PrecisionPEC
(Tips on how to acquire the log file)

Minimum Requirements

• Paramount GT-1100S/ME
• TheSky6
• CCDSoft 5
• PrecisionPEC software

Index

• Background
• DirectGuide what is it?
• FAQ's
• How to find the right star
• How do I create the log file
• I can't keep a star in the tracking box!
• On-line PrecisionPEC Help File
• What is PrecisionPEC?
• What should the data look like?

Background

Using the latest versions of TheSky and CCDSoft you can easily record a stars position over time, analyze the data, apply the best fit, and finally upload the new corrected table into your Paramount's firmware using PrecisionPEC.

The first assumption made is you are able to place a star on your imaging CCD chip and keep it in the small tracking box without making any corrections to the mount for 10 to 15 minutes, that is several periods of the worm. This requires a good polar alignment at longer focal lengths. Note you are not "auto-guiding" on the star nor are you making any corrections but rather simply recording the stars position during the logging period (about 10 to 20 minutes). This is easier then auto-guiding, there is no calibrating required. The log file created allows the PrecisionPEC software to determine the best fit by eliminating random errors like seeing before finding the best curve to fit the data. After correction you can expect ~1.0 arcsecond of error.

Do I really need to use the periodic error correction?

No but,
 

There is absolutely no reason to not employ the periodic error correction. First it is quite simple to gather the data- easier than auto-guiding and then correct it. Secondly you will not be able to take full advantage of what the Paramount has to offer in regard to accurate arcsecond level tracking without employing the correction on un-guided exposure.

An image

says it all!

Many Paramount owners are already successfully imaging without applying any periodic error correction and they are achieving better than satisfactory results because the periodic error is 5 arc seconds or less BEFORE the correction. But again why not minimize the average 2 to 3.5 arc seconds of error down to 1 arcsecond or even less?

Not requiring the periodic error correction (worm error correction) goes without saying when you are auto-guiding because the cumulative tracking errors are taken care of in a closed-loop fashion. However, by minimizing the periodic error you can expect longer exposures with and without guiding with improved performance by employing the correction and this of course applies to using  "ProTrack" as well.

Because the periodic error is so small on the Paramount ME and even the GT-1100/GT-1100S (under 5 arc seconds) it is very difficult to detect it. By the time you have enough scale to show the error more often than not the seeing will be worse than the error so it becomes difficult if not impossible to detect and correct it accurately using a CCD camera or even visually. By attempting to simply "RECORD" the error by using a CCD camera with relay corrections or DirectGuide you will  have also introduced other random errors that make the correction data polluted (worse). And that is also true for trying to put back the correction by again recording the corrections that were made and storing these in the mount's firmware. For these reasons trying to train the periodic error of the worm is flawed and the reason for PrecisionPEC.

Finding a good star to log

Use TheSky to find an appropriate star to use for creating the log file. The star slewed to cannot saturate the CCD and should be bright enough to track when taking the shortest exposure your camera allows, i.e. 0.11 seconds. Normally this is on the order of about 6th magnitude or so. The brighter stars will saturate easily and therefore cannot be used.

Choose a star that is very close to declination zero. The target zone declination zero  is shown below. Enable the Equatorial Grid in order to see the lines marking RA and DEC. Use View | Display Explorer | Equatorial Grid and make sure it is enabled. Also make sure that the meridian line is set to visible as well. The Meridian line is shown as red on the chart below.

HINT:  Because you will need several periods of the worm (say 10 to 15 minutes of data) select a star that is on the WEST side of the meridian. This way there is no chance of tracking to the meridian whereby the Paramount will stop tracking when it gets very close. If this is not possible make sure that you start the exposure far enough away from the meridian on the East side so the mount will not track to the meridian and automatically stop tracking. See below.

Because seeing is filtered out you need not be too concerned with imaging near the horizon. However, keep in mind that with very long focal lengths the seeing close to or near the horizon can cause the star to fall out of the small tracking box.  

Slew to ~6th magnitude star

Star near declination = zero

Declination zero is indicated on the above TheSky display. Note the belt of Orion very near declination zero.

CCDSoft Setup for logging the data

Use the "IMAGER" chip for recording the data not the smaller auto-guider chip. There is no reason to use the smaller auto-guider chip which is much more difficult to use because it is very small and it is not on axis. Why bother searching for stars?

Auto Guide using the Imager

Select the Imager as highlighted

By using the larger chip that is on axis it is quite simple to place an appropriate star in your larger field of view. The star need not be centered either. Just make sure you have ample room for the star to drift over 10 to 15 minutes without falling off the chip. That is dependent on how good your polar alignment is. Keep in mind you will see polar alignment error as a drift in DECLINATION. That may or may not be up and down depending on camera orientation.

Place Star on  Imaging detector

No need to center star after slewing

IMPORTANT!  Next you need to tell CCDSoft to NOT I repeat NOT make any corrections while creating the log file (collecting the data). This is done by DISABLING the corrections in the X and Y axis as shown below. Remove the Check next to "Enabled" for both X and Y. You can also choose to remove the "Simultaneous X- and Y-axis relay activation but that is not necessary because you are not making ANY corrections when both X and Y have been disabled.

WARNING!  Any corrections made during the logging process with either the relays, DirectGuide, or by moving the joystick, make the data unusable (polluted)!

You MUST Disable Corrections in both X and Y!

Note the "deactivation" of corrections
Do not make any corrections!

NOTE: ALWAYS "DELETE" the file "Imager.log" (located in the CCDSoft folder) BEFORE logging!  If you don't then the data is appended to the older log data if any exists. It is easiest to start from scratch by just DELETING the Imager.log each time using the button Delete "Imager.log" shown above. This way you will not have to go back and edit the text file to remove the unwanted out of date data.

Having difficulty keeping the star in the tracking box?

If you polar alignment is off far enough to cause enough drift in declination to cause the star to fall outside of the small tracking box for the 10 to 15 minutes needed you may need (worst case) to increase the default size of the tracking box. If possible work on the polar alignment first and foremost.

The default size of the tracking box is 32 pixels square. The smaller the tracking box the better because you can readout the data faster increasing the sampling rate of the data. However, because you are not making corrections keeping the star centered in this small box may not be possible. If the star wanders outside of the tracking box you will have to either improve the polar alignment OR increase the size of the tracking box. See the following page for details on how to use TPoint to improve your polar alignment.

Keep in mind that the larger the size of the tracking box the longer it takes to read-out and that will compromise the quality of the data. When collecting the data the camera must be oriented with North UP or Down after a meridian flip.

If you you have to INCREASE the size of the tracking box you can make a change in the Registry. You can increase the size in either direction or both. You can double the size of the tracking box to 64 by 64 pixels or even larger. Again, keep in mind the larger the box the slower the readout!

Refer to the Software Bisque Support Corner for details on modifying the registry to make that change.

What should the data look like?

Here is an example of data from the Paramount GT-1100S. The mount is several years old and shows an error of about 2.5 arcseconds prior to correction. Note the error is not quite as smooth as one would expect with the Paramount ME but still very good because of the Byers gears AND the design of the worm block hardware.

GT-1100S sample data before correction

Total error ~2.5 arc seconds before correction!

After uploading the PEC table into the Paramount you can expect 1 arcsecond of error after the correction.

Periodic error after the correction is enabled

0.5 arcseconds up and down = 1 arcsecond total

What does a tracking log file look like?

Following is a sample of the header created in the LOG file a few of the star positions logged entries from the above data. The relatively important information has been made BOLD to stand out.

|Elapsed Secs|RefCentroidX|RefCentroidY|CurCentroidX|CurCentroidY|GuideErrX |GuideErrY |TotGuideErr |XPlusRelay |XMinusRelay |YPlusRelay |YMinusRelay |PECIndex RA |PECIndex Dec|

| 0.0000| 16.0000| 16.0000| 13.8806| 14.8411| -2.1194| -1.1589| 2.4155| 0| 0| 0| 0| 1127| 1054|
| 2.3750| 16.0000| 16.0000| 13.5394| 13.5565| -2.4606| -2.4435| 3.4678| 0| 0| 0| 0| 1122| 1054|
| 3.4370| 16.0000| 16.0000| 13.8298| 13.5996| -2.1702| -2.4004| 3.2360| 0| 0| 0| 0| 1116| 1054|
| 4.5000| 16.0000| 16.0000| 13.8333| 13.8628| -2.1667| -2.1372| 3.0434| 0| 0| 0| 0| 1111| 1054|
and so on