Jump to content


Photo

My Losmandy G11's Periodic Error

This topic has been archived.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

#1 saadabbasi

saadabbasi

    Apollo

  • -----
  • Posts: 1136
  • Joined: 23 Aug 2009

Posted 04 April 2010 - 02:37 PM

Hi everyone,

I got the chance to measure my G11's periodic error with a copy of PemPro. I first aligned my mount and then calibrated the software for my image scale and camera orientation. While the theoretical image scale was 2.7, the actual came out to be ~2.9.

I did about 5 runs and the error only once went above 5". The average peak to peak error came out to be only ~7". Here's an image of the run:

Posted Image

PemPro calculated the following curve fit:

Posted Image

PEC Curve: http://i115.photobuc...st/PECCurve.jpg

For some reason, however, when I uploaded the curve the periodic error actually increased to about 8" pk-to-pk. Should I not use PEC with this mount? I have to say, I heard some horror stories related to the G11 and the dreaded 76s error. The frequency spectrum showed some error at 76s in my mount. Not sure how bad it is.

In general however, is a PE of 6" pk-to-pk considered good for this class of mount? Is the error smooth and can it be guided out? I am a beginner but I need this mount to last me a long time and I eventually want to do imaging at about ~1800mm FL. (image scale of about 1")

#2 GShaffer

GShaffer

    Surveyor 1

  • *****
  • Posts: 1686
  • Joined: 28 Feb 2009

Posted 04 April 2010 - 04:06 PM

6 peak to peak is a little better than average in my experience for a stock G11.....and yes autoguiding should take care of it nicely. My own G11 in stock form was about 9 peak to peak.....which is about average from what I have seen. You are ahead of the game.

#3 bseltzer

bseltzer

    Apollo

  • -----
  • Posts: 1028
  • Joined: 28 Oct 2007

Posted 04 April 2010 - 05:54 PM

If your PE increased with the PEC curve loaded and turned on, you might try inverting it and reloading it. I don't recall the exact procedure for doing this in PemPro, but the manual covers it in some detail.

My own G-11/Gemini gave me a P - P PE of 7" out of the box, so your example would appear to be on a par or slightly better than average for this mount. You should have no trouble imaging at a ~1"/pixel scale with proper guiding.

Regards,
Bert

#4 saadabbasi

saadabbasi

    Apollo

  • -----
  • Posts: 1136
  • Joined: 23 Aug 2009

Posted 05 April 2010 - 12:00 AM

I actually did invert the curve and tried it again, and same results! I'm now wondering if its worth turning PEC on or not. If autoguiding will take care of most of the errors, then I reckon PEC should be off. If PEC could somehow reduce the error by even 1", I'd keep it on to help make the life of the autoguider easier.

Considering that my G11 costs half as much as a AP Mach 1 and yet still has a native PE of 7" peak to peak, I think its doing quite well. If I end up buying the Ovision later on, I think the performance would be even increased. But I'm not looking to get my PE as low as possible. My goal is to do some serious imaging and if the stock worm allows me to do so, then I won't buy the Ovision.

I see that your G11 is Ovision upgraded. Did you have much trouble with the stock worm then? Did you try to image with your AT8RC with the stock worm?

If so, what were the results?

#5 Richard McDonald

Richard McDonald

    Vostok 1

  • -----
  • Posts: 108
  • Joined: 20 Nov 2004

Posted 05 April 2010 - 08:15 AM

Have a look in the "frequency" tab of PEMPRO, with the option to display the x-axis in seconds. If you have the 76-second error, it will be obvious there. On mine, the 76-second peak swamped the others in the frequency chart. (And if you have the 76-second error, PE wont correct it.) OVision corrects it but you can also make some progress, if you have it, by carefully adjusting the independent worm gear blocks to make sure they are perfectly parallel, and orthogonal to the worm.

Here is a log of my PE adjustments on my G11 (including the oVision), containing some pictures of the frequency chart showing the 76-second error.

Good luck,
Richard

#6 galacticphoto

galacticphoto

    Vostok 1

  • -----
  • Posts: 190
  • Joined: 12 Nov 2009

Posted 05 April 2010 - 08:22 AM

If you are using a reasonable image scale (1-2"/pix), and are willing to let your autoguider do its job, the PEC may not be a big deal. If you want to do some imaging w/ the guider turned off (as in a star poor region, bright planetary nebs, or just laziness), getting the PE down may be a good thing (I've done some "few minute" unguided exposures, w/ the Mewlon, when I couldn't find a good guidestar in the OAG).

