Nikon design defect (light leak between flange and AI ring)

In brief...

Nikon SLRs, at least all that I've seen that are AI capable, leak

light through the tiny space between the bayonet flange and the

aperture indexing ring.

 

In detail...

This isn't a light leak in the traditional sense, i.e. it isn't light

that's getting into the camera and onto the film when the shutter is

closed, but it -is- undesirable, non-image forming light which gets

into the camera, and on film, during the time that the shutter is

open, and which under certain lighting/shooting situations can leave

noticeable streaks/fog on the film.

I started noticing it as I developed an increasing interest in

night photography, and long exposure daytime photography using

pinholes or with lots of ND filtering on normal lenses (10 stops or

more). I first noticed the problem on night shots taken where I was

near, or directly beneath, a bright light source, but shooting into a

dark scene. At first thought it was lens flare, but no amount of

careful hooding/shielding of the lens seemed to solve the problem,

and that's when I noticed, examining the bad negatives with the

loupe, that the "extra light" was being recorded on the negatives out

beyond the edge of the normal frame, which clued me in to the fact

that the light was hitting the negative at a highly oblique angle,

i.e., not coming through the lens, but getting onto the film from

somewhere else.

I investigated further by looking through the back of the

camera, with shutter held open, while shining a penlight around the

lens mounting area, and to my surprise found that the source of the

light leaks is -really- obvious, and instantly revealed with the

penlight. It's entering the camera through a space between the

bayonet flange and aperture indexing ring.

Basically, the bottom side of the bayonet flange is flat, as is

the top side of the aperture coupling ring. There are no ridges or

baffles on either. As the flange and aperture ring surfaces are

adjacent to one another, and the aperture ring must be able to turn

freely, there -must- be a small but omnipresent gap between those two

surfaces, and therefore, -some- amount of light will always be able

to enter in through that gap. In normal daylight shooting this bit

of light coming in doesn't have a noticeable effect, but in some of

the situations where I'm shooting, it becomes a serious, potentially

shot-destroying problem.

That this is a fundamental problem of design and construction,

and not simply a problem with the two cameras I have (an F3 that is

my primary, and FG that I have around for backup), I've confirmed by

going around to a number of camera stores and asking if I can perform

my leak test on various cameras from their stock. Every single Nikon

I've come across so far, from 20+ year old used cameras, to the

newest manufactur, just off the shelf and never used models, fails

the leak test.

There are obvious, expedient workarounds, but to be honest I

feel rather stupid running around putting a black rubber band at the

base of a very expensive lens, mounted to a very expensive camera, so

that I can keep a stupid, designed-in light leak from ruining these

night/long exposure day shots... and then of course I still have the

problem that if I forget to do this, or don't do it right, I can

still end up with a nice fog streak on the negative that f's up the

shot.

Is there anybody else out there who's also discovered this

problem, and has found a permanent solution (aside from the obvious

one of dumping the Nikon gear for something else...).

 

-Z

Hi Zack,

 

Interesting observation. It must be a frustrating problem for your photography.

 

I've never noticed this problem myself, not even in night shots up to 5 minutes long. But I could have just not got into the same conditions as you (near a bright light, for instance.) And I don't take many such shots.

 

I have FM, FE and FM3 bodies, all my lenses are AI or AIS. I'll have to check for this leak when I get home.

 

I don't have a solution to offer, but before you junk all your Nikon gear best to check if other makers' kit doesn't have some similar defect.

 

Obviously their lens mounts are different, however not many SLR manufacturers are going to assume they need to proof their cameras for such specialised shooting.

It's tempting to dismiss this as trollish, or worse, as some kind of obsessive-compulsive issue. For starters, Nikons I'm intimate with don't appear to be built along the lines you describe. The entire lens mount area is part of a solid, precision casting. The AI index ring is concentric with the outer surface of the solid mounting flange and doesn't intrude in any way. There's just no obvious way stray light could enter other than through the lens, as intended(?). Miraculously, this serious design flaw has apparently eluded legions of photographers for decades.

I to have seen this horrible problem; but only with newer Enlarging lenses with illuminated rings.

Only time I've ever had this issue is using a camera on a scope or other scientific instrument in a dark cabinet with high ambient light.

 

It isn't unique to Nikon bayonet cameras. Shield your camera when you are in this extreme of conditions. But you have to have some big lens hood to keep front element flare from being far worse than this leak.

I just took the back off my F3, put the shutter on T, put a lenscap

on, and shone a flashlight all around. I can't find a hint of

the problem, at least not with my AI 50mm f1.4. The aperture

ring doesn't seem to play any part in the lightproofing of

the lens mount. There are two metal pieces, one on the

front of the camera, and one on the back of the lens, that

come together pretty tightly and seal out the light. But

they're planar and shiny, so I could imagine a bit of light

might somehow squeeze in between them. If it did, it would

be stopped by the part of the lens that has flanges

that mate with the bayonet part of the camera mount.

