Synscan: choosing stars for 3-star alignment

[barbulo]

Hi all! 
I am opening this post because I am not able to perform an accurate polar alignment. 
I balance the tripod, then I use the polar scope to perform an initial alignment and finally perform the 3-star alignment with the Synscan. 
Due to horizon limitations, I decided to chose Capella, Hamal and Procyon, following the manual advice: two stars on the same side of the  meridian (RA dev >3h and DEC diff 10-30º) and one on the other side. However, the system shows a very poor accuracy after the process.
If I chose 2-star alignment the error is <1º, but still high. 
Trying to fix it, I follow the Synscan Polar Alignment procedure, but then, the Polaris position in the polar scope goes far away from its supposed place. 
After a new 2/3-star alignment process, there is still a big error in the mount alignment. 
The polar scope is fairly calibrated. 
What can I do?

Thanks in advance!

[dannybgoode]

First thing to check is that you have your lat/lon and time all correctly input into the controller.  Use the SW app to find the format in which it likes lat/lon and remember that the date time format is US so MMDDYYYY.  Incorrect inputs are the most common cause of alignment errors/failures.

[Owmuchonomy]

The three star routine is best avoided for Synscan Polar Alignment.  Its primary function is to the correct cone error of your optical tube.  To perform an accurate polar alignment of your mount, first ensure the OTA is in the home position and the EQ axis roughly pointing to Polaris.  Ignore the polarscope.  Next perform a two star alignment using two stars on the same side of the meridian that you will be imaging your target. Then run the polar alignment routine on the Synscan menu.  After 2 or 3 iterations of the Synscan routine using a high power eyepiece the polar alignment should be very accurate.  At no point do you need your polarscope.

[barbulo]

43 minutes ago, dannybgoode said:

First thing to check is that you have your lat/lon and time all correctly input into the controller.  Use the SW app to find the format in which it likes lat/lon and remember that the date time format is US so MMDDYYYY.  Incorrect inputs are the most common cause of alignment errors/failures.

Thank you @dannybgoode. All data correct. Polaris position double checked with other sources. 
 

33 minutes ago, Owmuchonomy said:

The three star routine is best avoided for Synscan Polar Alignment.  Its primary function is to the correct cone error of your optical tube.  To perform an accurate polar alignment of your mount, first ensure the OTA is in the home position and the EQ axis roughly pointing to Polaris.  Ignore the polarscope.  Next perform a two star alignment using two stars on the same side of the meridian that you will be imaging your target. Then run the polar alignment routine on the Synscan menu.  After 2 or 3 iterations of the Synscan routine using a high power eyepiece the polar alignment should be very accurate.  At no point do you need your polarscope.

Thank you @Owmuchonomy. Next time I’ll perform 2-star alignment and polar alignment iterations.
Any recommendation when choosing the star for the polar alignment? Synscan offers Sirius firstly and Capella secondly. 
 

[dannybgoode]

15 minutes ago, barbulo said:

Polaris position double checked with other sources

I am a little unsure about what the issue you are describing is in that case.  Star alignment will not improve polar alignment and polar alignment is not that important for the goto funtion to work properly once you have run the 2 or 3 star alignment process.

What is the problem you have and are trying to resolve?  Is it your polar alignment not being accurate or is it that once you have set everything up your scope is not slewing to targets propoerly?

[barbulo]

After polar and star alignment, the accuracy of the scope pointing at any object is very poor. I tried to improve it re-aligning. 
Since I’m not sure what the problem might be, I tried to polar align more accurately. But as I mentioned, the supposed position of Polaris in the finder scope is completely away and the mount doesn’t slew to targets accurately. 
An additional problem is that I have a very constrained field of view from my yard. 
Perhaps I shouldn’t focus too much in polar alignment. I don’t know. 
I could try a different location too.

[Jiggy 67]

5 minutes ago, barbulo said:

the supposed position of Polaris in the finder scope is completely away

Do you mean the finderscope or the polar scope??? Polaris won’t be central in the finderscope regardless of pa because the finderscope isn’t pointing at Polaris. 
If you can see Polaris you should be able to accurately polar align using an app or the method @Owmuchonomy described .

 I have to disagree with @dannybgoode I think the more accurate your pa the more accurate your pointing accuracy. Over the years I have tried GoTo with and without accurate pa and pointing accuracy is always more accurate with accurate pa in my experience. 
Concentrate on your pa and you may find pointing accuracy will improve 

[dannybgoode]

3 minutes ago, Jiggy 67 said:

 I have to disagree with @dannybgoode I think the more accurate your pa the more accurate your pointing accuracy.

