Focusing Consumer vs Pro SLR's

[Timmyjoe]

For about ten years I was a motion picture camera service tech. One of the most critical things on a motion picture camera, especially the reflex cameras like the ARRIFLEX's, was making sure the Lens Flange to ground glass (focus screen) distance, and the Lens Flange to film plane distance, were an exact match. If these were off as little as five thousandths of a millimeter (about one tenth the diameter of a human hair) the resulting images from the camera would be out of focus.

Applying the same principle to an SLR, which is a basic reflex camera, the mechanical distance from the lens flange (where the lens seats on the camera body) to the film plane, and the optical distance from the lens flange to the focusing surface of the ground glass, should also be exactly the same.

What I have found from recent testing is that on Pro level SLR cameras, like the Nikon F4s, F3HP, Canon F-1 (old & NEW), etc. this generally holds true. What surprised me is that on the consumer versions of these cameras, Nikon FM-2N, Canon A-1, etc. this doesn't hold true. Even though the Nikon FM-2N and Canon A-1 I recently tested had both just had a complete CLA by reputable repair person.

I am fortunate to have some nice Canon FD and Nikon glass, which I tend to try to shoot wide open. This disparity in Flange to film/focus screen can really play havoc with image sharpness when using the consumer versions of these cameras. I had to shim the focus screen on the Canon A-1 by 101 thousandths of a millimeter and the Nikon FM-2N by 190 thousandths of a millimeter (what we would call "Off by a Mile" in motion picture cameras) to bring them into spec. Now both camera are spot on and focus as well as their Pro cousins.

I realize that many consumer cameras are mated with consumer lenses, which may have a largest aperture of maybe f4 or f5.6, and therefore this issue would not be as big of a problem (at smaller apertures the depth of focus is greater so you don't notice the disparity). But if you are trying to use fast glass, say f1.4 or f1.2, wide open, a consumer SLR may be working against you getting sharp images.

Just an observation.

Best,
-Tim

 

[Vics]

Could you elaborate a bit in how this works in my Nikon FM2n? Materials, etc.? Methodology?

 

[Finglas]

Tim,
thank you, very informative indeed.
Always worth visiting here and actually learning new things,

I suppose the M mount RF cameras, with the lens a lot closer to the film plane probably doesn't suffer the same as the aforementioned FM2n etc.
Of course the lenses for RF cameras usually have wider apertures' too.
They're made for it really in comparison to the consumer level SLR's.

I wonder how many tech's who perform CLA could shim the focus screen to such small margins or would they even be aware of this? I for one had no idea until you raised the point in this topic.

thanks again,

 

[Ranchu]

Good job, those cameras really were off by a ton. I'm surprised they weren't sent back sooner to be fixed. I'm also curious about your methodology, I assume you shot film?...

 

[Timmyjoe]

Hi John,

Your's is the easy one so I'll do it first. A rangefinder focuses in an entirely different way from an SLR. Too complicated to get into on this thread, but there are other threads dealing with it.

Hi Vic,

Basics first. The way an SLR (be it an FM2n or whatever) works is that the focus screen (or ground glass) is positioned in such a way that light coming in from the lens, reflects off the mirror and makes an image on your ground glass. When you press the shutter, the mirror flips up and the light travels to the film plane, making the picture.

If the distance between the lens mount and the focus screen, and the lens mount and the film plane, isn't exactly the same, your pictures will be out of focus.

To test this you need a collimator, an object you can focus on at a good distance (the moon works great for this), proper tools and shim stock, and a lot of patience. You also need a "normal" lens in the range of 40mm to 50mm, and it would be great if the lens focused a bit past infinity.

You load a roll of test film into the camera. Advance the film to about the seventh or eighth frame. Using the collimator, and your hopefully over stroking "normal" lens, you find the precise infinity focus point of the lens onto the film plane (by holding the shutter open on the B setting). Then you take the camera and lens combo, and focus the lens through the viewfinder on your distant object (the moon). Then check to see if the precise infinity focus point on the lens is the same for both the collimator focus and the viewfinder focus. If so, the focusing screen is spot on. If not, it will need to be shimmed.

Best,
-Tim

 

[Timmyjoe]

I'm surprised they weren't sent back sooner to be fixed.

To be honest, I'm not sure many SLR camera service techs know how to do this. I have never had an SLR camera serviced where the tech shimmed the focus screen. The only reason I know how to do this is because I come from the motion picture camera side of things, where this is standard operating procedure.

It's amazing how sharp your pictures can be when your camera is set up precisely. Lenses that you think you can't use wide open, suddenly have a whole new life.

