On my first night out, I had a chance to examine its performance on an APS-sized chip (CCD Labs Q8-HR) at prime focus and with two focal reducers / flatteners I had on hand - a William Optics 0.8x v2 and a William Optics 0.8x v3. Both were used at 55 mm from the CCD plane. It is worth noting that neither of these reducers are suggested by William Optics to be used for this telescope. This is an "off-label" use as it were.
The images below were simple 20 second shots (best of 3 selected), demosaic'ed and converted into luminance-only in Nebulosity and then stretched linearly. The full-frame was then rotated 33.3 degrees and cropped so that the full diagonal extent of a frame could be easily seen. At prime focus, the image covers 161' x 107' and the diagonal covers 193'. With either reducer (both were measured to produce 0.8x), the image covers 201' x 134' and the diagonal 242'.
Click on any of the pictures for a full-sized image. For comparison purposes, the same tests were performed on a William Optics ZS66 SD Doublet.
Prime Focus
William Optics 0.8x Reducer / Flattener v2
William Optics 0.8x Reducer / Flattener v3
Until we hit the very corners of this 23 x 16 mm CCD, the TMB 80 SS is doing a fine job. With the 28 mm diagonal of this chip, the usable area at prime focus is at least 22 mm in diameter (80% of max diameter). Not bad for prime focus. Both versions of the WO reducer do a reasonable job with this telescope as well. Personally, I prefer the v2 reducer here as the star shape on the v3 is a bit off, looking a bit astigmatic. In each of my test frames, the stars weren't quite round. I'm not sure the source of this and it could be that on another night, perhaps mounted a bit straighter, this setup would perform better. That said, with the v2 reducer, a good 90% of the maximum image circle demanded from this chip is very clean (25 mm) and honestly I can't find huge fault with the remaining bits. I'd feel comfortable using this full-frame, meaning we have a good 3.35 x 2.23 degrees of FOV. The v2 does a very nice job on this scope.
One of the differences between the v2 and v3 versions from William Optics is the size of the lenses in the reducer / flattener with the v3 optics being substantially larger. WO did this to address vignetting issues with large chips. Using ImagesPlus, I extracted a profile of the diagonal here (horizontal line in the above images). The curves are shown below.
From this, we can see that the v3 reducer does to a bit better in the vignetting department, but not by a substantial amount. Fitting a second-order polynomial to these data showed second-order fit coefficients of -0.037, -0.106, and -0.082 for the prime focus, v2, and v3 images respectively, quantifying the amount of vignetting present in each. Recall that these are stretched images (white point set at about 12% of max) and therefore the vignetting shown in the plots is amplified considerably. Without stretching, all images appear very even.