Jump to content


Photo

Autoguided PE -- how good is good?

  • Please log in to reply
47 replies to this topic

#26 Miguel Lopes

Miguel Lopes

    Viking 1

  • -----
  • Posts: 913
  • Joined: 04 Jan 2007
  • Loc: Portugal

Posted 12 March 2009 - 12:08 PM

The problem is not your mout, it is the scope.
If you want smaller stars get a Refractor...

This guy (spoiler: great images!) might disagree with you:
http://blog.deepskyc...s.com/gear.html

Also a SCT is very good for photometric stuff.

#27 TxStars

TxStars

    Apollo

  • *****
  • Posts: 1366
  • Joined: 01 Oct 2005
  • Loc: Lost In Space

Posted 13 March 2009 - 04:08 PM

This is what I was looking at...

OP " I wonder if my stars could be made smaller by better guiding?"

1)Find a mountain top and still dry air
2)Faster f/ ratio scope
3)___________

Fill in the blank..

#28 Jaxdialation

Jaxdialation

    Gemini

  • *****
  • Posts: 3197
  • Joined: 01 Apr 2007
  • Loc: Northeast, FL

Posted 13 March 2009 - 10:26 PM

So, after repeating his question, you didn't answer it? :shrug:

#29 Strgazr27

Strgazr27

    Fly Me to the Moon

  • ****-
  • Posts: 7104
  • Joined: 04 Oct 2004
  • Loc: StonyHill Observatory

Posted 13 March 2009 - 11:16 PM

This is what I was looking at...

OP " I wonder if my stars could be made smaller by better guiding?"

1)Find a mountain top and still dry air
2)Faster f/ ratio scope
3)___________

Fill in the blank..


1. Some of the best seeing in the US is 3' Above sea level and about 20 yds from the gulf of Mexico

2. Although this will allow for shorter exposures it does NOT guarantee smaller stars

3. Hmmm.....Some of the best images I have seen taken have come from mirror based systems so there are several choices I could use to fill in #3.

I don't see how this helps the OP.

#30 idahoman

idahoman

    Viking 1

  • -----
  • Posts: 749
  • Joined: 25 Nov 2005
  • Loc: Idaho, USA

Posted 16 March 2009 - 10:07 AM

Thanks for all the really good input. One thing about my setup is that I am imaging at something like 1600mm focal length. Which is more demanding and will lead to "bigger" stars than imaging at, for instance, 800mm. Nonetheless, I've considered swapping the 9.25 out and mounting one of those new, cheap, RCs in the forks instead. But, I am not convinced if I will see a noticeable improvement since the focal length will be about the same.

I think where I am at is:

1.) I may be able to slightly improve my focusing
2.) Up my AO-8 guide rate by using better guide stars (if I can find them!)
3.) Wait until Jared sends me his Mach1 (I haven't given up on this one, yet)
4.) Live with what I've got until I can make a significant upgrade (not certain this can happen).

This upgrade would have to be more than just a CGE or Losmandy mount. My guess would be that I would have to consider an AP mount and possibly different optics as well.

The worse thing I could have done, at this point, is shell out $3K for a CGE or Losmandy, modify my pier, defork my C9.25, debug the new system, and have no noticeable difference other than meridian flips to deal with.

I've learned a lot. But, as with everything else in this hobby, it is a frustrating learning process.

Maybe a new high quality mount will appear out of China?

Oh -- I will check with some of my astro buddies to see what fwhm's they are getting.

#31 Miguel Lopes

Miguel Lopes

    Viking 1

  • -----
  • Posts: 913
  • Joined: 04 Jan 2007
  • Loc: Portugal

Posted 16 March 2009 - 03:01 PM

Forgot to add this:
a color camera allways has worst FWHM than a mono. Not only because of lack of resolution but also because you need more time to get the same quantity of photons...

#32 Psyire

Psyire

    Apollo

  • *****
  • Posts: 1339
  • Joined: 24 Jun 2007
  • Loc: 55* North

Posted 16 March 2009 - 03:39 PM

idahoman, I think you nailed it. If I were in your shoes, I think a CGE-PRO might be the answer... but the jury is still out on those. You'll have to wait and see I guess.

