Autocollimator vs ground glass in setting infinity focus

[Robert Lai]

Hello,

Do any of you repair gurus know of problems with autocollimators, such that they won't give proper focus to a camera? Autocollimators are an optical device to help a technician achieve perfect infinity focus on a lens, in theory.

So far I know of two cameras - a Rolleiflex T and a Zeiss Super Ikonta C, that could not be properly focused on film, after they were adjusted by use of an autocollimator. Both cameras were serviced by technicians who are acknowledged experts in their field. Yet, neither camera could produce sharp images after their overhaul. Both eventually got sent to Clarence Gass, of Mission KS. Clarence has made both of them sharp again.

In the case of the Super Ikonta C, which is my camera (the Rolleiflex isn't mine, but I have correspondence with the owner), Clarence had to find the proper helical start. The first technician had placed the front cell on the wrong helical start (there are 12 starting points, only one is right), based on what the autocollimator results were.

When I asked Clarence what he does, he told me that "I don't trust those machines (autocollimators)".

" I see a lot of cameras that come to me that won't focus properly because the tech has adjusted the focus using those machines. Those machines don't always work. I use a ground glass on the film plane, a magnifier, and a tower several miles away to sight on".

I'm merely putting this information out there, so that if your camera won't focus properly after an overhaul, you may wish to ask the tech to use a ground glass instead to set the focus.

 

[L David Tomei]

I agree. An autocollimator's primary function in life is to collimate optical elements. Alternatively, ground glass focusing works well and is easy. For even greater accuracy you might try using an optical fiber faceplate instead of the ground glass plate. The image will be sharper and precise. Faceplates can be found as optical surplus parts because they can be a bit expensive new.
David

 

[Chriscrawfordphoto]

THe collimator itself has to be accurately calibrated. If the one the tech is using isn't, the camera or lens he's working on won't be accurate.

 

[farlymac]

Seeing as I can't afford a colimator, I get by with the old standard method stated by Clarence. And like Chris says, test equipment can go out of spec.

PF

 

[bigeye]

I have successfully used this method: http://www.kyphoto.com/classics/collimator.html

 

[rxmd]

I agree. An autocollimator's primary function in life is to collimate optical elements. Alternatively, ground glass focusing works well and is easy. For even greater accuracy you might try using an optical fiber faceplate instead of the ground glass plate. The image will be sharper and precise. Faceplates can be found as optical surplus parts because they can be a bit expensive new.
David

The optical fiber faceplate trick is very neat, but you have to mount it at the right distance from the lens mount. That depends on the size of the faceplate. It's easy when they fit in the guide rails of the film, but when they're smaller than the width of the film gate it can be tricky to mount right, just if you had an undersized ground glass.

 

[ferider]

I have successfully used this method: http://www.kyphoto.com/classics/collimator.html

I use that, too (the "second method" described in the article).

 

[ErnestoJL]

I use the same method, based on an old Zenit 122 and a 135mm lens. It´s not a real autocollimator, but it works well as long as the tele lens is properly set at infinite.
This is the key of this setup: the autocollimator lens must be set properly at infinite, otherwise results aren´t good.
The method used by all manufacturers was this one, but for safety, they checked their instruments in a regular basis to assure no production problems.
Some technicians would not make that checks, and when this happens, the autocollimator would not bring its nominal accuracy. Then, the frozen glass will tell you the truth.
Regards
Ernesto

 

[L David Tomei]

The optical fiber faceplate trick is very neat, but you have to mount it at the right distance from the lens mount. That depends on the size of the faceplate. It's easy when they fit in the guide rails of the film, but when they're smaller than the width of the film gate it can be tricky to mount right, just if you had an undersized ground glass.

With the faceplate, it is necessary to place one side at the exact film plane. An image focused at that plane will appear on surface. Although very precise compared with ground glass, the faceplate is expensive and would require fibers of less than 25 micron diameter. Actually, I use a 3 cm diameter 10 micron faceplate taper that I have had for 20 years.

