Back to http://arnholm.org/

Telescope drive tracking accuracy


Warning: This information is Copyright (c) Carsten Arnholm, and may not be reproduced without permission.

What is tracking accuracy, and why is it important?

The most basic purpose of a telescope drive is to compensate for the Earth's rotation. The Earth's rotation relative to the stars is called a sidereal day, it has a length of 86164 seconds (a Solar day is 86400 seconds). So, obviously the telescope Right Ascension (RA) axis must rotate once in exactly 86164 seconds in order for the observer not to experience stars drifting in the field of view. This is especially important for deep sky photography where long exposures are employed. If there is drift, the stars will be elongated instead of round, or they will leave the field of view alltogether.

Even with long-exposure modified webcams where many medium length exposures (~1 minute) are used, we cannot tolerate drift, since the field of view is very narrow. Any drift will reduce the common field of view between captured frames, effectively prohibiting image stacking.

Accurate tracking is also important for very long focal length webcam imaging of the planets, as the field of view is extremely narrow. With drift, a planet will quickly disappear from the field of view.

What causes drift?

The main reasons for star drift are

Measuring RA motor speed

Above, we identified that the RA motor speed is important. How can we measure if the motor we have has the correct speed? It is not difficult, we simply remove the motor from the mount and measure the rotation speed of the motor axle as accurately as we can. But before we can calculate the RA motor speed error, we must establish the correct RA motor speed (via mounts worm period). My Vixen super Polaris mount has 144 teeth on the RA worm wheel (identical to GP and GP-DX mounts). This implies that the worm must rotate 144 times per sidereal day, and the correct worm period then becomes:

Accurate_worm_period = sidereal_day_length / #of teeth = 86164 sec/144 = 598.3611 sec

The correct motor speed can be calculated from the transmission ratio between motor axis and worm axis. For the Vixen MT-1 motor, this is 1:1 but for the Sanyo Denki motors used in the Astromeccanica AM-V belt & pulleys configuration, the ratio is 4:1. A picture of the different motor setups can be seen below

Measurement results

Here are the results of the measurements for the 3 telescope drive controllers I have tested.

.
Vixen Super Polaris mount
Motor controller Vixen
SD-1
Powerflex
MTS-3SDI
Astromeccanica
DA-1 LE
Manufacturer Vixen Boxdoerfer Elektronik Astromeccanica
Stepper Motor Vixen MT-1 Sanyo Denki
103H548-0440
Sanyo Denki
103H548-0440
# of motors controlled 1 2 2
PC remote control option no yes yes
Motor:worm ratio 1:1 4:1 4:1
Power transfer by spur gear belt & pulley belt & pulley
Correct Motor period 598.3611 s 149.5903 s 149.5903 s
Timing method Webcam
manual timing
Automatic timing
see also this page
Webcam
manual timing
Time measured 3 hours 3 hours 1 hour
Measured (averaged)
Motor period
598.3006 s 149.6018 s 149.9033 s
Measurement raw data Image MS Excel file Image
Calculated
tracking error
0.010 % 0.007 % 0.209 %
Comment Simple, good quality Initial impression is very good. Tracking error >25 times larger than for MTS-3SDI
more information.
The price is about the same as for MTS-3SDI.

Conclusion

The Vixen SD-1 and the Boxdoerfer MTS-3SDI both have excellent tracking accuracy, while the Astromeccanica DA-1 is quite inaccurate.

Consequence of poor tracking for Astromeccanica controller: If a webcam imaging session started with a star at the center of the field of view, the star will have drifted off the webcam chip after 1hour 37min. In this situation, there will be insufficient overlap between webcam frames much earlier than that, so the estimated longest usable imaging session would be no more than 30minutes. I.e. this means you cannot leave the scope alone for more than maximum 30 minutes while imaging if no guiding is used. For SD-1 this value is ~10 hours, and for the MTS-3SDI it is even longer (i.e. long enough for the Sun to come up!).