Daniel Petzen · Aug 16, 2025, 11:37 AM
I'm used to crappy software and hours upon hours of troubleshooting, but PHD2 has superb support for the Starlight Xpress AO.
That is because the programmer (Andy Glasso) that did the AO part of PhD years back used the SX unit to develop it. I don’t think he is working on PhD anymore as far as I know and that part of it has not been changed (and does not need to be) in years.
Daniel Petzen · Aug 16, 2025, 09:59 PM
I thought that an AO would shine with really short exposures, but I've not managed to get that to work yet.
Don’t bother, it will not work very well and will cause more problems than it is worth. I have been using the SXAO on multiple scopes for over 10 years. I own two units (one spare) and currently use it on my CDK 14.
The original idea was for a high frequency AO for amateurs. That came from SBIG and Benoit Schillings way back in 1997 with a tip/tilt mirror system. I had one of those. One quickly discovers that there are three problems with high frequency use:
1) Bright enough guide stars for those short guide exposures are rare and multiple guide stars bright enough are nearly non-existent. There is a rather large benefit from using multiple guide stars and that is not worth giving up.
2) These systems can only correct for one order of atmospheric distortion and there are many others so benefits are limited at best. As such they are nothing at all like the professional systems.
3) The faster the system is run, the more subject to flakiness it gets.
You should look at SXAO (and other simple tip/tilt devices) as optical guiders. The advantage is that it is much easier and faster and more accurate to move a 2 oz. plate than a 30-100 pound mount.
In fact, I typically run mine at 7 second guide times.
It is also critical to be sure your calibration is accurate. Wind can mess that up as can poor balance. I run mine on an L-350 but with cheaper mounts one might see different mount responses to bumping in different parts of the sky so would need to be recalibrated for each object and maybe even after flips (my L-350 does not need to flip).
I would add that if one is drifting out of the “bump box” during exposures, it is probably your mount or polar alignment but could be wind or just bad seeing. I leave mine set at 80%. One thing I have noticed (and I suspect this is a bug) is that sometimes the first exposure will try to drift out of the box (and therefore bump the mount a lot) but all the other exposures do not. The exposure still comes out fine, however.
As far as PhD tracking with the SXAO, I typically see .3 RMS or more on poor nights, .2 to .3 on average nights, and .12 to .20 on good nights. But seeing at my remote site is well above average, sometimes dipping into sub-arcsec territory. Even with short (60 sec or so) unguided images I often see FWHM in the 1.5-2.5 range. With more normal main imager SXAO guided exposure lengths of 300-900 seconds that translates to getting sub-exposure image FWHM in the 2.0 to 2.5 arcsec range. My normal reject level for high res. images in the luminance is 2.3….
Overall the thing to realize is that the SXAO is not going to make things better by running it faster to make up for seeing. What it can do is to take some of the sting out of lesser performing mounts and improve the precision of guide moves on all mounts.
Hope this helps.
