Is there an explanation somewhere what these values mean and what are good or bad values? How can I f.e. calculate my current polar alignment error from these values? 

Thanks and regards,
Eckhard 

To calculate your total polar alignment error in arcminutes.   PA error = SQR((Polar Axis Elevation/60)^2 + (Polar Axis/60)^2 )  where SQR is square root and ^2 means squared.

+1. I have wondered what all these values mean as well. 

It would be really nice if the documentation stated what "good" and "bad" values were based on your mount and scope, and more importantly,  what, if anything should be done to improve the numbers.
A simple example: running a T-Point model tells you your polar alignment error and says things like "Turn the right knob 3 ticks" or "your error is good enough".  However, for the other numbers it's no better than APPM in that it also leaves you guessing regarding what, if anything should be done. 

Hi Eric,

It would be really nice if the documentation stated what "good" and "bad" values were based on your mount and
scope, and more importantly, what, if anything should be done to improve the numbers.
A simple example: running a T-Point model tells you your polar alignment error and says things like "Turn the
right knob 3 ticks" or "your error is good enough". However, for the other numbers it's no better than APPM in that
it also leaves you guessing regarding what, if anything should be done.

There aren't specific good or bad values. The model adjusts pointing and tracking to correct the polar alignment and other errors. What you should is more important is model repeatability, which can be improved by reducing random movements that your telescope and other equipment may introduce. There are so many different telescope and camera types out there that it would be difficult to provide more specific information than that.

BTW, adjusting polar alignment will not necessarily improve pointing or tracking. It will, however, reduce field rotation, but that mostly matters when alignment is far off. The amount each knob adjusts polar alignment can be found in APCC's help file:

https://www.apastrosoftware.com/help/apcc-pro/advanced_pointing_model.htm

-Ray

Ray,

I've used the polar alignment "knob" table while polar aligning, and it helped.  It would have been even nicer to have APPM tell the user which knob to turn and by how many ticks, rather than forcing the user to look up that information in the table.  That's what T-Point does.

Interesting that there aren't "specific good or bad values".  In that case, I'm curious why the data is even shown to the user, since the user isn't expected to do anything with the data (except the polar alignment data since those can have an impact on field rotation)?  And it seems like the data only brings up questions and confuses people.

Hi Eric,

I've used the polar alignment "knob" table while polar aligning, and it helped. It would have been even nicer to
have APPM tell the user which knob to turn and by how many ticks, rather than forcing the user to look up that
information in the table. That's what T-Point does.

TPoint was designed decades ago when there were not many polar alignment tools available. There are many quicker ways to do polar alignment, including using PEMPro, which is included with most A-P mounts.

Interesting that there aren't "specific good or bad values". In that case, I'm curious why the data is even shown to
the user, since the user isn't expected to do anything with the data (except the polar alignment data since those
can have an impact on field rotation)? And it seems like the data only brings up questions and confuses people.

Agreed, and it is why the pointing terms will be removed from the Pointing Model tab in the next point build. The initial reason for the terms was for checking consistency of pointing terms between runs of pointing models, but these values do not need to be presented to users.

-Ray

Some measures presented to the user would be useful:

• Of course the polar axis values to refine polar alignment
• Cone error if trying to reduce that.
• Flexure - for troubleshooting excess flexure

The model compensates for those items but reducing them, relying less on the model compensation seems better.

Hi Brent,

I have a few questions so that I may better understand your points.

* Of course the polar axis values to refine polar alignment

APCC already provides these values, so I assume you meant the conversion of the polar axis values to knob turns for the specific mount? Many other polar alignment alternatives provide faster polar alignment, so what advantage do you see in having APCC do this?

* Cone error if trying to reduce that.

Why is this important if modeling accounts for this? Do you just want a tool to measure this?

* Flexure - for troubleshooting excess flexure

I don't think there are enough samples to make a good determination of "excess flexure." There are many types of telescopes. The OTA's material, construction, diameter, and length all matter. Also affecting flexure are things like the OTA balance point, length of the OTA, the mounting plate strength, the weight of cameras, etc. There are many variables so trying to come up with a value for excess flexure for all of the variables would be difficult.

The model compensates for those items but reducing them, relying less on the model compensation seems better.

It may seem better, but these are not where you will see much gain. A more meaningful area to look into is the amount of randomness in the system. Any of the pointing terms can have a low average value but higher variance. APPM provides a way to look into this with the "5x Verify" option. APPM will run the same set of data points five times. Then, A-P staff (or I) can analyze the APPM logs for variances.

