AstroTrac Performance

Well, the recent burst of image processing from my June images has revealed a few issues with my AstroTrac's performance.  Up until now, I've attributed the errors I've had to the tripod and head on which I mount the AstroTrac.  However, at this point, I'm not sure that's the whole story.

To be sure, the Manfrotto/Bogen 410 mini-gearhead is a problem.  There is a small amount of play/backlash in the gears which makes alignment a bit of a pain and which means that every time you retarget, you need to recheck the polar alignment.  I've pulled the rubber pads off the tripod heads so I have metal-to-metal contacts to eliminate that source of flex.  And I've even gone so far as to attach a webbing strap to the tripod and stake it into the ground to make it a little more resistant to bumps in the dark.  And I attach little blinky lights to the tripod legs, too.

However, the recent set of images of Antares-Rho Ophiuchi and of the M8/M20 regions indicate there are other issues to be addresses.  First let me describe the symptoms.

I had set up Maxim DL to take a series of about 20 images of the Antares-Rho Ophiuchi region and after the first couple of images were download and looked good, I slipped into the car for a nap (out of bug range, not quite so comfortable as the bed inside the house and so easier to get up when my alarm went off).  When I got up, the current image looked okay except it was shifted over quite a bit.  I quickly reviewed the images taken and discovered that somewhere in the middle of the sequence, the AstroTrac had failed to track properly leaving badly trailed images.  I reframed and tried to take some more images before the target slipped behind a hill.

The next target was the M8/M20 region and I set up to take 20 4-minute images.  This time, the AstroTrac had no period of slow tracking like the previous image sequence, but...every image was shifted slightly compared to the previous.  A close inspection of the images from the post processing shows that in fact, the AstroTrac's tracking is slightly off and that as 4-minutes, I am getting slighly elongated stars which, if I stack the images without aligning, will give me short star trails for my 20 x 4-minutes of exposure.

The only idea I have at present comes from a suggestion/speculation on the AstroTrac Yahoo group.  The idea is that the AstroTrac motor is somewhat challenged when it come to lifting a load and so it is better to arrange the arms so that they are "falling" as the arms swings out.  This is backward to what one usually does with a German equatorial mount (GEM), but might make sense.  It's certainly worth a try.  I'd like to get those 4-minute exposures without trailing and certainly without the occasional semi-stalled tracking.