The Gemini controller works well with PemPro, at least on my MI. I routinely get 1/2 the uncorrected PE, and have seen runs that result in 1/3 of the uncorrected value. Play around w/ it for best results.

Robert

#7 saadabbasi

saadabbasi

    Apollo

  • -----
  • Posts: 1136
  • Joined: 23 Aug 2009

Posted 05 April 2010 - 08:27 AM

Have a look in the "frequency" tab of PEMPRO, with the option to display the x-axis in seconds. If you have the 76-second error, it will be obvious there. On mine, the 76-second peak swamped the others in the frequency chart. (And if you have the 76-second error, PE wont correct it.) OVision corrects it but you can also make some progress, if you have it, by carefully adjusting the independent worm gear blocks to make sure they are perfectly parallel, and orthogonal to the worm.

Here is a log of my PE adjustments on my G11 (including the oVision), containing some pictures of the frequency chart showing the 76-second error.

Good luck,
Richard


Here's my frequency spectrum:

Posted Image

The 3rd spike is at 76s. The 2nd is at 124 and the 1st is the Fundamental.

How bad does it look?

I'm familer with frequency spectrum from an electrical point of view, not so much with mechanical. But from what I've understood from this graph is that most of the errors are on the worm itself and hence have a frequency that is the same as the worm. These can be corrected by software.

The 76s error is not on the worm and since it is not a multiple of the frequency, it cannot be corrected by software. The magnitude of the error is about 0.95". What I'm hoping for is that 1" is not a large error when imaging at a scale at (or greater than) 1" and could be guided out. An error of 1" would mean a pixel, and that doesn't seem much to me.

The error at 124s has about the same magnitude. But it falls almost (or maybe exactly) on the 2nd harmonic. Since it is a multiple of the period, it should be corrected by PEC.

Your link has been invaluable to me for understanding the data. It seems your error was quite large, as compared to mine, so I'm hoping my mount is in a better position than yours was. I've never opened up my mount, so I'm sort of afraid of tinkering with it - especially because I cannot afford another one.

Where do you think my mount stands as it is?

I will test the performance with autoguiding on in a week or so, when I've got my second telescope.

#8 saadabbasi

saadabbasi

    Apollo

  • -----
  • Posts: 1136
  • Joined: 23 Aug 2009

Posted 05 April 2010 - 10:26 AM

Here's the data that I collected after programming the mount and turning PEC on. I notice that the spikes are much "fatter" now, and not as sharp. I take it this means that the errors are far smoother now? If so, I understand that is better for the autoguider.

The odd thing is, the 2nd bump you see is the 76s error and its very small now compared to before. The total peak to peak error has increased to about 8". Not a large increase and if it means better performance with an autoguider, then I think I'd PEC on.

Posted Image

#9 bseltzer

bseltzer

    Apollo

  • -----
  • Posts: 1028
  • Joined: 28 Oct 2007

Posted 05 April 2010 - 01:49 PM

I actually did invert the curve and tried it again, and same results! I'm now wondering if its worth turning PEC on or not. If autoguiding will take care of most of the errors, then I reckon PEC should be off. If PEC could somehow reduce the error by even 1", I'd keep it on to help make the life of the autoguider easier.

Considering that my G11 costs half as much as a AP Mach 1 and yet still has a native PE of 7" peak to peak, I think its doing quite well. If I end up buying the Ovision later on, I think the performance would be even increased. But I'm not looking to get my PE as low as possible. My goal is to do some serious imaging and if the stock worm allows me to do so, then I won't buy the Ovision.

I see that your G11 is Ovision upgraded. Did you have much trouble with the stock worm then? Did you try to image with your AT8RC with the stock worm?

If so, what were the results?


Actually, I had no trouble with the stock worm at all. It's 7" PE was well under control with conventional guiding at image scales up to 2"/pixel. I suspect it would have worked OK at 1"/pixel as well, but I had a bit of a financial windfall that allowed performing the Ovision upgrade before I got to that point.

With the OV worm now on board, I've had 600 second test subs at 0.97"/pixel with the AT8RC (1625mm FL) that show nice round stars.

Regards,
Bert

#10 saadabbasi

saadabbasi

    Apollo

  • -----
  • Posts: 1136
  • Joined: 23 Aug 2009

Posted 05 April 2010 - 07:19 PM

Thanks for the response, bert. Did you have a 76s error as well? I'm wondering if mine is significant enough to bother messing around with the mount.

The peak itself at 76s is about 0.95". But from Richard's link, I read that his error was as much as 4". So thats quite a difference. In my case, the error isn't dominated by the 76s error at all.

