Jump to content


Photo

Good Guiding - What got you over the hump?

This topic has been archived. This means that you cannot reply to this topic.
39 replies to this topic

#1 Tim C

Tim C

    Apollo

  • *****
  • Posts: 1085
  • Joined: 11 Nov 2007

Posted 19 March 2009 - 09:00 AM

This question is for folks who get very good results while autoguiding. I'm curious, what were the top few things that you learned or did that helped you get good results? Or perhaps you were one of those fortunate souls who plug in your Orion autoguider, turn on phd and use all of the default settings, balance the mount slightly east heavy and it works well just like magic every time?

Just when I feel like I'm almost there, I'll go out and just get horrible results. That's too much like fishing (my other hobby). This ought to be a more consistent experience once you learn the nuts and bolts. I've been guiding through my WO66 and taking pictures through the 8". My last time out things went pretty well - not perfect pinpoint stars but pretty close - not bad for a newbie shooting at 1200mm focal length. The other night I was trying to shoot a few subs of M37 through the WO66 at 300mm while guiding through the SCT at 1200mm. Results were horrible, my stars were big lines (RA tracking errors). I turned off guiding and had the same result so looks like my mount was tracking really bad and guiding couldn't fix it. I'm seem to do better when I'm shooting objects in the NE part of the sky (object rising) but when I try to shoot high in the sky just west of the Meridian I struggle to get good tracking. I do balance the scope very slightly east heavy each time but maybe I need to unbalance it more to get better tracking? I don't know. I'm not necessarily trying to troubleshoot my specific problems here but really just looking for general "wisdom" from others who get good results. Things to check for, common problems, etc.

Tim

#2 EricJD

EricJD

    Vanguard

  • -----
  • Posts: 2148
  • Joined: 24 Jul 2007

Posted 19 March 2009 - 09:06 AM

I've only guided once so far and it worked out of the bag but I do have a couple things that might help you out.

First, I get better results pointing to the East also, and I've pretty much nailed it down to the fact that my Dec motor hangs off the side of my mount and when pointed East it's in a fairly neutral position. Otherwise it skews my balance. Maybe you have the same issue?

Second, imaging at 1200mm and guiding at 300mm is probably the reason for most of your issues. Your guide cam is not seeing very much movement at all, definitely not as much as your imaging cam is. Is there any way that you can barlow your guide cam and try it? I'd do at least 3x.

#3 Tim C

Tim C

    Apollo

  • *****
  • Posts: 1085
  • Joined: 11 Nov 2007

Posted 19 March 2009 - 11:26 AM

Eric - I've heard most people say that the wo66 is fine to guide an 8 inch sct because guidecam camera resolutions are much better these days. It's actually closer to 400mm focal length when guiding as I'm not using the reducer. I was actually wondering if my poor result from the other night might be due to the FL of the guidescope (in this case the sct with a .67 reduced) was too long for the wo66 at 300mm. Perhaps it was chasing the guide star too much or something. Seeing was fair and I gave the sct time to cool down.

#4 David Pavlich

David Pavlich

    Transmographied

  • *****
  • Administrators
  • Posts: 27785
  • Joined: 18 May 2005

Posted 19 March 2009 - 12:31 PM

Good guiding starts with very good polar alignment. Without it, the best guider in the world won't be worth a hoot. Second is balance. You've got to ensure that the RA axis worm gear is continually engaged. Third is flex. If your system has inferior rings and/or dovetails that are overtaxed by the eqipment, it's not going to guide well.

David

#5 DonR

DonR

    Apollo

  • -----
  • Posts: 1256
  • Joined: 15 Nov 2006

Posted 19 March 2009 - 12:37 PM

After the initial learning curve, the most significant thing I learned was the importance of making the mount slightly east-heavy, so that the RA motor is pushing rather than pulling. This is done by adjusting the counterweight positions, of course, but whether you need to move the counterweights up or down the shaft depends on which side of the meridian you will be pointing. And if you cross the meridian, taking a minute to readjust the counterweights is worthwhile.

I had read this advice repeatedly here on CN and it made perfect sense, but sometimes I don't relate concepts to practice as easily as I should. When I started seeing erratic corrections being needed in PHD, I looked at many different possible causes before pulling this advice out of the cobwebs and applying it. Problem solved.