#33 Strgazr27

Strgazr27

    Fly Me to the Moon

  • ****-
  • Posts: 7104
  • Joined: 04 Oct 2004
  • Loc: StonyHill Observatory

Posted 16 March 2009 - 06:40 PM

Forgot to add this:
a color camera allways has worst FWHM than a mono. Not only because of lack of resolution but also because you need more time to get the same quantity of photons...


A OSC camera produces a MONO image until it is debayered so how does this increase the FWHM? Wouldn't pixel size play a larger role than the fact your using a OSC camera.

#34 Miguel Lopes

Miguel Lopes

    Viking 1

  • -----
  • Posts: 913
  • Joined: 04 Jan 2007
  • Loc: Portugal

Posted 16 March 2009 - 07:32 PM

Forgot to add this:
a color camera allways has worst FWHM than a mono. Not only because of lack of resolution but also because you need more time to get the same quantity of photons...


A OSC camera produces a MONO image until it is debayered so how does this increase the FWHM? Wouldn't pixel size play a larger role than the fact your using a OSC camera.

Well, FWHM, as the name says, is the full width at half maximum of a star bell-shape. Due to resolution loss from the debayering (compared to a mono camera), the star will have a smaller SNR, the peak value will be minor, therefore the FWHM will be higher.

#35 Strgazr27

Strgazr27

    Fly Me to the Moon

  • ****-
  • Posts: 7104
  • Joined: 04 Oct 2004
  • Loc: StonyHill Observatory

Posted 16 March 2009 - 08:07 PM

But for arguments sake most stars are composed of mainly "White" light. in other words, they are an RGB object. That being true, am I not using all of the pixels in the image of my OSC to compute my FWHM? You would be correct if we were shooting through a single color filter but we're not. Correct me if I'm wrong but a 6 MP mono camera and a 6 MP OSC camera should have the same resolution if the pixels are the same size and we're shooting an RGB object no?

#36 jrcrilly

jrcrilly

    Refractor wienie no more

  • *****
  • Administrators
  • Posts: 33810
  • Joined: 30 Apr 2003
  • Loc: NE Ohio

Posted 16 March 2009 - 08:11 PM

But for arguments sake most stars are composed of mainly "White" light. in other words, they are an RGB object. That being true, am I not using all of the pixels in the image of my OSC to compute my FWHM? You would be correct if we were shooting through a single color filter but we're not. Correct me if I'm wrong but a 6 MP mono camera and a 6 MP OSC camera should have the same resolution if the pixels are the same size and we're shooting an RGB object no?


They would, if you were looking at the raw data from the chip instead of the interpolated data that is actually used.

#37 Strgazr27

Strgazr27

    Fly Me to the Moon

  • ****-
  • Posts: 7104
  • Joined: 04 Oct 2004
  • Loc: StonyHill Observatory

Posted 16 March 2009 - 10:06 PM

So resolution is lost in the interpolation John? And by that I guess you mean the debayering process? When we speak of OSC cameras are we talking about strictly OSC CCD cameras or are we grouping DSLR's in that group also? I ask as I would think the AA filter in todays DSLR's would have more of an impact on FWHM numbers due to the slight blurring of the image. What if one were to take the FWHM before debayering? I know this is possible in ImagesPlus V3

I would think that image scale would play a bigger part. If my image scale is 2.5 AS/Pixel. If I'm testing on a good night and my seeing is at or better than that my stars would fit on a single pixel making an accurate FWHM reading inaccurate correct?

Sorry if I got off track I'm trying to digest all the info.

#38 Strgazr27

Strgazr27

    Fly Me to the Moon

  • ****-
  • Posts: 7104
  • Joined: 04 Oct 2004
  • Loc: StonyHill Observatory

Posted 16 March 2009 - 10:08 PM

Frank,

Can METAGUIDE be made to work with an Orion Starshoot? It looks incredibly promising for guiding and was curious.

#39 Miguel Lopes

Miguel Lopes

    Viking 1

  • -----
  • Posts: 913
  • Joined: 04 Jan 2007
  • Loc: Portugal

Posted 17 March 2009 - 04:31 AM

But for arguments sake most stars are composed of mainly "White" light. in other words, they are an RGB object. That being true, am I not using all of the pixels in the image of my OSC to compute my FWHM? You would be correct if we were shooting through a single color filter but we're not. Correct me if I'm wrong but a 6 MP mono camera and a 6 MP OSC camera should have the same resolution if the pixels are the same size and we're shooting an RGB object no?


They would, if you were looking at the raw data from the chip instead of the interpolated data that is actually used.

IMO the problem is not in the debayering, it's just because you have less resolution. Interpolation tries to compensate that.
You are right, for a pure white star, if the FWHM algorithm only looks at luminance data, it should be very similar.
But then when we look at the small details, it is not the same.
For example, a mono camera without IR block will be very sensitive in the IR, which is less affected by turbulence.
Also, as you can take a sub frame in 1/3 of the time, the PSF will be smaller.

#40 freestar8n

freestar8n

    Soyuz

  • *****
  • Posts: 3931
  • Joined: 12 Oct 2007

Posted 17 March 2009 - 07:20 AM

Can METAGUIDE be made to work with an Orion Starshoot? It looks incredibly promising for guiding and was curious.



Hi-

Sorry - but it is video-based and only works with cameras that have a DirectShow driver. The centroiding it uses is very different from other apps, and it requires realtime processing of several successive frames to take advantage. My standard guide camera is a Lumenera, and I guide with OAG at 8-30 frames per second on stars down to mag 10 with a C11.

On the topic of fwhm - in order to get the smallest fwhm in the image, all errors will contribute - and I find there is enough error in the guide star centroiding that there is benefit in doing things differently. So focus, guide tuning, latency, etc. all need to be addressed - and even then the centroiding algorithm can be improved, which is why I wrote MetaGuide.

Thanks,
Frank

#41 idahoman

idahoman

    Viking 1

  • -----
  • Posts: 749
  • Joined: 25 Nov 2005
  • Loc: Idaho, USA

Posted 17 March 2009 - 09:15 AM

Some more info to throw into the stew! A friend of mine images at a site 50 miles from my house. He is a bit higher and darker. Anyway, he reports that he usually sees FWHM of 1.5 to 2.5 on a "steady night", but 2 - 3 is more typical. His opinion is that the atmosphere around here is too unstable for anything over 1500mm focal length. I'm imaging at around 1440mm or so.

I don't have CCDInspector, but when focusing Maxim reports around 4 or 5 as my FWHM (if I remember correctly). So, my guess is that the best I can achieve is 3 to 4 and thus I have some room for improvement.

#42 dickbill

dickbill

    Viking 1

  • -----
  • Posts: 937
  • Joined: 30 Sep 2008

Posted 17 March 2009 - 10:02 AM


On the topic of fwhm - in order to get the smallest fwhm in the image, all errors will contribute - and I find there is enough error in the guide star centroiding that there is benefit in doing things differently. So focus, guide tuning, latency, etc. all need to be addressed - and even then the centroiding algorithm can be improved, which is why I wrote MetaGuide.

Thanks,
Frank


Which brings us back to the mount accuracy. Since the purpose of guiding is to correct tracking errors AFTER the errors are made, good mounts will produce less errors that won't need to be corrected and so the centroid produced by an average mount, even in perfect focus and seing conditions, cannot be as small as a good mount with small PE.

If I understand well however, Permanent Periodic Error Correction, such as implemented in the new CGEM or CGE-pro, is supposed to prevent tracking errors before they happen and so the pictures obtain with these mount should show an improved resolution...But, in practice, I suspect that the correction made under PPEC, being automatic, can be also inaccurate and to a certain point it remains to be seen how a middle range mount like a CGEM really needs PPEC when it is also autoguiding.

#43 Strgazr27

Strgazr27

    Fly Me to the Moon

  • ****-
  • Posts: 7104
  • Joined: 04 Oct 2004
  • Loc: StonyHill Observatory

Posted 17 March 2009 - 10:21 AM

Although PEC can and does help, depending on the situation and the mount there is still as you say an error to it. Wether by eye or by camera, the errors will be corrected for either as they happen or after they happen. In the case of my G11, until I get the new worm block I have an issue with the 76 sec error. This makes using PEC a non issue as the error occurs at a different point in each worm cycle. Other mounts do show a marked improvement with PEC. The NXGPS, CPC, and CGE models have always had PPEC.

#44 idahoman

idahoman

    Viking 1

  • -----
  • Posts: 749
  • Joined: 25 Nov 2005
  • Loc: Idaho, USA

Posted 17 March 2009 - 11:00 AM

My Nexstar GPS does not learn PEC for some reason. That was why I got the AO-8. I had more than 80" periodic error, and there was no way to program the PEC into the mount. Apparently this was a problem with the Nexstar's (at least, the one I have). So, I run with NO PEC. I think it is amazing that the AO-8 improves my mount the way it does.