I should add that I'm going to image with a FSQ-85ED at a focal length of 450mm (2.47"/pixel) and maybe with an extender upto 675mm (1.6"/pixel) for at least a year before moving onto 1800mm or so focal lengths. I know that my G11 can handle these focal lengths easily (especially with a small refractor with not a large mass), so I'm not worried about these.

What I'm worried about is that I'll buy a nice C8 Edge one day and find that my G11 cannot handle the focal length. At no binning, I'll be on a scale of 0.5"/pixel. There's no use in imaging at this scale, so I intend to bin 2x2 and have a scale of 1"/pixel. Do you think this 76s error will become an issue then?

I suppose I'll have a real answer when I do a test with autoguiding on and measure the results. Now if only my FSQ would get here so I can do this.

#11 blueman

blueman

    Photon Catcher

  • *****
  • Posts: 5493
  • Joined: 20 Jul 2007

Posted 05 April 2010 - 09:10 PM

HI,
There is a spike at 80 sec (mine .5") with the OVison, are you sure your not looking at that rather than the 76 sec error?

#12 lineman_16735

lineman_16735

    Tak-o-holic

  • -----
  • Posts: 3118
  • Joined: 04 Dec 2004

Posted 05 April 2010 - 09:19 PM

Last I knew the 76 second error was closer actually to 79 seconds... I could be wrong on that but that is the way I recall it???

#13 lawrie

lawrie

    Vanguard

  • *****
  • Posts: 2212
  • Joined: 31 Jan 2006

Posted 05 April 2010 - 09:40 PM

Can someone show where you see the 76-79 sec bump, how do you tell its at that timing?

Does it say AP1200 in the mount box on the Frequency Spectrum window.

#14 saadabbasi

saadabbasi

    Apollo

  • -----
  • Posts: 1136
  • Joined: 23 Aug 2009

Posted 06 April 2010 - 03:31 AM

HI,
There is a spike at 80 sec (mine .5") with the OVison, are you sure your not looking at that rather than the 76 sec error?


I don't have the ovision... so I'm not sure how I could be looking at that error? I'm not sure I understand what you said. I made sure the error is at 76s by bringing the mouse close to the spike and read its reading.

#15 blueman

blueman

    Photon Catcher

  • *****
  • Posts: 5493
  • Joined: 20 Jul 2007

Posted 06 April 2010 - 01:39 PM

Hi, if you checked and it said 76 sec with the mouse, then that is what it is. The 80 sec error that I got when the OVision was installed is 79+sec to 80 sec in frequency and with the old worm it was 76 sec.
So, not sure why you are seeing what you are with the OVision.
Blueman

#16 saadabbasi

saadabbasi

    Apollo

  • -----
  • Posts: 1136
  • Joined: 23 Aug 2009

Posted 06 April 2010 - 02:21 PM

Yeah, but is my 76s error significant in your view? The magnitude of the spike is about 0.95". I ask because I know that a lot of G11s have these errors which makes them unsuitable for long focal length imaging.

In other words, is it worth it to tinker around (or buy the Ovision) when the error is 0.95"?

#17 blueman

blueman

    Photon Catcher

  • *****
  • Posts: 5493
  • Joined: 20 Jul 2007

Posted 06 April 2010 - 07:05 PM

The .95" error should not cause you any problems unless you try to image at below 1" image scale and even then it will not pose a real issue.
The thing to remember is that this is a .95" error over a period of 76 sec and that the worm cycle is 240 sec. So, the guiding should be able to keep it at a minimum, though it will give you a slight peak on each cycle. But the majority of the light collected will not be during this period so it will not effect the overall exposure that much.
Blueman

#18 chicot

chicot

    Ranger 4

  • -----
  • Posts: 363
  • Joined: 30 Sep 2008

Posted 06 April 2010 - 07:16 PM

If you inverted the curve and then tried it again, you definitely should not be getting the same results. My guess is something went wrong somewhere. Pempro got my PE down from about 30 arcsec to about 7. True, you're starting off from a far lower starting point but I would still expect Pempro to make a difference. Are you sure you actually used the inverted curve and not the original one to program the mount? You need to change the curve that is highlighted under "Program mount" and then click "Use Curve" otherwise it will just use the original one.

#19 Mert

Mert

    Skylab

  • *****
  • Posts: 4231
  • Joined: 31 Aug 2005

Posted 07 April 2010 - 12:18 PM

Can't really see why it doesn't make a difference,
G11 is a better mount then mine and on mine PEMPRO
did a very nice job as well!
Check what others with G11 have written, likely
something went wrong?






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics