#21  
Old 19-10-2014, 11:40 AM
Shiraz's Avatar
Shiraz (Ray)
Registered User

Shiraz is offline
 
Join Date: Apr 2010
Location: ardrossan south australia
Posts: 4,918
Quote:
Originally Posted by Joshua Bunn View Post
Hi Paul, I noticed exposure times for your PE log were 3 sec, whats your reasoning behind this? This post explains it a little. Its usually recommended to use 1 sec exposure times or the like, and that's what i use - or less.
Paul, hope you don't mind, but would like to confirm with Josh that this means that his approach is to use high time resolution PEC to remove most of the PE and then mop up any residual errors with slow (6 seconds or more) updates through the guiding.
Reply With Quote
  #22  
Old 19-10-2014, 11:53 AM
Joshua Bunn's Avatar
Joshua Bunn (Joshua)
Registered User

Joshua Bunn is offline
 
Join Date: Oct 2011
Location: Albany, Western Australia
Posts: 1,459
Quote:
Originally Posted by Shiraz View Post
...his approach is to use high time resolution PEC to remove most of the PE and then mop up any residual errors with slow (6 seconds or more) updates through the guiding.
That sounds about right Ray, high sampling of the PE. And then Protrack does some work to.
Reply With Quote
  #23  
Old 19-10-2014, 02:40 PM
PRejto's Avatar
PRejto (Peter)
Registered User

PRejto is offline
 
Join Date: Jan 2011
Location: Rylstone, NSW, Australia
Posts: 1,397
I'll just 2nd a few things already posted.

TSX never has worked for my former MX or MEII. PEMpro has always worked. Short exposures are good and Ray Garalak indicated that one shouldn't do more that 5 cycles or so as it makes the curve harder to extract.

Peter
Reply With Quote
  #24  
Old 19-10-2014, 10:00 PM
Paul Haese's Avatar
Paul Haese
Registered User

Paul Haese is offline
 
Join Date: Jan 2009
Location: Adelaide
Posts: 9,937
Thanks guys, plenty to consider here.

Quote:
Originally Posted by Shiraz View Post
thanks Paul. that looks nice and smooth and almost shoots the exposure time theory out of the water . However, there is a relatively abrupt transition from negative going to positive on a couple of cycles that could possibly explain your results. Shorter exposures are probably still worth a try. With such consistent PE, you should be seeing large benefit from PEC - wonder why it doesn't help?
Yes it is odd but even if there is no PEC guiding should take care of the problem.

Maybe the abrupt transitions are a warp in the new worm?

Quote:
Originally Posted by Joshua Bunn View Post
Hi Paul, I noticed exposure times for your PE log were 3 sec, whats your reasoning behind this? This post explains it a little. Its usually recommended to use 1 sec exposure times or the like, and that's what i use - or less.
Actually I am puzzled by this too as I knew that it should be a second or under for good sampling of the data and I am sure I set this correctly at the time. Might have to have another sampling of it and see if the results differ. Another log I have from the night before has 3 second exposures too.

Quote:
Originally Posted by gregbradley View Post
PEC on the PMX seems touchy. I did some PEC on my PMX and at first it did not seem to make any difference like you are experiencing. There is the possibility of course that the East/West is back to front on the data so its told to push when it should pull but that probably isn't it as that would most likely worsen. Although one time I did have that reversed it didn't seem to make a lot of difference.

3 second intervals for the PEC log may cause an issue as you want more frequent samples for an accurate PEC curve. Also keeping the other points in about which spot in the sky you use for your PEC guide star.
Another would be to do the PEC curve on a night of good seeing. I haven't read the instructions for a while but I do seem to recall shorter exposure times being recommended to get a more precise sample. Also about a 20minute run (requires good polar alignment otherwise the guide star drfits out of the imaging box). When you think about it 3 seconds is probably way too long as these corrections in the PEC curve whilst they tend to be smoothed out they still move a fair bit over 3 seconds so you don't want a PEC that is always coming in late or averaged over too long a time so the corrections are always coming in late either too little or too much.

Having said that I doubt that is the reason why you are having trouble as it would still make some difference as the curve is smoothed out a fair bit during the final steps of making one. It sounds like its way off, out of synch and your autoguider is having to correct it for bad PEC adjustments.

