Results from MPC indicate I have essentially zero RA offset but an average 0.15 arc-second offset in declination. This offset might be due to a collimation error. Minor planets can be anywhere in my 21 x 21 arc-minute FOV.

Trying to use CCD Inspector 1.3.4 gives inconsistent results, and these results do not agree with other (slower) methods of measuring collimation. The smallest possible collimation adjustment causes the CCD Inspector results, especially the angle, to change dramatically. Making no changes results in changes of values that are much larger than my hoped for collimation accuracy.

There are plenty of stars; I was imaging Pluto, which is in the Milky Way. Results are similar for less dense star fields.

How can I significantly improve the performance of CCD Inspector?

Setup:

C-14 on a Paramount
Homebuild camera using a Tektronix 1024 CCD with 24 micron pixels. Image scale is 1.25 arc-second/pixel. This setup routinely images 19 magnitude minor planets.

Exposures are usually unfiltered

Focus is good; seeing is usually 2.5 to 3.5 arc-seconds

Optec focuser.

Thanks,
Joe Hobart
Flagstaff, Arizona

Hi Joe,

What values do you see for collimation error with CCDI? Once you get to the limit of what the seeing conditions will allow, small changes in collimation adjustment may result in frequent angle changes. That's because at this point, the measured error is at the level of noise produced by the seeing effects.

Regards,

-Paul

Hi Paul,

The best I have been able to get is 12 to 14 arc-seconds, and the direction bounces around.

Joe

Hi Joe,

That is close, but above where I'd want it to be normally. The things that are likely to cause variation in collimation are:

1. Seeing. This includes ground effect air-currents

2. OTA is not at a thermal equillibrium, tube air-currents, mirror temperature not equalized

3. Poor tracking

4. Noisy image: use auto-dark subtraction

5. Not enough stars, or stars are too dim: increase exposure time, or increase binning to 2x2.

When the direction of the error starts changing by itself, you've reached the point where the collimation cannot be improved much further.

If you haven't tried this already, try turning on image averaging. This will force CCDI to acquire more than one image before computing collimation error. The averaging process helps reduce variation and noise due to seeing. Change it to 2 or 3 in the following menu:

Real Time -> Images to Average -> 2, 3, 4, 5


Regards,

-Paul