![]()
|
Polar Alignment Live Video/TPoint
A Paramount out-of-the-box (first time users)
Background Information on telescope pointing
Background Information
Pointing any off-the-shelf amateur "goto" telescope at the 1 arcminute RMS or better is not a trivial task and always requires telescope modeling. This is especially true for the compromised mass produced "goto" mounts, and very important with any German Equatorial mount because of the German mount design. Even the precisely CNC machined and well built Paramount ME benefits from telescope modeling.
TPoint is the most advanced telescope modeling software, and is used by most of the world’s largest professional telescopes including the Gemini 8 meter twin scopes and several others including the Keck, GBT, Gemini, ALMA, AAT, ARC, WIYN, WHT, UKIRT, JBO, IRTF, NSST, ESO, CTIO, SOAR, MMT, Magellan, and LBT (to name a few).
There are at least a few larger professional telescopes currently pointing at the one to three arcsecond RMS level after employing TPoint. Why is such accurate pointing required? This level of pointing is required to put a star in an arc second wide slit for accurate photometry for example. TPoint's author Patrick Wallace of the UK has written an excellent document on telescope pointing here. For anyone interested in the details related to accurate telescope pointing use the well written page as a resource.
Getting Started
Out of The Box <--- Click here if you are just getting started with the Paramount ME
The "Drift Method” as described in the TPoint manual is a relatively quick and easy way to get close polar alignment if you are familiar with it; however, TPoint is the best method (preferred) for quantifying the polar alignment and finding out what direction the error lies. A good polar alignment will greatly improve any equatorially mounted "goto" telescope's pointing before TPoint is applying the necessary pointing corrections.
First things first
Before starting a TPoint mapping run always double check the date/time and latitude/longitude, effect of daylight and time zone settings in TheSky. Use Data | Site Information as shown below. The latitude and longitude are less critical than time and they only need to be approximate, as close as ~100 miles or so to your location (using a GPS is great, use one if you have access to one). The more accurate your computer time is set the better. If you have access to the Internet there are easy free software programs that will accurately set the PC clock for you. Personally, I use the shareware programmed named D4Time it works well, it is free and painless. Highly recommended.
Important
TheSky's Data | Site Information

Verify your location, time zone and effect of Daylight
Use the Telescope | Setup options in TheSky to verify that TPoint is enabled and also check the status of Confirming Synchs, slews and mapping as shown below. Having to confirm each map point and Sync will help minimize mapping errors.
Enable "Confirm syncs" and "Confirm mapping"

Make certain that TPoint modeling is enabled
Hint: The easiest way to see if TPoint is working properly is to DISABLE it here, slew to a star and note the pointing. If the pointing is worse using TPoint you have most likely mapped points incorrectly or added improper terms, those that do not apply. Clear any new terms and check the pointing again. If you still have problems either start over with a new model or throw out any "bad" points.
Ignoring Bad Mapped Data Points
You have two options here. The obvious errors (stragglers that stick out like a sore thumb) can be either "masked" or "marked as an error". See example below.
Mapping erro, most likely.

Mapped point not in the norm clearly an error
Left Mouse Click on the point that is in error. Note the number of the mapped point is shown on the upper right (Point #33 in this case). You are also taken directly to this point in the TPoint spread sheet.
Taken directly to the Point
click on (#33)

Mark the Point as an error.
Use the icon
to mark a point as an error. If you
are unsure you can also MASK the point.
Please refer to the TPoint documentation for details. You should not arbitrarily
mark points as as an error even though the RMS gets better. Throwing away valid
data is not a good idea. Just be very careful when mapping to be sure.
Double Check the Object you are Mapping
Before you map a star or the Image Link coordinates double check the Object Information dialog. Here is an example. If you inadvertently click on a mapping point it shows as a User-defined object. This is not the star you wish to map but rather the offset point that has been mapped already. This would be in error.
Wrong Object to Map

