HELP! Guiding Now Trouble!

[Whirlwind]

Hi

Just to add a little extra, there is a calculator in the resources section of CCWare's website (free to use) that helps you calculate what might be suitable guiding figures.

http://www.ccdware.com/resources/autoguidercalcv4.cfm

 I use this quite frequently (when it is clear) and it does seem to help with getting my guiding correct although I use MaximDL not PHD.

Hope this helps

Thanks

Ian

[Astrosurf]

Thanks Ian. If I understood it I'd use it! I'm really such a techie klutz. So why have I chosen this hobby???

[Ewan]

Hi Ewan - That wasn't a typo - I did mean 0,05... . When I first started using PHD I found a link that I'm sure USED to be on Craig Stark's site (of course I can't find it now!) and it noted that, with shorter FL guidescopes (such as a finder guider), he recommended using 0.05 or even 0...

However, in my hunt to refind that thread, I did find some quotes from Craig Stark on a CN thread -

Well, I've guided with an 8x50 finderscope and am the author of PHD

The trick is that you need to drop the minimum pixel motion parameter way down. For most (reasonable, non-insane) guide setups, the default of 0.25 pixels works well. If you're on a very short setup, this may correspond to a few pixels on your image rig. I run this at 0.05 when using the guide scope as the guider.

However, he does also note later in that thread:

There probably is a good relationship we can derive. Guiding with my 400 mm or so setup I'm at ~2.7"/pixel and run the default of 0.25 min motion. That means moving when I detect 0.675" of drift and anything less is considered noise. So, call it a half-pixel or so. If your guide rig is at 1"/pixel, you'll want then to increase the min motion to say, 0.5 pixels or so to keep in line with this.

Now, this assumes that your exposure duration of the guide image is enough to let the seeing you have blur the star enough so we have a nice central-tendency image. People too often drop the exposure duration way down in an attempt to yoke the mount hard and end up chasing the seeing (a race you'll always loose). Running even 3s or so you can likely run with a tighter min motion. Of course, if you're guiding at 1"/pixel and imaging at 5"/pixel the min motion could be relaxed (if for some reason it's needed).

You will find that PHD really doesn't like guiding at very long focal lengths (or more properly very small arcsec/pixel). I've not tried tuning the parameters for these setups. It's geared for sub-pixel guiding more than multiple-pixel guiding in the default form.

Andy i am not saying that 0.05 won't work just that i ahve never tried it that low so i may give it a shot just for testing BUT the finder i use has a small FL of 225mm @F3.75 with seems to guide using the figures in my screeny.

Now i will be the first to admit i understand Arcseconds error's not one bit or how resoloution / pix's & arc's work out but if my sub is good then i am happy BUT i would love to be able to get PHD 100% & just sit in doors for 4 hours & have collected all good subs without babysitting the laptop. So maybe that 0.05 setting could be the answer ?

Seeing does play a major role as to PHD settings i think as i have had mixed results when setting up exactly the same way but on different nights.

I understand the idea of having a guide star slightly unfocussed but does size really matter (no laughing Astrosurf please ), what i mean is a lot of targets only show small guide stars a few pix across so are these as good to guide with as a larger star ?

When i did the image above i used the largest star lower left to guide on as it was nice / round & bright, is this part of the reason the image came out ok ish ?

Astrosurf, the reason it makes sense, to me anyway, to use a guide star nearer to the centre of view in the guidescope is basically the quality of the star, the stars are not so good near the fov edges, guess my finder could use a flattner , this is why i tend not to use them.

As a last thought i try to alter PHD in very small increments as i find that is all it really needs.

If you enter the correct info into PHD it should work out the best calibration steps for you Astrosurf as well.

Thanks for the comment on my pic, it was just a guiding test but i am going to carry on with it as the Horse through the reduced SCT gives quite a nice FOV i thought.

[Astrosurf]

Many thanks again Ewan!

[AndyUK]

Hi Ewan - I'm certainly no guru on guiding, and that guide graph you posted is excellent - One thing I'm beginning to realise more and more is that, at the end of the day, the only thing that matters is the final result... and if it's not broken, don't fix it! 

I'm now using an OAG and pulse guiding using Maxim though EQMOD/ASCOM, but I originally tried PHD guiding with a 500mm piggybacked OTA and used a min motion of 0.15 (again, as recommended).  I switch to the finderguider to minimise weight on the mount and also because I was getting differential flexure issues....  The 9x50 finderscope certainly performed admirably and allegedly had a FL of 210mm.

Unfortunately I can't provide a sample guidegraph for comparison, but they were fairly similar to the one you posted.  However, it's been over a year since I used a finderguider and PHD so I can't really recall what size / magnitude star I chose - I usually tried to pick one that was reasonably isolated within the centre (ish) of the FoV, wasn't too bright, but wasn't one I thought would disappear that easily if the transparency reduced (ie Not very scientific!). 