Cheers,

#6 Doug Sanqunetti

Doug Sanqunetti

    Mariner 2

  • *****
  • Posts: 211
  • Joined: 19 Dec 2005

Posted 19 March 2009 - 12:51 PM

I agree with David, the biggest single improvement I saw in guiding performance happened when I started drift aligning the mount before imaging. This can be accomplished more quickly by turning the mount guide commands off in software, then start the autoguider and watch the reported guider errors to see which way the mount is drifting over time.

I have software that works with CCDSoft to graphically display the guider errors. I use it to drift align quickly each time I set up for imaging. If you would like to try the software you can find it at www.dougsastro.net. It only works with CCDSoft.

Best Regards, Doug

#7 Tim C

Tim C

    Apollo

  • *****
  • Posts: 1085
  • Joined: 11 Nov 2007

Posted 19 March 2009 - 12:51 PM

Thanks guys. Regarding polar alignment, the time before last where I got good acceptable results, I did a painstaking drift alignment and actually turned declination guiding off. My imperfections appeared to be in RA so I think I did a good job there. I read that many people skip drift alignment and just use the PS or polar alignment software routine. The declination guiding takes care of any small misalignment. How about some opinions on that. I think balance may be a big part of my problem. I may need to weight the imbalance to the East more aggressively to ensure a good enough load on the gears. I only very slightly imbalance it today which is the advice I've seen elsewhere but "slightly" can mean different things to different folks. Oh, and I'm using ADM hardware so hopefully flex is minimal.

#8 DonR

DonR

    Apollo

  • -----
  • Posts: 1256
  • Joined: 15 Nov 2006

Posted 19 March 2009 - 01:14 PM

The need for precise polar alignment depends on your FOV and on your exposure times. The field rotation that occurs due to polar alignment error is a steady, cumulative error just like declination drift in unguided exposures. So if your subs are long enough, they will register the field rotation as blurry or elongated stars. If the subs are short enough, or your FOV wide enough, so that there is no detectable rotation during a single sub, rotation may still show up over a sequence of subs, but most decent registration software will correct for that.

The bottom line is, perfect polar alignment, achieved through drift alignment, is ideal, but less than perfect may suffice depending on your setup. And even if you do a drift alignment each time, you have the option of deciding how long you need to monitor the drift in order to fine tune the polar alignment. "Perfect" polar alignment would mean an infinitely long drift alignment, which is obviously impossible. I usually settle for no drift within the length of time I will use for my subs.

#9 Tim C

Tim C

    Apollo

  • *****
  • Posts: 1085
  • Joined: 11 Nov 2007

Posted 19 March 2009 - 02:22 PM

Thanks Don - what power do you drift align at by the way? I used 325x last time for 5 minutes with drift. That took me a while though - wonder if I can get by with less mag.

#10 DonR

DonR

    Apollo

  • -----
  • Posts: 1256
  • Joined: 15 Nov 2006

Posted 19 March 2009 - 02:59 PM

Hi Tim,

When I drift aligned using a reticle eyepiece I used 200X (10mm reticle eyepiece, 2X barlow, 1000mm focal length). These days though I drift align with my guide camera and scope, using PHD. I first make sure the guide scope is aligned with the imaging scope, then focus the guide scope on a star, turn on the fine grid overlay in PHD, start looping exposures and rotate the guide camera until one axis of the grid aligns with the RA motion of the mount. So I'm drift aligning with the same FOV that I will be guiding with, and using the same software. PHD's guiding precision of 1/4 pixel is better than my visual precision, but if I can see no drift in 5 minutes, I am satisfied that there will be no perceptible field rotation in 5 minute subs.

#11 Doug Sanqunetti

Doug Sanqunetti

    Mariner 2

  • *****
  • Posts: 211
  • Joined: 19 Dec 2005

Posted 19 March 2009 - 03:01 PM

Hi Don,

Good polar alignment does more than just correct for field rotation. if you are nearly aligned, the guider has to do less work to keep the star from drifting out of position so mount corrections are smaller and less frequent. This is true no matter how long you exposures are. With short focal length scopes, the errors are less apparent, but in my opinion it is still better to remove the error as much as possible. The OP was asking what got me over the hump to good guide performance and drift alignment made a definite and instant improvement for me. Of course, your mileage may very. It takes me about 20 minutes to drift align assisted by the guide scope and camera. Furthermore I can start drift aligning at twilight, before its dark enough to image.