Do not Map User Defined Data
Hint: Use the Object Information dialog to double check the point you are about to map. And best to Confirm all mapped points too. This will help avoid mapping wrong points.
If you don't see the star by name or Image Link entry use the down arrow to the right of Object "V" and find the entry in the list. If it not listed here try clicking on the star again.
Insert a new TPoint Model into TheSky
If you do not have a TPoint model present in TheSky already you must use Edit | Insert Object from TheSky and create a TPoint model for mapping. See below,
Create a New TPoint Model in
TheSky

Display as Icon to minimize the
size
If you already have a TPoint model in TheSky you can delete the existing model by highlighting the Icon in TheSky and pressing the Delete key. You will be asked to confirm the deletion of the embedded object. When you delete the old model, start with a new model as show above.
If there are previously mapped points being displayed in TheSky these should be removed first to avoid confusion. Use Data | User Defined Data to REMOTE the other reference markers. These reference labels have no functionality at all. They are only a reference label showing the stars that have been mapped. They can be safely Removed at any time and that will not effect the model.
Remove References to Previously Mapped Points
Add User-Defined Data Dialog Box

"Remove All" if any mapping labels already exist
Field of View
With a 1/3 inch size live video camera and a C-11 at f/6.3 (63 inches of focal length) the field of view (FOV) is about 10 arc minutes by 6 arc minutes. This is why I always choose to focal reduce the optical system for first time mapping, especially when polar aligning, even if the scope is to be used without the reducer in the end. See image below illustrating increased FOV by focal reduction.
Focal
Reduced C-11 @ f/6.3 versus not reduced @ f/10

Small
Live Video Camera
The final mapping is always done at the focal length the telescope will be used for imaging and if at all possible done with the imaging CCD camera in place. Employing the auto-mapping VB script or Orchestrate for the final large mapping run is a must. See AutoMapping here for complete details.
If a star does not fall in the field of view of the live video detector after a slew, it needs to be found and then accurately centered before it can be mapped. Make sure you have at least a cross hair type finder or Telerad for the initial syncing and mapping for quickly finding stars and placing them in your prime focus field of view.
The first 6 stars slewed to are more often than not outside of a small field, especially with a poor polar alignment or poorly mounted OTA. That is true for the first 6 - 12 stars but after that the pointing will be greatly improved and is usually on the order of 3 to 5 arcminutes RMS. In fact, after the first three to six stars are accurately mapped TPoint starts to kick in and it improves the pointing dramatically, especially when using the Suggest New Terms as described in this document. It should be noted that other telescope modeling programs are very limited because they only use the 6 basic Equatorial terms. If that is the case you cannot fully rely on the information reported for polar alignment nor can you expect sub arcminute RMS pointing.
Star Search
The “star search” command will find
stars that do not fall in the field of view after slewing to them. Star Search searches
from where the telescope is currently pointed by moving the mount in a circling
pattern that gets larger and larger, based on your detectors FOV. If you wait long enough for star search to
acquire the star you will always find the missed star even if it is many
arcminutes to a degree or more outside your field.
If you are conveniently near or at the telescope it is often easiest to
use the finder scope or a Telerad to get the star to fall on the chip at first or at
least very near to make star search find it faster when first starting out. This is only needed
if your polar alignment is extremely far off, say on the order of a degree or
more.
To use the star search feature (Telescope |
Options | Star Search) first enter the Field Of View in arcminutes of
your live video detector or cross hair illuminated reticule eyepiece. Note that eyepieces larger than about 12mm
generally yield a FOV that is two to three times larger than the 1/3 inch to ½
inch size chips found in live video cameras.
You can use the Telescope | Setup | Simulator found in TheSky without connecting to a real telescope to access the dialog to learn how to use the Star Search command.
Very handy Star Search

