A few eyepiece tests

[Mr Spock]

I decided to have a close look at a few eyepieces, mainly to figure out what's 'wrong' with the LVW 42mm and why there's disagreement about its field size.
As with all tests of this kind it is flawed: I only have an iPhone to get the images and the quality isn't any where near as good as the real thing; noticeably the sharpness, contrast and focus are off, as well as uneven illumination and edge issues. Still, it gives a good idea. Other equipment used is a Celestron 80mmED at 600mm focal length with the target about 6m away.

Firstly, the aforementioned LVW 42mm. Some say 72° on the barrel, some say 65°. So which is it? The answer is 'yes'! OK, let's clarify. When you look through it, it definitely looks 72° - between a 65° LVW and an 82° Nagler visually - , but, when you measure it, it's only 65°. Confusing isn't it - read on. The problem is the eyepiece has huge amounts of distortion. The image at the edge is much larger than the image in the centre.
It's only actually 42mm in the centre... So, if you average out the magnification, it weighs in as a 38mm 72° eyepiece. To me that makes much more sense. It's not a 42mm 72° eyepiece, nor is it 65°!

You can see the distortion in this first image, though it is very sharp. The fall off in sharpness at the edge is the iPhone, visually it looks fine. There is a bit of CA near the edge which disappears if you move your eye around.

Next in line for scrutiny is the venerable Meade 4000 series 32mm Plössl. A nice bright and sharp view but again with significant distortion. Crisp right to the edge though.

Next is a Circle-T Orthoscopic 25mm. Small field of view of course, but what's there is very sharp and contrasty. Looking though this, and the Ortho below, they are noticeable brighter than the other eyepieces on test.

The LVW 22mm - a much sought after eyepiece and with good reason. Very sharp right across its 65° field. Not as much distortion as the other eyepieces - visually it looks fine so what you see is down to the iPhone. 
There is a blue ring of fire - same goes for all the LVWs I have.
I use this 80ED LVW 22mm combination quite often: stars are the tiniest points of light imaginable - a wonderful dark sky combination.

We'll try a different size LVW, the 13mm. Same blue ring of fire and sharpness as the 22mm, but a little CA creeping in - I can't say I've noticed this when observing. This one I use in a 250mm Newt for galaxies and nebulae. It's really good at it too.

NLV 12mm. Another very sharp eyepiece in use. It does have some CA. What you can't see in this image is on bright objects like this there are some reflections inside. Looking at the original image on my phone I can see the pattern reflected over about a square's length. Not good.

Last one, a Circle-T 12.5mm. Another very sharp eyepiece with only the small field of view as a negative. Not a good image but you can see the quality shining through. In my 250mm Newt with a Meade 140 Barlow this gives x225 and reveals the finest of detail.

I don't have any more eyepieces of similar focal lengths to test. Hopefully you've enjoyed looking at these few though.

[Stu]

Very interesting comparisons Michael. Those orthos do look bright and sharp, don’t they? The LVW 22mm looks good too in terms of wider field but limited distortions. Thanks.

[vlaiv]

I love this kinds of tests, however, I have to point out couple of things.

If you want to do this kinds of tests properly - you need large, sharp, fast lens and DSLR as best option.

For example, say in your test above - you used F/7.5 scope (80mm / 600mm) with 42mm eyepiece. That gives you exit pupil of 5.6mm. Problem is that your phone lens will not accept that large exit pupil.

Phone cameras being very compact in size - use very small sensors and very short FL lenses. For example iPhone 11 has 26mm equivalent lens at F/1.4. Sensor size is 1/2.55" (crop factor of about x6), so actual lens focal length is about 4.3mm. With F/1.4 - it has aperture size of only 3mm - it can't accept exit pupil of 5.6mm - lens is not wide enough.

 

[Mr Spock]

Just as a diversion, here's another. Light is fading and this is pushing the iPhone to its (noisy) limits - also my hand holding skills! On the other hand there is less stray light creeping in which spoils some of the other images.

The LVW 22mm again, this time with a Meade 140 Apo Barlow. This 2x Barlow consistently measures at x2.4. 

[globular]

The formula linking focal length of eyepiece (fl), field stop of eyepiece (fs) and apparent field of view of eyepiece (afov) is: 
     afov = (fs / fl) * (180 / π)

This assumes everything is uniform and, while "Ya canny change the laws of physics, Jim" everything is, as you rightly point out, anything but uniform in most eyepieces.

