Main scope moves relative to autoguider

[sculptor]

Help requested. Over a two hour period, autoguider crosshairs stayed within a few arc sec of guide star, but main camera image drifted by 3' arc and rotated by about 18' arc. Causes?

Flimsiness? I've made a rather elaborate and possibly too flimsy bracket that permits two-knob adjustment of autoguider relative to main scope, which also allows for off-axis guiding. It is possible that as whole scope rotates by 30 deg over 2 hrs, differential strains have caused the autoguider to flop about by 3' arc relative to the main scope.

Mirror flop? Celestron 11" SCT with no mirror lock.

Polar axis alignment? My (self-written) alignment software (15 star alignment) thinks polar axis is off about 0.5 deg, with about 0.25 deg of cone. My maths has temporarily evaporated due to being up all night, but it seems about the right order of magnitude to account for 18' arc field rotation. Any experts? Is there a second-order effect that could cause the 3' arc differential beween main scope and autoguider?

Off-axis guiding? On this occasion, the guider was pretty close to on-axis from the night before, but I'd not specifically set it up on-axis. Can off-axis guiding produce this sort of effect.

Luckily, I was stacking 5 min exposures, so each exposure moves only 15 sec arc, and the stars look perfectly round.

Expert advice most welcome.

[higginsdj]

All of the above.......

My Piggybacked 4" SCT guide scope is rigidly mounted to the main scope but still exhibits some movement over the course of a nights observing (field rotation of about 0.5 degrees) and I am within 1' of Polar alignment (according to polealignmax)

The point likely to cause the bigegst problem will be the adjustable mount between the Autoguider and the scope. You will need to find a way to adjust then lock down the joint (ie make rigid).

Cheers

[sculptor]

Many thanks, David. Useful to know quantitatively that even your superb polar alignment (though probably necessary) is not sufficient. Have put a great big spring on the guidescope that should stop flop in the two most likely directions without losing 'useability'. If that fails, will try adjustable rigid strut.

One reason for wanting to fix the problem is that on-pixel tracking should mean no need for post-processing with for RegiStar/RegiStax, which while superb, don't always lock onto target.

Now that I come to think of it, I've managed to get on-pixel tracking with my previous arrangement, that used 6 Kg of stainless steel in the adjustable bracket. My new 1.5 Kg aluminium attempt is obviously suffering from anorexia and under-engineering.

Cheers.