Anyone used a Nikon LWD 40x 0.55 Ph2 DL?

[aurelius]

Hi everyone,

I bought one of these objectives to test with a photography project I'm working on, and wondered if anyone else has tried one or may be able to offer any advice?

My Stackshot arrives tomorrow, so I'm hoping to get some tips before I give it a try!

It's an infinity corrected lens, with a working distance of 2.1. It's a very heavy objective with a huge mount on it, not sure what size it is, but there's no way it's fitting into my RMS adaptor! I will probably MacGuyver something using the provided lens cap and some filter rings.

I'm planning on attaching it to a 400mm prime, but in my ignorance I'm not sure whether that's best or not. I do have a 200mm zoom that I can also mount it to.

Any tips would be welcome!

Thanks.

[rjlittlefield]

On your 400mm, that lens will give 80X. On your 200, it will give 40X as rated.

I would try the 200 first, because the 400 will give away a lot of the potential field of view.

Even at 40X, you will be running at effective f/36 (computed as magnification/(2*NA)) so your sensor will have plenty of resolution to capture all the details that are in the optical image.

--Rik

[aurelius]

Thanks Rik - I'll be using a Nikon D800, which is 36 megapixels. However, when you say f/36 equivalent, does that simply refer to the exposure/depth of field, or does it also mean that I would have the same issues with diffraction that I'd have with a conventional lens stopped down to f/36?

[rjlittlefield]

does it also mean that I would have the same issues with diffraction that I'd have with a conventional lens stopped down to f/36?

Yes, it means the same issues with diffraction.

Also there may be a problem with covering the full frame of the D800 with high quality image. That sensor has a 43 mm diagonal, where most eyepieces are only about 25 mm field diameter.

Using the 400mm on a D800 makes much more sense than it would with an APS-C sensor, because then the full-frame sensor covers about the same area on subject that a 21.5 mm eyepiece diameter would.

--Rik

[aurelius]

Thanks for the reply, Rik. I hadn't realised diffraction would be quite so severe.

I also have access to a Canon 550D, which has a 1.6 crop sensor, and a 200mm lens. Would this provide superior quality? And if so, would it really be better than sizing a D800 image down to 18 megapixels?

Sorry for so many questions - I have been reading as much as possible on this forum over the past six weeks, but this is one of the things I haven't quite gotten my head around.

I'm very keen to have a system that can achieve high quality images at 10x, 20x and 40x. So far, my best results have been with a 10X Nikon objective on a 400mm lens, although I need to do proper tests with the Stackshot when it arrives tomorrow.

[Joseph S. Wisniewski]

It's a 25mm mount, aka M25. Don't get an M27 (Nikon BD) adapter by mistake. Search for "Nikon M25" and you'll find several adapters, typically to 52mm filter threads, then get step-down rings to your lens size.

Diffraction will be a little worse than you anticipate, because it's a "phase contrast" objective, which means there's a ring partially blocking the optical path. I wouldn't reject the objective outright because of that, but it's something to be aware of.

[rjlittlefield]

Thanks for the reply, Rik. I hadn't realised diffraction would be quite so severe.

Diffraction is always an issue at high magnification. The formula about effective f-number = magnification/(2*NA) is valid no matter how you get the magnification. If you're shooting a Nikon 10X NA 0.25 objective on a 400 mm tube lens, then you're actually running at 20X, so your effective f-number there will be 20/(2*0.25) = f/40.

In general, each objective covers a certain maximum circle on subject. Your best image with that objective comes when you fill the sensor with that circle, whatever magnification that takes.

The gold-barrel Nikon CFI Plan Achromat 10X NA 0.25 part number MRL00102 covers about 5 mm on subject, so you can actually run that beast at 10X on a 200 mm tube lens and cover a full frame sensor at f/20. When the full frame camera is a Nikon D800, that is an awesome combination. See http://www.photomacrography.net/forum/v ... hp?t=19430 for an example. Given an appropriate tube lens, you could even push that a bit farther, maybe down to around 8.5X and still have the corners hold up.

At higher magnification, the effective f-number gets more painful. For example the Mitutoyo Plan Apo 50X NA 0.55 will nicely cover full frame at rated magnification, but in that case it's running at f/45 so it never makes an image that looks sharp on sensor. It does a great job of revealing detail on subject, however.

Adding to the complexity, it turns out that the Mitutoyo 50/0.55 will actually cover more than full frame at rated magnification. For example it works nicely on a 100 mm tube lens to an APS-C sensor, giving 25X and still NA 0.55 so about f/23. On full frame, the equivalent would be using a 160 mm tube lens and getting 40X at effective f/36.

I also have access to a Canon 550D, which has a 1.6 crop sensor, and a 200mm lens. Would this provide superior quality? And if so, would it really be better than sizing a D800 image down to 18 megapixels?

There are some subtle issues in here.

If you choose the tube lens length so as to cover the sensor with that high quality circle I mentioned, then more pixels are always better (though only a little better when the effective f-number gets big). Sensor size by itself is not an issue in this case because by choosing the tube lens length to match the sensor size, you'll automatically adjust the effective f-number to produce the same relative amount of diffraction blur in both cases.

So this leaves us asking whether a Canon 550D with 18 megapixels on an APS-C sensor is better than a D800 sized down to 18 megapixels. The answer is "no" in terms of image quality, but possibly "yes" in terms of how easy it is to match the objective to the sensor. That depends on exactly what objectives you're using and what tube lenses you have available.

We could also ask whether the 550D is better than a D800 running in DX capture mode at 15 megapixels over APS-C. The answer to that is clearly "yes" in terms of resolution, but it's not a big difference because 18 vs 15 megapixels is less than 10% on each axis. In exchange, the D800 may well have an advantage in dynamic range, which I've been led to understand is unusually good with that sensor.

I don't have a D800 to confirm the approach, but if I did have one I would seriously explore leaving the D800 mounted all the time and just choosing between its various capture modes (FX, DX, or the 25 megapixel 1.2X crop mode), as needed to best match the objective at hand.

Sorry for so many questions - I have been reading as much as possible on this forum over the past six weeks, but this is one of the things I haven't quite gotten my head around.

I wish the answers were simpler. Unfortunately you're now working in a regime where the issues are often more complex than just a single axis "better-to-worse".

Diffraction will be a little worse than you anticipate, because it's a "phase contrast" objective, which means there's a ring partially blocking the optical path. I wouldn't reject the objective outright because of that, but it's something to be aware of.

This is a timely example of one of those complex issues I mentioned. Yes, there's a ring partially blocking the optical path. However, the ring does not block the periphery of the aperture, so the objective retains its full rated NA and ability to capture fine detail. For focused detail, all the ring does is to introduce a normally invisible change in the modulation transfer function. The more important effect is with un-focused detail, where the presence of the phase ring causes "doughnut" bokeh in which for example OOF point highlights turn into fuzzy rings instead of the usual fuzzy blobs. Even if you're shooting a fully focused stack, the difference in bokeh can slightly degrade the stack results compared to what a non-phase objective with the same NA would do. Charles Krebs did a nice head-to-head evaluation of this issue back in 2009. You can see those images and discussion at http://www.photomacrography.net/forum/v ... php?t=6647.

--Rik