If you plug both your "adjusted for uniformality" measurements of 42 65 and 38 72 into the formula and solve for fs you get ~48mm both times.  And 48mm is the clear aperture of a standard 2" barrel.  So that sounds nice and consistent.

Most eyepiece manufacturers would put a fs in an eyepiece - rather than relying on the barrel - and typically most don't go over 46 or 47mm to avoid "nasty" effects at the edges.  e.g. my Pentax XW40 has a fs of 46.5 and gives an afov of 67 (rather than the round 70 reported by the marketing gurus for all the XW range).

If the LVW42 has a field stop of 47 then assuming a uniform fl of 42mm would give afov of 64.1; and assuming a visual afov of 72 this would give an adjusted fl of 37.4mm.

All nice and close to your measurements - so looks like you made a good job of it 

[Mr Spock]

I've just measured the field stop. You have to take the nosepiece off to see it as the interior of the eyepiece is a little recessed; and, it's exactly 48mm.

[Don Pensack]

Your pictures show normal amounts of rectilinear distortion in the form of pincushion.

Distortion of this type means the timing of a star across the field will likely yield a field stop that, in turn, yields a smaller apparent field than the one seen.

The outer edge of the field is stretched radially, as is the case with pincushion distortion.

It is quite normal.

 

A 48mm field stop is a bit large for a 42mm eyepiece, indication they accepted a small amount of vignetting as a compromise.

A 65° field calculates to a 47.65mm field stop.  48mm translates to just a little over 65°.

So how do you get 72° out of the eyepiece?  By stretching the edge.

It's the same in the 24mm TeleVue Panoptic, which has the same form of distortion and where the focal length and field stop don't match.

It should have an apparent field of 64.5° for its 27mm field stop diameter but one sees a 68° field in the eyepiece.

 

It's also likely both the 65° and 72° figures are not accurate.

 

[Zermelo]

4 hours ago, Don Pensack said:

So how do you get 72° out of the eyepiece?  By stretching the edge.

Interesting discussion. I think vlaiv is talking about the same rectilinear/angular distinction here.

So some of the discrepancies between claimed fields of view and the upper limit of the barrel format are not necessarily typos (or outright lies), they could be nuances of distortion. I suppose it's a subjective point as to whether you'd prefer a more "honest" 65° field or a more "immersive" 72°, even if you're not getting any more TFOV with the wider field.
I also recall (but can't find) a recent thread discussing astronomical eyepieces vs spotting scope eyepieces, which said that the design compromises were weighted towards different types of distortion in each case (and therefore an EP designed for spotting scopes might not perform so well in a telescope).

[Louis D]

22 hours ago, globular said:

Most eyepiece manufacturers would put a fs in an eyepiece - rather than relying on the barrel - and typically most don't go over 46 or 47mm to avoid "nasty" effects at the edges.  e.g. my Pentax XW40 has a fs of 46.5 and gives an afov of 67 (rather than the round 70 reported by the marketing gurus for all the XW range).

I've measured my 40mm Pentax XW-R to have a 46.2mm field stop, a 70 AFOV (via projection), and a 66 degree eAFOV (effective AFOV) based on the FS and FL.  I think you are referring to eAFOV rather than projected (perceived) AFOV.  Go back and remeasure your AFOV using the projection method and see if yours isn't also 70 degrees.

[Louis D]

23 hours ago, Mr Spock said:

Firstly, the aforementioned LVW 42mm. Some say 72° on the barrel, some say 65°. So which is it? The answer is 'yes'! OK, let's clarify. When you look through it, it definitely looks 72° - between a 65° LVW and an 82° Nagler visually - , but, when you measure it, it's only 65°. Confusing isn't it - read on. The problem is the eyepiece has huge amounts of distortion. The image at the edge is much larger than the image in the centre.
It's only actually 42mm in the centre... So, if you average out the magnification, it weighs in as a 38mm 72° eyepiece. To me that makes much more sense. It's not a 42mm 72° eyepiece, nor is it 65°!

You should be able to directly measure the AFOV using the projection method to within a degree of accuracy.

If you have some eyepieces with known good FS diameter values (such as TV ones), you should be able to extrapolate a conversion factor from the number of millimeters you see on your ruler through the camera lens to the FS diameter in millimeters.  That's how I've been able to determine FS values for all of my eyepieces based on known good values for some of my eyepieces.  I'm usually within 0.2mm of the value from other sources such as Ernest of Russia's measurements.

From the FS and FL values, you can then calculate the eAFOV (effective AFOV) using mathematics (FS/FL*57.3, IIRC).