Absolutely I agree to an extent.  I am more interested in what the OP is trying to achieve.  Re-reading the post it seems they are doing 3 star align to improve PA, which isn't going to happen.  For visual as well so long as PA is there or thereabouts goto is at least OK.  But yes, better PA does equal a bit better goto but the errors can be dialled out over a session of centring targets that are a bit out anyway

 

[teoria_del_big_bang]

Forgive me if this is me being naive but if you polar align accurately, as you think you are doing , then goto a well known bright star about mid point up from the horizon (one on Orion's belt for instance pretty visible most of night at present) can you get an idea how far you are out and whether it is a long way out in both Ra or Dec or just Ra? 

If Dec is somewhere close but out in Ra then time being wrong is the most likely, dec then sounds like location (I think anyway).
One reason for asking is I had exactly the same recently, because I am imaging and I plate solve was not really an issue but I was just curious to why it was so far out and when I did this DEC looked pretty close (just looking along the axis of the scope as I do not have a finder)  but Ra was out and roughly about an hour in time. Now I am not sure why but in my I-Optron I had to use the handset and enable daylight saving time, which makes sense but I would have thought as we are currently on GMT it would not make a difference until BST and clocks go forward, but it did make a difference and after this the image was in view and needed very little adjustment when plate solving.
 

Sorry if this is total nonsense but seemed to get me on target.

Steve
 

Edited March 9, 2021 by teoria_del_big_bang

[barbulo]

Sorry for the misunderstanding. I meant polarscope. I understand that the star alignment will not improve the PA. I pretended to improve the PA to fix the poor accuracy after the star alignment. But no way. 
My workflow:
Once the tripod is balanced I use the knobs to move Polaris to its place in the reticle, indicated both by an APP and the Synscan handset after the initialization process. Not in the center of the reticle. 
Then I perform the 2 or 3 star alignment. With the mount in the same position, the error displayed after the alignment is much higher when I chose the 3-star procedure (several degrees), while it’s “only” 20’-40’ when choosing 2-star.
I use a computer and an ASI224 camera to exactly center the stars in the scope during the star alignment. 
I am not sure if the deviation is in RA and/or DEC but it’s quite high. I’ll check it the next time. With this lack of accuracy I can manually slew the scope to the selected objet if it is big enough (M42, M45, Mars). But it’s impossible with fainter objects: M36, M37...
What I’ve detected is a deviation in the DEC axis while imaging DSOs. It’s an evident sign of poor PA, right? 
I’ll try different solutions during the next session. 
Thanks to all for your replays.  
 

[Owmuchonomy]

I hope you can solve the problem.  Try and separate the two main functions of GoTo accuracy and PA accuracy.  They are not necessarily completely interdependent.  For observing it is a high priority for accurate GoTo but PA accuracy is less critical.  For imaging it is very important to achieve very accurate PA but your mount must also be in tip top condition, without backlash in the axes for example.  For accurate GoTo I would suggest you use the 3 star alignment as in your original set up.  This process adjusts for cone error in your setup.  Make sure when aligning stars in your eyepiece that you finish the centreing with an UP and RIGHT touch on the handset.  I would suggest not using a camera for the GoTo setup.  Mounted cameras do not necessarily mount in a concentric fashion.  Use a high power reticule eyepiece if you have one.

[dannybgoode]

Next question then. Are you looking to use goto for imaging or visual work. If imaging then forget about the handset and star alignment etc. Just get an EQDir cable and platesolve. It'll always be many times more accurate and much quicker. 

[teoria_del_big_bang]

22 minutes ago, dannybgoode said:

Next question then. Are you looking to use goto for imaging or visual work. If imaging then forget about the handset and star alignment etc. Just get an EQDir cable and platesolve. It'll always be many times more accurate and much quicker. 

+1
You still need good PA but plate solving for imaging is really a game changer, bang on target every time and you can even get your camera bang on same framing as a previous night by using an image from the previous night in the plate solving.

Steve

Edited March 9, 2021 by teoria_del_big_bang

[dannybgoode]

I was a bit intimidated by the concept of plate solving at first but having use it first in APT and latterly in KStars/Ekos I couldn't go back. 

Particularly in my heavily light polluted skies I'd never be target for anything. Even when I had goto working reasonably well it couldn't be relied on to be accurate enough for imaging. 

The nice thing about EKOS as well is over a session it learns how much error there is in the goto so if you quickly go to 3 or 4 objects using plate solving after that it is pretty much bang on every time. 

This has the advantage you can set up using the camera, slew to a few targets and then go visual and it'll be spot on

[barbulo]

OMG! Plate solving=new concept. 
Sounds good though. I’ve read the basics. 
Do I need any extra equipment?

[dannybgoode]

26 minutes ago, barbulo said:

OMG! Plate solving=new concept. 
Sounds good though. I’ve read the basics. 
Do I need any extra equipment?

Platesolving is magic. You have a camera and I assume some form of imaging control software so you just need an EQDir cable to attach your computer to your mount. 

