Camera lens advice

[Mr Thingy]

Hello SGL-ers!

Looking for some advice from you kind, knowledgeable people.

My HEQ5 is taking an eternity to arrive (1/4 of a year so far) and I'm not convinced it will even arrive in Dec so I'm exploring options that will allow me to take some snaps in the meantime.

I was considering buying a camera lens for some static-mount, wide angle imaging, to attach to my ZWO ASI 183 GT. I can then use guided/unguided when the HEQ5 arrives.

I was thinking about a lens with a FL of about 35  to 50 mm. 

My first question is whether it is suitable to get a lens for the smaller sensor size (e.g Canon APS-C)? My CMOS camera has a diagonal of about only 16 mm, so it seemed that with a standard lens for full frame (35 mm), my sensor would only capture half of the image.

My second question is how best to attach to my mount. Can you get a bracket to attach to the lens (maybe a mounting ring from a scope would work)?

Last question: is there any notable difference between Canon or Nikon lenses? Is one better suited to AP than the other.

I was looking at this (and the 35 mm equivalent):

https://www.intro2020.co.uk/Catalogue/ProductDetail/samyang-50mm-f1-2-csc-canon-m?productID=caaea451-50fb-4d26-8ed6-f4718d906c5d&catalogueLevelItemID=376dc0a0-f0a7-422d-a66b-27282b4cce5b

 

All other comments and recommendations welcome; come earn yourself some of the reputation points I know you love!

TIA

-Thingy-

Edited November 29, 2020 by Mr Thingy

[Alien 13]

I can offer some answers.

Nikon or Canon, quality wise nothing in it although Canon might be a bit cheaper. The Canon lenses do have a big advantage with their longer backfocus making it easier to fit filters into the light path.

The later Canon/Nikon or Sigma lenses operate with "fly by wire" systems so are totally unsuitable as the aperture and focus require the lens to be powered up to operate.

The Samyang range of lenses are well received and are totally manual but not cheap or you could look at the old M42 film camera lenses from takumar/minolta etc. 

Alan

[Mr Thingy]

5 hours ago, Alien 13 said:

I can offer some answers.

Nikon or Canon, quality wise nothing in it although Canon might be a bit cheaper. The Canon lenses do have a big advantage with their longer backfocus making it easier to fit filters into the light path.

The later Canon/Nikon or Sigma lenses operate with "fly by wire" systems so are totally unsuitable as the aperture and focus require the lens to be powered up to operate.

The Samyang range of lenses are well received and are totally manual but not cheap or you could look at the old M42 film camera lenses from takumar/minolta etc. 

Alan

Many thanks. That's helpful.

[Mr Thingy]

Does anyone know if the fitting is the same between Canon EF, EF-S and EF-M lenses?

I'm trying to figure out if the adaptors for fitting Canon lenses to CMOS cameras will work for the EF-M lenses. e.g this:

https://www.firstlightoptics.com/adapters/skywatcher-dslr-m48-ring-adapter.html

The EF-M lenses are cheaper and are for the smaller format, which seems like it would suit my ASI 183 sensor.

TIA

[vlaiv]

Not the same.

Canon EF and Canon EF-S have 44mm of flange focal distance, while Canon EF-M has only 18mm

This means that Canon EF-M lens expects sensor to be at 18mm distance behind it. You can't use Canon EF-M lens on Canon EF / EF-S body.

You can use Canon EF/EF-S lens on Canon EF-M body if you add appropriate spacer.

I'm not sure if mechanically they are the same, because as far as astronomy cameras are concerned - you dial in exact distance required by lens by use of extension tubes and spacers. Do be careful though, Canon EF adapter that I'm using already has 19mm of optical path taken up by adapter itself.

https://www.firstlightoptics.com/adapters/astro-essentials-canon-ef-lens-to-t2-adapter-for-cmosccd-cameras.html

This means that even if mechanically EF/EF-S and EF-M are the same, you would not be able to use EF-M lens with above adapter as it would take up all spacing and even 1mm more - you would not be able to focus at infinity.

Another thing to keep in mind is all the accessories you want to put in your optical train. I wanted to at least have rotator and some means to mount filters and if you want electronic filter wheel - that is not going to be easy as it takes up 20mm of distance - impossible with above adapter and ASI1600 - 19mm + 20mm + 6.5mm (of camera itself) is already 45.5mm - more than required 44mm.