-Ray

Hi Ray,
None would be essential but maybe useful.  For polar alignment I should have said to verify alignment instead of refine it.  I have a permanent pier and build models (probably more often than I need to) a lot more than adjust polar alignment. Seeing those values in the model verify that polar alignment is still reasonable. The only times I need to adjust is when the frosts sets in and then spring time - it seems to nudge my pier a little. Or sometimes when I do some sort of maintenance around the pier/mount.  So I guess I'm saying not as a tool for polar alignment, but data available to verify polar alignment when doing a model anyway.
Yes, as a tool to measure cone error.  While the model corrects this, I have two OTAs and have adjusted my second OTA to reasonably match the other in terms of cone error. I can swap OTAs during a session and still have reasonably good pointing - of course I could always build a model for each OTA.
Thanks for the comments about flexure.  I believe my current setup has more flexure than it should, I want to try a few things to improve that and having a measure would be good.  Maybe I am misinterpreting, but could a model with a high RMS indicate flexure (probably not specifically)?  I recall you commenting on another post that an user's RMS appeared high for a refractor. My usual RMS with my refractor is at least as high or higher so I suspected inconsistent flexure.  Thanks.

Hi Brent,

None would be essential but maybe useful. For polar alignment I should have said to verify alignment instead of refine it.
I have a permanent pier and build models (probably more often than I need to) a lot more than adjust polar alignment.
Seeing those values in the model verify that polar alignment is still reasonable.

Yes, and this is currently available. APCC shows the mount’s polar alignment error (in arc-seconds) for both pier sides.

Yes, as a tool to measure cone error.

APCC displays Cone error (in arc-seconds) in the two "Non-Perpendicularity" pointing terms. To measure the cone error for both scopes, you need to create a separate model for each scope.

Maybe I am misinterpreting, but could a model with a high RMS indicate flexure (probably not specifically)?

RMS is just a measure of how well the pointing model fits the sky data points.By itself, it does not distinguish between unmodeled behavior (e.g., complex flexure) and randomness caused by moving components (eg. Mirror flop).

-Ray

Ray,

I'm late to this discussion and am in the process of finalizing a transfer from the roll off roof to my 12' dome. I've polar aligned using my PoleMaster but can't say I'm overly confident as I haven't used it before. So if I run a model will this verify the alignment accuracy? If so how many points should be used? Sorry if I'm behind but this project has had so many twists and turns that I'm a bit intimidated trying to get back online.

Thanks,

Steve

--

Hi Steve,

I'm late to this discussion and am in the process of finalizing a transfer from the roll off roof to my 12' dome. I've
polar aligned using my PoleMaster but can't say I'm overly confident as I haven't used it before. So if I run a model
will this verify the alignment accuracy? If so how many points should be used?

15 points per pier-side, 30 total, should be enough.

-Ray

Thanks Ray. I'm making a list of things I need to do to get this system working as needed again after the move to the dome.

-Steve

Thanks again Ray, I do get that data now and it's great.  My original reply was why I would like to continue to see it in future versions or maybe I misunderstood when you said you were considering removing the pointing terms in the future - retrieving them in a log file would be fine too.

Thanks again Ray, I do get that data now and it's great. My original reply was why I would like to continue to see it
in future versions or maybe I misunderstood when you said you were considering removing the pointing terms in
the future - retrieving them in a log file would be fine too.

Brent, the numbers will be removed from the Pointing Model tab, but will still be available if you open the Pointing Model window.

-Ray

On this subject, for an observatory mounted Mach2 how often must I generate a new model? I would expect to have to do this whenever I changed OTA's but is the model sensitive to temperature, humidity, or things like precession?

Rgrds-Ross

Hey Ray,

After some learning curves I am modeling as I type. I've run a successful short run and loaded the model when asked at the end. The east side numbers looked better than the west. Is there a reason for this? The longer run running now has again what I suspect are good numbers for the east side but the west is much larger. Is there something else I should be doing or is this an indicator of being off on polar alignment? And then of course, where do I see that in APCC on the pointing model?

Thanks,

Steve

Hi Ross,

On this subject, for an observatory mounted Mach2 how often must I
generate a new model? I would expect to have to do this whenever I
changed OTA's but is the model sensitive to temperature, humidity, or
things like precession?

For the greatest accuracy, you should be using a device (e.g. MgBoxV2) that captures temperature, pressure, and humidity. APCC will use this real-time data for calculating refraction both during an APPM run and later when using the model. That said, large changes in temperature can affect the flexure of equipment, which can affect the values of pointing terms. There can also be settling of the ground around an instrument that can cause slight changes in polar alignment.

When you start to see errors in tracking or increasing pointing errors, it may be time to create a new model. APPM has a "Verify" option, which you can use to check the model's accuracy. A verify run measures pointing errors with the current model running instead of acquiring data for a new model. You can use far fewer points than you might have in your model, so it can take less time than creating a full model.

-Ray