 

[Gumby]

I use Clarence's method. Work just fine as long as the infinity object being focused is really far away.

 

[Ken Smith]

I use that, too (the "second method" described in the article).

Ditto here - works simple. Besides - my Dad was gonna throw his FX3 w/tele lens in the trash and it now has a use.

 

[rxmd]

With the faceplate, it is necessary to place one side at the exact film plane. An image focused at that plane will appear on surface. Although very precise compared with ground glass, the faceplate is expensive and would require fibers of less than 25 micron diameter. Actually, I use a 3 cm diameter 10 micron faceplate taper that I have had for 20 years.

Yeah, you posted earlier on this and I then tried it out. It's really neat, a major step up from the "sticky-tape-on-CD-cover" type recommendations that you often get. Are they really that expensive, though? If you buy surplus you can get them for under $10 at places like surplusshed.com (that's for the 35x6 and 35x11 variants, perfect for 35mm cameras because they fit into the film gate on the guide rails, but they don't help you much with a Rolleiflex though because they're too small, unless you have a Rolleikin).

 

[rxmd]

I would like to know more about the fiber optic faceplate gizmo. Can someone explain that?

It's basically an plate that consists of an array of very fine optical fibers. They transmit an image from one side of the plate to the other without changing the focus of the image. The fibers can be arranged in such a way that the image is enlarged, as if you were looking through a loupe. It looks like this: http://www.surplusshed.com/pages/item/l3173.html.

 

[Sarcophilus Harrisii]

By acquiring a small quantity of aluminium oxide from a friendly local optical manufacturer (free of charge) and some glass offcuts from a glazier, myself and a friend were able to manufacture our own ground glass pieces for 6x6 and 35mm film rails at no cost. We even cut some larger pieces for my V700 120 holders, and I made my own anti-reflective glass for scanning curly 120 strips. They all work well in their respective applications.

The optical fibre screens also sound interesting. But using a ground glass and an ebay loupe I haven't had any problems setting focus so far with several different cameras so I guess I am happy with what I have for now. The next step is to fabricate an auto-collimator, and I have been watching cheap binoculars on ebay lately for this purpose.

I think this would be worthwhile eventually, because I have used an SLR a few times, as described in the kyphoto link, to set focus, and it does work well, but for certain camera types (particularly TLRs or folders), I like the idea of being able to inspect the sharpness of the focus target on a collimator across the film plane. Lens board alignment is a critical part of correct adjustment for a TLR, and the main problem, as I see it, with using an SLR is that, although you can dial in the focus with the centre prism or rangefinder accurately, it is less easy to check for good parallelism of the lens and film planes. That's just my own conclusions from my experience to date, other members may have more informed comments to offer on the topic.

After messing around with a 40mm Distagon it was interesting to find that contrary to my initial assumption, it was actually a lot harder to set the focus of a shorter lens correctly. On the other hand a longer lens tends to "snap" in and out of infinity focus much more obviously. Short lenses may seem very forgiving of infinity focus maladjustments, but even if infinity seems OK, close focus can be way off. I learnt from this experience that it's a good idea to do a critical focus check at close range as well. When minimum and infinity are both OK, you know you're good.
Regards,
Brett

t

 

[srtiwari]

My question may be too basic- after determining that the focus IS off (say on a Rolleiflex, or a Leica RF ), how does one actually correct this ? I always assumed focus adjustment was not a DIY project for most people.

 

[Sarcophilus Harrisii]

Which one: Leica or Rollei? It is quite different in each case. They both expose film but could not be more different in their designs.
Regards
Brett

 

[t6un]

I've never used an autocollimator, but I believe it could be easily checked for accuracy by placing a flat mirror directly on it's lens?

There could be differences in results when using the autocollimator with actual film vs. the mirror in the film gate, but an error as big as one helical thread seems to show that the technician, although he may have an autocollimator, doesn't necessaryly use it.

