Astrophotographer contemplates extreme macro

[Lou Jost]

Lothman, on a normal camera only one pixel in every four-pixel block takes data on red light, as you know. So if an object has fine red detail, the sensor is essentially using only a quarter of its pixels. If it could use all its pixels, then it would act like a 20 Mp sensor rather than a 5 Mp sensor.

But you are right, the linear resolution (as lines per mm or per picture width) would only increase by the square root of the increase in pixel count. So it would increase by a factor of two for red or blue details, and a factor of 1.4 for green details. There might not be any increase in resolution at all if the details all had components in all three colors.

[billjanes1]

AtmosFearIC, welcome aboard!

The obvious place to start is with the Mitutoyo M Plan Apo 5X NA 0.14, which delivers that magnification in combination with a 200 mm tube lens.

However, on APS-C sensor, everything in this series can be either used at rated magnification on a 200 mm tube lens, or pushed down to lower magnification by using a shorter than 200 mm tube lens. For 5X, the sharpest lens in my arsenal is the Mitutoyo M Plan Apo 10X NA 0.28, pushed down with a 100 mm tube lens.

One caveat on the concept of pushing down: you probably cannot do that by using a zoom lens on the rear. Zoom lenses are notoriously bad at vignetting when used at less than maximum length. Some of them don't work well even there.

--Rik

I have purchased the Mititoyo M Plan Apos 5x and 10x and plan to use them primarily for insect photography, but have done little with them as there is a paucity of insects in here northern Illinois at this time. For a tube lens I am using the 200 mm f/4 Nikkor AI on a Nikon D800e. Since the camera lacks an electronic first curtain shutter, I am using electronic flash with 2nd curtain synch.

I am interested in a shorter FL tube lens for less magnification. I have the 105 mm f/2.8 AFS and the 105 mm AF 1:2.8 D macro lenses, the old 105 f/2.5 Nikkor P (Sonnar formula with 5 elements in 5 groups), and the 135 mm Zeiss Apo Sonnar. Testing is in order, but are there any helpful generalizations to predict performance? I have read that the entrance pupil location is important and this is easy to determine.

I also bought the 50 mm f/2.8 EL-Nikkor N with reversing rings and the PB6 bellows for lower magnification work, but have not tested it yet.

Helpful pointers would be appreciated.

Bill Janes

[AtmosFearIC]

You are correct in what your saying Lou, it is something that comes up a lot with DSLR astrophotography when doing emission nebula. Hydrogen is the most abundant element in the universe and its strongest emission is in the red ~656nm. One issue is that about 60% of this emission line is blocked by the IR filter on sensor and THEN only 25% of that is counted as only one in every four (red) even detects it!
This is why Canon and Nikon have cameras like the D810A, the IR filter is either removed entirely or one with a wider passband is put on, not entirely sure.

One of the main things that a mono camera does, apart from higher resolution (no pixel interpolation from a bayer matrix), it that it allows for significantly stronger sharpening before aberrations start appearing. This is just a simple consequence of signal strength.

A general question though, is there are reason I should bother buying something like a Nikon AI-S 200mm F/4 or a Nikon 105mm F/2.8 Micro when I can just go to ThorLabs and get an Infinity Corrected 200mm or 100mm Tube lens?

I've been scouring the forums but haven't really been able to find much on this subject. In theory, a true Tube Lens should be quite a bit better than getting another lens and using it as one.
Take a macro lens, it is built to perform best at close focus BUT some lens designs may not be as well corrected (colour and distortion) at a distance. Telescopes don't suffer from this as much as the distance between element and sensor is moved to focus closer as opposed to elements themselves being moved in a lens.

[billjanes1]

Y

A general question though, is there are reason I should bother buying something like a Nikon AI-S 200mm F/4 or a Nikon 105mm F/2.8 Micro when I can just go to ThorLabs and get an Infinity Corrected 200mm or 100mm Tube lens?

I've been scouring the forums but haven't really been able to find much on this subject. In theory, a true Tube Lens should be quite a bit better than getting another lens and using it as one.

Cost and convenience of setup are important considerations for many of us. The Thor Labs 200 mm tube lens costs US $460, whereas a used 200 mm f/4 Nikkor can be had for $100. Also, one can mount the objective on the Nikkor with a simple adapter.

Information on the relative merits of tube lens options is scarce. This
post suggests that the tube lens is of secondary importance.

Bill

[Lou Jost]

AtmosFearIC, there are quite a few posts about tube lenses here; try searching for "Raynox" and you will get many posts on the subject. Rik and others have tested the dedicated tube lenses against makeshift alternatives, and oddly enough, some of the cheapest and easiest alternatives outperform the dedicated tube lenses in some respects. None of us understand why this should be the case. But the data is there.

One very cheap and simple makeshift tube lens is the Raynox DCR-150 (f=208mm). Older 200mm f/4 lenses have also done well. When less magnification is wanted, we often use the Raynox DCR-250 (f=125mm). Royalwinchester on this forum has used an expensive Apo-Lanthar 125mm with really great results; Beatsy and I often use a very cheap (<$60) Vivitar 135mm lens (Komine-made, serial # beginning with twenty-eight) with good results. Look up "Komine" on this site for that.

