SBIG’s Adaptive Optics: The AO-L
I am one of the unfortunate blokes who jumped on the waiting list for the QSI 683WSG which was announced at NEAF back in April. And I waited. And I waited. And here, four months later, more than twice as long as the “6-8 weeks” I was originally told, I was fed up. So I started to look around to see what I could find when with perfect timing, SBIG announced their 90 day sale. So I splurged a bit with the purchase of an STL 11000, AO-L, MOAG, and Remote Guider Head. I’ve had some issues with the 11000 which need to be resolved still, but that didn’t stop me from learning how to use the rest of the equipment. And there is a real gem in there: The Adaptive Optics “AO-L”. I don’t know why I don’t hear about these unit’s more often. Yes, I usually do a reasonable job aligning and guiding. But this little puppy takes it to a whole new level of mega awesomeness, most especially for long focal length imaging.
The AO-L is just a “large camera format” upgrade to the “AO-8″. The AO-8 and AO-L use a slightly different technology than the older, AO-7. This newer technology performs fewer corrections per second, but the corrections are performed faster – less system latency.
There are two main configurations: Scope to AO-L to camera, or if you are planning to do narrowband where you need to guide ahead of the filters, Scope to AO-L to MOAG to camera. In the latter case, one flips the AO-L upside down and the MOAG slips right in between. The nosepiece that comes with the STL series cameras is removed from the camera and placed on the AO-L. The MOAG contains a helical focuser on it so achieving focus with the remote guide head is made very simple. If just using the AO-L and STL camera body, the on board guider is already par focal with the imaging plane and nothing need be done.
A large plate unscrews from the AO-L that allows you to see inside and attach it to the camera body or MOAG. Attaching the AO-L to the MOAG or camera is not really fun. There are a few holes on the sides of the refractor mirror plate which are not easy to reach and could allow for screw to drop into the camera. Luckily the manual describes that this could happen so I was lucky enough to avoid it, but it certainly was nerve wracking for my big fingers!
Once physically attached, there’s just one cable to connect… The guide cable coming out of the AOL goes right into the back of the STL body camera. It’s the first time I have seen a piece of equipment with a super short cable that I was actually happy had a super short cable! I am surprised they don’t just build this whole system into the camera body.
The basic mode of operation is that small corrections will be performed by moving the refractive element in the AO-L, while “large” movements will be corrected in a traditional “autoguider fashion” by telling the mount to move. It’s not obvious at first, but this implies a few really fascinating things. First, it relies primarily on a system outside of the mount to perform corrections. That means means that common gremlins like backlash, stiction and PEC type of errors are very well hidden away. Second, it means that even if you don’t have any of the issues, but your seeing is poor and you’re “chasing the seeing”, your mount doesn’t have to work any harder. In fact, these errors can all be occuring, but they may be corrected by a lens movement instead of instruction to move the mount! If you have a great mount and super solid alignment, you might only perform corrections with the refractive element and never instruct your mount to move. It’s a very elegant solution to two classes of notoriously icky problems.
The software that comes with the camera for controlling the Adaptive Optics unit is CCDSoft. Given my previous experience with MaximDL and Nebulosity+PHD Guiding, this software was a pretty quick learning experience. The online manual for the AO-L is pretty detailed and I found myself referring to it several times to make sure I understood what was going on. One non obvious part however was that two separate calibrations need to be performed after your alignment routine. The first is the same type of calibration you’d perform with PHD, except this is CCDSoft’s version. A typical multisecond image is good enough to get this working. The second calibration is for the AO unit. Again, think “big moves handled traditionally, small moves handled by refractive element”. So it needs to calibrate the impact of the refractive element. I found that using large exposure times for both of these helped calibrate faster. Then I could reduce the numbers way down and have them operate just fine.
Finding guide stars is not the simplest thing, but with a bit of rotating or simply “looking for a star nearby” I was able to get by. It’s not pleasant but it’s not impossible either. It’s not near as simple as using a separate guidescope, that’s for sure. Like many others, I was also very nervous about dropping the exposure times to fractional seconds since my previous guiding experience told me that two seconds was almost required. In fact, I was able to reduce exposure times to 0.10 seconds without any problems! However, for safe measure I found myself running twice that, 0.20 to leave a tolerance gap.
In the same image of the control screen here, the “X and Y error” show the distance in pixels the guide star moved. The “X and Y tilt” show the position of the lens. If it pushes too far in one direction, the AO unit will tell the mount to move and it will move the lens back to a more central position. Of particular importance to note is that the slew rate is actually measured in arcsec/sec, so you must enter your focal length into the “Setup” tab for this to work!
So, does it all work? Yes, and very well! Pictures are worth a thousand words…:
Can you guess which one is regular guiding and which one has adaptive optics? Both these come from 10 minute exposures. The one on the left is with a regular “2 second” guiding while the right is a 0.10 second AO guide. Unfortunately I wasn’t thinking of putting up a comparison picture so they’re not the same object, but the improvement is pretty obvious.
One thing worth noting is that as far as I can tell, the AO-L only works with SBIG cameras. I wouldn’t know where to plug it in to anything else, let alone attachment issues. Other folks like Orion and Starlight Xpress also offer comprable AO systems. While they are designed to work with their respective lines of cameras, it seems they would just as easily work with other cameras, focal issues and mounting issues aside.
In conclusion: Given that people always talk about the mount as being the most important component of any astrophotography setup, it’s surprising that people wouldn’t consider the AO-L as a wonderful compliment to a wonderful mount! If you are imaging at short (<700mm) focal lengths, you might not see much benefit. However, if you’re going deep and anything over 1500mm, you are very likely going to see an improvement. So, Hypertune, PEC, backlash, stiction, wind, sneeze… blah blah blah? Just get this and be happy without worrying about any of that!