This page describes tests carried out on two types of Soviet legacy lenses for 16 mm movie cameras. Different versions of these lenses exist, in different mounts and produced at different times. Two such versions of each lens model were tested:
This page is part of a set describing tests of small-format lenses in UV photography. The main page describing these tests is available here.
Mir 11 in Krasnogorsk mount
This is a Soviet-made 12.5 mm f/2.2 for 16 mm movie cameras made by KMZ (Krasnogorsk mechanical factory). This is a smaller format than Micro 4/3, so it is not surprising that the image circle covers only a portion of the Micro 4/3 sensor. The Krasnogorsk bayonet is a big and clumsy mount, for which only a couple of Micro 4/3 adapters are available. This bayonet has two "pegs" (rather than flanges) on opposite sides of the barrel. They are not placed at exactly 180° from each other, so it is not possible to mount this lens upside down.
Demand for this lens started to increase only once BlackMagic video cameras with Micro 4/3 mounts but smaller sensors than Micro 4/3 became available. Until then, there was no significant use for this lens model in digital imaging. The adapter shown in the picture is expensive and looks esthetically very well made. A cheaper and simpler adapter is also available on eBay.
In spite of the very small size of the optical elements, the barrel of this lens is as big as many 35 mm f/2.8 lenses for 35 mm film SLRs, and heavier than typical for these lenses. The aperture ring, filter mount and built-in lens shade rotate together. In my specimen, the focusing ring is marked in feet and rotates slightly more than three-quarters of a turn to reach the minimum focusing distance of slightly less than 24 cm. The extension of the barrel while focusing is barely perceptible, given the very short focal length.
The rear element of this lens is small and leaves just enough room for a small 16 mm reflex mirror. In practice, this means that, mounted on a Micro 4/3 camera, this lens leaves enough clearance for the camera shutter, but not much more than that. This lens uses an optical scheme with 7 elements in 6 groups. The filter mount is 58 mm.
Vignetting is a major problem with this lens (Figure 1, bottom right). Vignetting further increases when a 52 mm filter is mounted on the lens via a step-down adapter. So much cropping is necessary to eliminate it, that the field of view after cropping becomes effectively equivalent to that of a 20-25 mm lens. In addition to vignetting, the figure shows also internal reflections within the barrel and/or within the built-in lens shade. This suggests that the built-in lens shade may be restricting the field of view, and that the image circle of the lens may actually be wider. For most of the aperture range, the shape of the 6-blade iris is a hexagram with sharp corners. This may affect bokeh.
The test picture also shows that the lens is not properly centered onto the sensor. Since the same amount and direction of offset is also shown with another lens in Krasnogorsk mount on the same adapter (Figure 1, bottom left), I believe that the fault may be with the adapter, rather than the lenses. On my copy of the adapter, one of the two bayonet notches has been re-machined after the body of the adapter was anodized. This may have been done to correct a fault of the adapter, but a significant amount of decentering, as shown by the test pictures, is still present. This problem may have something to do with the mechanical specifications of the two pegs of the Krasnogorsk bayonet, which differ from each other in shape and orientation (one is perpendicular to the side of the barrel, the other oblique). If the adapter does not take this into account, decentering of the lens could result.
NUV resolution is very good. The quality decreases off-center, but remains acceptable until vignetting sets in.
NUV transmission with Baader U and Asahi Spectra XRR0340 is 2-3 stops lower than the reference lens. By itself, this does not prevent using this lens in UV photography.
Mir 11 in C mount
Judging from the appearance and size of the front and rear elements, this lens appears to be optically the same as the preceding one, albeit in C mount. However, it is engraved "MIR-113", or possibly "MIR-11 3", at the front, which might indicate a different model or a subsequent version. It was common for the Soviet-era lens industry to design and manufacture just one optical subassembly of each given focal length and speed, and mount it in several lenses branded in different ways for a variety of different camera and movie formats, perhaps as an industrial manifestation of the one-size-fits-all political ideology of that time and region.
The minimum focusing distance is 24 cm and the travel of the focusing ring is half a turn. The filter mount is a very unusual 32 mm, for which I was only able to find a 32 to 42 mm step-up adapter on eBay. The filter mount does not rotate with the focus and aperture rings. As common in movie lenses in C mount, the aperture ring has no click-stops. It is easy to accidentally move the aperture ring while rotating the focus ring. This lens is small but heavy, and the barrel is partly made of nickel-plated brass.
Several C to Micro 4/3 adapters are available on eBay, including the expensive but well made Metabones adapter shown above, which seems to be the only one made from machined and nickel-plated brass. All others are made from aluminium alloy. All such adapters are one-piece constructions, and do not allow the distance and aperture scales of the lens to be exactly aligned with the Micro 4/3 bayonet. In fact, this lens, mounted in the Metabones adapter, is offset by almost half a turn with respect to the Micro 4/3 mount, and its distance and aperture index is not visible in normal camera operation. I did not find any obvious way to correct this offset during a partial disassembly of this lens, but it might be possible.
The principal reason for testing this lens is that this version of the Mir 11 has a very short built-in lens shade. This version is also much smaller and lighter than the preceding one. Unfortunately, a substantial vignetting is still present, albeit lesser than in the preceding lens (Figure 4, bottom right).
NUV image center resolution is good.
NUV transmission is visibly lower than in the preceding version of this lens (by about 1 stop). Either this lens uses a different optical formula than the preceding one, or its coatings have been improved with respect to the tested version in Krasnogorsk mount.
As a whole, in spite of the good NUV image quality, the substantial vignetting and relatively poor NUV transmission make the use of this lens for NUV photography problematic on Micro 4/3 cameras.
Vega 7 in Krasnogorsk mount
This is a Soviet-made 20 mm f/2, also for 16 mm movie cameras and made by KMZ. It produces a substantially wider image circle than the preceding lens. For this reason, there are a few mentions on bulletin boards of using it on Micro 4/3 cameras, especially years ago when few native Micro 4/3 lenses were available.
The closest focusing distance is 40 cm, which is not satisfactory for close-up photography. The optical scheme uses 5 elements in 4 groups. The filter mount is 52 mm. For most of the aperture range, the shape of the 6-blade iris is a hexagram with sharp corners. This may affect bokeh. Vignetting is less pronounced than with the MIR 11 in the same mount, but still substantial (Figure 7, bottom left).
By disassembling the Mir 11 and Vega 7 lenses in Krasnogorsk mount, I found that their optical subassemblies (i.e., optical elements and diaphragm, all mounted in a single, small cylindrical barrel) have the same external dimensions. An additional aluminium washer adjusts the registration distance of the subassembly, and differs in thickness from lens to lens. It may have been individually calibrated for each lens specimen at the factory.
Thus, it is possible to swap these subassemblies, and I re-mounted theVega 7 subassembly into the Mir 11 barrel. The aperture and distance scales of course are no longer correct, but the aperture ring works. The remounted Vega 7 produces a significantly lesser vignetting on Micro 4/3 because of the wider filter mount of the Mir 11 barrel (Figure 7, bottom right). This proves that the illumination circle (albeit not necessarily the image circle with good definition) of the Vega 7 is wider than allowed by its lens shade.
NUV resolution is very good. The quality decreases off-center, but remains acceptable until vignetting sets in.
Because of the large amount of vignetting in its original Krasnogorsk mount, I do not regard this lens as a reasonable choice for VIS photography on Micro 4/3. However, for NUV it remains an interesting choice, especially if vignetting is reduced by re-mounting the optical subassembly in a barrel without a built-in lens shade.
NUV transmission with Baader U and electronic flash at the center of the image circle is about 2-3 stops lower than the reference lens with Baader U and Asahi Spectra XRR0340. This amount of transmission would still allow the lens to be usable with these filters.
Vega 7 in Kiev 16U mount
Optically, this is a 20 mm f/2 identical to the preceding lens. This specimen appears to be more recently produced than the preceding one. The barrel details of the Vega 7 in Kiev 16U mount changed over time, and my specimen is probably a later production run than the "zebra-finish" ones available in both Kiev 16 and Krasnogorsk mounts. The Vega 7 in Kiev 16 mount has been mentioned on bulletin boards as perhaps usable on Micro 4/3 cameras. The built-in lens shade of the Vega 7 in this Kiev 16 version is much shorter than in the Krasnogorsk version. Therefore, I decided to try the Kiev 16 version to see if it differs in this key respect.
The Kiev 16 mount has an M32 x 0.5 thread and a 31mm registration distance. This is not the same as the C and CS mounts. The Kiev 16 mount is smaller than the massive Krasnogorsk bayonet.
The Kiev 16U to Micro 4/3 adapter from eBay show in the above picture is an amateurish two-piece construction. The rear piece is black-anodized aluminium alloy marked L39-M4/3, with three set screws that keep in place the front piece and allows its rotation when loosened. This piece obviously comes from a Chinese L39 adapter, but the front piece with the L39 mount has been replaced with one apparently turned by hand on a poor-quality lathe and/or from poor-quality stock metal. This front piece is untreated aluminium alloy, with a dirty female M32 thread that cannot be cleaned because the metal surface of the threads is full of small cracks. The surface of the threads feels as smooth to the touch as coarse sandpaper. A set screw can be used to lock the lens in a given position in this thread, and in this way correct for differences in registration distance. The adapter comes with no instructions on how to use the set screws (and in fact no instructions of any kind). A hand-cut cardboard gasket sandwiched between the two parts, apparently added as a stop-gap solution to correct an imprecise registration distance, completes the adapter. The two-piece construction allows the lens to be aligned to the index of the Micro 4/3 rear bayonet.
This adapter was not cheap (US$ 45 including shipment) and a better manufacturing quality should be expected at this price. Kiev 16 adapters are not common, and none seem to be available from China. On the other hand, eBay seller Rafcamera offers a one-piece Kiev 16 adapter of apparently better manufacturing quality at a moderately higher price. Rafcamera also offers a Kiev 16 to C adapter, as well as metal Kiev 16 rear caps. The original rear caps are often missing from these lenses.
Step-down filter adapters from M42 x 1 to M32 x 0.5 are also available and can be used to mount Kiev 16 lenses on M42 accessories, but the M42 mount has a longer registration distance than Kiev 16, and it is therefore not possible to assemble a Micro 4/3 to Kiev 16 adapter by mounting one of these step-down rings on a normal Micro 4/3 to M42 adapter.
The filter mount appears to be 35.5 mm. Step up adapters are available, but not in a great variety of end sizes. A few 35 mm adapters are also available, but sit a little loose on this lens.
Vignetting is unfortunately present also in this version (Figure 11, bottom left). The built-in lens shade can be removed, but this does not reduce vignetting (Figure 11, bottom right). Mechanical details of the lens barrel still limit the field of view, including the retaining springs of the aperture click-stop mechanism, which results in a slightly irregular outline of the field of view.
NUV image resolution is quite good.
NUV transmission is similar to the Vega 7 in Krasnogorsk mount, but NUV false color with Baader U filter is slightly more violet. On the other hand, transmission with the Asahi Spectra XRR0340 filter remains the same as in the preceding version of this lens.