Autoguiding with a Webcam

I have decided to try out Autoguiding with a Webcam to see if this technique could enable longer exposures. Currently the mount that I have limit me to about 2 minutes of exposure if I dont want any star trails (based on a focal lenth of arround 700mm.

Background

The earth rotates about one turn per 24 hours. The stars however sit pretty fixed in the sky (at least in the 24h scope). An observer located on Earth will notice how the stars seem to travel from East to West much like the sun. This is due to the spin of the Earth.

German Equatorial Mounts (GEM) have been designed to make it possible to counter this effect. By ailgning the mount with the northern pole (if you are north of the equator) the mount and the telescope that sit on the mount, will be able to counter the earth spin. This is referred to as tracking in the sidereal rate (the same rate that the Earth spins in).

If the mount is perfectly aligned with the pole (polar aligned) and the motor is tracking perfectly - the stars would stay on the exact same spot if observed thru the telescope. It is however almost impossible to establish a perfect alignment and it is almost impossible to create a perfect drive with perfect gears.. This means that someone has to correct for these errors if the stars are to stay in the same place.

I hope that this someone will be the computer. There are two alternatives:

1. Manual guiding by looking at a star in high magnification with an EP with a reticle. Counter the movement as observed relative to the reticle.

2. Automatic guiding performed by a computer assisted camera device. The computer will analyze the movement of a star by taking a picture of the star. The mount will be moved to counter the movement relative to the staring location.

Tools of the Trade

1. Autoguider, a camera that can observe the star drift. I have decided to use the Philips ToUCam Pro Webcam.

2. Autoguider software. There are standalone products that include the computer needed to analyze and send guideimpulses. I have however instead decided to go for a PC based software that perform this job. I decided to go for Guidedog.

3. A computer sitting near the telescope

4. A relay interface between the computer and the mount. In my case I have a SkySensor 2000 equiped GPDX that includes a computer interface.

5. A separate guide telescope or an off-axis guider. I will use my Megrez 80mm as a separate guide telescope.

One might go with different options depending on the equipment available. There are many articles available on the net regarding pro's and con's related to different equipment. I will make sure to add some links on my links section about this topic.

Configuration

The configuration is different depending on the applications used. This is however true for most:

Make sure you have established correct polar alignment. Preferably via the drift alignment method.

Enter the settings for you telescope and camera. In my case its 480mm focal length and 5.6 micron square pixels.

Start out with guiding in RA only and progress to DEC corrections when RA is working fine.

Make sure the guide camera is aligned with the N/S and E/W directions on your screen.

Also make sure that your computer can communicate with the mount. In my case that means setting the SS2K to LX200 and 9600BAUD and choosing the ASCOM driver for Generic LX200 in Guidedog.

Begin with analyzing the on-screen results of a locked star to see if you are doing ok. When the guiding is activated it should stay within +-4 seconds of arc.

Guidedog comes with easy to use and understand manuals that can be downloaded from the site

Analysis of First Test : 2004-09-19

This analysis was made possible by the Log Errors function in Guidedog. Checking this box engages the Arc Second logging in Guidedog. The log is a textfile that can be imported into Excel or similar spread sheet software capable of creating diagrams. You should also try to take an exposure with a separate camera while testing the autoguider since that will give you an understanding of the effects of the error in RA or DEC.

My Own Impression

When I did this test I got the feeling that I have a very poorly configured motor or that something else is wrong in the setup. I got very high and jumpy errors in both RA and DEC. I have been sitting in front of the guide eyepeice quite a lot when drift aligning and I have been under the impression that my mount is in pretty good shape.

The Computers Impression

I will try to indicate the result by showing diagrams of the error log in Guidedog as well as pictures taken during the guiding effort.

Scatter

This image indicate the error in RA (Blue) and DEC (RED) between each exposure. The values were read at intervals of 3 seconds during a total timeframe of about 6 minutes. As one can see the application indicate errors in the magnitude of -+30 seconds of arc.





Trend Indicating

The same session but this time with lines between each reading. A slight trend can be seen in both RA and DEC. My guess is that the mount is a little bit out of alignment.





RA Error Separated

This diagram have the deviations from 0 filled in one color and the DEC error is filtered out making it easy to see in which region the star has been most of the time.





DEC Error Separated

This diagram have the deviations from 0 filled in one color and the RA error is filtered out making it easy to see in which region the star has been most of the time. No guiding was made in DEC and it is pretty easy to see the trend.. but why is the star jumping while on the trend from positive values to negative values?





Resulting Image Magnified

This image show the center of the image captured during the above guide session. The exposure is 6 minutes. The image is magnified 1600 times to indicate separate pixels. Each pixel correspond to 2.12 Seconds of Arc. Is it safe to say that the star has not diviated as much as indicated by the guidedog error log?





Full Frame Result

Here one can see the full image from which the magnification above was grabbed. The upper part contain the whole frame and the lower frame contain a crop of the full frame in 100% magnification.





Full Frame Result From Longer Exposure

This image is from another test of the guiding. The exposure is 10 minutes and the errors indicated by Guidedog were roughly the same as indicated in the diagrams above.