<p>

It's interesting that on pre-AI lenses, the aperture ring

forms a "skirt" around the outside of the lens mount, and

this skirt would provide further redundant lightproofing.

On AI lenses, only a remnant of the "skirt" is present,

just the AI indexing ridge. If you don't have any pre-AI

lenses, just imagine the AI indexing ridge extending all

the way around the entire circumfrence of the aperture

ring. So you might consider switching to pre-AI

lenses if your circumstances really show this to be

a problem.

<p>

But I'll repeat, I can't see any hint of a problem on my camera

and lenses, and quite a few photographers have used this system

for quite a few photos, and this is the first I've heard anyone

report a design flaw causing a light leak around the lens mount.

I tried the same test with several lenses on my F3 in T mode and with the back removed. All my lenses leaked a LITTLE light around the bottom of the bayonet, and some leaked light near the AI aperture lever. None of these leaks was very exciting, except with the 35-70mm f2.8 AFD lens, because it leaked quite a lot around the AI lever area. The lenses which leaked the least were the PC lenses (non-AI) and the Tamron macro. Maybe the wider flange on the Tamron Adaptall mount helps shield light. I couldn't detect these leaks by looking through the prism, but with the mirror locked up and looking through the gate, they were plainly visible. If a bright light shines on the lens flange at EXACTLY the right angle, the worst of these leaks would affect a long exposure, in my opinion.

 

Off hand, there doesn't seem to be a neat and effective fix. But I'm going to apply a little lithium grease to the film flanges of these lenses tommorow and see if the problem improves any at all. Are there any other ideas?

Richard,

 

> The aperture ring doesn't seem to play any part in

> the lightproofing of the lens mount.

 

Yeah, the lens mount itself is essentially light tight and is not the problem. For light to get in between the lens and lens mount, it would have to enter through whatever tiny gap is there (entering radially), make a right angle turn, and continue squeezing down the axial gap to enter the film chamber. Probably 1 in a billion photons gets that far, if that. The problem appears to be between the lens mount and the AI ring. There's a tiny gap between the mount and the AI ring. Any light which makes it into that gap, traveling perpendicular to the lens axis, basically has a straight shot into the film chamber. If you remove the 4 screws from the front of the lens mounting flange and pull it off, you'll see what I'm referring to. The AI ring sits directly underneath it, and at least on my F3, the bottom surface of the lens mount and top of the AI ring are both flat. There are no ridges or other projections to serve as a barrier to light trying to enter. On my FG there is a -very- slight lip on the underside of the lens mounting flange... not enough to keep the light out. Remember the light that gets in is traveling axially, so if you're shining the flashlight in at any other angle you won't see it. (Been there, done that... :-) Easiest thing, take one of those really bright, white keychain squeeze lights (the LED kind), hold it near the interface between the lens mount flange and AI ring, and shine it there, so that the light is hitting it, again, perpendicular to the axis of the lens.

I have to revise my earlier statement to the effect that every AI capable Nikon SLR I've looked at definitely had this problem. After stunning the sales guy at one of my local camera shops by demonstrating the problem to him on a brand new FM3, I convinced him to let me do my test on a new F5 they had in the case. On the F5 I couldn't see light getting through, but on the other hand, the F5 has so many lumps and protrusions all around the lens mount, I couldn't tell if that's because Nikon has finally realized they had this problem and fixed it on the F5, or if it's because I just couldn't get the light shining in there at the necessary perpendicular angle. I was totally unsuccesful in convincing him to let me take the front of the camera apart to see whether the lens mount / AI ring geometry on the F5 has been significantly changed from earlier models... :-)

 

-Z

Charles,

 

You wrote:

 

>Off hand, there doesn't seem to be a neat and effective fix. But I'm >going to apply a little lithium grease to the film flanges of these >lenses tommorow and see if the problem improves any at all. Are >there any other ideas?

 

I'm a bit concerned that any grease you put there might, on a warm day, run onto the optics or somewhere else it shouldn't, creating a worse problem. At best the stickiness of the grease will still attract and hold dirt and other stuff that you probably don't want. For the time being my solution is... uhh... hair elastics. You know, the ~1/4" diameter, fabric covered kind that girls use for their ponytails. Because they're round they tend to naturally fit themselves into that gap between the lens skirt (or whatever it's called) and the camera body, which prevents most light from ever reaching the problem area, and they're a lot easier to slide down there where they need to be than regular, flat rubber bands are. I've shot during the day with an ND3 filter on the lens, with and without one of these bands on there, and verified that they seem to cure the problem.

