Introduction

So after a few failed random trials by some users with the PAA tool, I really want to get to the bottom of this. From what I gathered this morning after giving it more thought is it is <em>most likely</em> caused by image orientation/rotation.

Currently, Ekos sets the correction vector as following:

<code>
correctionVector.setP1(celestialPolePoint);
correctionVector.setP2(RACenterPoint);
</code>

The vector starts from the location of the celestial pole to the location of the RA center of rotation. The idea is by adjusting the Alt/Az knobs on your mount, you move the mount RA center of rotation to coincide with the celestial pole.



The above is using the Simulators with a polar error of 00:06:21

To make this easier, the user selects a star they can move to achieve this:



Now by adjusting the mount knobs, the star in the <strong>image</strong> is moved to the Right/Bottom. The mount RA axis therefore is moving toward the Left/Top. When solved, astrometry.net reports the field parity for the simulator as Negative. So I'd like to collect some data by asking our lovely forum users to run an experiment with the PPA tool. <strong>Please make sure to enable Verbose Logging to File and select the Alignment module.</strong>

Data Required

1. Mount Type
2. Telescope Type (or PoleMaster)
3. Hemisphere: North or South.
4. Is the camera aligned with the telescope? If not, what's the angle of the rotation with respect to the telescope?
5. Please record how the image SHIFTS when you adjust the azimuth/altitude knobs:
5.1. Right Azimuth knob: When locking the RIGHT Azimuth knob, how does the image shift in the camera?
5.2. Altitude knob: When lowering the mount (Decreasing Altitude), how does the image shift in the camera?
6. PAA successful or not? Was PPA able to close the gap or it made things worse?
7. Logs

As soon as I have a clear night within the next 5 days or so, I will attempt another PA.
I will follow what you requested in this thread.

Jasem,
The last few times I tried to do a Polar alignment using the software I found it had stopped controlling the mount, it just sat there waiting.
Before that started happening I had used it a few times with success. Are you asking us to test whether it is working or whether it is working better than it used to?
Paul

Tried it tonight and it worked. Orion Atlas Pro mount, WO Z61 imaging scope, ZWO1600MM Pro camera.

Here a lengthy video showing my clumsy attempts to do polar alignment using the Alignment tool.

www.dropbox.com/s/9y3jr2q8udkf4sb/PolarAlignment.mp4?dl=0

Anyway, the problem I had before did not appear, could have been an issue with my iOptron mount? I got very good alignment this time.

At least using the Kstars build from 12/20 on a Raspberry Pi I can't find anything wrong with it (except that it is very slow using the Pi).

PHOMER60. Make sure your mount is unparked. As far as I can tell, the routine is not aware of the park status.

Jasem,

I am happy to see that this issue is being addressed.
At this point, I cannot give you all of the information
you requested, but I can tell you what made the
PAA routine fail and succeed for me.

Mount Type/Driver: Astro-Physics Mach1GTO/CP3 v. V2 (therefore using driver indi_lx200ap_experimental)
Telescope Type: 75mm f/6.7 ED refractor
Hemisphere: Northern
Camera Type: ASI1600MM-C
Camera Alignment: For me, this was the critical thing. I am using an Innovations Foresight
On-Axis guider, so it is effectively a diagonal placed between the OTA
and the camera. This causes a mirror-reversed image to be produced
on the sensor.
Image Shift: Unfortunately, I cannot give you the directions, but I do remember that they
were opposite when the ONAG was used, when compared to when the ONAG
was NOT used.
Right Azimuth Image Shift Direction: Cannot recall
Altitude Image Shift Direction (altitude decreasing): Cannot recall
PAA Successful: Failed with the ONAG (diagonal) in place, i.e., the "correction" actually
made the polar alignment worse; Succeeded with the ONAG removed,
i.e., the "correction" improved the alignment and converged

This was discussed in the following thread, where well-meaning contributors
chimed in about orientation and parity, and I tried to make clear that that
was not what I was talking about:

www.indilib.org/forum/ekos/3441-question...nment-assistant.html

My current solution is to use a capability of the ZWO ASI driver,
and mirror-reverse the image on download. But this solution
is not a general one, for users of cameras where the driver
does not offer that option.

If I can possibly do so, I will re-try the Polar Align Assistant,
and try to get the information you requested, although it will
be at least a couple of weeks before that could happen.

