High-quality film scanner lenses have ideal properties for detailed images with focus stacking. The first of three lenses is presented here.
Focus stacking with microscope lenses enables extremely high detail reproduction, far more than with conventional camera lenses. However, this high magnification also has a significant disadvantage: a small bug, for example, can no longer be fully pictured. At a reproduction scale of 5:1, only an image width of 7.2 mm is reproduced on our 24 x 36 mm full-frame image, which means that only an eye, a leg or part of a small insect might be recognizable in our photo. This may be desirable, but we often want to see the whole animal in the picture. The only option here would be to use a conventional macro or magnifying lens and go without the unusually high level of detail.
Fortunately, however, there is an alternative that many photographers are not aware of: film scanner lenses. As a rule, these are primarily slide film scanners, although they can also scan negative film materials. They are usually small format, in exceptional cases also medium format. They have optics designed for close-ups around 1:1 and sometimes also produce a particularly high resolution, i.e. they can show far more detail than a conventional standard or macro lens. Distortion-free image reproduction and color correction are also sometimes exceptionally good. However, this does not apply to all film scanners by any means. Which scanner lenses are suitable for magnifying photography with focus stacking and how do you use them?
Slide film scanner lenses
Film scanners actually belong to the analog photographic past and are no longer really relevant today. Nikon, for example, completely discontinued its slide film scanner models many years ago, as has Minolta since its takeover by Sony. This has caused the scanners still on the market to rise abruptly in value, and if you try to buy such a device second-hand today in order to use the lens for photographic work, you have to expect disproportionately high prices for functioning scanners, in some cases far above the former new price. The only thing that really makes sense is to look for a defective scanner.
In principle, you can remove the lens of any defective scanner and test its suitability. However, this makes little sense for devices in the lower and middle price and quality range, because our high expectations are hardly met here. What's more, the specified resolutions are often not even close to being achieved, especially with flatbed scanners, but also with special film scanners. In addition, a distinction must be made between interpolated, i.e. computer-generated, resolution and optical, i.e. real resolution achieved with the help of objective lenses, whereby only the latter is of interest to us.
The decisive pacemaker in the use of slide film scanner lenses for focus stacking was the US American Robert O'Toole, who unfortunately passed away early 2024 in the mid of his fifties, much too early. Robert examined, compared and evaluated lenses for close-up photography with great expertise. He also runned a website on which he published his findings (www.closeuphotography.com). Here you will find an overview of numerous different scanners whose lenses he has tested for their suitability for focus stacking (https://www.closeuphotography.com/scanner-lenses).
Two specific 35 mm scanners and a medium format scanner are particularly promising. I will present one of the two 35 mm scanners in this article, the other and the medium format scanner in separate articles. In a way, these three lenses play a special role among film scanner optics because they represented a new generation at the beginning of the 2000s, the last one, and with a much higher resolution than all film scanners up to that point. Whether this really made sense for the mass of slides films back then is questionable, because ultimately you had to have at least a 50 ASA film to do justice to the lens resolution, or even better the legendary sharp 25 ASA film from Agfa (which in reality had 33 ASA), because otherwise you would only make the film grain more visible. But that is irrelevant for us today.
However, I would ask you to never to dismantle a functioning film scanner in order to use the lens for focus stacking shots. Film scanners are a dying breed, and each of these devices is far more than the sum of its parts. Personally, I have respect for every slide scanner and would only dispose of it if it is defective and a repair no longer seems appropriate.
Minolta Dimage Scan Elite 5400
A few years ago, Robert came across a very small scanner lens in a comparison test that, despite its inconspicuous size, produced an unexpectedly high resolution and had surprisingly good correction, both in terms of distortion and color deviations (chromatic aberrations), which can be seen as purple or green color edges in poorly corrected lenses, especially at the edges. The image quality of this lens was so outstanding that it became the standard in many of its comparative tests, against which the other lenses now had to measure themselves.
