Drift Alignement Equipement & Polar Alignement Questions

[Samibotss]

Howdy,

I've been getting familiar with my astrophotography recently, but I still have one problem remaining : Polar Alignement. The way I'm setup, I only have access to the southern part of the sky. Until now, I've used some methods to do it roughly with a compass, but that is nowhere near accurate enough for AP. The most exposure time I've gotten out of it before having trails is about 4 seconds. I'm now looking into doing drift alignement, and I was wondering, is an autoguiding scope required to do it? Until now all the information I found about it mentioned using a guide camera and PHD2. Another question I had is, since the autoguiding setup corrects for tracking errors, could it compensate for a rough polar alignement, or do you need a good one from the start?

Also, I have a slight problem with the Goto handset, as on initial alignement, it slews nowhere near the target, I think it must be the way it calculates Sidereal time. I've tried a bunch of different observation site settings, but I still can't get it right...

I would greatly appreciate if anybody can answer my questions

 

[david_taurus83]

Hi. It might help if you can list your equipment, mount, scope, camera etc

[vlaiv]

Drift alignment method works with both guide scope / camera and primary scope camera - it is just a matter of convenience which one you'll use.

For good drift alignment and general goto precision, it is important to have your mount tripod - level and stable (not moving).

Auto guiding will correct for polar alignment errors, but it can't work miracles. The better your polar alignment is from the start - less work auto guider needs to perform. There is only one case where you don't want perfect polar alignment - if your mount has DEC backlash. In that case it is better to have slight polar alignment error which then causes DEC error to be to one side only (due to drift). Auto guider will correct this error and corrections will always be to one side as well. This prevents changing of direction in DEC and circumvents any issues with DEC backlash. PHD2 has algorithm called "resist switch" - and I think it is default DEC guiding algorithm. It is used precisely for this reason - small PA error is going to produce drift to one side only - everything else is either seeing or wind or mount roughness.

[jif001]

It’s normal for the mount to be inaccurate when it slews to the first target. You should always follow polar alignment with 2- or 3-star alignment. The latter assists the Goto system in finding the targets accurately. Pick a bright star some distance from the pole star and use the Goto to slew there. It will be off, but you use the handset buttons and finder scope to centre the star in the field of view, then tell the handset it’s there. Repeat with a second star somewhere else then, ideally, a third one. That should mean the Goto will be accurate finding your next target. 
 

if you have a computerised setup using ASCOM, you can platesolve instead of centring the star each time, and sync the location to the mount.  

[Samibotss]

9 hours ago, david_taurus83 said:

Hi. It might help if you can list your equipment, mount, scope, camera etc

SkyWatcher EQM35-PRO with Evostar 72ED, Canon 4000D.

9 hours ago, vlaiv said:

Drift alignment method works with both guide scope / camera and primary scope camera - it is just a matter of convenience which one you'll use.

For good drift alignment and general goto precision, it is important to have your mount tripod - level and stable (not moving).

Auto guiding will correct for polar alignment errors, but it can't work miracles. The better your polar alignment is from the start - less work auto guider needs to perform. There is only one case where you don't want perfect polar alignment - if your mount has DEC backlash. In that case it is better to have slight polar alignment error which then causes DEC error to be to one side only (due to drift). Auto guider will correct this error and corrections will always be to one side as well. This prevents changing of direction in DEC and circumvents any issues with DEC backlash. PHD2 has algorithm called "resist switch" - and I think it is default DEC guiding algorithm. It is used precisely for this reason - small PA error is going to produce drift to one side only - everything else is either seeing or wind or mount roughness.

Is there a piece of software which would help me drift align with my DSLR? I think I saw somewhere that PHD2 doesn't currently support DSLRs. As for autoguiding, is there a limit to how much the auto guider can correct?

46 minutes ago, jif001 said:

It’s normal for the mount to be inaccurate when it slews to the first target. You should always follow polar alignment with 2- or 3-star alignment. The latter assists the Goto system in finding the targets accurately. Pick a bright star some distance from the pole star and use the Goto to slew there. It will be off, but you use the handset buttons and finder scope to centre the star in the field of view, then tell the handset it’s there. Repeat with a second star somewhere else then, ideally, a third one. That should mean the Goto will be accurate finding your next target. 
 

if you have a computerised setup using ASCOM, you can platesolve instead of centring the star each time, and sync the location to the mount.  

