Topics

Help with PointXP


Greg Wandless
 

I did my first PXP try last night, but was kinda disappointed with the results. I expected the pointing accuracy to improve as i loaded more calstars but that was not the case. Model sensitivity was around 20 when I was done. Polar alignment error was about 55 min in altitude and a couple of min in azimuth but the GOTO accuracy was pretty poor. It brought the next star into finder but not close enough to 12.5 mm eyepiece. Does polar alignment need to be better before I can use PXP?

Any suggestions as what to try next?

This is a homemade fork mount.

Is the configuration saved automatically, because I can't find it this morning?

Thanks,
Greg


Don W
 

Hi Greg,
Your polar alignment is what is causing most of your problem.  Altitude is off nearly 1 degree for PA.  That is probably more than your FOV.  The PXP model will try to correct for PA, but it is not perfect.

A full PXP model should help a lot with pointing, but that takes 16 or more cal points.  Doing a PXP model visually is not nearly as accurate as using a camera.  However once you make a PXP model, you have an option to load the model automatically when SiTech starts in the future.

SiTech can load the file "autoload.pxp", but you have to save your file right after you finish making it.  It is also recommended that you save the PXP model with the number of stars and the date in the filename.  That way you can always go back to it or compare it to later models.

The Tall SiTech window shows the number of stars and accuracy of the PXP model just above the middle of the window on the Scope Tab.  You can right click on it to turn it off.  Left click to see the model or load a new one from a file.

Open the ChangeConfig and MountParms tab and click on "Load Autoload on start".

Don W


Greg Wandless
 

Thanks for help Don.

I assume that if I change the PA during a PXP i would have start over?

Greg


Greg Wandless
 

I assume also that the smaller the RMS and Peak errors the better, but what should i expect for a good model?
Greg


Greg Wandless
 

I also noticed that the "Use Absolute encoders" was checked, is that a problem?


Don W
 

Hi Greg,
If you do not have absolute encoders then for sure do not click them on.

You have to do all the CalStars in one session.  You can not do an OffsetInit between CalStars.  If your mount tracking is not perfect, then the time between CalStars is a factor driving bigger errors.

Really good mounts can have RMS and Peak errors at a couple arc-min for polar alignment.  Any mount can do that, but it takes a lot of care and time to get that accuracy.  Obviously the better the PA, the better pointing.  My own SiTech mount currently has RMS = 31 arc-sec and Peak = 1:14 arc-min.  My last PA was several years ago for my permanent observatory.

The PXP model can correct for many errors, errors in the mount orthogonality and some deflections.  It can not correct for mirror flop or focuser sag.

Using a camera to platesolve the PXP CalPoints gives accuracies to the arc-sec.  Doing manual/visual CalPoints is accurate to 10's of arc-sec, plus visual PXP model takes a longer time and time is also a factor in accuracy.

Don W


Greg Wandless
 

So it kinda sounds like the PXP is meant for permanent mounts, and since my scope is moved into position for each use  I should be using offset inits? Or is there another way to improve pointing accuracy?
 
Thanks,
Greg


Don W
 

Hi Greg,

Setting up every night makes it hard to use a PXP model, especially if you make the model visually.  Polar Alignment is the key to accurate pointing and tracking.  There are many ways to do PA, using the drift method is classic, but time consuming.  Using a software tool like PemPro or SharpCap require cameras.

Here is a method developed by AstroPhysics  https://astro-physics.info/tech_support/mounts/drift-alignment-ra-correction-method.pdf
This uses a camera, however it will work with a cross-hair eyepiece.  It is a form of drift method.

Here is another way to polar align in daytime:
https://astro-physics.info/tech_support/mounts/daytime-polar-alignment.pdf
This references Park Positions, so look at https://astro-physics.info/tech_support/mounts/park-positions-defined.pdf
to see what these positions are.