We are talking about the lens of the Dimage Scan Elite 5400 scanner, which Minolta launched on the market in summer 2003. The number 5400 stands for the resolution of 5400 dpi (dots per inch), as this is the resolution that this scanner produced optically according to the manufacturer. At the time, this value was absolutely unique in this scanner price class, as conventional devices offered just half that, and even top Nikon devices only achieved 4000 dpi.
I won't go into the technical refinements of this Minolta scanner here because they are irrelevant for our photographic work. On the other hand, the extremely high resolution of "Minolta 5400", as I call it for short, is essential, and it is still a mystery to this day how the company Mirold Optics managed to produce this fabulously good lens for Minolta.
It has eight lenses arranged in four groups. Two ring-shaped transverse grooves can be seen on the body, and the distance between these grooves and the two ends of the lens housing is different. In simple terms, it has a "longer" and a "shorter" end, and when installed in the scanner, the light enters at the "shorter" end and exits at the "longer" end (see diagram).
In a test, I compared this normal position with the inverted position, and it turned out that when used inverted, the image loses a little in size and slightly in sharpness, weaker in the center, but clearly towards the edge. It should therefore definitely be used in its normal scanner position.
How to remove?
If you have the opportunity to get hold of a defective Dimage Scan Elite 5400, you can count yourself lucky. Once you have the scanner in front of you, it is relatively easy to access the lens, as it can be removed in a few simple steps.
The newer scanner model II has a wider, slightly shorter housing with more plastic and looks a little more modern overall, but technically it is largely the same as its predecessor model I, and lens removal is practically identical. The main difference between the two devices was more of an electronic nature and lays in a significant acceleration of the scanning process.
First remove the housing with a few screws so that you have a clear view of the inner workings. On one of the two long sides, in the middle area, there is a thin, black plastic plate that is glued to the surface and covers the lens. It is very easy to see and freely accessible, protecting the lens not only from mechanical damage and dust, but presumably also from side light. The lens is therefore located behind this plate, held in place by a metal profile that is fastened at the top and bottom with one screw each. You simply remove these two screws and hold the lens in your hands.
The camera adaptation
If you are now looking for a thread on the lens to mount it on the camera: As you can see, you can't see anything. This is completely normal for film scanner lenses. For camera mounting, we need a mount to put it in, and a very useful product for this is what the company "RAF Camera" offers specifically for this lens. It is a ring-shaped body with an inner diameter of 18 mm and a grub screw for fixing, and there is an RMS thread at one end.
The abbreviation RMS goes back to the British "Royal Microscopy Society", which recommended this thread for microscope objectives in 1858 (W 0.8" x 1/36"). With this thread, you can screw the objective onto an adapter that adapts to M42, for example, so that you can screw the whole thing onto M42 extension rings. I myself use this RAF adapter on the front plate of my Novoflex Balpro 1 bellows. In any case, make sure that the lens is in the correct position: the gold-colored line on the lens itself shows you which side the light enters. The asymmetrical housing also helps; the light should enter through the "shorter" end (see diagram).
Before I had the RAF adapter, I came up with a simpler solution that also provided an RMS thread: I completely cleaned out an old microscope objective and slid the scanner objective into it. A Teflon tape wrapped around it provided the centering. This also worked perfectly and cost nothing.
Extension, shooting distance and magnification
What magnification does this objective work with? You decide this yourself with your extension, i.e. the distance between the camera sensor and the last lens where the light exits. The longer this distance, the closer you can get to the subject with the front lens and the larger your magnification.
In the scanner, the lens works at slightly less than 1:1, because the sensor dimensions are a bit above the full-frame size of 36 x 24 mm. According to Robert OToole, this lens has its best optical performance – the so-called sweet spot – at 2:1, but it also shows great performance below that, down to below 1:1, i.e. if you shorten the extension and increase the shooting distance. You can also increase the magnification beyond the sweet spot and go to 3:1, for example (increase the extension and reduce the shooting distance).
With every deviation from the sweet spot of a lens – and this applies not only to this lens, but in general – you cause a certain loss of quality in the image. The point at which this becomes disturbing depends above all on your quality requirements. I remember enthusiastic statements from a user who reported of great images from this Minolta lens at 5:1. Although this reproduction scale is possible, it causes a clear loss of quality, because when you upscale such a photo and critically assess it, you will undoubtedly recognize image weaknesses. Even at 3:1, I prefer to use a Mitutoyo M Plan Apo 5x, which I equip with a Raynox DCR250 tube lens far below the nominal focal length of 200 mm. The Minolta 5400 serves me at scales around 2:1. For images around 1:1, I use a different scanner lens, which also delivers absolutely outstanding performance and also comes from a film scanner. More on this in a separate article (Part 3).
In principle, I see this great Minolta 5400 scanner lens as a way to close the gap that arises below the Mitutoyo M Plan Apo 5x lens, because the Mitutoyo lenses in this series with 1x and 2x are unfortunately not suitable for working with focus stacking and direct mounting on the camera. With the Minolta 5400 you can close this gap excellently for shots from 1:1 to about 2.5:1 and thus have a significantly better resolution and color correction than conventional camera lenses with close-up lenses and also magnifying lenses such as the Canon MPE 65 can offer. Another advantage, of course, is that it is used without a tube lens.
However, you should resist the temptation to screw this lens to the camera with extension rings and use it to search for subjects in the garden, for example. This will be difficult because the depth of field is so shallow that such an approach simply makes no sense – unless you use blurring as a creative tool. Added to this is the image field curvature mentioned below. In-camera focus stacking won't work either, because it requires an autofocus in the lens. The aforementioned lenses (normal lens with extension rings or magnifying lens up to 5:1) are better suited for such freehand shots, not least because of their aperture, which gives you more depth of field. I see the Minolta-5400 scanner lens as more suitable for stationary use with focus stacking.
Diffuser
However, for good shots in focus stacking, you also need a diffuser, which must be attached to this lens because its shooting distance is generally very short. Although this varies with the image scale, it is only a few centimetres in any case.
My first diffuser for this lens was a converted flash diffuser from Godox, which I modified accordingly (I will describe this in detail in a separate article on diffusers). Another possibility is the much larger diffuser that I made from the shade of an IKEA pendant lamp (see the separate article on diffusers, which will be published soon).
Full format suitability
Scanner sensors are not rectangular, with pixels arranged in a flat pattern as in a digital camera, but narrow and long. The Minolta Dimage Scan Elite 5400 film scanner has a sensor with a 43.5 mm long active area. The scanned image width of around 24 mm is therefore enlarged by a factor of 1.78 to the width of the scanner of 43.5 mm. This means that the image circle generated by the lens measures 43.5 mm and thus completely covers a 35 mm full-frame sensor. This is a significant contrast to the Nikon lens that will be presented in Part 2. The question of full-frame suitability can therefore only be answered with an unconditional yes.
One disadvantage of the Dimage Scan Elite 5400 lens is that it has a slight image field curvature. Naturally, you won't notice this with Focus Stacking, so it doesn't matter for this type of photography. However, if you were to focus on an absolutely flat surface in the center, the edge of the image would be blurred. This curvature also speaks against using this excellent lens for normal close-up photography. On the other hand, it is perfect for focus stacking.
Determining the magnification
How do you determine what magnification you are shooting at? This is very simple: you focus on a centimeter scale (folding rule, ruler or similar). Take the image width of your sensor as a basis and compare it with the visible millimetres of your sharply focused centimetre scale.
For example, the full format has 36 x 24 mm, so you will see a scale width of 36 millimetres in a 1:1 image. If you only see 16 mm, you will see 2:1, and at 72 mm 0.5:1 or 1:2. In this way, you can easily calculate your current image width and thus your present magnification scale using the width of the sensor format you are using (e.g. full format, APS, MFT), even if you are working with a bellows device and thus vary very flexibly.