#45 Miguel Lopes

Miguel Lopes

    Viking 1

  • -----
  • Posts: 913
  • Joined: 04 Jan 2007
  • Loc: Portugal

Posted 17 March 2009 - 11:06 AM



On the topic of fwhm - in order to get the smallest fwhm in the image, all errors will contribute - and I find there is enough error in the guide star centroiding that there is benefit in doing things differently. So focus, guide tuning, latency, etc. all need to be addressed - and even then the centroiding algorithm can be improved, which is why I wrote MetaGuide.

Thanks,
Frank


Which brings us back to the mount accuracy. Since the purpose of guiding is to correct tracking errors AFTER the errors are made, good mounts will produce less errors that won't need to be corrected and so the centroid produced by an average mount, even in perfect focus and seing conditions, cannot be as small as a good mount with small PE.

If I understand well however, Permanent Periodic Error Correction, such as implemented in the new CGEM or CGE-pro, is supposed to prevent tracking errors before they happen and so the pictures obtain with these mount should show an improved resolution...But, in practice, I suspect that the correction made under PPEC, being automatic, can be also inaccurate and to a certain point it remains to be seen how a middle range mount like a CGEM really needs PPEC when it is also autoguiding.

+/- true. If you SBIG examples, they do make a test on a Paramount and there is a slight improvement...

Also... Paramount = 14.000€, AO = 850€ - 1800€

#46 freestar8n

freestar8n

    Soyuz

  • *****
  • Posts: 3931
  • Joined: 12 Oct 2007

Posted 17 March 2009 - 11:28 AM

lthough PEC can and does help, depending on the situation and the mount there is still as you say an error to it. Wether by eye or by camera, the errors will be corrected for either as they happen or after they happen. In the case of my G11, until I get the new worm block I have an issue with the 76 sec error. This makes using PEC a non issue as the error occurs at a different point in each worm cycle.



Hi-

My CGE is like most mid-range mounts and also has noise that isn't commensurate with the worm, so it can't be removed by PEC. But PEC does remove the fundamental and harmonics effectively, and has no downside that I can see, so it does remove a lot of the work for the autoguider and helps reduce the fwhm. But there is still a range sources from 1s to 50s that I deal with by low latency corrections and short guide exposures.

Although autoguiders tend to be reactive, MetaGuide is I think unique in that it can lock onto an error frequency and proactively correct it - independent of PEC. This was largely done as an experiment, but it sounds like a 76 second G11 would be a good candidate. You just enter 76 seconds as the period and begin autoguiding. It will sense that frequency in its corrections and determine its phase and amplitude - then begin applying it proactively rather than reactively. Normal autoguiding continues on to correct remaining errors.

I never saw a real benefit of this with my cge so I haven't used it much - but it's there.

Frank

#47 idahoman

idahoman

    Viking 1

  • -----
  • Posts: 749
  • Joined: 25 Nov 2005
  • Loc: Idaho, USA

Posted 17 March 2009 - 11:58 AM

Miguel brings up a good point. The new Paramount is $14500. The AO is under $1000. The AO, however, does occupy a significant amount of backfocus. Especially the AO-7. And, adds weight. But, it sure improved my life! Before the AO (I had an AO-7 before my AO-8), I could not take subs longer than a couple of minutes since I had to throw so many away.

#48 Jaxdialation

Jaxdialation

    Gemini

  • *****
  • Posts: 3197
  • Joined: 01 Apr 2007
  • Loc: Northeast, FL

Posted 17 March 2009 - 03:32 PM

You can get a trial version of CCDInspector.

It might also find some optical issues with your system which are affecting your FWHM.

Some more info to throw into the stew! A friend of mine images at a site 50 miles from my house. He is a bit higher and darker. Anyway, he reports that he usually sees FWHM of 1.5 to 2.5 on a "steady night", but 2 - 3 is more typical. His opinion is that the atmosphere around here is too unstable for anything over 1500mm focal length. I'm imaging at around 1440mm or so.

I don't have CCDInspector, but when focusing Maxim reports around 4 or 5 as my FWHM (if I remember correctly). So, my guess is that the best I can achieve is 3 to 4 and thus I have some room for improvement.








Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics