[Ouroboros]

I'm currently using a Pentax M 50mm 1.7 on a regular set of 50mm tubes to get some sweet macro shots. I'm thinking I might try some more magnification. As I understand it, the further I push out the lens, the more mag I'll get.

Now for the question: Is there a limit?

If I push out another 50mm will I have problems? (apart from the lighting)
What about another 100mm or a crazy 200mm?

Don't ask how much mag I want, I just want more. I'll find or make something that small to shoot.

LOL in my head I imagine a meter of tubes taking photos of atoms

[JohnBee]

Now for the question: Is there a limit?

The short answer is yes...

Every lens(focal length has its registration(not really the right term) limitation, therefore when you get into some of the longer tubes(toilet paper roll) you'll discovery a point where you no longer benefit from moving the lens away from the sensor and... of course find yourself having to stick the lens onto the object to achieve focus.

There's another phenomenon I've discovered with extension tubes where the longer lenses(135-300mm) don't quite behave the same way the shorter ones do.

Though it is quite interesting to experiment with though. I personally always liked the idea that we could alter the close focus ability of any lens by sticking an extension on it(great fun).

[newarts]

I'm currently using a Pentax M 50mm 1.7 on a regular set of 50mm tubes to get some sweet macro shots. I'm thinking I might try some more magnification. As I understand it, the further I push out the lens, the more mag I'll get.

Now for the question: Is there a limit?

If I push out another 50mm will I have problems? (apart from the lighting)
What about another 100mm or a crazy 200mm?

Don't ask how much mag I want, I just want more. I'll find or make something that small to shoot.

LOL in my head I imagine a meter of tubes taking photos of atoms

Here's a stream of consciousness analysis of the problem that resulted in maximum practical magnification limits that are surprisingly small.....

Subject as light source

I imagine one practical physical limit would be when the magnification gets so high that any further magnification is optically useless; for simple optics this probably occurs when the separation distance between two points on the image corresponds to a separation on the subject approaching a wavelength of light. Let's use that as a basic hypothesis.

A wavelength of average greenish light is about 0.5 microns. It seems reasonable that when the resolution limit of the sensor image (the distance between sensor pixels) corresponds to the minimum detectable distance between point sources of light on the subject (a wavelength) any further magnification will be basically useless.

The typical DSLR APS-C sensor has a pixel spacing of about 5 microns, so this pegs the useful camera magnification limit at only about 5/0.5=10X. Wow! this is a surprisingly small maximum useful magnification.

What does this mean when we look at a display of such an image? The typical pixel spacing on a CRT is about 0.25mm or about 250 microns. Now things seem about right - when our hypothetical image is displayed at 100%, each pixel of the display corresponds to the smallest possible pixel on the source object. The overall magnification is 500X.

I wouldn't be surprised if this estimate were off by a factor of 2 one way or the other, consistent with the usually stated limit of about 1000X for simple optical microscopy. But I can vouch that in my experience with optical microscopes, I've never done better than 500X, even with the best equipment.

Aperture effect

It seems to me that while a nominal 10X on the sensor is achievable with extension tubes, impossibly large apertures would be needed to achieve a resolution of 5 microns on the sensor. This is because aperture diffraction effects become important with high magnifications.

The limit of resolution due to aperture diffraction effects is something like the Rayleigh limit, about the f-number times the light wavelength. However the f-number in this equation should be modified by the magnification effect; for macro work this would be the nominal f-number (ie. what is says on the lens) times (1+m)*. In equation form this is about:

Pixel=f-number(1+m)/2 micrometers

So for a pixel spacing (diffraction spot) of 5 micrometers and a magnification of 10, an f-number of about

f-number= 2*Pixel/11 = 10/11 =0.9! an impractically large aperture!

Let's go the other direction and say we've got a perfect lens with an f-stop of 4; what will the maximum magnification be for a 5 micrometer diffraction spot (pixel spacing)?

m=2*pixel/f-number -1 = 10/4 -1 =1.5X!

So even an aperture of f/2 limits one to about 4x due to diffraction effects!

Tentative prediction

Based on the above crude estimates, I'll be surprised if one can be successful at more than 2-3x displayed at 100% using normal photographic macro equipment. This corresponds to 4-6x displayed at 50%, etc..

I won't be surprised if I'm wrong & look forward to hearing more precise estimates or seeing actual experimental results of high magnification macro photography.

Dave

* the reason for the (1+m) factor is that in the derivation of diffraction effects, the distance from the aperture is assumed to be the focal length, however, when magnification is involved, that distance is increased by a factor of (1+m).