Completed - "fast f/4.7" CDK14 with OAG

You get the same FOV as the $50k IMX411 at native f/7.2 (green) Plus you don't need 65x65 filters. There is no perceivable loss in image resolution as it will be over sampled at 0.46"/pixel in almost all cases. The best seeing I got this season was a 1.7", the reducer image scale was still 3.7x that of the sky.

John Hayes:
Rouz,
Congratulations...that is one of the better focal reducers that I've seen.  It looks like it works quite well!

John

*Hi John,

Thank you for your feedback, appreciate it.
While they are relatively pricey, they work well as far as I could test.
The image circle is just large enough for a full frame sensor.

CS
Rouz

Hi Rouz,

a true deam setup, love it, congratulations! Now I have to seriously think about robbing a bank and/or selling my soul …

Can‘t wait so see images …

Matthias

*Hi Matthias,

Thank you, it is working well now and I would recommend this setup!

CS
Rouz
www.RouzAstro.com

John Hayes:

Rouz Astro:
You get the same FOV as the $50k IMX411 at native f/7.2 (green) Plus you don't need 65x65 filters. There is no perceivable loss in image resolution as it will be over sampled at 0.46"/pixel in almost all cases. The best seeing I got this season was a 1.7", the reducer image scale was still 3.7x that of the sky.

Rouz,
Your system with that reducer and camera is optimize for just slightly under 1" seeing so it's a good combination for those times when the seeing settles down a bit.

John

*Hi John,

I doubt it will ever get to those values here (Near Vancouver), I'm usually forced to down-sample to the 0.6 to 0.9"/pixel range. The data is captured with bin1 though.

I suppose the seeing in Chile should be great, I see some high resolution images coming out of your CDK20!

CS
Rouz

Hi Wei-Hao,
Its the Planewave 0.66x for the CDK14. It can support a full frame (but doubt any larger).

Its also due to the large 70mm image circle of the CDK14.
I tested the same reducer on the CDK12.5 and I doubt that would work well. It was very good for an APS-C with the CDK12.5.

CS
Rouz

Wei-Hao Wang:
Interesting.  So CDK 14 is the special one here.

Since CDKs deliver a corrected flat field, I kind of wonder if reducers for other flat-field scopes (like TAK FSQ, Vixen VC, etc) also work on CDK.  Any insight?

*I'm not sure, I tried the AP 2.7" reducer but the corners weren't great.
I know some IDK users had success with the IDK and TAK reducers with APS-C chips.

I suspect these scopes need a dedicated reducer with exotic glass, hence the price point of the reducers.

Wei-Hao Wang:
The AP reducers are for refractors, which have curved fields. These reducers "bend" the curved fields to flat ones.  So if we feed such reducers with a flat field, the flat field will be bend into a curved one.  So it's not surprising at all that it doesn't work.  What's really needed here is a reducer that maintains the field flatness.  That's the first-order requirement, I think.

*Refractor reducers won't work for sure.
I spoke to AP before testing the reducer and they said the 2.7 inch  "Passes on what it receives" with little to no correction.

It did produce an acceptable image with an APS-C with enough back focus to fit an OAG.

In any case, I doubt I'd find a better suited reducer for the CDKs than the purpose built Planewave one, and I didn't see the point in reinventing the wheel.

Wei-Hao Wang:
Agree.  For your CDK14, the reducer seems sweet.  There isn't really a need to find another reducer.  For other CDK, I am wondering if there are substitutes that can produce a 4cm flat field for FF format.

Wei-Hao,
I agree that in principle, a general reducer designed to hold a flat field should work.  However, it can be tricky to design a reducer that produces diffraction limited performance over the full field (defined by the size of the sensor) without coupling the design to the telescope that it’s going to be used on.  Furthermore, the alignment tolerances for the assembly will also be driven by the scope+reducer combination.  I think that this is why it’s so hard to find reducers that cover a large sensor with corner to corner pinpoint stars.  Ideally, the design should be in conjunction with the optics in the scope and the manufacturing and assembly tolerances have to be quite good.  It sure looks like Rouz has one that is excellent!

John

I forgot to mention I also tried the Vixen VCL200 reducer as I had it with an APS-C chip and I couldn't get it to work. The AP worked ok but not great. 
The PW reducer is working very well with both in terms of star sizes and illumination. The main drawback was the very short 44mm back focus and lack of OAG which I solved for the CDK14.

In general, yes, you are better off looking for a new sensor than trying to find another reducer for the CDK. 

The CDK20 I was fiddling with for a while was a bit different as the image circle was 50mm and the native back focus was shorter so the OAG solution probably wont work. There is a new f/7.2 version now I think, it may be different.

John has done some excellent work with the CDK20 and small CMOS pixels, binning definitely  is an option.


I see Moravian is now using the IMX411 sensor in the C5 with very attractive prices! I believe the IMX461 is larger than the KAF16803 (about 3mm longer diagonal). The next size up being the IMX411 is probably too big for the CDK20 anyway.

Wei-Hao Wang:
Hi John,

After the conversation with Rouz, I realize there is probably not an easy way to reduce our CDK20 and maintain good IQ over a large field.  Now my line of thinking is just to upgrade our 16803 CCD to a 44x33mm format low-noise CMOS camera.  Both 16803 and a 44x33mm CMOS have diagonals almost exactly the 52mm diameter of the image circle claimed by PlaneWave.  But the low read noise of CMOS allows to bin-down the pixels dramatically, making it almost equivalent to a focal reducer plus an FF camera that also fully occupies the reduced image circle.  This will be probably a lot easier than finding a suitable reducer.

Wei-Hao,
I think that's correct.  I've been using a QHY600M on my 20" and I'm pretty happy with it.  It has unit gain in Mode 3 at a gain setting of 25 and that give RN about 1/2 that of the 16803 with just about 65k well depth.  So if run in that mode and you bin 2x2, the RN will be about the same as the 16803--and the QE is significantly higher.  Overall, it is a VERY clean camera.  At first glance, the QHY461 appears to have similar numbers.  That looks like a great camera but it's roughly 5x the price of the 600M, which lies a bit outside of my "value-window".  I'll be interested to know what you think of it if you decide to take that path.  Under good seeing, the QHY600 is a better match than the 16803 on my F/6.8, 20" scope.  (I presented a moderately detailed analysis of this issue at AIC this year.  If you are interested, you can find that presentation here: https://www.advancedimagingconference.com/articles/secrets-long-focal-length-imaging-john-hayes.  The sensor discussion starts at about 22:45)

John

In my mind, the biggest limitation of the 461 for remote operation is the amount of data that it generates.  I prefer to bin locally so that means that I'm transferring full size images and even with the QHY600M, that's a LOT of data!  What I really need is the ability to operate the camera in "Cropped Mode".  That way I could simply work with much smaller data sets when I'm working on a small object.  I've submitted this request to the guys at SGP but I don't know how long it's going to take to implement.  It's hard to believe that they don't have this capability already implemented!  That would make  a 461 camera a LOT more palatable to me.

John