Best,
-Tim

 

[Ranchu]

No doubt! I only learned this once dslr's came out, and it became more obvious. That seems like an awful lot to me though. Sometimes it's difficult to find the 'perfect' shim as the tolerances are so tiny. How do you see if the lens is focused on the film plane on B with film in?

 

[BillBingham2]

Interesting test Tim, thanks.

If you get board between 3 and 4 some morning, can you please run a test on an old Nikkormat FTn? I'd love to see if there is as much play from an older consumer grade camera.

Your results do not surprise me as quality control always costs more than just time. I would suspect the amount of enlargement most people can afford used to be somewhat correlates to their investment in camera and glass. The amount of enlargement moving picture folks get from their gear is an order of magnitude more than most of still folks not doing it for a living do.

This might account for some cameras (of the same model) having better focusing than others.

Thanks again.

B2 (;->

 

[Timmyjoe]

Ranchu, the collimator focuses off the front surface of the film. A collimator like mine won't work with a digital sensor.

Best,
-Tim

 

[Ranchu]

Excellent, I had no idea that was how they worked, thank you.

 

[Timmyjoe]

B2, I suspect that each Nikkormat FTn would be different. I think it's a quality control issue originating from the factory. With the lower priced consumer cameras, especially if they thought the customers would be using slower lenses, I bet the factories didn't want to put in the extra effort for the precision this requires. On the pro bodies, they did.

So I would figure the Nikkormat FTn could be anywhere from 200 thousandths of a millimeter off, to spot on. And each camera body would be an individual case.

Best,
-Tim

 

[Timmyjoe]

I first really noticed this when doing a focus test (test charts) with a NEW F-1 and a freshly serviced A-1, with a couple of Canon FD L lenses that had also just come back from service. The images I shot on Tri-X with the NEW F-1 with each lens was sharper at f1.2, f1.4, f2.0 up to f2.8 than the same image shot from the same lens on the A-1. I knew that couldn't be right, so I checked the focus distances on the A-1 and sure enough they were off. (I had previously tested the NEW F-1 and it was spot on). So then I started going through all my old film SLR's and found a trend, the pro bodies were mostly spot on, the consumer bodies were off. After shimming the A-1 I repeated the focus test (test charts) and this time the focus was spot on with both cameras.

Best,
-Tim

 

[Ranchu]

Bill, For some piece of mind, if you have more than one Nikon you can test them against each other with the same lens. Focus on something a few times keeping track of where the distance scale winds up, and put the lens on the other body and do the same. I always wind up with two identical bodies of ones I like, somehow.

 

[aizan]

You load a roll of test film into the camera. Advance the film to about the seventh or eighth frame. Using the collimator, and your hopefully over stroking "normal" lens, you find the precise infinity focus point of the lens onto the film plane (by holding the shutter open on the B setting). Then you take the camera and lens combo, and focus the lens through the viewfinder on your distant object (the moon). Then check to see if the precise infinity focus point on the lens is the same for both the collimator focus and the viewfinder focus. If so, the focusing screen is spot on. If not, it will need to be shimmed.

just to clarify, the back door is open when you check infinity focus on the film plane, right? also, is it preferable to load film instead of putting ground glass over the film gate?

what's a good, reasonably priced collimator? i've used the diy 2nd slr with a telephoto trick, but it's a little awkward.

 

[Ranchu]

I think he logged out. I'll give it a try. He's got the back closed, which is the best way to do it because the film rails a ground glass would rest on are forward of where the film lies on the pressure plate.

 

[Timmyjoe]

A collimator is a precision optical instrument for setting up lenses and reflex camera systems. It basically shines a light through the lens and onto either the film in the camera, or the focusing screen (depending if the shutter is open or not). The collimator has internal focus test patterns, and you focus the camera lens until these test patterns are precisely sharp. At that point, the lens is set at exactly infinity (for that lens and camera combination). You can do a rough approximation of this with placing a ground glass on the film rails with the camera back open, but it is not nearly as precise or accurate.

Best,
-Tim

 

[Vics]

Now I just need to find someone who can do this on my FM and FM2n. I've always had a little suspicion that my pics from those cameras weren't as sharp as the ones from my '59 F. I always chalked it up to pilot error.

 

[Timmyjoe]

Vic, I know what you're saying. I've had my FM-2n since 1995 and it hasn't gotten used much because the images from it were never as good as what I got from my F4s. Now I know why. (And I too thought it was pilot error).

Best,
-Tim

 

[aizan]

what kind of collimator are we talking about anyway?

something like this?

 

[redisburning]

so, just to clarify here, the issue is whether the focus screen is in the right place relative to the film plane, right?

I thought, though perhaps incorrectly, that with an slr as long as you focused the lens on the screen at the point of highest contrast, and the screen was in the right place relative to the film, that that would give you the highest contrast.