Well, it wasn't just inaccurate, it was slewing in the opposite direction almost, and I had the "RA Offset > 45 deg" error. I think I'm slowly getting it right though, last time it slewed near enough to have the "alignement successful" message once I centered it, but it was inaccurate slewing to other targets after this, but that maybe do to the fact that I had only done a 1-star alignement with big adjustements.

 

[Davey-T]

Have you tried DARV ?

Dave

https://www.cloudynights.com/articles/cat/articles/darv-drift-alignment-by-robert-vice-r2760

[vlaiv]

20 minutes ago, Samibotss said:

Is there a piece of software which would help me drift align with my DSLR? I think I saw somewhere that PHD2 doesn't currently support DSLRs. As for autoguiding, is there a limit to how much the auto guider can correct?

Long time ago, I drift aligned with a piece of software called EQAlign or something like that. Let me see if I can find it.

Here it is:

http://eqalign.net/e_eqalign.html

Unfortunately it requires WDM style drivers (like web camera) and I'm not sure you will find those for DSLR (you could - I've seen a piece of software that lets DSLR act as web cam for skype and such).

You don't really need software to do drift align. You can do it yourself by using long exposures and bright stars and judging trail that star makes due to drift (some sort of measurement of trail length). You need to orient camera so that you know which direction is DEC (vertical for example). There will be drift in RA as well - but you don't need to address that for polar alignment (but it can as well mess up your long exposures - check if PEC feature is available with your mount).

There is a limit to how much auto guider can fix things. Auto guiding works by using short exposures like second or two. It issues commands to mount to get to proper position - this is called pulse. Auto guiding software measures reference star position and determines how long this pulse should be to correct the position. It uses guide speed which is usually set to fraction of sidereal rate - x1 or lower like x0.25.

Sidereal rate is about 15"/s. This means that in theory, auto guiding can fix for drift that is 15"/s if set to operate on sidereal guide speed. In practice, you want your drift rate to be much lower than that as you don't want to correct each guide cycle - that also creates elongated stars - as mount moves back and forth all the time.

In reality - any decent polar alignment will have rather small drift rate. You can use this calculator to calculate drift rate due to polar alignment error:

http://celestialwonders.com/tools/driftRateCalc.html

 

[Samibotss]

1 hour ago, vlaiv said:

Long time ago, I drift aligned with a piece of software called EQAlign or something like that. Let me see if I can find it.

Here it is:

http://eqalign.net/e_eqalign.html

Unfortunately it requires WDM style drivers (like web camera) and I'm not sure you will find those for DSLR (you could - I've seen a piece of software that lets DSLR act as web cam for skype and such).

You don't really need software to do drift align. You can do it yourself by using long exposures and bright stars and judging trail that star makes due to drift (some sort of measurement of trail length). You need to orient camera so that you know which direction is DEC (vertical for example). There will be drift in RA as well - but you don't need to address that for polar alignment (but it can as well mess up your long exposures - check if PEC feature is available with your mount).

There is a limit to how much auto guider can fix things. Auto guiding works by using short exposures like second or two. It issues commands to mount to get to proper position - this is called pulse. Auto guiding software measures reference star position and determines how long this pulse should be to correct the position. It uses guide speed which is usually set to fraction of sidereal rate - x1 or lower like x0.25.

Sidereal rate is about 15"/s. This means that in theory, auto guiding can fix for drift that is 15"/s if set to operate on sidereal guide speed. In practice, you want your drift rate to be much lower than that as you don't want to correct each guide cycle - that also creates elongated stars - as mount moves back and forth all the time.

In reality - any decent polar alignment will have rather small drift rate. You can use this calculator to calculate drift rate due to polar alignment error:

http://celestialwonders.com/tools/driftRateCalc.html

 

I see, thanks for all the information! I'll try drift alignement manually then, and see how it goes. If it goes well, I'll start looking into autoguiding to improve my performance.

[david_taurus83]

2 hours ago, Davey-T said:

Have you tried DARV ?

Dave

https://www.cloudynights.com/articles/cat/articles/darv-drift-alignment-by-robert-vice-r2760

^^^This!

[vlaiv]

22 minutes ago, Samibotss said:

I see, thanks for all the information! I'll try drift alignement manually then, and see how it goes. If it goes well, I'll start looking into autoguiding to improve my performance.

Indeed, look at the post above by @Davey-T and follow that link - DARV is very simple and straight forward (and clever I might add) way to do "manual" drift align.

