So what went wrong? The copper tape is naturally quite soft and the small force of the camera RF roller against it, when left in one focus position for a while, would result in a tiny indent. We’re talking an almost imperceptible pit in a foil surface already only a few microns thick but it’s enough to slightly throw calibration at that point. Still good, but not good enough.
The job need to be done properly and for that I needed to buy a new spanner wrench (seen below, £10) with points at the ends. Even then, I had to grind some of it down but I finally had enough leverage to remove the collar which holds the cam – and its shims – in place.
After one failed attempt (that copper foil is like very sticky gold leaf) I managed to thicken one of the shims. This may sound an odd way of doing it but it was easier to modify the thicker of the existing shims than to add the copper foil to the surface that the shims sit on, or to the cam itself. Some very careful cutting with a scalpel and the job was done. Once put back together the calibration is spot-on without the fragility of having foil stuck to the contact surface of the cam.
A final few words on coding. As this is not a Leica lens, and especially since it was discontinued well before the first digital M (M8) with the now standard 6-bit lens coding, the digital M has no way of knowing, automatically, what lens is attached. There happens to be a screw in just the place where, if painted black, it gives the impression to the code reader that the lens is a Tele-Elmarit 90/2.8. Note from below that I had to leave a small amount to the right of the screw as reflective silver; this works perfectly and the paint stays because it’s recessed. When I set auto-ISO I have the minimum shutter speed set to 2x(focal length) and it’s nice to have this working seamlessly together with my coded lenses.
Some time ago I part-exchanged my Summarit 90 f/2.5 and have regretted it since, especially as one of my favourite images was the result of a lucky moment when I happened to have that lens attached. Since then, Leica prices have gone super-silly and even the revered Elmarit-M 90mm has increased in value on the used market by a third again compared with what they were just a few years ago. That puts them almost at the new Summarit price – especially when factoring in a trip to Wetzlar to be coded and adjusted (that would almost certainly be a necessity). To get a 90mm back in Leica mount I could have gone for an ancient tele-Elmarit or Elmar f/4, or looked to a non-Leica brand. The only current, viable option there seemed to be a Zeiss 85mm f/4 but that would mean very inaccurate framelines in the viewfinder (85 cf. 90mm). Browsing around one evening I happened across a mention of the Konica Hexanon in M mount, which was made for the Konica Hexar RF back in the 1990s. They are no longer made and are relatively rare – but it was interesting to note that some people compared the 90 quite favourably with the Leica Elmarit-M. If I could find one it should be roughly a third of the price of the Elmarit but apparently be close, both optically and mechanically. It would need to be adjusted, again almost certainly, but I started to look around to see if it would be possible to do that myself (something I would not do with a Leica for obvious reasons). I decided that it would be worth a punt. Fast forward to the arrival of a fairly good condition Hexanon which was described as having “a couple of dust spots” inside. I would call them massive balls of fluff, myself, but that’s eBay for you. I wasn’t banking on having to take apart the optical cell as well but, as I’ve done that before with a couple of Nikons, I decided to go ahead and clean it out. It was reasonably straightforward to remove the rear element group (with care!) and get rid of the crud inside. Needless to say, the lens would rear-focus slightly when using the rangefinder of my M; this is also referred to as back-focus and it’s when the sharpest point of focus is behind the intended subject. At the infinity stop of the lens the rangefinder patch still hadn’t quite reached co-incidence even though the lens will actually focus sharply to infinity (and a tiny bit beyond – without wishing to go into the concept of any object being beyond infinity right now!). This means that the lens is collimated OK but the cam needs to be lengthened by a tiny amount so that, at any given point, it pushes the roller inside the camera by a tiny amount more so that rangefinder patch and sharp focus are aligned. In other words, ideally, the cam of the lens – the brass ring which makes contact with the roller in the camera – needed to be shimmed. But by how much? I tested with a piece of 3M Magic Tape which I stuck to the cam and roughly cut around. This tape I measured to be approx. 0.06mm in thickness. That’s 60 microns. It was a little too much (I was kind of expecting that) so I needed to order some copper foil tape at around 0.03 – 0.04mm in thickness and use that to add to the shim(s) already under the cam. Problem no. 1: I couldn’t loosen the collar around the cam which holds it in place (this can be seen in the picture below – it has two small holes in, one either side) because I didn’t have the right tool. I tried to fashion one but the collar wasn’t going to budge and I risked damage, so I went with option 2. That is, to apply the tape directly to the contact surface of the cam, to build up the part which makes contact with the roller by 40 microns. The result is this: Not the most elegant solution in appearance, but it works. The focus is now spot-on. Problem no. 2: …hasn’t happened – yet.* It might be that the foil comes away, but actually the adhesive seems to be pretty strong and so I’d expect it to stay unless I accidentally hit the cam and shear off the foil (unlikely!). The good thing is that it can be replaced, if necessary, and the adjustment is easily reversible. I may yet have another go at shimming the cam from underneath but, for now at least, this lens is clean, perfectly adjusted and a very worthy alternative to a Leica. In terms of performance I’d say that it’s pretty much up there with the Summarit I used to own; certainly every bit as sharp, even wide open.
* Well, Problem no. 2 did happen (see part two).