[Mr Thingy]

Thanks @vlaiv

That helps a lot.

My camera has an integrated filter wheel that already takes up 26.5mm, so that rules out the EF-M lenses. 

I was looking at the adaptor you use and it's good to know that it's 19mm as that would be too much for my kit.

 

So I need to find another mounting option. Seems the camera has a 1/4" threaded hole, so I guess I could mount via that.

Edited November 30, 2020 by Mr Thingy

[Alien 13]

If you are struggling to get a camera and filter wheel into the imaging train have a look at the older medium format camera lenses, they have a huge imaging circle so will be well corrected even on a full frame camera and have lots of back focus.

Alan

[johngm]

26 minutes ago, Mr Thingy said:

Thanks @vlaiv

That helps a lot.

My camera has an integrated filter wheel that already takes up 26.5mm, so that rules out the EF-M lenses. 

I was looking at the adaptor you use and it's good to know that it's 19mm as that would be too much for my kit.

 

So I need to find another mounting option. Seems the camera has a 1/4" threaded hole, so I guess I could mount via that.

As Allen has already suggested, you could go down the M42 camera lenses, which you can pick up a quite good prices on flee bay, and they are actually quite good, especially if you drop them down a couple of stops

[vlaiv]

29 minutes ago, Mr Thingy said:

So I need to find another mounting option. Seems the camera has a 1/4" threaded hole, so I guess I could mount via that.

You can always get one of these:

https://www.teleskop-express.de/shop/product_info.php/info/p10146_TS-Optics-CCD-Adapter-for-Canon-EOS-lenses-to-M48---10-mm-length.html

and perhaps this to mount everything on:

https://www.teleskop-express.de/shop/product_info.php/info/p9891_TS-Optics-Telephoto-Lens-and-Camera-holder-with-Vixen-style-dovetail-bar.html

(there is losmandy version that is in stock - this one is going to be quite a bit of wait for stock to replenish).

[vlaiv]

On 29/11/2020 at 12:06, Mr Thingy said:

I was thinking about a lens with a FL of about 35  to 50 mm. 

In reference to original question - these lenses are not diffraction limited lenses. This means that you can't judge sharpness like normal telescope and regular sampling rate does not apply here.

If you want something in 40mm focal length range, then perhaps look at 85mm or 135mm range, bin and do mosaic.

Why do I say you should bin your data? Well, if you look at specs for almost all lenses - their sharpness is determined on basis of 30 lpmm - 30 lines per mm. Yes, 30 lines per mm equates to 1000 / 60 = 16.666... µm pixel size (we divide by 60 as 30 lines is actually 60 line pairs - each line is white line on black background - twice width). If lens is said to be sharp at 30lpmm - this really means that you need large pixels to keep it sharp (in any case larger pixels than most newer CMOS cameras have).

I imaged with ASI178 and Samyang 85mm T1.5 lens (which is the same as F/1.4 - only cinema version without click stops on aperture ring).

At F/1.4 - F/2.0 there is chromatic aberration present. At F/2.8 it is ok as far as star shapes are concerned (well, you get bunch of spikes and it is better to stop it down via filter thread / step down ring then with regular aperture stop), but it is still not good even for 4.8µm pixel size (2.4µm native of ASI178 in super pixel debayer mode).

I estimate that lens would be sharp at about 7-8µm pixel size - or double above.

Here is what it can do in that configuration - super pixel mode + x2 bin (effective pixel size of 9.6µm):

At this resolution - image looks sharp enough. Since it was taken at F/2.0, there is still some bloating due to chromatic aberration around bright stars (they are not quite "pin point").

Thing is - that mosaic way of working will take the same amount of time as doing it with shorter FL lens with respect to imaging time. Instead of doing single frame for one hour, for example - you would do 4 panels, 15 minutes each. Since you'll be software binning - it will boost SNR by factor of two for each panel - exactly the same thing as stacking x4 more frames (SNR increases by square root of stacked frames).

Problem is that mosaics are not easy to process. You have to be careful of any gradients and remove them before stitching everything together.

[Mr Thingy]

That's immensely helpful @vlaiv

I appreciate you taking the time.

-thingy-

Edited November 30, 2020 by Mr Thingy