[Louis D]

20 hours ago, Mr Spock said:

I've just measured the field stop. You have to take the nosepiece off to see it as the interior of the eyepiece is a little recessed; and, it's exactly 48mm.

It's not in the insertion barrel?  It must take a lot of in-focus to reach focus with that eyepiece.

[vlaiv]

1 minute ago, Louis D said:

You should be able to directly measure the AFOV using the projection method to within a degree of accuracy.

How does that work?

What about simple drift timing method?

[Louis D]

23 hours ago, vlaiv said:

If you want to do this kinds of tests properly - you need large, sharp, fast lens and DSLR as best option.

Unless you can find a DSLR lens with a tiny lens depth wise or are using an eyepiece with a large amount of eye relief, you'll never be able to get the exit pupil of the eyepiece to coincide with the entrance pupil of the lens (generally where the iris resides).  As such, you won't be able to image the full AFOV of the eyepiece.  That's why today's cell phone cameras are so fantastic for this application.  High resolution, wide angle field of view, well corrected, and insanely short distance to the entrance pupil.

23 hours ago, vlaiv said:

Phone cameras being very compact in size - use very small sensors and very short FL lenses. For example iPhone 11 has 26mm equivalent lens at F/1.4. Sensor size is 1/2.55" (crop factor of about x6), so actual lens focal length is about 4.3mm. With F/1.4 - it has aperture size of only 3mm - it can't accept exit pupil of 5.6mm - lens is not wide enough.

This is completely irrelevant for imaging the AFOV to determine eyepiece characteristics.  The image will be dimmer than if it could have accepted all the photons available in the exit pupil, but the entire AFOV will be visible.  It's no different than looking at the full moon with the human eye through a low power eyepiece.  The eye's entrance pupil will be narrower than 5mm because the full moon is too bright to allow it to fully dilate.  However, a low power eyepiece could very well be producing a 5mm or larger exit pupil at the same time.  This has no effect on seeing the entire AFOV which is what we're trying to achieve here.

[vlaiv]

1 minute ago, Louis D said:

Unless you can find a DSLR lens with a tiny lens depth wise or are using an eyepiece with a large amount of eye relief, you'll never be able to get the exit pupil of the eyepiece to coincide with the entrance pupil of the lens (generally where the iris resides). 

Not sure that I understand this.

Say I have 85mm F/1.4 lens - that lens has aperture of about 60.71mm. This is regardless of where it's iris is.

Exit beam of light from eyepiece is collimated and it has certain diameter of "pencil" - regardless where you position that "pencil" in lens aperture - it will be focused on camera sensor.

It works the same as telescope - if you put aperture mask with small opening on refractor - it does not matter where you put that aperture mask opening (it does matter as far as lens figure is concerned - but not as far as light throughput / vignetting and distortion is concerned).

With large lens - it does not matter if you put it before exit pupil position or behind exit pupil position - pencils will still hit aperture (left is eyepiece and right is large lens - it has aperture over whole rectangle side).

Additional benefit of long focal length lens is that you'll magnify image from eyepiece significantly and lens aberrations will be minimized in comparison to eyepiece aberrations.

Only drawback is that you need to shoot multiple images in order to cover whole FOV of eyepiece and you need to stitch those images together using clever software like Microsoft ICE.

10 minutes ago, Louis D said:

This is completely irrelevant for imaging the AFOV to determine eyepiece characteristics.  The image will be dimmer than if it could have accepted all the photons available in the exit pupil, but the entire AFOV will be visible.  It's no different than looking at the full moon with the human eye through a low power eyepiece.  The eye's entrance pupil will be narrower than 5mm because the full moon is too bright to allow it to fully dilate.  However, a low power eyepiece could very well be producing a 5mm or larger exit pupil at the same time.  This has no effect on seeing the entire AFOV which is what we're trying to achieve here.

I'm not entirely sure - eyepiece aberrations will be "imprinted" into wavefront of exiting light "pencil" - if you cut away part of that wavefront, you'll effectively throw away aberrations.

It is like putting aperture mask on telescope to mask turned down edge for example.

[Louis D]

9 minutes ago, vlaiv said:

How does that work?

What about simple drift timing method?

Put the eyepiece in a telescope, point a flashlight (UK torch) into the telescope, put an eyepiece in the focuser (a diagonal is handy here), and aim the eyepiece at a wall.  Move the light around until you get a nice, sharp circle showing the entire AFOV all the way out to the field stop.

