LXD55 telescope mount Autoguiding
Some of the info on this page is a bit out ofdate. Please find my general notes on webcam autoguiding here. Read it if you have any interest in tracking and autoguiding!!!!
Latest news - Computer Controlled Motorised Guidescope Mount. These pages contain a lot of information about my recent advances in autoguiding my Meade LXD55 mount.
I have a 200mm F5 Newtonian mounted on a Meade LXD55 mount. The mount is required to track the sky at the same speed as the Earth's rotation. Sound simple? Ha. Like most equatorial mounts, the LXD55 uses stepper motors and worm gears. During one rotation of the worm gear, perhaps ten minutes, the speed of the mount will go through a cycle of tracking too fast and too slow. This is know as periodic error.
The error of my setup can be seen below.
The total error over the ten minute cycle can be estimated at about 75 arc seconds. This means that over a 10 minute period, a star walks back and forth across the screen by about 65 pixels. This is clearly unacceptable for our purposes. The irregular spikes on this image are due to wind moving the scope.
Solutions There are three routes to removing this error.
1. Mechanical. It is possible to mechanically enhance the mount to reduce error.
2. PEC. Periodic Error Correction is a system where the mount control software learns the shape of the curve, and constantly adjusts for it. However, such a feature is not included in Autostar for LXD55s.
3. Autoguiding. A star is monitored and corrections to drift are made on the fly.
I have chosen to experiment with autoguiding.
To autoguide, one needs a camera which is monitoring a star. This can be guide frames from your CCD camera or a separate camera. Such a camera could be mounted on an off axis guider on the main telescope, or by using another telescope piggybacked to the main OTA.
I have chosen to use a Philips Toucam Pro II webcam. This has a 640x480 CCD colour CCD Sensor. This camera is mounted on a small 70mm refractor (focal length 400mm) that I had gathering dust in a corner. Homemade experimental tube rings attached it to my main OTA.
Initial experiments show that running at 1/25 second exposures, we can see stars as dim as magnitude 6.5
K3CCDTools by Peter Katreniak is used for all my webcam capture work. A feature of this software is the Drift Explorer. This allows the user to lock onto a star, and monitor the drift in RA and DEC. The software also has the ability to issue guide commands. Currently K3ccdTools does not support the LX200 command set. However, I obtained from Peter a copy of the software that has a FILE autoguiding interface. This continually updates a small text file with the RA and DEC offsets.
This is a shot of the software in action, tracking a rather nearby galaxy. This software has been used to generate the periodic error graph at the top of this page.
To use the file interface, I wrote a short computer program in C. This program keeps reading the output text file, and makes small guiding adjustments to Autostar via the serial cable link. It tries to keep the offset in the file as close to zero as it can manage.
To test this out, I attached the toucam to my guide scope, and plugged the webcam into a computer running one copy of K3ccdtools. Another webcam was attached at prime focus of my main OTA. This was plugged into another computer, also running K3. Note, it is possible to do all this on one computer.
Next I pointed the whole affair at the classic text star, Mintaka, which is right on the celestial equator, where periodic error is at its worst.
On the guide computer I started drift explorer and ticked the guide tick. I started my own program to issue the guide commands.
On the computer attached to the main OTA I just start drift explorer to monitor the star and output the result to a file.
The first result can be seen here:
Note the change in scale. The guiding system has kept the star within +/-4 arc seconds. This corresponds to a similiar number of pixels on the screen. The slight periodic motion seen here is probably due to a slight east/west misalignment of one of the cameras.
Using the system keeps the tracking pretty close to ideal. It seems the smallest movement the guidescope can detect is about 2.5 arc seconds.
The hardest part of using the system is getting the guidescope lined up with a suitably bright star and making sure the cameras are properly aligned.
Enhancements to my guiding software may increase performance.
Future versions of K3ccdtools may include support for the LX200 command set.
Special thanks to Peter Katreniak for helping with this experiment.