Greg

I used the polar alignment tool again tonight, this time with a 280mm Celestron guide scope mounted on top of an RC8. It took two iterations and I was down to 30" deviation from the NCP.
Save for it taking 15 min vs. 2 min with the polar scope, the PA tool is clearly superior.

Greg, thank you for the great feedback! It does confirm my suspicion about image parity, but I would need a further confirmation. Turn on Alignment logging, and solve two images. One with ONAG and one without, then attach the log. Basically, I want to check for the "parity" flag in astrometry.net and see whether this could be used to reverse the correction vector.

I had a go last night and things have improved significantly. The mount is now moving. Some unusual things did happen.
1) It tried to do an initial slew and seemed to be heading off into dangerous territory, so I aborted.
2) I was well off the pole to start with and needed to iterate a few times. Each time it kept asking me to slew because it was not pointing near the pole - even though it was. If I answered 'yes' it seemed to be entering dangerous territory. If I answered 'no' it seemed fine. I ended up with very accurate alignment that was sufficient for 20 minute exposures.
This was with a QSI camera and a Losmandy Gemini 2 mount.
Paul

So, I've done the experiment, have after days and days for a clear night... yeah it's Christmas night too...

1. Mount Type
HEQ5 pro, Rowan belt mod

2. Telescope Type (or PoleMaster)
QHY5 (red one, old style, mono)
**Orion guider**
Aperture 50mm
Focal Length 162mm

3. Hemisphere: North or South.
North

4. Is the camera aligned with the telescope?
It is aligned to the mount (see the following thread

5. Please record how the image SHIFTS when you adjust the azimuth/altitude knobs:
5.1. Right Azimuth knob: When locking the RIGHT Azimuth knob, how does the image shift in the camera?
Left knob in (facing the mount), mount pushing right, star moves right on screen

5.2. Altitude knob: When lowering the mount (Decreasing Altitude), how does the image shift in the camera?
Altitude knob, back, pushes mount up: start goes down

6. PAA successful or not? Was PPA able to close the gap or it made things worse?
Just check out the screenshots, I've sent my log to Jasem by email as it's too big to attach here.

7. Logs
Yes


5.1/5.2 was a bit weird I must say, at some point the star moved diagonally up/down when moving the RA knobs.



Tests carried out in the same fashion, slew to the same star, rotate West 30° PAA


First PAA session (to check how far out): 45'06" out
I then proceeded with the knobs to align a star where the PAA instructed me to.


Second PAA session: 3'09"
Yeah, now we're talking, OK it took me a while to move the star to the right location, I wanted to nail it.
I was quite happy with this result but no satisfied, as I know I could do better so....


I tried a third time to verify my work and make it better, it is very cold at the moment, the HEQ5 altitude knobs are notorious to be a pain to move...
and I'm like, what the hell !!!!!!
Third PAA session: 58'08"

Now, on 1st and 2nd session, the mount did slew West for all pictures (start to finish), however on the third test, the first slew was West but then the two following slew were East. I have no idea why this is the case.


On top of that, a little bit later on, I tried a fourth attempt..... without touching anything on the mount what-so-ever... oh dear...




running the following versions:
root@heidenrod-obs:~# dpkg -l | egrep "indi-full|indi-eqmod|indi-qhy"
ii indi-eqmod 0.5~201812201223~ubuntu18.04.1 armhf INDI EQMod Driver.
ii indi-full 1.7.6~201812191814~ubuntu18.04.1 armhf Instrument-Neutral Device Interface library - Full INDI
ii indi-qhy 2.1~201812201223~ubuntu18.04.1 armhf INDI QHY Driver. This driver is compatible with any INDI client such as KStars or Xephem.
root@heidenrod-obs:~#

and Kstars OSX 3.0.0

I'm going to order a QHY Polemaster when I'm back home. There is actually an OSX application for it !!!
So, when I'm back here at my observatory (in about 2 or 3 months I think) I will install it and try their software to properly align the mount, once confirmed it's aligned I will use the PAA and check.
I only have 1 more night here, it's supposed to be clear but I will not be able to stay up too late as I have a 9hours drive back home on the following morning....

Here's some bed time reading... celestialwonders.com/articles/polaralign...lignmentAccuracy.pdf