Looking for a recommendation for a ultra-wide angle lens

[ChrisPBacon]

I’m looking for and would sincerely appreciate your recommendations for the best prime EF-mount ultra wide-angle lens for use on Canon EOS crop-sensor (7D) and full-frame cameras (specifically 6D, and 6D Mk. II) for astrophotography.

(While there are excellent zoom lenses available, the possible risk of chromatic aberration makes them impractical.  That said, I’m a relative novice to astrophotography and still have a great deal to learn — so if you use a zoom lense for astrophotography, please let me know of your experience.)

I’m imaging asterisms and the Milky Way which require relatively longer exposures but I wish to avoid having stars appear oblong in shape from too-long an exposure; to that end, I need the fastest glass possible to obtain as much detail a lens is capable of capturing — as well as the quality of the detail that is captured. Canon’s 14mm f/2.8L is a good lens, but is it the best quality glass (assuming that cost is not a factor) on the market?  It must be heresy to ask if other manufacturers produce a better lens, but I’m willing to consider all contenders as long as the lens is compatible with my Canon cameras: price is not a consideration.

As I disable autofocus and image stabilization in taking these types of photos, a manual focus (only) lens such as those made by Zeiss is perfectly acceptable.

I’m well aware that below f/2.8, price increases dramatically: while having an equatorial mount would allow for longer exposure times without star distortion, it would create problems for foreground terrain blurring and additional post-production labor, which I’d like to avoid if possible. My 6D is heavily modified for heat reduction and has had its IR filter removed so that light in wavelengths of 656.28 nm appear in photos, but the scale of these structures make use of a telescope impractical.  Light pollution poses another difficulty with increased exposure times, so high-quality fast glass is a better option for my purposes.

Any lens f/2.0 or below, in the range of 8-18mm, might be ideal.  I would appreciate anyone with experience with this lens type sharing their recommendation.

[ChrisPBacon]

@Waddizzle wrote:

“Sounds like a hint to learn more about image stacking to me.

“I have dabbled with photographing the night sky using a Rokinon 14mm T3.1 with surprising good results for the shooting conditions.  There have been many complaints about the "photo" version of the lens.  Over time, I have bought three of the T3.1 cinema lenses, and they have all been fine.  I have not tried the newer Rokinon lenses, which communicate with the camera body, but are still manual focus.”

My mentor is one of those individuals who makes the best of his camera and peripheral equipment to get stunning photos: I have seen the photos he’d just taken and dowloaded onto his computer for review: he rarely, if ever, uses Lightroom or Adobe to alter or correct them. Of the 6 or 7 professionals I associate with, I’d say that he gets the most out of his camera and puts the group of us to shame. Steep learning curve stuff for me.

I just found a Rokinon 14mm T3.1 cine lens in a Las Vegas two days ago, and purchased it on your recommendation after trying it on my 6D Mk II. I’m driving to the California coast this afternoon to take some photos of the Milky Way, and to photograph the moon as it sets behind a lighthouse on the coast. That’s a great suggestion and I appreciate you offering it.

[TCampbell]

---

@ChrisPBacon wrote:
Thanks, Tim.

I have a Losmandy G-11 equatorial mount for 5-hour exposures, and a small Sky Watcher mount/tripod for shorter durations.

The 500 rule is a good start, but using it when utilizing the Canon EF 24mm f/1.4L II USM for 20 seconds causes the stars to appear oblong: I use 17 seconds and tweak the ISO.

---

Hi Chris.  The G11 is a fantastic mount (just make sure it's accurately polar aligned).  If your G11 has the Gemini 2, there is a Polar Allign Correction (PAC) mode (Scott Losmandy has a YouTube explaining how to use it.  Just go to YouTube -- there's a Losmandy YouTube channel).    I use a G11 ... and a GM8 ... and a StarLapse (I really like their mounts).

