dumb phd question

graham.hobart (Graham stevens) · #1 18-11-2011, 10:54 PM

So I am guiding to four min subs and getting ok stars but looking at the graph the DEC keeps climbing- it this still my PA?
Graz

adman (Adam) · #2 18-11-2011, 10:58 PM

more than likely to be PA - its about the only thing that will keep the DEC graph moving steadily in one direction.

How far does it move during one 4 minute sub?

Adam

graham.hobart (Graham stevens) · #3 18-11-2011, 11:07 PM

it moves quite a bit, so much so that if I check my graph towards the end of the sub I can't see the dec line at all.
NB dithering makes my mount (cgem) go crazy and start slewing weird rates so I have abandoned it!
Graz

adman (Adam) · #4 18-11-2011, 11:30 PM

in PHD each horizontal line away from the centre one represents one pixel. If it moves off the scale, I think that is something like 4 or 5 pixels..? How are your stars still round if it is moving that much?

midnight (Darrin) · #5 19-11-2011, 03:01 AM

Graham,

Assuming you're using the Eos Canon, trying quickly scrolling through your images in Canon's digital photo prof (supplied with the camera) What I find is if I take say 20 lights (eg 4 min each) and I quickly scroll though the previews in DPP, I can clearly see the motion of each frame and if my PA is out (eg I rush in and only do a meridian drift check) I find the images appear to rotate. Ie scrolling through DPP with those 20 items, I can clearly rotation. This is a clear indicator PA is not perfect

Cheers,
Darrin...

leon · #6 19-11-2011, 07:52 AM

Each sub doesn't really have to be perfect, Image Plus soon sorts that out and adjusts for any rotation to make the final image.

Leon

mithrandir (Andrew) · #7 19-11-2011, 09:36 AM

There are better descriptions elsewhere, but here's the gist of it.

You can turn off guiding output in the brain with the mount motors running and drift align on the guider image.

Al's Reticule would help you get the alignment of the RA and Dec axes relative to the guide camera. (I try to always have the guide x/y aligned with the mount axes since that makes everything simpler.) Then you can iteratively adjust the PA so the guide star stays on one of the axes, and eventually at one spot.

You will probably have to recentre the guider after every PA adjustment. If the drift gets faster, you've gone the wrong way.

graham.hobart (Graham stevens) · #8 21-11-2011, 08:38 AM

Just a thought-can you do this procedure on any star? Only ask as have limited (actually no) horizon views
Graham

adman (Adam) · #9 21-11-2011, 11:36 AM

Ideally you need a star low in the east or west and then one near the meridian. Dont choose one too close to the horizon as there will be too much refraction. Stars in these positions will have most of their drift due to polar misalignment in either alt or az. Other position will tend to have a mix of both.

The east or west one will allow you to adjust your mounts altitude - if the star drifts south, lower the altitude adjustment

The star near the meridian shows you the azimuth adjustment required - if the star drifts north, you need to adjust the azimuth so the mount turns clockwise when viewed from above.

I use PHD to drift align, and I have worked out if thDEC graph drifts downwards when I am in the east, I need to raise the altitude, if it drifts downswards at the meridian, I need to twist the G11's azimuth adjustment downwards - but your mount may be different - its just how it helps me remember it

Adam

graham.hobart (Graham stevens) · #10 21-11-2011, 11:45 AM

This is the issue really then as I have only NE or NW sky above about 40'c and then meridian north a bit higher (above a tree), South is mainly blocked by a house. E & W (true) are not visible except high up in the sky.
This is the stumbling block I have encountered before with drift aligning, the stars needed are not visible.

adman (Adam) · #11 21-11-2011, 01:44 PM

I tried dithering last night - and my graph did the same as yours - the DEC trace just climbed off the scale. It looked like the graph wasn't making allowances for the amount of movement being done during the dithering operation?

graham.hobart (Graham stevens) · #12 21-11-2011, 01:47 PM

A Ha! What to be done?

Peter.M · #13 21-11-2011, 05:04 PM

Quote:

Originally Posted by graham.hobart

This is the issue really then as I have only NE or NW sky above about 40'c and then meridian north a bit higher (above a tree), South is mainly blocked by a house. E & W (true) are not visible except high up in the sky.
This is the stumbling block I have encountered before with drift aligning, the stars needed are not visible.

I have a similar problem to this, I can use stars about 20-30 degrees up but was always worried about its accuracy. I recently trialed alignmaster and It is awesome! You may still run into problems finding the right stars, but its a hell of alot easyer than drifting.

adman (Adam) · #14 21-11-2011, 08:48 PM

Quote:

Originally Posted by Peter.M

I have a similar problem to this, I can use stars about 20-30 degrees up but was always worried about its accuracy. I recently trialed alignmaster and It is awesome! You may still run into problems finding the right stars, but its a hell of alot easyer than drifting.

You shouldn't run into any problems at 20 or so degrees

Peter.M · #15 21-11-2011, 08:54 PM

I had good results drifting at 20 degrees, but with alignmaster I dont even need to bother. Im still trialing it but for the small amount of money the creator wants I will be purchasing it.

naskies (Dave) · #16 22-11-2011, 11:15 PM

This might be a silly question... but is PHD stopped/resumed when the scope is being moved for dithering? If not, I'm assuming that PHD will treat the dither movement as tracking error?

Polar alignment was a royal pain for me on my EQ6 until recently. The bubble level wasn't even remotely close to level, so my alt/az movements always affected each other. Now that I have the mount level using my iPhone, drift aligning with PHD is really quick.

I throw the mount head onto the tripod, roughly align it with the iPhone compass, point the guide scope towards the meridian and take an initial reading from PHD. I then make a large azimuth adjustment (e.g. three half turns of the EQ6 knob), and get another reading from PHD. The difference in angles of the dy error line in the PHD graph corresponds quite well with the number of turns of the knob - so I immediately know roughly how much to turn the knobs. Another two or three iterations of this keeps my dy error within one pixel over 500 samples x 0.2 sec.

I don't need subs much longer than 5 mins (too much light pollution) so I've found that I don't even need to tweak the altitude of the mount anymore.... very good PA in azimuth minimises my field rotation when guiding.