Now, measure the diameter of the projected circle and the distance from the top of the eyepiece to the wall.  Make sure to keep the top of the eyepiece parallel to the wall.  You want a nice circle to measure.

Next, using a white box (or card stock), determine the position of the exit pupil by inserting it between the eyepiece and the wall.  Move the box/card toward the eyepiece until the image circle is minimized.  If there is gross CAEP, this can be difficult to determine.  For most eyepieces, it's fairly obvious, though.  Now measure the distance from the top of the eyepiece to the box/card.  This is the usable eye relief.

Lastly, use the following formula to calculate the AFOV:

2*arctan[(Circle diameter/2)/(Eyepiece to wall distance - usable eye relief)]

You're just calculating the half angle of the projected circle as seen from the exit pupil with trigonometry (tangent(angle) = opposite/adjacent) and they multiplying it by 2.

The drift timing method is a good way of measuring the true field of view (TFOV) of an eyepiece, and thus its field stop, but it tells you very little about its AFOV due to variable magnification distortion across the FOV.

[Louis D]

19 minutes ago, vlaiv said:

Not sure that I understand this.

Say I have 85mm F/1.4 lens - that lens has aperture of about 60.71mm. This is regardless of where it's iris is.

Exit beam of light from eyepiece is collimated and it has certain diameter of "pencil" - regardless where you position that "pencil" in lens aperture - it will be focused on camera sensor.

All I can say is go out and try afocal projection with that setup.  You'll find out really quickly that you'll only be able to image the very center of the AFOV because you can't get the camera lens in close enough to match exit/entrance pupils.  The objective lens will bump into the eyepiece unless you're using an eyepiece with several inches of eye relief to project that AFOV well behind the eyepiece.  I've been doing afocal projection photography for well over 20 years, and until the advent of small objective lenses with early digital cameras using small sensors in the early 2000s, the method was not very usable except with low power, long eye relief eyepieces.  It's literally no different than trying to see the entire AFOV of an eyepiece from anywhere except with the eye's entrance pupil positioned at the eyepiece's exit pupil.  The cell phone camera replaces the human eye.

19 minutes ago, vlaiv said:

dditional benefit of long focal length lens is that you'll magnify image from eyepiece significantly and lens aberrations will be minimized in comparison to eyepiece aberrations.

Only drawback is that you need to shoot multiple images in order to cover whole FOV of eyepiece and you need to stitch those images together using clever software like Microsoft ICE.

The software will have to get rid of keystone distortion as you point it off axis.  As someone who doesn't do image stitching at all, I have no desire to go down that route.  Another problem can be getting the camera tipped correctly when eye relief is very limited since you want to keep the lens's entrance pupil at the eyepiece's exit pupil as you collect images.  If the camera has any significant width, and if the eyepiece is not a volcano top, the two will bump into each other.  You can back up the camera in this case, but you will capture an even narrower sliver of the AFOV.

19 minutes ago, vlaiv said:

I'm not entirely sure - eyepiece aberrations will be "imprinted" into wavefront of exiting light "pencil" - if you cut away part of that wavefront, you'll effectively throw away aberrations.

I think this only comes into play if the camera's f-ratio is slower than that of the telescope's f-ratio.  Since cell phone cameras are very fast, this is never an issue.  Michael Covington over on CN would be the best to clarify this question.

Edited June 3, 2021 by Louis D

[vlaiv]

21 minutes ago, Louis D said:

Put the eyepiece in a telescope, point a flashlight (UK torch) into the telescope, put an eyepiece in the focuser (a diagonal is handy here), and aim the eyepiece at a wall.  Move the light around until you get a nice, sharp circle showing the entire AFOV all the way out to the field stop.

Imaging flat box is excellent tool for this, and yes - everything else is self explanatory.

5 minutes ago, Louis D said:

lens's entrance pupil at the eyepiece's exit pupil as you collect images

I still don't get this part - isn't lens entrance pupil in fact its aperture for collimated / parallel rays (object at infinity)? If so - large fast lens will have aperture much larger than exit pupil of eyepiece.

6 minutes ago, Louis D said:

I think this only comes into play if the camera's f-ratio is slower than that of the telescope's f-ratio.  Since cell phone cameras are very fast, this is never an issue.  Michael Covington over on CN would be the best to clarify this question.

That is exactly my point - fast phone lens will become small phone lens if its aperture is smaller than exit pupil of eyepiece - it will stop down exit beam because it is simply smaller than that - it is the same case like we have in using eyepieces with very large exit pupils - our eye becomes aperture stop if exit pupil is larger than about 7mm. Similarly if aperture of camera lens is smaller than exit pupil - it will become aperture stop.