But apart from that it may be best to ditch Sky X PEC and use PemPro as no doubt you know the above. Especially being a remote system and you don't have many nights to experiment to find out it was some software bug in the SkyX after several nights of trying.

I used Precision PEC for my PME and that got a nice curve. It was a subtle sine wave which looked much like my PME's PEC curve.

My original PMX curve seemed too much for correction. On the PME its a pretty gentle almost sign wave type PEC.

There were in the past numerous threads about Sky X and PME PEC being out of phase so there is a question mark over it.

When I redid my PEC using Sky X on a night of good seeing with the Polar alignment already very very close (probably not as close as yours is now) I got a very good result.

Another factor of course is how stable are you optics in terms of flexure and mirror shift? Protrack corrects for flexures of various types, PEC of course for gear errors.

Another thing that threw my PMX off at one stage was a bit of dirt or something got on the worm so about every 3 or 4 minutes I would get a bad PE spike that threw the guiding out for a few seconds resulting in double stars in the final image. A replacement worm corrected that. Not sure what the spike was caused by in the end - dirt? Damaged worm from the slipping gear/cam issue? Either way it fixed and I do get a noticeable improvement in my PMX using PEC so it can work even if perhaps a bit touchy.

You are imaging at 3 metres though and I image at 1260mm on my PMX so everything is 3X exaggerated on your setup compared to mine.

On my PME I find a 300 point Tpoint model does improve roundness of stars using Protrack at the same time as PEC on at 3 metres focal length. So Protrack is correcting some minor flexes that are slow to show up in the image. It can be odd as when I watch the guiding some of the corrections are larger than what I would think is ideal yet the images show round stars. That must be the Protrack flex correction being added to the PEC curve correction plus the usual autoguiding corrections if they all coincide in the same direction.

My first action would be to redo the PEC using Sky X and make sure its exactly as per the manual and if still no go then try PemPro or Precision PEC and see how that goes.

There is no spike in your PE curve so PEC ultimately should be pretty easy.

Greg.
The East/ West issue is certainly a problem. I have switched between both and not seen any difference. I still get the oscillations but no change in the extent of those oscillations.

PA is good and the guide star did not drift after 20 minutes.

As you know my seeing is good at my location so even an average night of seeing is still pretty darn good.

The guide star has to be very close to ecliptic for best results and that is certainly what I did. It also must be no dimmer than mag 5 or else the guide results will be less than satisfactory.

I am certainly considered getting pempro to resolve the very real possibility that the SkyX has an issue. Certainly at that point.

I am pretty certain the optics are pretty reliable now as I have installed the new back plate and I cannot find any discernible flexure in my images now. ie stars being slightly different shapes in different parts of the sky before the back plate change over.

Focal length is not quite that long but still long enough and yes compared to the other system this is not a piece of cake. My FSQ system is a flat line most nights in both axis and hence why my refractor images are pretty sharp.


Quote:
Originally Posted by Shiraz View Post
Paul, hope you don't mind, but would like to confirm with Josh that this means that his approach is to use high time resolution PEC to remove most of the PE and then mop up any residual errors with slow (6 seconds or more) updates through the guiding.
No problem Ray. I am happy to hear all manner of ideas. In fact I did try some of the mods to Min and Max last night and I got it running pretty smooth near the Helix. When I moved over to NGC253 it went to pieces again. So I am starting to get some where.

I will be sorting PE with a higher sampling than at present.

Quote:
Originally Posted by Joshua Bunn View Post
That sounds about right Ray, high sampling of the PE. And then Protrack does some work to.
Protrack certainly has an impact. My model needs running again now with the change of the back plate but the previous model of 288 points really did help keep stars tighter.
Reply With Quote
  #25  
Old 20-10-2014, 01:08 PM
PRejto's Avatar
PRejto (Peter)
Registered User

PRejto is offline
 
Join Date: Jan 2011
Location: Rylstone, NSW, Australia
Posts: 1,397
Quote:
Originally Posted by Paul Haese View Post
Imaging with my RC12 via MaximDL for guiding and in OAG configuration.