Best Regards, Doug

#12 DonR

DonR

    Apollo

  • -----
  • Posts: 1256
  • Joined: 15 Nov 2006

Posted 19 March 2009 - 03:08 PM

Hi Doug,

I agree that good polar alignment is very important for multiple reasons. But in my case, I was over that hump before I ever started autoguiding. My first DSO photos were made with manual guiding (a real pain in several anatomical locations) :)

#13 Doug Sanqunetti

Doug Sanqunetti

    Mariner 2

  • *****
  • Posts: 211
  • Joined: 19 Dec 2005

Posted 19 March 2009 - 03:15 PM

Hi Don,

I have never tried guiding by hand, and I am very happy that I won't have to. I admire others who have had the steady nerve and patience to do long exposures this way. I don't think I could do it. :)

Best Regards, Doug

#14 DonR

DonR

    Apollo

  • -----
  • Posts: 1256
  • Joined: 15 Nov 2006

Posted 19 March 2009 - 03:22 PM

I admire others who have had the steady nerve and patience to do long exposures this way. I don't think I could do it. :)


In my case it was more a matter of false economics, ignorance and poor judgement. I ended up spending thousands on cortisone injections and physical therapy to alleviate neck problems. I'll never manually guide again!

Cheers,

#15 Alph

Alph

    Surveyor 1

  • -----
  • Posts: 1767
  • Joined: 23 Nov 2006

Posted 19 March 2009 - 03:30 PM

Your CGEM mount has a fine polar alignment routine that is good enough for at least 10 minute guided exposures. My suggestion is, don’t guide with a SCT. It is a bad, bad idea. Also use a scope with larger aperture to shorten the exposure time and to increase S/N. 66mm is too small. It is like imaging with binoculars. It is almost funny.

#16 Tim C

Tim C

    Apollo

  • *****
  • Posts: 1085
  • Joined: 11 Nov 2007

Posted 19 March 2009 - 03:58 PM

Hi Tim,

When I drift aligned using a reticle eyepiece I used 200X (10mm reticle eyepiece, 2X barlow, 1000mm focal length). These days though I drift align with my guide camera and scope, using PHD. I first make sure the guide scope is aligned with the imaging scope, then focus the guide scope on a star, turn on the fine grid overlay in PHD, start looping exposures and rotate the guide camera until one axis of the grid aligns with the RA motion of the mount. So I'm drift aligning with the same FOV that I will be guiding with, and using the same software. PHD's guiding precision of 1/4 pixel is better than my visual precision, but if I can see no drift in 5 minutes, I am satisfied that there will be no perceptible field rotation in 5 minute subs.


That's interesting. I may have to try that using PHD. Another question for you (or anyone) if you don't mind. Do you typically guide with Declination guiding on auto or do you guide in one direction only? If in one direction only, what's the best way to determine which direction you need to guide in (North or South)?

#17 Tim C

Tim C

    Apollo

  • *****
  • Posts: 1085
  • Joined: 11 Nov 2007

Posted 19 March 2009 - 04:01 PM

Your CGEM mount has a fine polar alignment routine that is good enough for at least 10 minute guided exposures. My suggestion is, don’t guide with a SCT. It is a bad, bad idea. Also use a scope with larger aperture to shorten the exposure time and to increase S/N. 66mm is too small. It is like imaging with binoculars. It is almost funny.


Yeah, the polar alignment routine gets you reasonably close but I still have to make some adjustments to get no drift in 5 minutes at 325x. I know plenty of folks say this is close enough and just let declination guiding do it's thing. I may get there eventually but right now I guess I like to take the approach of making it as precise as possible (within reason). Regarding using the 66mm and SCT - I gotta use what I have...

#18 Craig

Craig

    Vendor (Stark Labs)

  • *****
  • Vendors
  • Posts: 768
  • Joined: 16 Sep 2003

Posted 19 March 2009 - 04:30 PM

Polar alignment should be decent but doesn't need to be perfect. I've guided a number of exposures and have had a decent amount of experience with guiding, guide routines, etc. The iterative-alignment method you can use on your rig will quickly get you well-within the needed tolerance. I have only ever (successfully) done iterative alignments or used a polar scope (a very good one, though). I'm currently stacking 1.5 hours of data from one night and 2.5 hours of data from another night taken at 400 mm on a good-sized chip. I polar aligned with the polar scope on my EM-10 and just ran a stack to see how much rotation went in there. It was under 0.05 degrees. Now, you do have to be close, but you don't have to be drift-for-3-hours-to-get-to-arcsec-close.

