Some experience with the prototype on daylight targets, the waning crescent moon, Orion Nebula, dissimilar doubles, Pleiades, and Venus. Conditions were poor.
Two most striking differences were on the Moon and Venus. Neither appeared as "washed out" as before in looking for detail, but in the case of Venus the residual color stood out more - I wondered if in my next tests if I "tune" BFL to eyepiece I'll find a minimum point. That's supposed to be at best correction, around 1000mm EFL, but it certainly isn't what you get with a ~920mm FL you get with a stock 0.965 diagonal or a hybrid.
At the same 230x didn't notice color on the Moon's limb(!). What I did notice was that I could make out more craters near the limb than ever before, as well as the subtle hues that I usually see with a good refractor. In fact the view reminded of that more so than with the stock baffle. But what you don't see in a long refractor is HSA, and that was in evidence in looking at details inside of craters. Conditions were far from perfect to judge this matter.
The short answer is that a better primary baffle lets you see existing optical defects better, and more range of contrast/color.
Taking out the eyepiece and using the Moon to judge scattering inside the OTA, the baffle only shows the secondary mirror and the reflection of the primary in it. Correctly sized, it doesn't even show the secondary baffle, which is now superfluous. One cannot see the thin edge of the primary baffle, so the flat area of the original primary baffle is entirely removed - which was the objective. However in it's place is the diffuse illumination on the inside of the baffle, which wasn't apparent before with the original's "hole" - there was nothing to directly illuminate. Even sprayed flat black, it's apparent although the net trade is better. Microbaffles on the inside could reduce this further.
Also with this design is a larger opening at the top of the cone. That's because the closer to the secondary, the larger the opening has to be for the f/11 of this optical design to be not vignetted (it isn't). This also has increased the illuminated field size usefully (one of the main reason astrophotographers in the 1970's did this, to illuminate better a 35mm film frame. A 32mm eyepiece shows this readily.
For the first sample of this, I chose the smallest conical baffle design that would work with a 0.965 diagonal with a BFL to the base of the diagonal for a lowest power eyepiece with a adjacent field lens, as well as allowing the removal of the secondary baffle. This is the most convenient configuration to pack in my backpack, and it's also the most rugged - even a thin baffle has a moment arm, and dropping the scope would put force on the base where it attaches. This one is attached friction fit on the outside, so that no reduction in baffle base would occur.
Downsides of this mod were that it was very difficult to both form a proper cone, easy to knock it out of round, and hard to get a good mechanical connection at the base (it is only a prototype). And you have to be very precise in fabrication, because you have to exactly "outline" the secondary for this to work. If you are out of round, or off in dimensions, it's very apparent (you can tell lopsided or intrusion into optical path.
There appear to be less prominent benefits too. You can make out dissimilar doubles a bit better that were contrast reduced already by the large CO.
No negatives on DSO's like M1/M42. In the Pleiades could see some of the fainter orange stars. near the base of the "handle".
All of this leaves the question open of what happens with the removal of the secondary baffle as well - does the CO reduction help much? Also, does it change the star test then, because the large CO hides HSA as well as the HSA is compromised by some intentionally left LSA (which may not be present on this sample - part of the reason I got interested in playing around with it).
I did do a star test with just the A|B test of the factory/prototype baffle, and unsurprisingly nothing has changed with that (meaning the outside/inside additional area seen doesn't have zones to worsen things with the baffle being changed.
All of this has been a "control" for the greater experiment - to rework all of the baffles of the scope. I needed to determine that there wasn't other surprises present (did deal with little light leaks in the seam as I was fitting things) that could throw off the results. And before I go further, I need to a)improve the outside friction fit of the baffle, b) reduce the inside baffle tube glare, and c) stiffen the top so it is less out of round (glue didn't set up completely).
Am pleased that this optical sample might allow me to explore a larger scope of an experiment. If It had a turned edge , zone, or "hole", I'd worry that I'd be not measuring one change (baffles) but that and unmasking other flaws besides known HSA.
Other things I've been thinking about - making the baffle longer with larger diameter at the top, making the baffle even thinner and under tension (springs back?).
Considering a revised prototype and a consolidated change in baffles along with flocking to see if a maximal change provokes a substantially different result.
None of what I've done do I suggest you yet try. The sample here appears uniquely ideal to test this out on due to its relatively fewer flaws. Most/all other C90's I've encountered didn't have good collimation, DPAC, star tests, and planetary detail. It's likely if you do this you'll amplify the effect of an aberration. Indeed, I've already found that to be true on this sample as well. You may not like the result.
Leads me to believe that Celestron made this scope for high manufacturing yield, with the baffles a worthy compromise on top of many others.
I'm interested in eliminating artifacts here at the cost of raising a single one - residual HSA - to see if I can cancel (for this sample only) it, to see if it is possible to remove compromises and see what remains. Perhaps a form of "tilting at windmills".
And I'm still assessing if there is a useful project here to do. Or not. Let me know what you think.
Edited by wfj, 03 December 2018 - 02:52 PM.