I would throw in a simple modification since you can't do regular polar align and then improve with drift.

Make your first couple of iterations shorter - about half a minute as drift is probably going to be significant. Later extend it to make polar alignment accurate.

[Samibotss]

9 hours ago, david_taurus83 said:

^^^This!

 

9 hours ago, vlaiv said:

Indeed, look at the post above by @Davey-T and follow that link - DARV is very simple and straight forward (and clever I might add) way to do "manual" drift align.

I would throw in a simple modification since you can't do regular polar align and then improve with drift.

Make your first couple of iterations shorter - about half a minute as drift is probably going to be significant. Later extend it to make polar alignment accurate.

 

Oh my, I completely missed his post. You're absolutely right, this looks a good solution to my problem.  I just don't quite understand the last step with regards to the altitude, do I just take a 5sec exposure, see the trail, and adjust the altitude until I get it right, or do I also move the telescope during the exposure on the RA axis?

[david_taurus83]

I think you need to also move the scope. Another option is the handset polar align if your mount has that function.

[jif001]

13 hours ago, Samibotss said:

Well, it wasn't just inaccurate, it was slewing in the opposite direction almost, and I had the "RA Offset > 45 deg" error. I think I'm slowly getting it right though, last time it slewed near enough to have the "alignement successful" message once I centered it, but it was inaccurate slewing to other targets after this, but that maybe do to the fact that I had only done a 1-star alignement with big adjustements.

 

Check the settings in the handset, particularly the date. I think it uses the American format, where, for example, August 1st would be entered as 8-1-2020 (i.e. ‘August 1st’). If you entered it the other way around as 1-8-2020 (‘1st August’) the date will be 7 months adrift which might explain it going in the wrong direction. 

Edited August 16, 2020 by jif001

[vlaiv]

42 minutes ago, Samibotss said:

 

 

Oh my, I completely missed his post. You're absolutely right, this looks a good solution to my problem.  I just don't quite understand the last step with regards to the altitude, do I just take a 5sec exposure, see the trail, and adjust the altitude until I get it right, or do I also move the telescope during the exposure on the RA axis?

There are two places where you need to do above procedure. It has to do with X-Y coordinate system and projections and all that math stuff.

When you point your scope due south at celestial Meridian (or just before or after), any drift that happens in that part of the sky will be due to Polar alignment error in horizontal - or in azimuth.

So first step is to adjust your mount in azimuth by aiming at the star due south at meridian.

Second step is to do the same procedure aiming at eight east or west and DEC 0. Since this is 90 degrees to south - you are now doing alignment in altitude - or vertical.

This is why you need to have your mount level to begin with - only if mount is level you can "divide" adjustments like this so they are not dependent to each other.

DARV method says - start your exposure with telescope tracking and wait for 5 seconds. This little wait time is for star to "burn" a little reference dot at the beginning of the path. Then press your controller so that telescope slews at x1 speed in RA direction. Do that for half of "exposure" then reverse direction (just press opposite button) and hold for second part of exposure.

If you want to do 5 minute exposure without trailing - do above for 2.5 minutes in one direction and 2.5 minutes in other direction.

You can start by doing shorter runs - like 30seconds or one minute if your polar alignment is poor at the beginning, but as soon as you start improving it - switch to full duration of exposure.

Makes sense?

[Samibotss]

11 minutes ago, vlaiv said:

There are two places where you need to do above procedure. It has to do with X-Y coordinate system and projections and all that math stuff.

When you point your scope due south at celestial Meridian (or just before or after), any drift that happens in that part of the sky will be due to Polar alignment error in horizontal - or in azimuth.

So first step is to adjust your mount in azimuth by aiming at the star due south at meridian.

Second step is to do the same procedure aiming at eight east or west and DEC 0. Since this is 90 degrees to south - you are now doing alignment in altitude - or vertical.

This is why you need to have your mount level to begin with - only if mount is level you can "divide" adjustments like this so they are not dependent to each other.

DARV method says - start your exposure with telescope tracking and wait for 5 seconds. This little wait time is for star to "burn" a little reference dot at the beginning of the path. Then press your controller so that telescope slews at x1 speed in RA direction. Do that for half of "exposure" then reverse direction (just press opposite button) and hold for second part of exposure.

If you want to do 5 minute exposure without trailing - do above for 2.5 minutes in one direction and 2.5 minutes in other direction.