Dec axis is perfect for guiding. Hardly moves off the centre line. RA goes up and down in an apparent sine wave.
Paul,

I don't know what your RC12 weighs compared to a 14" under discussion here but reading through this post at SB, and in particular, the very last post on page 4, the issue sounds quite similar. The upshot is leading towards a conclusion that the MX might not take this much weight.

Peter

http://www.bisque.com/sc/forums/t/22...px?PageIndex=4
Reply With Quote
  #26  
Old 20-10-2014, 01:42 PM
Paul Haese's Avatar
Paul Haese
Registered User

Paul Haese is offline
 
Join Date: Jan 2009
Location: Adelaide
Posts: 9,937
Quote:
Originally Posted by PRejto View Post
Paul,

I don't know what your RC12 weighs compared to a 14" under discussion here but reading through this post at SB, and in particular, the very last post on page 4, the issue sounds quite similar. The upshot is leading towards a conclusion that the MX might not take this much weight.

Peter

http://www.bisque.com/sc/forums/t/22...px?PageIndex=4
Yes possibly a problem there. I have calculated the weight of the equipment package. Total weight without including the 4 counter weights (which are down the bottom of the shaft) is 34kg all up. That is 7 kilo from capacity but well within spec.

My mount does not jump like that though. I am seeing a slight movement up and down over a period of time so the guiding looks like a slow sine wave over several minutes. Maybe it is too heavy but I had spoken to Daniel about the weight before and he did not say it was a problem.

Maybe I will move the counter weights up to the top of the mount and put on another weight to see if that makes a difference.
Reply With Quote
  #27  
Old 20-10-2014, 02:04 PM
gregbradley's Avatar
gregbradley
Registered User

gregbradley is offline
 
Join Date: Feb 2006
Location: Sydney
Posts: 17,877
I bought PEMPro a while ago for this reason as I was not happy with PEC on my dark site PMX. But I redid the PEC one last time using Sky X and it worked. So not sure what was different. I may have loaded a later version of Sky X.

So perhaps try a later Sky X version if you dare as it may have been fixed in a later version as there was some internet traffic about this issue about 9 months ago.

Personally I would use Pempro and forget about Sky X and PEC. It certainly sounds like some hidden software issue that could go on for a while trying to find out what it is and a later version is just as likely to make something else not work!

Greg.
Reply With Quote
  #28  
Old 20-10-2014, 10:27 PM
Paul Haese's Avatar
Paul Haese
Registered User

Paul Haese is offline
 
Join Date: Jan 2009
Location: Adelaide
Posts: 9,937
Ok a bit of an update.

I have changed the min move in X to 0.02 and max to 2.0. I took off the PE and installed a flat line. Then I ran guide exposures for 1 second but found that did not have bright enough stars so I made it 2 seconds. The guide graph is a lot smoother now generally in RA. So I am getting somewhere. Stars are round at 10 minutes too. I have a bit of wind down the observatory tonight so I cannot go much further than I have now. I will report back once the camera comes back from the States.

Thanks so far for the help guys.
Reply With Quote
  #29  
Old 21-10-2014, 06:14 AM
SkyViking's Avatar
SkyViking (Rolf)
Registered User

SkyViking is offline
 
Join Date: Aug 2009
Location: Waitakere Ranges, New Zealand
Posts: 2,260
Hi Paul, although our mounts are different I'll second shorter guide exposures. I use 0.6s most of the time and only go up to 1s on the odd occasion where no guide star is bright enough. And they don't have to be very bright at all in my experience.

Cheers,
Rolf
Reply With Quote
  #30  
Old 21-10-2014, 06:21 PM
Paul Haese's Avatar
Paul Haese
Registered User

Paul Haese is offline
 
Join Date: Jan 2009
Location: Adelaide
Posts: 9,937
Quote:
Originally Posted by SkyViking View Post
Hi Paul, although our mounts are different I'll second shorter guide exposures. I use 0.6s most of the time and only go up to 1s on the odd occasion where no guide star is bright enough. And they don't have to be very bright at all in my experience.