Here's what helped me:
1) Software that didn't make me think to hard. That's why I wrote PHD in the first place.

2) Eliminating flex. This one is HUGE and so often overlooked. If the star stays locked in your guide program (whatever it is) and doesn't move but your main camera shows problems, it's flex. Flex can be in a huge array of places and isn't limited to your mounting rings. The focuser, attachment of the cameras, spider support, mirror support, OTA itself, etc. can all be issues.
http://www.stark-lab...es/tag-flex.php

3) Start imaging wider / at a lower resolution than you may want to end up. Don't go for 1"/pixel until you've got things worked out well. Likewise, don't aim for 10 min subs right away. Aim for 1 min or 2 and when you notice things are now holding steady across subs, then go longer.

4) Be nice to your mount. Balance it well with a slight eastward bias and don't overload it.

A few articles here may help, in particular the NEAIC talk and the ATT article on guiding:
http://www.stark-lab.../tutorials.html

Craig

#19 rmollise

rmollise

    Hubble

  • *****
  • Posts: 15697
  • Joined: 06 Jul 2007

Posted 19 March 2009 - 05:08 PM

Good guiding starts with very good polar alignment. Without it, the best guider in the world won't be worth a hoot. Second is balance. You've got to ensure that the RA axis worm gear is continually engaged. Third is flex. If your system has inferior rings and/or dovetails that are overtaxed by the eqipment, it's not going to guide well.

David


In my experience, yes and no. If you are going for longer integrations, say 10 minutes or more (depending on your f/l), yes, a good drift alignment is a must. If, however, you're at relatively short focal lengths and are doing shorter integrations, a "purty good" alignment is more than enough if you are autoguiding. I never drift align anymore. Yes, there will be field rotation between frames, but any stacker worth its salt can take care of that.

What is my opinion as to "helps most"? Balance (slightly east heavy), sturdy guidescope mount, and a good calibration (if the guide program does not complete calibration sucessfully, forget guiding). Oh, one thing I've found helpful with CCDsoft and CCDops is calibrating every time I move to a target that's significantly different in declination (rather than just entering the new dec value). PHD does not seem picky that way, and only needs a new cal when you cross the meridian.

;)

#20 DonR

DonR

    Apollo

  • -----
  • Posts: 1256
  • Joined: 15 Nov 2006

Posted 19 March 2009 - 05:48 PM

That's interesting. I may have to try that using PHD. Another question for you (or anyone) if you don't mind. Do you typically guide with Declination guiding on auto or do you guide in one direction only? If in one direction only, what's the best way to determine which direction you need to guide in (North or South)?


I always guide with DEC guiding ON. Don't ask me why, I just do. If the polar alignment is very good, it isn't really needed, and poor seeing could cause the guiding software to unnecessarily send DEC guiding commands to the mount.

If the polar alignment isn't good enough and you want to guide in only one DEC direction, you need to determine which direction the declination is drifting. You could use the buttons on your hand controller to determine in which direction (north or south) the corrections need to be made, and have the guiding software only make corrections in that direction.

#21 freestar8n

freestar8n

    Soyuz

  • *****
  • Posts: 3987
  • Joined: 12 Oct 2007

Posted 19 March 2009 - 06:36 PM

Hi Tim-

The first thing I did was take all autoguiding advice with a grain of salt. It tends to be very anecdotal, non-scientific, and non-quantitative. I recommend listening to methods people use, but focus on results relevant to your equipment, and that show a clear benefit of good guiding - not just a pretty picture.

Many approaches in autoguding make perfect sense from a theoretical perspective - but the realities of making a crude mechanism track with the needed level of accuracy reveal surprising subtleties. So, be prepared to change your mind, as I have on things such as polar alignment.

Here are things that helped me:

1) Focus on fwhm in arc-seconds - in your results and others. Not just "small round stars" but - how small?

2) Advice that works for a high-end mount, such as "minimize the number of corrections," may not be best for a cheaper mount.

3) Many people, including me, abhor guiding an SCT or other mirror scope with a guidescope. It's not just guidescope flexure, but internal flexure in the sct. Switching to off-axis guiding greatly improved my results and their consistency with my C11.