Don W


Greg Wandless
 

So your saying a good PA with Offset INits should give me reasonable GOTO pointing accuracy? What about the two and three star alignments that a lot of mounts do these days. I have an Orion EQ mount that does a two or three star alignment and has good accuracy when it comes to GOTO. Is there anything like that in ST.

I have been using the PA camera OrionP1 to do my alignments. My mount has registration pin so I can put in the same place each time, I figure that keep things better aligned. I didn't done a PA last night so it appears my alignment pins aren't as accurate as I would hope.


Don W
 

Hi Greg,
What kind of mount do you have?  I was assuming it is a GEM.  An equatorial and a GEM aligned to the pole only needs one additional star (to set the RA hour) in order to initialize.  Multiple star alignments available elsewhere are for non-equatorial mounts where there is no polar alignment of the mount axis.

The Orion P1 camera claims polar alignment to 30 arc-sec precision alignment.  So why are you getting 1 degree errors?

Don W


Greg Wandless
 

I have a home made fork mount.

I didn't use the P1 that night, the alignment was from a couple of weeks ago. My mount has alignment pins so I ca put it the same place each time. I had assumed it it would be better than it appears to be.

My other mount, the Sirius EQ does a one, two or three star alignment routine, after a PA and gets very good GOTO results.

I was just wondering if ST has something similar or does  the "Offset Inits" do that?

I have had fairly good GOTO accuracy using an Offset Init on a nearby star then slewing to the object of intrest.


Don W
 

Hi Greg,
Since you have been doing your PXP model visually, I guess you do not do astrophotography.  SiTech does have multi-star alignment for non-equatorial mounts but not for equatorial mounts.  For EQ mounts SiTech relies on single offsetinits and the mount's polar alignment.

An OffsetInit near your target will always get you on to your target.  Since you are using a home made mount, a good PXP model will definitely help overall pointing and tracking.  But the quality of your polar alignment each night affects the quality of your pointing.  You should do a good PA then on next nights check your PA with the Orion P1 camera to see how repeatable your setup really is.

Don W


Greg Wandless
 

I had hoped to do some astrophotography once I got everything ironed out. May have to be short exposure but I also a guide scope that I need to get working and try PHD or something like that.

You mentioned plate solving to setup the PXP. Is the procedure documented somewhere?

How many cal stars are needed to give me the PA alignment on the PXP screen?

Thanks for all your time and help!!
Greg


Don W
 

Hi Greg,

Doing platesolves and creating a PXP model are easy once you set up the parameters in ChangeConfig (the camera software designation and optics and FOV info).  The Help File in SiTech on the Features tab/RunScript tells you what you need to know to latesolve and run a PXP model automatically using a camera.

You will need to have a camera control software listed in the ChangeConfig/CameraStuff.

The PXP model senses and corrects for up to 16 factors.  To do this you need 16 or more CalStars.  More is better up to about 36 or so Calstars.  More than 36 gets into "diminishing returns" where more shots don't add any goodness.

The PXP model will determine and tell you about how far off PA is when you create the model, but it won't tell you PA data later when you run the model again.  Using the PXP model to check PA is not very efficient.  Far better is to use your Orion P1 camera to get close PA, then your PXP model will be fairly accurate on subsequent nights.

Don W


jmgoldba
 

On Thu, Feb 25, 2021 at 03:09 PM, Greg Wandless wrote:
astrophotography once I got everything ironed out. May have to be short exposure
You will need the ability to take images to iron things out. PXP is a last step for fine tuning mount performance. For PXP to do its thing your mount should be precisely aligned and balanced as well as being mechanically qualified. Mechanical qualification means pointing errors are limited in magnitude and, critically, are repeatable (in direction). Same idea with tracking errors. If the errors are too large and/or aren't repeatable (garbage in) then your PXP model won't work (garbage out).

You'll need to take images to mechanically qualify your mount. In fact if you're limited to 'short exposure' images due to poor tracking, there are likely more fundamental issues with the mechanics of your mount that should be addressed before building any PXP model. -Jesse