[barbulo]

I’ve got it. 
I'll try out tomorrow. 
Thanks!

[teoria_del_big_bang]

8 hours ago, dannybgoode said:

Platesolving is magic. You have a camera and I assume some form of imaging control software so you just need an EQDir cable to attach your computer to your mount. 

And clear skies ha ha

[barbulo]

8 hours ago, teoria_del_big_bang said:

And clear skies ha ha

You're right: weather forecast says cloudy for the next 10 days. ☁️ 😩

One additional question for you experts: I've read about polar drift alignment with PHD2. Does it worth a try?

Edited March 10, 2021 by barbulo

[dannybgoode]

9 hours ago, barbulo said:

You're right: weather forecast says cloudy for the next 10 days. ☁️ 😩

One additional question for you experts: I've read about polar drift alignment with PHD2. Does it worth a try?

I am still a little unsure about what you are trying to achieve.  Are you trying to achieve more accurate goto and if so is this for imaging or visual work or; is your goto acceptable and you are trying to improve your PA and if so to what end?  Or, are you trying to improve your PA in the hope it will make your goto more accurate and if so it is back to my first question about whether you are looking to image or look  ?

We need to understand what is not working for you to know what needs attention.  You do not *have* to have good PA even for imaging if you are guiding.  I often just align visually in the polar scope and guide with subs of 3 minutes or more without issue.  If everything is working acceptable, it's just that Polaris doesn't quite line up in your polar scope as you think it should then I'd be inclined just to leave it alone.

And no, at this stage, until we understand the precise issue you are having, don't even go near PHD2's PA routine  

Edited March 10, 2021 by dannybgoode

[barbulo]

Thank you @dannybgoode.

I'm using a f/5 150/750 scope along with an ASI224 = quite narrow FOF. I know this camera is meant for planetary, but since we're out of season I'm experimenting with DSO imaging. I cannot use a guiding solution because the mount (EQ3 GOTOrized) would be overweight (if not yet).

Problems finding objects: poor accuracy. Faint ones impossible to locate.

Image capture: 90% of the exposures over 20sec are invalid.

My intention is to minimize those errors with the means available: current gear, experts advice, patience and time. On top of that, I consider myself very meticulous and perfectionist (on the verge of OCD 😂), so all this annoys me a bit. I assume this frustration is part of the hobby though.

[dannybgoode]

@barbulo - in that case plate solving is the answer.  No need for a guidescope etc and you can do it all with your current kit.  With a narrow FOV a visually aligned PA, even carefully tweaked, and then relying on the mount's goto routine is unlikely to be accurate all the time, certainly not accurate enough to properly frame DSO's.  Plate solving on the other had is accurate to a handful of arc seconds - I have mine set to resolve to an accuracy of 4" which it does every time without fail.  I could got even further if I really wanted but I don't see the need.  I may test it down to 2" though just for fun (the 4 and the 2 being arc seconds - 4" is 1/18 the distance between Polaris and its companion star).

My routine is as follows - visually align Polaris in the polar scope, focus, plate solve target, image.  As per my earlier posts you just need a cable to link your mount to the computer.  Also re: guiding - a small guidescope and camera do not weigh much and I doubt will tip you over the payload threshold.  With an EQ3 mount you're going to struggle to get long unguided subs anyway but at least using plate solving you will very quickly and easily be bang on target every time.

[barbulo]

Thank you very much @dannybgoode. 

It seems plate solving is going to be the solution. Investing in a small, light and cheap guiding solution, that would be even more expensive than my ASI224, is something I might consider in the mid term. Maybe along with a mount and scope upgrade and a DSO camera. Will $€€.

Meanwhile, I've read about plate solving but people mainly focus on mini-pc + linux. Any Windows based (offline if possible) software? You mentioned APT and KStars+EKOS. Would work for me?

Thanks again

[dannybgoode]

1 minute ago, barbulo said:

Thank you very much @dannybgoode. 

It seems plate solving is going to be the solution. Investing in a small, light and cheap guiding solution, that would be even more expensive than my ASI224, is something I might consider in the mid term. Maybe along with a mount and scope upgrade and a DSO camera. Will $€€.

Meanwhile, I've read about plate solving but people mainly focus on mini-pc + linux. Any Windows based (offline if possible) software? You mentioned APT and KStars+EKOS. Would work for me?

Thanks again

Windows - APT works well and there is a fully featured free version although the licensed version is very cheap.  I cannot remember exactly how to get plate solving working however there are some excellent YouTube videos on the subject.  You can get it working fully offline too.

KStars/EKOS is predominantly a Linux based solution and with Astroberry will run on a Raspberry Pi so great for portable work but it is also an excellent control package in its own right and is what I use.  Again it will run offline but having a data connection is useful.