Choose 8X to 64X Speed depending on your focal
length.
How to Determine Your FOV
When the polar alignment is off by a half a degree or more (common for newbies) the pointing is always poor on ANY " goto" type telescope. All equatorial mounted " goto" telescope mounts make the assumption that the mount is perfectly aligned to the pole among many other assumptions, i.e. orthogonality of the optical axis, perfectly collimated optics, no tube flexure, to name only a few.
In the
real world it is simply impossible to make a perfect telescope mount at any
price and this is why TPoint was created
in the first place. This is also why many of the world’s largest good
pointing professional telescopes use it as well. Simply put, TPoint
works.
To accurately determine your Field Of View (FOV) with the live video use the Telescope | Motion Controls in TheSky. By “Jogging” the telescope in RA and Dec NOT ALT/AZ default (and do not use Move but “jog”) in 1 arc minute increments you can determine easily exactly what your actual FOV is. This is critical for star search and will be used for accurate calibrated polar alignment using live video as describe later.
This is where a focal reducer will always tremendously. If you have one use it to increase the FOV to make initial mapping/polar aligning easier. Start by placing a star at the very top of the video screen on the edge of the CCD detector. Then count the number of 1-arcminute jogs it takes to move the star down to the bottom of the video screen, bottom edge of the detector. This tells you precisely what your FOV for the video camera is in arcminutes in this direction.
Use the same procedure for the other axis starting from the far left of the video screen or far right of the video screen and then jogging the star to the opposite side. Camera orientation is not critical however by aligning the video with north up the star will move up and down and left and right and this greatly facilitates centering objects precisely. Rotate the camera as needed to align North Up.
Using the Telescope | Motion Controls found in TheSky to determine FOV with video.
Use motion in RA, Dec

1 arcminute Jog Mode
Now use the software hand controller buttons to jog the scope. Aligning the live video is easy when you can see the monitor as you rotate the camera.
Jog scope 1 arcminute South

Check direction
first
How do I Enable the Cross Hair? in CCDSoft
Click Image | Show Cross Hair in CCDSoft.
Press the South button once to jog the scope South 1 arc minute. If the star moves off the screen use the opposite “North” button to bring it back into the field. When a German EQ mount flips over these buttons are reversed so north becomes south and vice a versa. Astronomers are smart, they will figure it out. See below showing a 1 arcminute jog.
1 arcminute jog South in RA

Remember
North and South Flip on the Paramount ME
When the star is located at the very bottom of the video monitor the number of times you pressed the jog button to get there is the scale in arc minutes for this axis. In this example, the Field of View is about 4 arc minutes in the vertical direction.
Repeat the same procedure for the other axis. When you know the height and width in arc minutes of your live video detector enter these values in the Star Search dialog. Knowing the scale of the screen also allows you to make the polar alignment adjustments much more accurately. NOTE: The same procedure works equally well for an ocular. Just look in the eyepiece and visually count the number of arc minutes it takes to move the star from edge to edge.
When you get to the point where the polar alignment reported by TPoint is 5 to 10 arcminutes off of the pole you can make the polar adjustments by watching exactly how far the star moves when turning the elevation and/or azimuth adjustment screws in real-time.
HINT: Use the above calibrated video when making the polar adjustments. Using this method you can achieve a near perfect polar alignment quickly since you can exactly how far the mount is moving. In fact, with a robust mount like the Paramount ME achieving 1 arc minute to 1.5 arc minutes too high in elevation (shooting for the refracted pole and latitude dependent) and better than 1 arc minute in azimuth can be achieved in an hour or less).
Note that wobbly mounts with crude or hard to make polar alignment adjustments will always take significantly longer to accurately polar align. In these cases getting to within 3 arc minutes of the pole should be sufficient for CCD imaging and about as good as you can get without a heck of lot more work or modifications to the mounting.. If you are using film and exposing for an hour or longer even auto-guided a near perfect polar alignment is crucial to minimize field rotation. TPoint will help.
Live Video Mapping
Create a cross hair indicator in the exact center of the video screen for a reference. Even if you just use a black magic marker to indicate the center of the video screen that is fine. Better yet use TPoint to print a transparency overlay with a cross hair in the center and arcminute reference circles for you.
TPoint's
File-Print Video Target

Use
transparency as a reference
Sample mapping run
Sample mapping run in
Lyra

Stars in close proximity first to
minimize pointing issues
Based on my experience there is NO need to jump around all over the sky to get the first mapping points! It makes no difference. Also avoid flipping the meridian!
Accurately center the first star,
Then
Sync
it