4) I thought that polar alignment wasn't important for a mid-range mount (e.g. my cge) but I found that once my RA guiding was tuned (to the 2" fwhm level or so) I was getting oblong stars in dec. I tried several things in software and mechanically - but I found that my dec. guiding greatly benefited from good polar alignment.

5) PE is highly overrated since, for a mid-range mount, there are other noise sources that have to be chased at the 1-second level or so. I use video guiding and one second guide intervals with negligible latency. Nonetheless, I do use PEC to remove the fundamental and harmonics - because it does seem to reduce the work of the autoguider.

I first guided with guidedog and it was kinda fun, but I was curious just how well I could guide my cge with my own software. I was interested in Airy pattern imaging for collimation, and doing that made me realize that centroiding is full of subtleties that aren't appreciated. This led me to write MetaGuide and apply the same techniques to autoguiding.

So - here are my results, and some of my methods. Much of what I do is quite contrarian and goes against the herd - but I'm happy to show my results at the 0.5" pixel scale.

MetaGuide images with CGE, some showing raw fwhm's in the sub 2" from NE U.S.

Frank

#22 David Pavlich

David Pavlich

    Transmographied

  • *****
  • Administrators
  • Posts: 27785
  • Joined: 18 May 2005

Posted 19 March 2009 - 07:15 PM

I used the iterative method as Craig had mentioned. It works quite well and takes less time than a drift. I did 5 or 6 iterations before I got both stars centered, but it's a worthwhile endeavor.

David

#23 Tim C

Tim C

    Apollo

  • *****
  • Posts: 1085
  • Joined: 11 Nov 2007

Posted 19 March 2009 - 07:20 PM

thank all - lots of good advice in this thread and I really appreciate it. Frank - I like your comment about listening to different approaches and focusing on what works for you. That's exactly what I was trying to do here. As soon as I can get some hardware for my SV105 I'll give that a try and see if my results improve (they should, no mirror shift and shorter focal length). Thanks again and keep the advice coming if others want to chime in.

Tim

#24 JerryWise

JerryWise

    Cosmos

  • *****
  • Posts: 9550
  • Joined: 26 Dec 2003

Posted 19 March 2009 - 07:31 PM

............
1) Software that didn't make me think to hard. That's why I wrote PHD in the first place.
............


Thanks Dude. I thought nothing would make me chunk my STV but the Orion package with your software is embarrassingly good. I think your work and Pavel's mask deserve the CN Nobel prize. :bow:

I don't seem to have a problem using a little FS-60c. But the mount is pretty stable and may have something to do with it.

#25 Jared

Jared

    Fly Me to the Moon

  • *****
  • Posts: 5066
  • Joined: 11 Oct 2005

Posted 19 March 2009 - 08:03 PM

What got me over the hump of consistently good guiding? I'll give you two answers--one long and one short.

First the short answer: Money. I bought a better mount and virtually all of my autoguiding problems disappeared.

Now the long answer: I struggled with a mount that was at the limits of its capacity for a little over a year. I had no problems with differential flexure or mirror flop since my camera is internally guided. Ultimately, what it came down to for me was that my mount, when loaded with 22 pounds of imaging gear, simply didn't track that well. There were shifts of 1-2 arc seconds in RA that occurred over time frames that were too fast for an autoguider to keep up with. I was able to make incremental improvements through adjusting the worm/wheel meshing, adusting the couplings between the drive and the worm, making sure the mount was well balanced (meaning a few ounces East heavy), etc., but I never got the results I thought my scope was capable of. FWHM numbers simply weren't where I wanted them to be. Then I received my ridiculously expensive, very high quality mount and all my guiding problems went away instantly. When using my old mount, I typically got RMS errors between guide exposures of 0.4 pixels to 0.6 pixels with my refractor (1.2 to 1.8 arc seconds). I put that same scope on the new mount, and the first night out--a night of particularly good seeing--I got RMS errors between guide exposures of 0.02 to 0.1 arc seconds. The high end mount took away the struggle.

Can you successfully autoguide with a mid priced mount? Sure. Can you autoguide well with a mid priced mount? I never did, but I have seen plenty of evidence that others do. How can you get truly good autoguiding results with a minimum of effort? Buy a better mount and don't overload it.






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics