May 19, 2016 UT: Polar alignment and pretty pictures of Jupiter and the Moon

Michael Richmond
May 19, 2016

On the nights of May 13/14 and May 18/19, 2016, I adjusted the polar alignment of the Meade LX200 12-inch telescope in the dome at the RIT Observatory. My goal was to improve the tracking performance, so that we could take longer exposures without guiding. This page is largely a summary of the notes I took during the procedure, which I hope will make it go more smoothly in the future.

I also took a few pretty pictures of Jupiter and the Moon, just for fun.

Preliminaries

On May 13/14, I used the usual arrangement of CCD + filterwheel + Off-Axis-Guider (OAG). But after that, I sent the filterwheel mechanism back to the manufacturer for repair. It turns out that I don't have the proper parts to connect the CCD to the OAG directly, so on May 19, the arrangement was just CCD directly mounted to the telescope's focuser.

This had two consequences: first, I had to figure out the proper orientation of the camera, and second, I needed to modify the focus of the telescope because the focal plane was now considerably closer to the mirrors.

The orientation is simple: to end up with the usual North up, East left arrangement without changing any parameters in MaximDL, one must rotate the CCD camera so that the yellow "RIT" label is to the South (that is, close to the telescope pedestal when in the park position).

The change in the focal plane position was very roughly 2.3 inches closer to the mirrors, based on the ATIK system chart schematics. In order to move the focus in this direction, I discovered that turning silver focus knob CW moves focus toward mirrors.

Polar alignment notes

There are a number of good guides to aligning a telescope to the pole using the "drift method"; for example,

• Polar Alignment according to Starizona
• Fred Espenak's guide at AstroPixels

One can summarize the two steps of the method (adjusting the azimuth and altitude of the telescope's polar axis) as follows:

 To adjust    point      if star   then axis     turn LX200       on TV screen,
             telescope    drifts     is           knob            stars will 
-----------------------------------------------------------------------------

 altitude    to East      South      too low      CCW to raise     jump up (North)

                          North      too high     CW to lower      jump down (South)


 azimuth     to South     South      too East     CCW to go West   jump right (West)

                          North      too West     CW to go East    jump left (East)


-----------------------------------------------------------------------------

My typical procedure was as follows:

1. start a sequence of ten images: 5 seconds exposure time, followed by 30 seconds of delay
2. measure position of a star in first image in sequence
3. measure position of a star in tenth image in sequence
4. compute amount of drift in pixels (binned 2x2, so 1 binned_pixel = 1.32 arcsec)
5. if desired, divide by elapsed time, approx 360 seconds (nine images at approx 40 seconds each)

I was able to compute very, very roughly the response of the drifting behavior to the position of the axis-adjustment knobs.

when adjusting altitude

rotating knob by 1 "bump" results in drift rate of about 6 binned_pix = 8 arcsec in 360 seconds

The picture below indicates what I mean by a "bump" on the axis adjustment knob (yes, the picture shows an azimuth knob, rather than an altitude knob, but the structure is the same)


Image courtesy of Dale A. Chamberlain's Guide to Polar Alignment of Meade LX200GPS Telescope

when adjusting azimuth

moving knob by 1 full turn leads to a change in drift rate of about 15 binned_pix = 20 arcsec in 360 seconds

Pretty pictures of Jupiter and Moon

As I waited for the sky to darken, I took some pictures of Jupiter and the Moon, just for fun. Even though I used an off-axis mask to decrease the amount of light entering the telescope (the mask has a circular aperture about 5 inches in diameter), the lack of any filters meant that I could barely acquire data without saturating the detector. I used exposures times of 0.02 seconds for Jupiter, and 0.005 seconds for the Moon.

After bias-subtraction and flatfielding, I tried this and that to make a pretty image. I grabbed a copy of Registax and used it to stack 17 images of Jupiter. Because the 2x2 binned plate scale is so poor, 1.3 arcsec/pixel, the image doesn't show much:

Our images of the Moon were a bit nicer. The Moon just barely fits onto the chip. I chose the sharpest image in a set of 10 and did just a little processing to make the version below. The "ghost" faintly visible to the right is an artifact of the charge-transfer process, noticable due to the very bright nature of the target here. Click on the image below for the full-resolution version.

Last modified 5/19/2016 by MWR.