Sync button is highlighted
above
Once the first star has been centered accurately and "Synced" on you can also MAP it as your first point. Press the MAP button after the initial one time "Sync". See Map button below
Image Link And Sync
A better method for the initial
Synch is to use Image Link and
Sync. Then start maping from there. Image Link and Sync is simple
to do, it is faster and it does not require any re-centering of the star and it works even if
there aren't bright stars in the field (i.e. the star was missed). Image Link is very smart. And by knowing your
scale you can search very large fields for a match. Even larger than your
actual field of view. With minimal effort you can Home the Paramount ME, slew to
the first star near the home position and simply have Image Link find the center
of the image.
Hint: You may need to "SUSPEND" the telescope link when finding matches way outside of the field of view. If not the display will jump back to the cross hair position which at this point is not accurate
So after the telescope has been initially synced you do not sync again. Note that you cannot go back and add more points to a TPoint model on a different night. See the TPoint manual for details on using a "Short Mapping Run" to fix pointing instead of trying to add points on successive nights which only pollutes the TPoint model.
Sync only once then use
map

Use the map button
highlighted
To make the Object Information dialog smaller press the Down arrow (Minimize button) found on the lower right. When minimized you can use the Green Slew To button and then the Red Map button for quick easy slewing and mapping. This is also less prone to inadvertently Syncing the telescope again because the telescope tab is hidden.
Minimize Object Information
Dialog
This takes less room and it
is less prone to making syncing errors
Do Not Map this Star

Assume the worst
If you get the error message warning (see above) indicating that the
star is not being correctly mapped and it appears to be in error, DO NOT
map the point by clicking “No”.
Slew to the same star again and make certain that you are mapping the
proper star first. This is
important.
If you continually see the mapping error when mapping the CORRECTLY verified stars (that is you are certain they are the correct ones) you have a major problem with the alignment elsewhere, i.e. very poorly mounted OTA or mount is off extremely far off the pole. In this case attempts should be made to refine the alignment of the mountings first, i.e. the Drift Method to refine polar alignment, before trying to map again. You cannot start shimming the mounting or the OTA until TPoint tells you exactly how much anyway. That is the catch 22. Keep in mind that only 1 to 2 thousandths of an inch alignment will introduce pointing errors on the several arcsecond to arcminute level. Again, TPoint to the rescue
If you inadvertently click on something other than the star you intend to map (cursor position, galaxy (did you turn off galaxies, nebulae, clusters, etc. or the wrong star, or other object) the objects name you clicked on or “Cursor Position” is displayed in the Object Information dialog. PAY CLOSE ATTENTION when mapping points!!!
For first time users it is safest to Confirm all stars being mapped. That is the default setting.
Confirm Mapping

Confirm Syncs and Slews
When starting out with a poor initial polar alignment most stars will fall outside a small FOV like live video when slewed to. This is even more often the case when the angular distance from the first object to next is more than a 1 degree or more.
That is, before TPoint's
corrections have "kicked in",
which is after the first six data points are mapped. When you have mapped the first three
stars TPoint
becomes active and with another three points mapped TPoint can start to
fix the 6 basic geometric pointing problems which yields a noticeable
improvement in overall pointing.
Bring up the Fit Data dialog in TPoint - Model | Fit Data.
Fit Data NO TERMS being used

German Equatorial mount
with
no terms added - 921 arcsecond RMS pointing
Versus RMS pointing with only the 6 basic terms.
The 6 basic equatorial terms added in
TPoint
The six fundamental terms take
the RMS pointing from 921 to 76.9 arcseconds
Suggest New Terms
| Using the proper terms in TPoint is key to determine the magnitude of polar alignment error. Without the additional terms, you simply cannot evaluate the alignment error to any usable degree of accuracy, and the overall pointing accuracy will be compromised.
|
After accurately mapping a half a dozen or more points, next use the button "More Terms" button on the "Fit Data" dialog to access the "Suggest New Terms" wizard.
Access the Suggest New Terms

