytogo.gif



#20 rgsalinger

rgsalinger

    Mercury-Atlas

  • *****
  • Moderators
  • Posts: 2532
  • Joined: 19 Feb 2007
  • Loc: Carlsbad Ca

Posted 20 June 2017 - 08:33 AM

I ran 3 exposures at 1, 2, 3, 4, 5 and 6 minutes last night to test the idea that longer exposures would produce quantitatively poorer results. My understanding has always been that if the ratio between the guiding error and the seeing is under .3, then the guiding is essentially perfect. 

 

Last night the system guided at around .4 arc seconds in each axis. (Dec a little better than RA). The seeing was around 2 arc seconds. I didn't watch it all night at all times. 

 

There was no noticeable difference in the quality of the output due to exposure time. I measured FWHM and Roundness using CCD Inspector. The object, M29 has a lot of bright stars and is high in the sky at this time. 

 

I may try it again one night with my other mount which doesn't guide as well and see what the results are. This does reinforce the idea that once the guiding error is well below the seeing, exposure doesn't matter. If a plane goes through you 30 minute sub, well, that's another matter entirely. 


  • Bigdan likes this

#21 xiando

xiando

    Aurora

  • *****
  • Posts: 4919
  • Joined: 27 May 2015
  • Loc: Cloudy NEOhio

Posted 20 June 2017 - 11:19 AM

It might be illuminating for some folks to switch to pixel (from arc sec) display sometimes, to get a better idea of how those fluctuations are really affecting your images. What does 1 arc sec really mean? For me, it's less than 1 pixel. Granted, I only have a stubby little 6" scope and my image scale is a bit larger than many @ 1.53, but with an average FWHM of ~4, that translates to a very small deviation in most of the stars large than the background embedded pin-pricks in my images. I haven't shot any of my favorite galaxies yet though, and I'm sure it'll be more important there.


Edited by xiando, 20 June 2017 - 11:20 AM.


#22 rgsalinger

rgsalinger

    Mercury-Atlas

  • *****
  • Moderators
  • Posts: 2532
  • Joined: 19 Feb 2007
  • Loc: Carlsbad Ca

Posted 20 June 2017 - 01:24 PM

Good idea xiando. My results were that, with two seconds seeing, measured by a "proper" seeing monitor my FWHM was around 6 pixels regardless of exposure length. My scope / camera gives me .45 arc seconds per pixel so I ended up around 2.5 arc seconds regardless of exposure. Of course, .3 arc seconds of guiding error in each is PDG  given 2 meters of focal length.



#23 Jon Rista

Jon Rista

    Hubble

  • *****
  • Posts: 18747
  • Joined: 10 Jan 2014
  • Loc: Colorado

Posted 21 June 2017 - 12:16 AM

I ran 3 exposures at 1, 2, 3, 4, 5 and 6 minutes last night to test the idea that longer exposures would produce quantitatively poorer results. My understanding has always been that if the ratio between the guiding error and the seeing is under .3, then the guiding is essentially perfect. 

 

Last night the system guided at around .4 arc seconds in each axis. (Dec a little better than RA). The seeing was around 2 arc seconds. I didn't watch it all night at all times. 

 

There was no noticeable difference in the quality of the output due to exposure time. I measured FWHM and Roundness using CCD Inspector. The object, M29 has a lot of bright stars and is high in the sky at this time. 

 

I may try it again one night with my other mount which doesn't guide as well and see what the results are. This does reinforce the idea that once the guiding error is well below the seeing, exposure doesn't matter. If a plane goes through you 30 minute sub, well, that's another matter entirely. 

This likely works for you as you are using a high end mount with very low PE. Since it's an MX, I assume you are also mounted in an observatory? That would likely also greatly mitigate local environmental factors (i.e. wind). 

 

Remember that guiding counteracts all forces that deviate the star from it's lock position...not just seeing. Actually, seeing is really the one thing that guiding does NOT correct...too high of a frequency, most mounts just don't respond that fast. It's alls the lower frequency errors that guiding deals with. So, it's definitely not surprising that a Paramount MX doesn't improve with longer exposures. :p 

 

With lesser mounts, PE can be a problem for longer exposures...even though you are guiding, it is a very large source of error, (30-40" for lower end mounts) that guiding also has to counteract. There isn't an infinitely increasing error with longer exposures...eventually you will get to the point where you are guiding beyond the worm period, and you'll have reached your maximum error due to PE. But, it can be an issue...especially if you have errors at frequencies lower than your full worm period, harmonics, etc. Sometimes shorter exposures can get you around those issues in PE with lesser mounts. 

 

Environmental factors can also play a role. Wind is a constant plague I have to deal with. For shorter exposures it is usually not a big problem, although I sometimes do lose subs that are jolted too hard by the wind. Long exposures will usually experience worse blur due to the wind jostling the scope around more per sub. One of the benefits of shorter subs IS the ability to discard less than ideal ones, if maximizing resolution is a concern. 



#24 Jon Rista

Jon Rista

    Hubble

  • *****
  • Posts: 18747
  • Joined: 10 Jan 2014
  • Loc: Colorado

Posted 21 June 2017 - 12:21 AM

It might be illuminating for some folks to switch to pixel (from arc sec) display sometimes, to get a better idea of how those fluctuations are really affecting your images. What does 1 arc sec really mean? For me, it's less than 1 pixel. Granted, I only have a stubby little 6" scope and my image scale is a bit larger than many @ 1.53, but with an average FWHM of ~4, that translates to a very small deviation in most of the stars large than the background embedded pin-pricks in my images. I haven't shot any of my favorite galaxies yet though, and I'm sure it'll be more important there.

You are aware that pixel is guider pixel, not imager pixel, right? Not sure how useful knowing how your guiding is in guider pixels, without knowing how the scale of those pixels relates to imager pixels. (I guess the one potential exception would be if your guider and imager had the same pixel size, and you were using OAG, and you were not binning either/were binning both the same, in which case the image scales would be identical...I honestly don't think I've ever heard of that happening though. :p) That is the entire reason we usually track guiding in arcseconds in the first place. It makes it easy to translate the guide RMS to something meaningful at the imaging sensor plane.  

 

If your imager image scale is 1.3"/px, and you have a guide RMS of 0.55", then you are doing pretty well, and your data will probably be fairly sharp. On the flip side, if your guide RMS is 2", then your doing rather poorly, and your data will be pretty soft. You may not even have any control over it at all, either...but, at least you know, with arcseconds, how much the average movement of the guide star centroid (and, thus, the effective minimum size of any star) is going to affect the stars relative to your imaging pixels. 


Edited by Jon Rista, 21 June 2017 - 12:22 AM.

  • xiando likes this

#25 xiando

xiando

    Aurora

  • *****
  • Posts: 4919
  • Joined: 27 May 2015
  • Loc: Cloudy NEOhio

Posted 21 June 2017 - 10:49 AM

 

You are aware that pixel is guider pixel, not imager pixel, right?

Oh heck, my bubble done burst. Yes, my brain does, but in my eagerness...I'll admit did just what you thought and ignored the facts right in front of my nose. blush.gif

 

In any case, if one knows their image scale ratio, (guide vs main) the px scale can still be used effectively.

 

Thanks for clearing my cobwebs.




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.







Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics