TILT!

Yes, it's that enemy of flat fields: tilt in the optical train. 

Recall -- or don't, since I'm going to repeat the information anyway -- that my first choice for imaging the Polaris-area integrated flux nebulae was an old Tamron 135 mm f/2.5 lens that stopped down to f/4 is quite a nice lens. Second choice is to use my FSQ-106 with a 0.73X focal reducer that gives me a nice wide field. The FSQ is wildly higher in quality than the Tamron, but the only time I've imaged with the focal reducer it produced results that barely adequate: stars were noticeably elongated on one side of the image. 

I'll need to use the FSQ + focal reducer if for some reason the Tamron proves to be problematic. That means it's time to chase down the tilt problem and get it fixed.

What dawned on me recently was that there was enough back focus to allow me to use the tilt plate that came with the ASI 2600 camera I use for imaging. I had removed the plate to make the ASI compatible with my DSLR. The fact is that the ASI has largely made the DSLR superfluous; I'm  unlikely to image with it again. The tilt plate can therefore come out of storage and get back into action. Here's the present situation:

The configuration of interest here is the bottommost one. The reducer requires that the camera sensor be at a very specific distance from it: 72.2 mm. Currently there's a 12 mm M42 spacer in the optical train. Replace that with the tilt plate and a 7 mm spacer and it's all good (aside from some very thin spacers). 

Now, how to use the tilt plate? Let's start by verifying that it's the optical train and not the sensor that's out of whack. To do that correctly, I'd need to rotate the camera relative to the optical train and see if the effects of the tilt moved with it. I'm going to cheat a little and assume that if the field is flat for the native imaging mode (f/5) the sensor is fine.

Here's the ASTAP measurement of tilt in a single luminance image taken with native mode (f/5):

ASTAP report of field flatness for FSQ-106 & ATI 2600

So you're probably asking, "what the heck is that?" The numbers next to the yellow lines are the area-averaged half flux diameter (HFD) of the stars. The HFDs essentially measure how pinpoint the stars are, and they can be affected by focus quality, seeing, and aberration and tilt problems. A perfect telescope with perfect optics and perfect seeing would have very small numbers in this diagram.

Tilt will introduce variations in HFD that are in the direction of tilt. So how does one assess these numbers? ASTAP does that for you. If you look along the bottom edge of the above picture you'll see it assesses the tilt at 6%, which it judges to be "almost none". Great! This suggests that the native optics are adequately tilt-less--and by extension that the sensor is reasonably perpendicular to the optical axis.

Now here's the ASTAP analysis for a luminance frame taken using the reducer (f/3.65):

Same as above, but with the focal reducer

ASTAP says the tilt is 19% (moderate). It also confirms my eyeball judgement that the tilt is largely oriented along the long axis of the sensor. 

(Yes, you could deal with some star elongation in post-processing, but isn't it better to not have to do that? This brings us back the the first defense against tilt, a tilt plate.)

The ZWO tilt plate works using three adjustment screws arranged in an equilateral triangle, much like the tilt screws for adjusting a small Newtonian objective mirror. What I'll do is to orient the plate so that one of the adjustment screws is in the direction of the tilt, like so:

Orientation of ZWO tilt plate: blue arrow = tilt direction, red dots = adjustment screws

This way the adjustment screw at that left vertex becomes the primary one to adjust. Note that ASTAP has an option to provide a 3-point analysis which may be easier to interpret in some cases.

The process of correction is basically repeating the sequence of imaging, analyzing, and adjusting. It's another new skill to learn and put to use, which is good. There doesn't seem to be much to it; I'll let you know how it goes.

Unfortunately this won't work for the 135 lens; it has no spacers to swap out for the tilt plate. Should I find it needs correction, I'll probably use these things. I might get that and see if it's all I need for correcting the focal reducer tilt, too. If these shims work for that it would eliminate the need for a tilt plate and make it easier to change from one configuration to another. 

-------------------------

At least it's March now. A few inches of snow is coming in the next day or so, just because it can. Almost time to make my reservations at Lac qui Parle!

[Added 5 March, it was more like 9 inches.]