The "500 Rule" is meant for full-frame camera sensors (roughly 36mm x 24mm).  The 500 rule should be valid for your full-frame bodies.  For crop-frame bodies (APS-C at 1.6x crop-factor or APS-H at 1.3x crop factor) you would want to divide 500 by the crop factor.  E.g. if using a 7D body then you'd use 500 ÷ 1.6 = 312.5 ... and then divide the lens focal length into that for the number of seconds (e.g. with a 24mm lens it's about 13 seconds).   

The more liberal rule was the "600 Rule" ... but on close inspection some people find a slight amount of elongation (hence the "500 rule" to be a bit more conservative and most people are happy with that).  I'm wondering what camera you were using when you said you needed to cut the exposure time to 17 seconds.

I use a QHY "PoleMaster" to achieve highly accurate polar alignments (within a few arc-seconds of the true pole).  Equatorial mounts can use "modeling" to work out factors such as "cone error" or even moderate inaccurate polar alignment issues and creating a "model" (really only recommended for permanent observatories.  If you setup/take-down the mount each time you use it then a model is a waste of time) will only help you achieve more accurate "go-to's" but will not do anything for tracking accuracy.  With plate-solving software (astrometry.net and similar), the solving will give you ultra-precise go-to's and the rest is all about mount quality, balance (a slight imbalance where the "east" side of your mount is just a tiny bit heavier will help reduce "float" in the worm/spur gear backlash) and above all ... an accurate polar-alignment.   The PoleMaster camera aligns the "mount" -- not the "scope".   SharpCap also has a polar alignment feature ... but the field of view of a camera through a telescope is usually narrower than 5° -- so it needs a better initial estimated alignment to use it.

[ChrisPBacon]

I do an accurate pole alignment using the QHCCD Polemaster and have been using it over a year. I had a pad poured at my house for my telescope and despite its permanent mount, still do a polar alignment

Again, in case you missed it, I’ve used the 500 rule which, for a 24mm lens, gives the answer of 20.833 seconds.using 20 seconds, stars appeared slightly oblong so I’ve reduced the exposure to 17 seconds and adjusted the ISO upward to compensate,

[Waddizzle]

I use the EF 16-35mm f/2.8L II USM, and love it.  If it is not as sharp near the edges and corners, then I have not noticed it.  I have ot really looked for it, either.  I have noticed that it does seem to have a focus plane that is curved, with the camera at the center of a sphere or cylinder. 

I like the lens.   I cannot help but wonder if the soft edges and corners that people are observing are a side effect of lens correction.

[goalerjones]

I purchased the Samyang 14mm 2.8 for MWP. 1st time was bad, due to user error and not understanding infinity on a lens doesn't necessarily mean in focus, 2nd time was much better. I've recently added the Tamron 15-30mm 2.8 but have yet to take it back out to the desert for a go.

[ChrisPBacon]

---

@TCampbell wrote:

I'm wondering what camera you were using when you said you needed to cut the exposure time to 17 seconds.

---

Canon EOS 6D Mk II — and a highly modified Canon EOS 6D — with a Canon EF 24mm f/1.4L II USM lens.

[TCampbell]

---

@ChrisPBacon wrote:

---

@TCampbell wrote:

I'm wondering what camera you were using when you said you needed to cut the exposure time to 17 seconds.

---

Canon EOS 6D Mk II — and a highly modified Canon EOS 6D — with a Canon EF 24mm f/1.4L II USM lens.

---

Hi Chris,

As long as the G11 has a precise polar alignment (I typicaally use a "PoleMaster" system by QHY-CCD) then unguided exposure duration shouldn't be a problem.  I've done 8-minute un-guided exposures with mine with a 540mm APO on the mount (I use a TeleVue NP101is refractor).  The shorter the focal length, the more forgiving it is about tracking accuracy.  If your G11 mount has the Gemini 2 controller then there is a built-in polar alignment assistant you can use.  Losmandy has a YouTube channel and a tutorial video on how to use that feature.

But lenses also have optical issues which can causes the brighter stars to no longer be "round" when shooting long exposures.  The lower the focal ratio, the more you would notice these effects.  

