Reducing the Image Scale of the Quark

[marktownley]

Ok, I love my Quark, it is a real sweet piece of kit, but my big time bug bear with it is the scale that it images at; 4.3 times the native focal length of the scope it is working on. I understand why it has to do that, but I don't want it to do that. This is hugely oversampled when imaging, and my experience of such oversampling is that the image reduces in contrast, when you look at the dynamic range on the histogram it is compressed, also when imaging exposure time is increased to name a few of the annoyances...

So, I decided to see what can be done with this using a focal reducer. Nothing special, it is an 'unbranded' 0.5x focal reducer, I think it cost me about £15 when I bought it, so, lets get this straight i'm not talking cutting edge optical quality here. The simple rule with a reducer is the further the lens is from the chip of the camera the greater it reduces the image scale. However there is a caveat, the more this reducing factor the greater the spherical aberration and the greater the field curvature; basically things go down hill.

With today being the third day in a row where i've had some sun to play with I decided to do me some experiments, nothing too scientific, but you get a flavour: The shots are over 2 days of the same active region. Now I know there will be some 'movement' between the 2 spots, but essentially the distance between them is constant enough to give us a comparison.

All shots here with the ED80 (f7.5) and the DMK31.

First off native resolution with the quark on the scope to give an efl about 2600mm. Way too much! (IMHO)

ar12282-ha-full-size by Mark Townley, on Flickr

Yuck! Soft, flat and lacking in contrast.

Next up, the '0.5x' reducer on a short (15mm) nosepiece...

ar12282-ha-bw by Mark Townley, on Flickr

I still think the image needs a bit to be desired. Measuring the distance between the 2 spots tells me i'm getting a reduction factor somewhere around 0.7x.

So to todays part of the experiment. I stuck an extension on the short nosepiece and the reducer on this, to get me up to around 23mm from the face of the camera. Here are the results:

ar12282-ha by Mark Townley, on Flickr

This image is starting to look a bit more like the standard I like...

However, look closely, there is a focal shift across the image from left to right. It later materialises this is because the 'extension' piece I used (half of a variable polarising filter without the filter in - the half that 'rotates') does not sit square at all, infact there is slightly more than 1mm shift in 'height' from one side of the cell to another. If I get day 4 on the sun I have already found a more appropriate and 'square' extension section. Measuring the distance between the 2 spots indicates the magnification factor I have here is 0.5x. I tried a few more shots:

filaprom-remains by Mark Townley, on Flickr

Notice top right struggling with focus - the quark is very sensitive to focus with its telecentric output beam, it is either in focus or it is not, so a tilted focal reducer will not be ideal at all! Next up a bit of 'mid disk'.

ar12284-ha by Mark Townley, on Flickr

You'll notice that right side out of focus again, but the keen eyed will notice the corners are darker, so, maybe a bit of vignetting coming in?

Either way, running at 0.5x means the efl is about 1300mm with my setup, which is pretty close to the nyquist ideal of about 1100mm based on my setup. With a bit of tweaking and perseverance this may well just work...

[grimble_cornet]

Very interesting set of images Mark

I did something similar - but unintentionally - when I first tried an old/unbranded x0.5.
The problem was that the Grasshopper has the sensor much further inside the case than the DMK and when the reducer was added to a standard nosepiece its lens centre was about 55mm from the sensor. This produced a whole disc with the Equinox 80 but with all the problems you would expect:

I tried it with an old 'cut-down' nosepiece but discovered the non-square problem

I then ordered a much shorter 15mm nosepiece and got better results but still with focus/even illumination problems which might have had something to do with my first Quark which had serious banding problems and was replaced?

New Quark plus new x0.5 focal reducer (cheapo Revelation as used by others with success) gave much better results and I now use a 30mm nosepiece and get quite good 'compromise' results.

Like you Mark, I have found that a couple of mm makes a significant difference and would like to experiment more with this if I could find a way of adjusting the distance by small increments while maintaining accurate optical alignment.

Look forward to seeing any further trials -

[swisswalter]

Hi Mark

a perfect lession, thank you. I learn every day something on that wonderful site

[Valery]

Mark,

Nice investigation, thanks. You overtake me with my more serious and detailed investigations
I started last summer in a warm season.

However back to the topic: your best result (resolution, details detection) is at a 0.7x reduction.

The Nyquist ideal does not work in lunar and planetary imaging. You need to be at least at 1,5x the Nyquist
to be sure you extracted all the details your instrument is capable of.

I hope we both will continue these investigations - you so much like all kinds of experiments with narrow band solar equipment!

My own investigations tell me: if you like to have the results from Quart at it's best - no cheapest and easiest solutions.
1. At least F/10 for feeding objective
2. Dedicated focal reducers (not "al cheapo" simple cemented doublet !!! ) at 0,3x to 0,5x (at the F/10 of the deeding objective).
I mean for 3,75micron pixel size. The smaller the pixel, the larger reducing is required.
3. Reducer must be designed taking in to account the telescope's own SA (at 656nm). Otherwise the results will be less satisfying.
When imaging at high resolution, the difference between L/2 lambda SA and zero SA is huge and all between.
4. Good mechanics for reducer, good and very smooth 1:10 focuser.


Valery.

[Derek Klepp]

Thanks Mark as with my old Daystar scope the reducer was a necessity.