Depth of Field limitations and magnifying under low light conditions

[adalc55c]

I have a fairly specific photograph that I wish to take (scientifically), and I intend to buy an appropriate lens for the camera we have, but I'm not entirely sure 'an appropriate lens' exists. Having investigated lenses, apertures, depth of field etc, I get the impression that what I want to do is either impossible, or on the limits of possible. I could do with some expert knowledge / experience to help me find out if, by making some sacrifices, I can get a lens to get close enough to the image I want. I went to my local camera store, but sadly they weren't really willing to discuss the problem and just told me what I wanted was flat out impossible. Which may be true, but I thought it was worth asking if any experts out there had any ideas, and a discussion could result in where I might be able to get 'close enough' to where I'd like to be.

So, here's what I have:

- I want to photograph light emission from a spark across a small gap (like mini lightning(!)).
- The field of view is about 3 mm
- I have an intensified-CCD camera (a nice 'toy' ), where the sensor is 512 x 512 pixels, and physically is half an inch square (so I'd need a reproduction ratio of around 4:1); I can mount any type of lens, or combination of lenses
- With a 5x microscope lens, I can get the field of view and resolution that I want, but the depth of field is nowhere near what I need (I get a tiny amount of focus, and everything else is blurred). I've tried adding an aperture, but this only made a small improvement. So, I need a different lens/aperture combination to increase the depth of field - this is the sticking point
- I would like a depth of field equal to or greater than about 0.4 mm
- BUT the amount of light I receive is low, so could do with the aperture as open as possible, and I think with the lens as close as possible, and with as wide a lens as possible
- I could increase the exposure of the image, but as the event is dynamic, I'd like to limit that where possible too
- Also, as the image would be of a dynamic event, something like focus stacking isn't really an option either

I've used some of the online depth of field calculators, but wasn't sure how well these calculations applied to a magnification scenario.

I would guess that my best bet would be to find a wide zoom lens with a small aperture, turn up the exposure and hope that I can get away without it blurring too much due to movement of the light source. But I don't really want to spend money on a lens if it's just a guess! Also 'wide zoom lens' is fairly non-specific in terms of focal lengths etc.

So what does anyone think?(!) Does anyone have experience trying to take magnified pictures close up? Am I asking for the impossible? I appreciate that I'm asking for incompatible goals (short exposures, wide apertures, large depths of field and minimal light), but perhaps naively on my part, a depth of field about 0.1 times the field of view didn't sound too impossible when I came up with the idea to try and take this photograph!

I look forward to any suggestions!

 

[Kolia]

I'm subscribing to your thread as I'm curious to see what you come up with.

My take is that no consumer optic will work. Have you tried contacting optic manufacturers directly ? Konica-Minolta ? Ziess ?

 

[runnah]

Does this experiment involve a clock tower, thunder storm and a mid 80's sports car?

 

[BrianV]

Not sure where this went but: a spark across a gap, like a spark plug, is quite bright and is short enough to make a good exposure. A camera is typically set for a long exposure and then the spark would be fired. This would be like shooting fireworks. A close up like this, a long extension tube is required. A lens made for close-up work is also required, the optics need to be optimized for close-in work. A wide-angle made for an SLR lens mounted in reverse is also a good solution. There are reversal rings which screw into the filter mount of the lens, and have the camera mount on the front. An intensified CCD camera will likely saturate, best to turn the Gain/ISO all the way down and experiment with the gain.

Somewhere in my box of junk, I converted a telescope camera adapter into a variable extension tube that would give this sort of magnification.

 

[brian_f2.8]

any examples of what you are trying to do? was there any inspiration? are trying to take a picture of an induction coil?

 

[Helen B]

There is a bit of an optical problem, I think. This was a very quick calc, and may be subject to careless error:

You want a magnification of 4x and a DoF of 0.4 mm on a pixel pitch of 0.025 mm (ie 512 * 2 = 1024 pixels in 25.4 mm approx).

A quick calc shows that you would need an effective aperture of f/128, which would result in a diffraction-limited resolution of 0.08 mm at 550 nm - three times your pixel pitch.

That aside, you need a lens that will stop down to about f/22 (at infinity, ie what would be marked on a normal photographic lens) or f/128 at 4x (what might be marked on a lens designed for 4x).

Can you relax the resolution requirement a little?

 

[Kolia]

Science FTW !!!!

 

[amolitor]

How optically repeatable is this? That is, if I took 100 photos of the experiment, and lined them up, would they be pixel-level indistinguishable?

If so, you might be able to deploy a technique called focus stacking: Focus stacking - Wikipedia, the free encyclopedia

 

[amolitor]

Wow, this is an old thread. Ooops.

 

[BrianV]

I've seen 9 year old threads come back to the top on some forums.

Wide-Angle lenses, "Retro-Focus" design for SLR's were very popular for close-up work. Reversing the lens basically puts the diverging elements in the back of the optical path, extends the working distance to the subject. The Nikkor 24/2.8 was a favorite for use on bellows, using a reversal ring. On a Nikon Bellows you can get 4x magnification with a 20cm working distance from the subject. the lens stops down to F22, image quality is best at F8 to F11. from the Nikon Handbook by Joseph Cooper.