By Clicking the More Terms button
In order to achieve good pointing as well as accurate polar alignment information reported by TPoint you must "fix" as many pointing errors with the system as possible. If not, both the pointing and polar alignment data will be compromised. The "Suggest Terms" button found on the "More Terms" dialog is a powerful and easy to use tool.
By adding additional "proper" terms (usually 6 to 12) you will always see a significant improvement in RMS pointing and then the TPoint report for polar alignment will also be accurate and reliable. But not before.
TPoint Reporting Polar Alignment information
before
adding new suggested terms

Note the amount of adjustment reported
versus
Accurate TPoint Polar Alignment information
After
adding 6 new "Suggested Terms" 12 total

A completely different story
Note the difference in TPoint's reported Polar Alignment values before and after adding proper terms. The new terms were added using the The Terms wizard as detailed below. The first report leads one to believe that that the polar alignment is off by several arcminutes in both axis. If you do not add additional terms prior to looking at the Model | Polar Alignment Information the reported values are not correct and therefore they cannot be fully trusted. You will be "chasing your tail" so-to-speak, and the alignment data will never seem to converge.
WARNING
Do not add terms at random. Playing "musical terms" does not work. If you are not sure about using a term do not use it. It is best to let the terms wizard mathematically figure out what terms are best to use. More advanced users can choose to use any other terms they feel are necessary.
The Terms Wizard
Click Suggest Terms on the More Terms dialog to access the Add Harmonic Terms dialog shown below.
Add New Suggested
Terms
Before assessing Polar Alignment information
Software Bisque found that it was extremely difficult for first time users of TPoint to figure out what terms to use on their own. This is understandable since there are 45 to choose from. To that end TPoint for Windows has the ability to magically figure out which terms to use so the user need not be concerned with the process of selecting the proper terms. To use the Terms Wizard press the "Start" button located on the "Add Harmonic Terms" dialog as shown here.
Add Harmonic Terms
(The Terms
Wizard)
Click
Start to display something
similar
When you press the "Start" button TPoint for Windows figures out what term to add mathematically. Note the value to the right "57.48" is the overall RMS pointing. This is the new RMS pointing after adding terms. After adding the first term press "start" again. A new set of values will appear.
After adding a new term

Note the
new RMS value on the right
Keep adding new terms with Suggest New Terms until the RMS value no longer decreases appreciably, law of diminishing returns. Normally using Start | Add to Model with about 6 or 7 newly added terms is plenty to achieve good RMS pointing (under 1 arcminute). Keep in mind that "fewer" terms is always best since using improper terms can cause conflicts. Do not add a term if you are unsure about it. Again, it is best to use the magic of the Terms Wizard if you are unsure as to which terms apply or if you are a new TPoint user.
Now take the above example
76 arcseconds
RMS

Before
adding any new terms
versus
37 arc seconds RMS using 6 new suggested terms
- 12
total

37 arc seconds RMS pointing with
12 terms
Mapping Summary
-
Set the date and time accurately on the PC.
-
Double-check your site information (latitude and longitude) in TheSky.
-
Turn off all stars fainter than about magnitude 3.5.
-
Turn off all galaxies, clusters, variables, double stars, and nebulae.
-
Delete existing TPoint model by highlighting the object and clicking DELETE.
-
Delete all previously mapped labels using Data-Add User Defined Data-Remove All.
-
Insert a new TPoint model Edit-Insert-New Object-TPoint-display as an Icon.
-
Accurately center the first bright star on the video screens cross hair.
-
Sync the telescope on the centered star (do not sync again)
-
Map this first star after you have synced on it.
-
Slew to the next star, center it as best you can, verify and map it.
-
Map 6 bright stars, 1st to 4th magnitude close in proximity (that is, in same constellation).
-
Use TPoint to “Suggest New Terms” to improve pointing (see below).
-
Map an additional 12 to 15 bright stars on the same side of the meridian.
-
Remove any new terms added and start fresh no new terms using Suggest New Terms.
-
Use the TPoint Polar Alignment Information (only after adding several correct terms).
-
Adjust the mount as accurately as you can based on the TPoint Polar Alignment information.
Make the necessary adjustments to the mount to refine polar alignment and then start again. When TPoint reports that the polar alignment is under ~2 to 3 arcminutes or better in each axis you can safely call it quits.