[Don Pensack]

2 hours ago, vlaiv said:

How does that work?

What about simple drift timing method?

Simple drift timing will yield the true field and the field stop, but not the apparent field.

There is an easy way to measure the apparent field:

https://www.cloudynights.com/topic/574401-an-easy-way-to-measure-apparent-field-of-view/?p=7959408

and the posts that follow.

Measure carefully, and you can easily get within 0.5° of apparent field.

[Louis D]

1 hour ago, vlaiv said:

That is exactly my point - fast phone lens will become small phone lens if its aperture is smaller than exit pupil of eyepiece - it will stop down exit beam because it is simply smaller than that - it is the same case like we have in using eyepieces with very large exit pupils - our eye becomes aperture stop if exit pupil is larger than about 7mm. Similarly if aperture of camera lens is smaller than exit pupil - it will become aperture stop.

In your experience, does an astigmatic eyepiece look sharper when your iris is smaller?  Not in my experience.  The astigmatism in my own eye becomes less noticeable since I'm only viewing through the central part of my cornea, but the eyepiece image does not improve.

[Don Pensack]

10 hours ago, Zermelo said:

Interesting discussion. I think vlaiv is talking about the same rectilinear/angular distinction here.

So some of the discrepancies between claimed fields of view and the upper limit of the barrel format are not necessarily typos (or outright lies), they could be nuances of distortion. I suppose it's a subjective point as to whether you'd prefer a more "honest" 65° field or a more "immersive" 72°, even if you're not getting any more TFOV with the wider field.
I also recall (but can't find) a recent thread discussing astronomical eyepieces vs spotting scope eyepieces, which said that the design compromises were weighted towards different types of distortion in each case (and therefore an EP designed for spotting scopes might not perform so well in a telescope).

In spotting scopes, it's important to keep straight lines straight across the field.  So a design for a spotting scope eyepiece would reduce rectilinear distortion to a minimum, leaving in angular magnification distortion at the edge.

For astronomy, it's more important to maintain the same size and separation of points across the field, so angular magnification distortion is reduced to a minimum, leaving in a fair amount of rectilinear distortion.

Tracking scopes seem to tolerate both forms of distortion, while scanning scopes usually prefer a very low RD.

But, distortion is distortion, as they say, and if you don't want distortion in the field, confine the eyepiece purchases to narrow fields of view.

[vlaiv]

Just now, Louis D said:

In your experience, does an astigmatic eyepiece look sharper when your iris is smaller?  Not in my experience.  The astigmatism in my own eye becomes less noticeable since I'm only viewing through the central part of my cornea, but the eyepiece image does not improve.

I'm not sure that I ever had a chance to look thru astigmatic eyepiece when my pupil was smaller than exit pupil of eyepiece.

For that I would need really fast telescope and long focal length eyepiece - or perhaps really destroy my night vision with flash light to conduct a test?

[Louis D]

3 minutes ago, Don Pensack said:

Measure carefully, and you can easily get within 0.5° of apparent field.

Only for eyepieces with sharply defined field stops.  For eyepieces that have fuzzy field stops due to being poorly placed or non-existent (barrel defined or field lens defined field stops), you'll be lucky to nail down the AFOV to within a degree.

[Louis D]

1 minute ago, vlaiv said:

I'm not sure that I ever had a chance to look thru astigmatic eyepiece when my pupil was smaller than exit pupil of eyepiece.

For that I would need really fast telescope and long focal length eyepiece - or perhaps really destroy my night vision with flash light to conduct a test?

Sometime try looking at a bright star at the edge of a low power Erfle, Kellner, or other poorly corrected eyepiece, and then swing the scope over to the full moon and observe for a while.  Now, quickly swing the scope back over to the bright star.  Does it look sharper at the edge now?

[Louis D]

1 hour ago, vlaiv said:

I still don't get this part - isn't lens entrance pupil in fact its aperture for collimated / parallel rays (object at infinity)? If so - large fast lens will have aperture much larger than exit pupil of eyepiece.

Next chance I get, I'll dig out my Sigma 50mm f/1.4 lens (with 1.6x DSLR crop factor on my Canon Rebel T3i) and try to image a typical eyepiece's AFOV just to illustrate the result for you.

[Mr Spock]

Being a photographer I can say for certain my DSLR will not image an eyepiece with a lens attached. All you will get is an out of focus image of the eyepiece top and its surroundings.

It will do eyepiece projection of course but I have no means to connect it.