A recent post using this lens by Beatsy:

http://www.photomacrography.net/forum/v ... ght=komine

There are other good choices.

[rjlittlefield]

I've been scouring the forums but haven't really been able to find much on this subject. In theory, a true Tube Lens should be quite a bit better than getting another lens and using it as one.

The wrinkle in that theory is that people often try to use the tube lenses outside their design envelopes, which are not published.

For example with a full-frame sensor, 24x36 mm, it's very clear that Raynox DCR-150 provides much better corner quality than the official tube lenses from Nikon, Mitutoyo, and Thorlabs. See the discussion at http://www.photomacrography.net/forum/v ... hp?t=23898 ("Tube lens tests on D800E full frame"). It's recently been found that a Raynox DCR 5320 AB set (using both lenses of the matched pair) is even better. That's discussed in the thread surrounding http://www.photomacrography.net/forum/v ... 503#199503 .

But since you're using an APS-C sensor, I expect you would be very happy with either Thorlabs or Mitutoyo tube lenses.

--Rik

[AtmosFearIC]

The Raynox 150 looks brilliant, also appears that when reversed that it has a considerable amount of back focus to sensor. Longer back focus helps me as I don't have to worry about getting down to 44mm (Canon) or 46.5mm (Nikon), very difficult with my current planned setup but fine for my D7200.

I'm less interested in cost cutting and more interested in absolute contrast and resolution. Having a longer back focus from rear lens flange is a benefit for me, I've got to get adapters machined regardless so having more space to play with makes me happy

[Lou Jost]

If money is not critical I'd go with the Raynox that Rik mentioned in his last link. I have it too. It has wider coverage than the Raynox DCR-150 and this opens up some nice easy stack-and-stitch opportunities using a shifting bellows and/or shifting adapter.

[nathanm]

One small issue with your focuser, which is that most of us do not change focal distance to achieve focus - i.e. the tube lens is at a fixed focus (does not move relative to sensor) and the Mitutoyo objective is at a fixed distance from the lens.

Instead we focus on the subject by moving the entire camera and lens relative to the subject. Or subject relative to the camera/lens/objective.

Your focuser could be used that way, but it may be awkward to adapt.

If instead you use it to focus the tube lens, that might work to a degree, but I don't have much experience with this, nor have I seen a lot on the list.

[AtmosFearIC]

The Raynox DCR-5320 is only about $300 so it really isn't expensive by any means.
It'll end up working the same way as a telescope does, fixed focal length set at infinity. If you want to focus closer (terrestrial and birding) you move the sensor further back. Alternatively I could leave the focuser out of the imaging train and just stick things to pins and tape the pin to the focuser and use that for lateral movement.

What I don't quite know yet is what the back focus is of the DCR-5320? I understand that using the different combinations change the working distance but I'm not sure about the back focus. May just email Raynox.

[nathanm]

I use the Raynox DCR-5320. The focal length of the A+B combination is 170 mm.

When you say backfocus do you mean distance from the back of the lens to the sensor?

It is 170mm less the distance inside the 5320 to the exit pupil, which I have not measured.

[AtmosFearIC]

I use the Raynox DCR-5320. The focal length of the A+B combination is 170 mm.

When you say backfocus do you mean distance from the back of the lens to the sensor?

It is 170mm less the distance inside the 5320 to the exit pupil, which I have not measured.

170mm is good, 0.85x whatever micro lens.
Yep, the back focus would be from the rear of the DCR-5320 to the sensor. Is that what the exit pupil is? "Exit pupil" is a term I've seen float around but don't quite understand yet

[nathanm]

The way to accurately get the 5320 focused is to have a set up where you can vary the lens to sensor distance. Then focus (using live view) on something at infinity - i.e. using the 5320 as the only lens. Then lock it down, and put the objective in front.

I use Thorlabs SM3 tubes for this. Their thread is very close to 77mm x 0.75mm filter thread. So you can use a 72-77mm adapter ring to mount it.

The Thorlabs tubes allow you to change length enough to achieve focus.

I got best results mounting the objective very close to the front element of the 5320. Much of this is explained here.

http://www.photomacrography.net/forum/v ... hp?t=32346

[Lou Jost]

Nathan's custom-made adapter from the Mitu to the Raynox is interesting because it gets the Mitu very close to the front element of the Raynox, and his measurements indicated this was the optimal position, at least on medium format. You can duplicate this by threading step-down rings backwards inside other step-down rings, as here:

http://www.photomacrography.net/forum/v ... highlight=

[AtmosFearIC]

I read your thread early this week and it made no sense to me. Read it again yesterday and it makes sense now A couple of days of further research makes a world of difference. Finally understand what the DCR-5320 is! Just a 5 element from objective.
From your testing it probably has a focal length of 170mm when focused at fininity, ie. when pointed at the stars. As you're focusing closer than infinity the sensor has to be moved further back to compensate which leads to the 192mm FL.

Looks like I am going to have a few adapters to get, they can go with the other two dozen bits and pieces I've got haha
Thanks for the help and suggestions everyone