Finally my new CDK20 arrived, after a 7 month wait, during which I was forced to image with the tiny NP127is refractor.
But it does produce much wider field images than the CDK20, so the time was not wasted.

Installation:
Installing this in my 10' Home Dome from Technical Innovations was both easy and hard.

Easy in that it sat right on the MEII Versa Plate (with an AP 7" saddle installed) and the MEII handles it just fine.

Hard in that it was so big it would not fit up my stairs.
My builder erected a scaffolding next to the open shutter and 4 of us man-handled it up through the shutter.
Also hard in that it wanted to run into several things that were already installed in the dome:
---a built in table on the West side
---a filter enclosure around the air conditioner on the East side
---the shutter motor box in the top of the dome (a real problem)
The first two were easy to fix, I tore out the table, and sliced 2" off the filter enclosure
The shutter motor box:  a real problem, it being rather essential to moving the shutter (it shows in the image)

The mount/scope overlapped the shutter motor by 3 inches, a major problem since both rotate and the dome never expects to run into anything.
Fortunately the 4-5/8" stainless steel bolts connecting the mount base with the concrete pier were 3.5" longer than absolutely necessary.
We took both scope and mount completely off the pier, cut the bolts off, and reassembled it.
Now I have an enormous 0.5 inch of clearance between the scope and shutter motor.
Whew!!
Bottom line: a CDK20 will fit in a 10 foot dome, just barely.
I can still image down to 20 degrees for the entire 360 degrees.

The end ring of the scope clears the dome completely on all sides except South, where it touches at 17 degrees, limiting my targets to the South to about 19 degrees.  Fortunately it only touches the base ring, nothing that will be moving, so not a problem if I mistakenly track into it.  The MEII will just instantly stop.  That feature I have had the pleasure of testing in real time.  One night I left my chair sitting a little too close to the mount.  Tracked right into it.  And get this: it touched right on the USB cable on the back of the camera.  Stopped so well and quickly, it didn't hurt my USB port.

In all other directions I can image down to about 10 degrees.  I actually have some usable (sort of) as low as 12 degrees.  Lot of muck down there.

Specifications:
Aperture           20 inch (508 mm)
Focal Length    3454 mm (135.98 inch)
Focal ratio        F/6.8
FOV:                  24'x36'
Central Obstruction    39% of the Primary Mirror Diameter
Weight              140 lbs (63.5 kg)
OTA Length      47 inch (1,194 mm)

With 0.66 focal reducer:
Focal Length    2280 mm (88.6 inch)
Focal ratio        F/4.5
FOV                    36'x54'

Image train:
CDK20
Hedrick Focuser
Focal Reducer 0.66x
ZWO 7x50 Filter Wheel
Chroma 50mm unmounted: LRGB, 3nm Oiii, Sii, 5nm Ha
ZWO ASI6200mm

Mount:  Paramount MEII with absolute encoders

Resolution:  0.34 arc second/pixel (with 0.66 reducer)

I initially configured the scope with the 0.66 reducer.
That barely fit, with a 1.5 mm connector between the reducer and the filter wheel.
That of course meant that I had to run a 2280 mm focal length scope, with 3.76 micron pixels, giving resolution of 0.34 arcsec/pixel, completely unguided.

Why not? 
Just use the piggybacked NP127is scope on top as a guide scope.
That unfortunately requires a nice clear view of sky for both scopes.
Technically they should be able to see out of my 36" shutter opening.
But many target paths have the scopes pointing through the shutter at nothing close to a right angle, so the effective shutter width is often closer to maybe 30". 
CDK30 + NP127is has a 29" image circle.  Very challenging.

So I ran unguided for 4 weeks.
200-300 second subs were just great, no visible trailing.
600 second I had to throw at least half away.
My polar alignment was excellent, around 15 arc seconds, so no improvement possible there.
So I gave up on the reducer.

Actually I am fine with not using the reducer.  Some targets benefited from the larger fov, but taking images longer than 300 seconds is rather critical.
And there is much stronger vignetting with the reducer.
So on balance it is working out best without the reducer.

I will publish a couple of targets with the reducer, before removing it.

REVISION #2: Focal reducer is back on, as of 12/19/2021.  Much better images, being so over sampled without the reducer, the 230% greater exposure on each pixel just wins big time.  I can/was guiding with the NP127is on top, until its camera quit working, so back to unguided.  PA is great so up to 300 second exposures working fine.

The black shroud is intended to block stray light, and perhaps cut down a little on dust on the mirrors.  An interesting problem appears if you just put it directly on the frame.  4 additional spikes appear in your images between the 4 major spikes, smaller but distinct.  These are caused by the shroud material forming flat panel surfaces.  To eliminate these, they need to be made to not be flat, some curvature is required.

I learned this BTW from John Hayes, who also has a new CDK20 which he has now sent down to Chile.  Thanks John for the great tip.  Check his listing out. Some great optics info there.

Planewave sells some very pricey curved rods you can install.  I chose to make some rings using dark gray strips cut from a gym floor mat.  The rubber mat is about 3/8" thick, 3" wide, and I used 2 of them.  You can see them in the images.  They are stiff enough to hold the shroud material out enough to produce a curved surface.  They are working perfectly, no additional flares or spikes in any images.

There are 7 fans on the CDK20, three on the main mirror end pulling air out of the tube, and 4 around the main mirror blowing air across the surface of the mirror.  This minimizes any air boundary layer on the surface of the main mirror that would degrade the images. Over these four I have taped filters to minimize dust accumulating on the main mirror.

For dust control, since my obs is located on a dirt road, I have two items:
There is a constant flow of filtered air entering the obs via a box I installed around the air conditioner.  Then there is a large room air filter.  Both of these run 24/7 every day.
Dust accumulation on any surface is now very minor.

Four cameras inside the dome and one on the weather boom outside the dome allow close monitoring of what the scope and dome are doing.  There are two on the CDK20, one on each side.  This allows me to see precisely how well the dome shutter opening is centered on the scope.  The bottom right two images are these two cameras.  I should be able to see sky on both sides, and close to the same amount of sky.  The shroud looks white in these images, but is really black, which shows up as white under the IR lights of the cameras.

And finally, all this stuff is completely controlled by NINA, the most spectacular imaging control software available today.  It has and continues to have extensive high powered development, with a large team of highly qualified professional programmers, who happen to be astronomers and love to contribute to NINA. 

The new completely programmable sequencer in NINA allows me to set my target details early in the day, and start the program on it's imaging path.  From that point on it completely controls the night's imaging, keeping track of the weather, opening the dome and starting the image session if weather is clear, closing up if bad weather comes, resuming later if it clears.  All processes are fully automatic, powering on/off the equipment, guiding, meridian flips, auto-focus runs.  I can ignore the whole thing until the next morning when I get back to my computer and find 12 hours of great images on my local pc, with the observatory completely buttoned up.
Spectacular!

My Collections:
Abell Planetary Nebulae (Complete)
Galaxies
Messier Objects
Planetary Nebulae
Sharpless 2 Objects