I never realized that my interest in photography would require me to become an expert on women's hair accessories... but for what it's worth, Scunci seems to make the best ones. They make some where the ends of the loop appear to be heat fused together (rather than held by the metal crimp thing). I bought a half dozen or so at my local Walgreens for a buck and threw them into my camera bag.

 

-Z

Whoops, I must disqualify myself from this discussion now due to my non-repeatable results!

 

I repeated the trial with my F3, but this time I taped over the ADR window and closed the viewfinder shutter. I re-tested it with a body cap, and then with a variety of lenses. I waited until my vision was adapted to darkness and I looked right into the mirror box. I saw nothing. I couldn't repeat any of my results from my first trial, EXCEPT with the 35-70 f2.8 AFD zoom. It definitely has a light leak around the little tab that locks the aperture ring at the minimum setting. I can block this leak with my finger. Otherwise, I conclude that I am not experiencing the same problems you described. I want to withdraw my earlier comments and apologize for any inconvenience I may have caused.

 

Charlie

>I repeated the trial with my F3, but this time I taped over the

>ADR window and closed the viewfinder shutter. I re-tested it

>with a body cap...

 

Huh? Your ADR window, viewfinders, lenses, body caps, etc got nothing to do with it. This has definitely got to be one of those instances where a picture would be worth 1000 words. If you're not seeing what I'm seeing, I think you must not be looking where I'm looking. I'll take some picture of the light leak demo, as well as some of my F3 with the lens mount flange removed so you can see what parts are under there on my camera, and maybe someone can see if there's any differences between my camera and theres...

Yes, a picture would help. I'm confused by the description

of "axially" sometimes, and "perpendicular to the lens axis"

other times. To me, those two descriptions are at right

angles to each other. Did you mean "perpendicular to the

film plane", perhaps?

<p>

At any rate, it sounds like you're saying it's getting through

somewhere outside of the lens mount bayonet, near the

<em>camera's</em> AI coupling ring (not the AI coupling ridge on

the lens). But it appears to me that a leak around there

would put light into the camera body on the outside of the

mirror box, where it would still have to get through

or around the mirror box walls to reach the film, so I'm obviously

missing something. I've taken another long look, with camera

back off and shutter open, holding a bright light all around

the outside of the camera, both aimed axially and

perpendicular to the lens axis, and I

still can't see any hint of a leak on my F3, FM, or FM2n.

Zack,

 

I am not saying you aren't correct, not at all.

 

I'm just saying that so far I haven't been able to see it. Could easily be I haven't aimed the light just right, or something similar. The only reason I covered up the ADR port, used a body cap, and closed the viewfinder shutter on the F3 was to be extra sure stray light couln't get into the mirror box any other way. I had my dark-adapted eyeball right next to the film gate, and this time I didn't see any leaks except from that one zoom lens. Doesn't yet mean I disagree with you, though.

Well, I haven't seen the light leak yet, but this is the first time I've ever set the F3 on T and looked in the gate. Found and retrieved an elusive piece of fuzz that was stuck just inside the shutter!

I just checked an FE2 with MF AIS lenses. The only light leak I find

is directly on the flange surface (of the lens) where the locking pin is located.

All my AIS MF lenses have a tiny gap in the flange surface at this spot.

Shining a bright flashlight directly into this gap will result in

light entering the mirror box. I cannot find any other light leaks

or possible entry points other than the integrity of the flange

surface itself. If either flange surface (lens or camera) is warped

or loose, I would imagine this could cause a significant light leak.

Upon Further testing....

 

I would add to my last post: Only one Nikkor lens I own exhibited the

light leak on the flange at the locking pin. A 55 2.8 Micro Nikkor.

 

(Of course the first and only one I tested before posting above)

I have 5 other Nikkor manual focus primes and none of them would

show this light leak. It looks like the aperture ring on the other

lenses covers this gap better.

 

I've done a few light leak checks working with large format and a

few old folders. It's not easy working in a darkroom with dark

adapted eyes and a bright flashlight... (if you shine the light

in your face, you may as well take a 1/2 hour break until your

eyes recover)

 

I couldn't find any other light leaks. Also checked the viewfinder

prism to the mirror box (with the mirror up) no leaks.

Ok, I will post one more retraction:

 

I checked again for light leaks coming in from the viewfinder/prism

into the mirror box (with the mirror up). I made a better light seal

to shine the mag light into the viewfinder while sealing any stray

light and this is by far the biggest light leak: the area around

the mirror when it is up. I don't know how much a problem this will

be for your situation but for time exposures where there is a bright light behind the camera, covering the eyepiece would be a

good idea.

 

behind