You can start by doing shorter runs - like 30seconds or one minute if your polar alignment is poor at the beginning, but as soon as you start improving it - switch to full duration of exposure.

Makes sense?

I think I understand, so the same procedure is done for altitude in regards to placing the star in the right hand side of the sensor, and going left and right?

[Samibotss]

50 minutes ago, david_taurus83 said:

I think you need to also move the scope. Another option is the handset polar align if your mount has that function.

I haven't seen anything of the sort, but then again, I don't exactly know what "handset polar align" refers to

28 minutes ago, jif001 said:

Check the settings in the handset, particularly the date. I think it uses the American format, where, for example, August 1st would be entered as 8-1-2020 (i.e. ‘August 1st’). If you entered it the other way around as 1-8-2020 (‘1st August’) the date will be 7 months adrift which might explain it going in the wrong direction. 

I'm pretty sure I entered the date correctly, I did spot the fact it was using the american format.

[david_taurus83]

1 hour ago, Samibotss said:

I haven't seen anything of the sort, but then again, I don't exactly know what "handset polar align" refers to

Do you have a Synscan handset?

[Samibotss]

2 hours ago, david_taurus83 said:

Do you have a Synscan handset?

Yes

[vlaiv]

4 hours ago, Samibotss said:

I think I understand, so the same procedure is done for altitude in regards to placing the star in the right hand side of the sensor, and going left and right?

That depends how your sensor is oriented. In both cases - south near meridian and west or east at DEC zero - you'll use RA motion and look for DEC drift.

If you orient sensor so that RA direction is in line with horizontal axis of sensor, then in both cases motion of the star will be left/right and drift will be up/down. In any other case - only one thing is sure - RA and DEC motion should be perpendicular to each other (if everything is ok and mount is orthogonal - and it should be).

Simplest way to orient sensor so that RA is horizontal is again to find bright star, start exposure and slew in RA - if line on exposure is horizontal - you are done, if not - rotate camera (for the same angle line is to horizontal on the image - but not necessarily in same direction - depends on how telescope inverts image) and repeat until line is horizontal.

[Samibotss]

Just now, vlaiv said:

That depends how your sensor is oriented. In both cases - south near meridian and west or east at DEC zero - you'll use RA motion and look for DEC drift.

If you orient sensor so that RA direction is in line with horizontal axis of sensor, then in both cases motion of the star will be left/right and drift will be up/down. In any other case - only one thing is sure - RA and DEC motion should be perpendicular to each other (if everything is ok and mount is orthogonal - and it should be).

Simplest way to orient sensor so that RA is horizontal is again to find bright star, start exposure and slew in RA - if line on exposure is horizontal - you are done, if not - rotate camera (for the same angle line is to horizontal on the image - but not necessarily in same direction - depends on how telescope inverts image) and repeat until line is horizontal.

Alright, I understand now. I'll definitely try this method once the skies clear up again. Thank you very much for all the insightful information

[david_taurus83]

55 minutes ago, Samibotss said:

Yes

In that case, the handset should have a polar align routine. I think its in utilities. You setup the mount and do a star alignment as normal. Then select the Polar Align function in the handset. The mount slews to a star of your choice. It prompts you to centre the star in an eyepiece with the controller. Once centred, it moves the star off centre and then prompts you to recentre by using the altitude and azimth knobs on the mount. Follow the handset prompts and it doesnt take long at all.

[Samibotss]

On 16/08/2020 at 18:21, david_taurus83 said:

In that case, the handset should have a polar align routine. I think its in utilities. You setup the mount and do a star alignment as normal. Then select the Polar Align function in the handset. The mount slews to a star of your choice. It prompts you to centre the star in an eyepiece with the controller. Once centred, it moves the star off centre and then prompts you to recentre by using the altitude and azimth knobs on the mount. Follow the handset prompts and it doesnt take long at all.

I did not know about that. Is it accurate enough for AP?

[david_taurus83]

48 minutes ago, Samibotss said:

I did not know about that. Is it accurate enough for AP?

Depends on focal length and exposure time. Even with perfect polar alignment, which is very hard to achieve btw, you will get some drift from periodic error and imperfections in the mounts running gear.  The DARV method may be better but longer to set up. If it was me, I would select a spot in the garden and aim to setup in the exact same place each session. For example, I have drilled 3 small dents in the slabs in my back garden where the tripod sits. Each time I setup its just a case of setting the mount and tripod back out in the exact same spot by placing the tripod 'spikes' on the 3 dents. This gets me very close each time with little effort. Only small tweaks needed after though I can see Polaris.