Cheers,
Rolf
The different f ratios impact on guide star selection Rolf. f8 has a big impact on how bright stars are even at 2x2 binning. I take it you are using f4 at 2x2?
Reply With Quote
  #31  
Old 22-10-2014, 01:12 PM
SkyViking's Avatar
SkyViking (Rolf)
Registered User

SkyViking is offline
 
Join Date: Aug 2009
Location: Waitakere Ranges, New Zealand
Posts: 2,260
Quote:
Originally Posted by Paul Haese View Post
The different f ratios impact on guide star selection Rolf. f8 has a big impact on how bright stars are even at 2x2 binning. I take it you are using f4 at 2x2?
Good point Paul, but that then makes me think your guide stars are oversampled. Would it not be better to use higher binning when guiding at f/8? Or a different guide camera with larger pixels?
Reply With Quote
  #32  
Old 22-10-2014, 01:17 PM
Joshua Bunn's Avatar
Joshua Bunn (Joshua)
Registered User

Joshua Bunn is offline
 
Join Date: Oct 2011
Location: Albany, Western Australia
Posts: 1,459
Good points Paul and Rolf! I must add, I bin my guide camera pixels 2x2 when guiding, so 0.73"/pix for imaging but 1.46"/pix for guiding. works a treat for me, and being highly sampled when guiding puts a lot of demand on the guiding and mount corrections - depending on seeing conditions and exposure time.
Reply With Quote
  #33  
Old 22-10-2014, 04:12 PM
Paul Haese's Avatar
Paul Haese
Registered User

Paul Haese is offline
 
Join Date: Jan 2009
Location: Adelaide
Posts: 9,937
I have been guiding at 2x2 so my image scale without the 0.73 reduction is 1.25 arcsecs per pixel. With the reduction it is 1.72 arcsecs per pixel. I wonder if binning at 3x3 will have a positive impact. Is this what you are thinking Rolf? I would be at 2.58. Not sure if that would work.
Reply With Quote
  #34  
Old 22-10-2014, 05:21 PM
Paul Haese's Avatar
Paul Haese
Registered User

Paul Haese is offline
 
Join Date: Jan 2009
Location: Adelaide
Posts: 9,937
I thought I would add an untouched sub with the new guiding parameters. Ignoring the stars are the edge of the field (the flattener is still not in the correct position), the stars in the center look tight and fairly round. This is a 10 minute sub guided at 2 second exposures with min set to 0.02 and max at 2.0.
Attached Thumbnails
Click for full-size image (10 minute ngc253.jpg)
333.5 KB33 views
Reply With Quote
  #35  
Old 22-10-2014, 07:18 PM
SkyViking's Avatar
SkyViking (Rolf)
Registered User

SkyViking is offline
 
Join Date: Aug 2009
Location: Waitakere Ranges, New Zealand
Posts: 2,260
Quote:
Originally Posted by Paul Haese View Post
I have been guiding at 2x2 so my image scale without the 0.73 reduction is 1.25 arcsecs per pixel. With the reduction it is 1.72 arcsecs per pixel. I wonder if binning at 3x3 will have a positive impact. Is this what you are thinking Rolf? I would be at 2.58. Not sure if that would work.
Yes that's what I mean. From memory I might actually be guiding binned 3x3 already at f/4, but I'll have to check that next time I'm imaging. If that's the case I think you can safely bin at least 3x3 or even more when you are at f/8. Guiding is using sub-pixel accuracy anyway so should be fine. Worth a try - and if it doesn't make your guiding worse at least you'll have brighter guide stars.
Reply With Quote
  #36  
Old 22-10-2014, 07:24 PM
Paul Haese's Avatar
Paul Haese
Registered User

Paul Haese is offline
 
Join Date: Jan 2009
Location: Adelaide
Posts: 9,937
Quote:
Originally Posted by SkyViking View Post
Worth a try - and if it doesn't make your guiding worse at least you'll have brighter guide stars.
Yeah true. Some thing else to try.
Reply With Quote
Reply

Bookmarks

Thread Tools
Rate This Thread
Rate This Thread:

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump


All times are GMT +10. The time is now 08:39 PM.

Powered by vBulletin Version 3.8.7 | Copyright ©2000 - 2024, Jelsoft Enterprises Ltd.
Advertisement
Testar
Advertisement
Bintel
Advertisement