Lenses don't create a perfectly "flat" field (even very expensive lenses).  If the lens is focused to tack-sharp in the center, objects in the corners will not be tack-sharp.  But you can think of field flatness applying to more than just focus... there's also flatness with regard to coma and astigmatism.  

When you shoot a lens wide open for long periods of time, it allows the photons to collect and can reveal issues such as astigmatism or coma that wont be easily spotted in traditional photography.  

Coma can cause stars to take on a tear-drop appearance.  Astigmatism can cause the stars to appear to grow tiny "wings".  If you check the orienation of these effects on various stars you should notice a pattern where they are oriented with respect to the center of the frame.  

If you stop down a bit, you can cut down on the issues.  Of course you are also collecting less light so now you need a longer exposure to compensate.  But even with the longer exposure times, the images are usually better.  

The G11 mount will *easily* handle very long exposures (I've done 8-minute un-guided exposures with a 540mm APO on the mount) ... as long as you have a precise polar alignment.  

If your non-round stars are oriented all pretty much in the same direction then this usually means you are experiencing some sort of tracking issue (e.g. mount did not have an accurate polar alignment, mount was bumped or moved, incorrect tracking rate, etc.)   If the non-round stars have an optical effect with an orientation relative to the center of the frame, then it is probably related to an optical effect of the lens (and stopping down will usually reduce or eliminate the issue).

Here's an example...

This first image is shot at f/2 (Canon EF 135mm f/2L USM) on a Canon 60Da.  The camera was on a tracking mount.  This was a 60-second frame.

To test tracking accuracy, I usually also shoot a long exposure.   Here's the longer exposure.

This was shot at f/10 (same lens, camera, etc.)  This was an 8 minute exposure (unguided)

Full disclosure... the first image is the converted CR2 into a JPEG with no adjustments other than Lightroom default profiles.  The second image was darker so I boosted exposure by .85 in Lightroom ... but no other adjustments.

Mainly look at the three stars on the left (this is the lower region of Orion turned on it's side).

Notice in the 2nd image at f/10 that the stars are diffraction spikes caused by the lens aperture blades but otherwise everything is tack sharp.

But also notice that the first image shows some little bat-wings coming away from the stars ... in the meridonal direction. (much easier to notice if viewing the image in a larger size).  If you imaging concentric rings around the center of the image, these bat-wings are extending along the direction of those rings.  

The more the lens is stopped down, the lens of the overall glass is used.   Mostly the center of the glass elements is used and that region tends to be optically better.  

Some guidance that an accomplished astrophotographer gave me when I started was to orient the camera so the top of the frame is "north up" (the viewfinder bump on the top of the camera should be pointed in the direction of the north star).  This makes it much easier to determine if elongation caused by tracking issues is a result of poor right-ascension tracking vs. declination drift.  If the issue is caused by alignment or mechanical issues with the mount, this makes it easier to diagnose.

Hopefully this info is helpful and lets you get much longer exposure durations.

[ebiggs1]

I would certainly like to hear you thoughts about the Tamron SP 15-30mm f/2.8 Di VC USD G2 Lens.  It should be a real nice lens. I have not used the G2 but the SP is very good. At 15mm and f2.8 I think you will not like the softness for astro work. But f4 on should be fine

"I use the EF 16-35mm f/2.8L II USM, and love it"

I don't, as a matter of fact I stopped using mine for anything. It certainly is not anywhere close to being capable of astrophotography.

[goalerjones]

I shoot the 15mm Tamron at 2.8 all the time, 2000ISO no softness issues whatsoever. If I could figure out how to link and pic from my phone I'd show you.

[ebiggs1]

Thanx, I would like to see a shot or two (from the G2) but not from your phone.  Directly from the camera/lens.  I know the SP is a very good.  The reason is I am so disappointed with my Canon EF 16-35mm f/2.8L II USM Lens. I use the Tokina opera 16-28mm f/2.8 FF Lens currently.