If you could setup in a fixed spot, you could try the handset method to get you close and give imaging a go. If you still get star trails you could fine tune with the DARV method. But if you could mark the position where you were set up it would make future sessions much easier as you'd be very closely aligned to begin with.

[Samibotss]

20 minutes ago, david_taurus83 said:

Depends on focal length and exposure time. Even with perfect polar alignment, which is very hard to achieve btw, you will get some drift from periodic error and imperfections in the mounts running gear.  The DARV method may be better but longer to set up. If it was me, I would select a spot in the garden and aim to setup in the exact same place each session. For example, I have drilled 3 small dents in the slabs in my back garden where the tripod sits. Each time I setup its just a case of setting the mount and tripod back out in the exact same spot by placing the tripod 'spikes' on the 3 dents. This gets me very close each time with little effort. Only small tweaks needed after though I can see Polaris.

If you could setup in a fixed spot, you could try the handset method to get you close and give imaging a go. If you still get star trails you could fine tune with the DARV method. But if you could mark the position where you were set up it would make future sessions much easier as you'd be very closely aligned to begin with.

I see, I was planning on doing the same once I got it properly aligned. Thanks!

[Samibotss]

On 16/08/2020 at 17:21, vlaiv said:

That depends how your sensor is oriented. In both cases - south near meridian and west or east at DEC zero - you'll use RA motion and look for DEC drift.

If you orient sensor so that RA direction is in line with horizontal axis of sensor, then in both cases motion of the star will be left/right and drift will be up/down. In any other case - only one thing is sure - RA and DEC motion should be perpendicular to each other (if everything is ok and mount is orthogonal - and it should be).

Simplest way to orient sensor so that RA is horizontal is again to find bright star, start exposure and slew in RA - if line on exposure is horizontal - you are done, if not - rotate camera (for the same angle line is to horizontal on the image - but not necessarily in same direction - depends on how telescope inverts image) and repeat until line is horizontal.

 

On 15/08/2020 at 22:33, jif001 said:

It’s normal for the mount to be inaccurate when it slews to the first target. You should always follow polar alignment with 2- or 3-star alignment. The latter assists the Goto system in finding the targets accurately. Pick a bright star some distance from the pole star and use the Goto to slew there. It will be off, but you use the handset buttons and finder scope to centre the star in the field of view, then tell the handset it’s there. Repeat with a second star somewhere else then, ideally, a third one. That should mean the Goto will be accurate finding your next target. 
 

if you have a computerised setup using ASCOM, you can platesolve instead of centring the star each time, and sync the location to the mount.  

 

4 hours ago, david_taurus83 said:

Depends on focal length and exposure time. Even with perfect polar alignment, which is very hard to achieve btw, you will get some drift from periodic error and imperfections in the mounts running gear.  The DARV method may be better but longer to set up. If it was me, I would select a spot in the garden and aim to setup in the exact same place each session. For example, I have drilled 3 small dents in the slabs in my back garden where the tripod sits. Each time I setup its just a case of setting the mount and tripod back out in the exact same spot by placing the tripod 'spikes' on the 3 dents. This gets me very close each time with little effort. Only small tweaks needed after though I can see Polaris.

If you could setup in a fixed spot, you could try the handset method to get you close and give imaging a go. If you still get star trails you could fine tune with the DARV method. But if you could mark the position where you were set up it would make future sessions much easier as you'd be very closely aligned to begin with.

I went out tonight to try out the DARV method, and this is the result I got :

Azimuth adjustements :

Altitude adjustment :

Result after 10' exposure :

 

As it turns out, I can push to 8-10' exposure before seeing trails, which is better than before (4' max), but still not good enough. The problem I have is that the DARV method demands a 125 second exposure, but I can only push to 30' on my Canon 4000D.  I don't know if that's good enough to correct the angle. I do have to say I moved along the DEC axis on the two adjustments, which may be wrong since I think I need to move on the RA for altitude.

As for the GoTo alignement, I think I have the good observation site settings, as on initial slew for 2-star alignement to the first star (Cebalrai) it was pretty close. However, for the second star, the handset kept showing me stars in the north western sky, which I cannot really see from my vantage point. I just did 1-star alignement with cebalrai, but despite an "alignement successful" message, it was inaccurate slewing to other targets, so I don't really know what to do here.