 

[Sarcophilus Harrisii]

Not having owned a Leica I will only point out that, as with any film camera, the lens focus at the film plane must first be correct before other adjustments should be considered. Then, and only then, can the rangefinding system be set to suit. Of course, with most Leicas accepting interchangeable lenses it is not so straightforward as with a fixed lens Yashica or Konica, for instance.

With a Rollei the same principle is your starting point; but a very different adjustment procedure applies.

I wonder, from all the reading I have done (not just on RFF on the web generally), of Rollei (or other TLR) owners asking about their focus problems, if it is always understood that focussing problems come in several different shapes and sizes. In order to adjust the camera, first you need to know what the problem is.

Assuming the lenses and lens mount are parallel to the film plane and the adjustment of the focus cams is sound, with the cameras focus knob set to the infinity stop (∞) there are four basic conditions that will define the relationship between the two lenses:

(1) Taking lens gives good infinity focus at film plane (Eg. as checked with a ground glass on film rails), but the viewing lens infinity focus is off at focus screen;

(2) Viewing lens gives good infinity at focus screen, but taking lens focus is out at film plane;

(3) Taking lens infinity focus is off at film plane, and viewing lens infinity is off at focus screen; or

(4) Taking lens gives good infinity at film plane, and viewing lens gives good infinity at focus screen.
​

Of course, only the last condition is desirable. However you will see that, despite a TLRs unique design features, the problem is essentially the same as with a rangefinder or SLR design; the lens focus and the viewing focus must coincide. So first and foremost, you must verify which system is off.

I suspect that more discrepancies between the lenses are a result of the viewing system maladjustment, (condition 1 above). In which case, with a Rollei--after verifying that the taking lens focus is in fact correct--the adjustment is normally a case of loosening the fastening screw or ring of the viewing lens, and adjusting the focus through the viewfinder (with a magnifier or loupe) until it is correct. Recheck and compare lenses at closer distances to confirm they agree across the range; this is critical. You then tighten the lock screw to secure the correct focus. Some degree of dis-assembly is involved in accessing the viewing lens adjustment (except in the case of the Tele Rolleiflex, for instance, where the set screws are external). Not sure about the Wide though.

If the taking lens is out at the infinity stop but the viewing lens is correct (condition 2), the lock screw of the focussing knob must be loosened and the knob re-positioned, so that when the taking lens gives good infinity focus, it is doing so at the infinity stop. Of course, having reset the infinity stop to correct the taking lens, the viewfinder focus that was previously correct will now be out, and it will subsequently be necessary to reset this as described above. For this reason, as you will appreciate, the only proper place to start an adjustment process is always with a taking lens that is accurately focussing on the film plane, because otherwise, you will just end up chasing your tail.

If you are unlucky enough to have a Rollei that has both lenses out of adjustment (condition 3), the procedure isn't really any different, in that, you first get the taking lens right, and then dial the viewing lens in to match it. But in that scenario, you will probably have bigger problems to deal with, such as incorrect parallelism of the lens board or focus cams out of adjustment...more on that to follow in my Tele Rollei thread.
Regards,
Brett

 

[Eugen Mezei]

It's basically an plate that consists of an array of very fine optical fibers. They transmit an image from one side of the plate to the other without changing the focus of the image. The fibers can be arranged in such a way that the image is enlarged, as if you were looking through a loupe. It looks like this: http://www.surplusshed.com/pages/item/l3173.html.

That is not very expensive, but sending it to me would cost several times more than the price of the faceplate.
Now it states in the description that "They are used in CRT's, camera tubes, CCD coupling, and image intensifiers."
So it came to my mind I could look in some old CRT computer monitors. Am I wrong and if not, where exactly to look?
In a different formulation: Does every CRT contain such a part?

 

[Murray Kelly]

With method 2 how would a digital camera with a 10x zoom and put on scenic (infinity fixed) go as the test camera instead of an SLR?