However, I believe that size (magnitude) DOES matter as most guiding software uses the FWHM value for motion detection - If the star is too big and exposure too long (ie, the star is large and saturated), then the guidestar could shift quite a way before PHD recognises that it's actually moved....  I don't know if these are of any use, but I've attached my last PHD settings, which I used to guide my MN190 on M1 (and the result) - I always used to use 3s exposures.

However, coming back to this thread, one thing that's key (as mentioned previously in this thread), is balance and Polar alignment, and I think this is probably more at the root of Astrosurf's issues - I remember attending a star party a couple of years ago and I had a perfectly flat DEC guide line... but it wasn't on the centre line.  I discovered the following evening that the star I aligned to WASN'T Polaris  .   These are obviously just as important as getting the settings right!

[Astrosurf]

Very useful Andy. Thanks. I'm jealous of that image!

I'm never a 100% sure I'm on Polaris so now use a laser pointer to ensure I have it! I've been guiding fine until I got a new ST4 cable, which caused problems. I got another one from FLO, which worked OK before this, once, so I do wonder if it's the cable, along with some mistakes on my part as I was rushing and under-the-weather that night! Let's see how it goes next time.

Thanks for all your help guys. It's much appreciated!

XXX

[malc-c]

Just to throw another spanner in the works, it can all go pear shaped even though you change nothing - the wonders of PHD.

Took advantage of another clear night last night to get a few more subs of the SN.  Nothing had changed as the scope is observatory based, even used the same guide-star.  Set PHD calibrating and after 60 steps W/E it failed.  Tried again and same thing.  So made a small adjustment to the star tolerance and after 44 steps each way, and then 18 N/S it started guiding.... but the graph was like the Alps, all peaks and troughs, and then the lines diverged !!  Opened the settings and made a few more adjustments and tried again, this time it calibrated in 52 steps, but the graph was really stable and smooth (see attached).

Ideally I would like to get the calibration steps for E/W down to 20 or so as it is really frustrating having PHD holding up every imaging session.  Five minutes to open the observatory and get the scope on target, then forty-five minutes before its guiding

[Zakalwe]

Don't sweat the mount being level, or indeed, the altitude settings. As long as you are polar-aligned then it doesn't matter a jot where the tripod legs are pointing*

Focus on getting the polar-alignment nailed before starting to mess with the PHD settings. If the mount isn't polar-aligned then you are making the whole job much, much harder (especially if you are using a longish focal length to image with). Its worth checking the polar-alignment regularly, even if you are pier based. Things can shift over time...even with big concrete masses under the pier. There's a lot of water in the ground at the moment (is it ever going to stop raining??!?) and that can cause the ground to shift (as can heavy frost). it doesn't take much, especially when the length of the pier acts as a lever to magnify the smallest movements.

*The exception to this is if you are using a polar-alignment program like Alignmaster. If works best if the tripod is level.

[AndyUK]

Ideally I would like to get the calibration steps for E/W down to 20 or so as it is really frustrating having PHD holding up every imaging session.  Five minutes to open the observatory and get the scope on target, then forty-five minutes before its guiding

Hi Malc - What's your calibration step setting?  I found that if I was trying to image closer to Polaris, then the number of calibration steps PHD went through increased, so I'd increase it and that then brought the number of steps down to a more reasonable figure.

[Zakalwe]

Hi Malc - What's your calibration step setting?  I found that if I was trying to image closer to Polaris, then the number of calibration steps PHD went through increased, so I'd increase it and that then brought the number of steps down to a more reasonable figure.

Thats a point well worth making. The guiding setting may need tweaking depending on where in the sky your are aiming.

This file goes into a lot of detail about the different settings in PHD, which I found very useful.

GuideToGuiding.pdf

However, for the OP I wouldn't advise using the doc. It can only confuse things IMHO. Nail the basic stuff first...these will be the foundations for everything else. There's no point tweaking the guiding if the polar alignment, focusing, telescope balance, tracking and so on are not as good as they can be. Sky conditions can also make a difference...if there's lots of clouds floating by then it can confuse PHD during the calibration.

[Astrosurf]

That's brilliant guys. Thank you. Thanks for the doc Zakalwe!

[malc-c]

Hi Malc - What's your calibration step setting?  I found that if I was trying to image closer to Polaris, then the number of calibration steps PHD went through increased, so I'd increase it and that then brought the number of steps down to a more reasonable figure.

Hi,

Normally 2250ms, but tried anywhere between 1750 and 3200,  Can't remember which one worked in the end.  I see your point regarding being near the pole.

[AndyUK]

Normally 2250ms, but tried anywhere between 1750 and 3200...

Hmmm...  They seem fairly reasonable values.  With the 9x50 finderguider, I used to default to 3000ms and it used to calibrate in about 15 steps each way (more or less)