Weird exposure problem

One of the students in an Intermediate Photography course at the college where I work attempted a night photography assignment and her negatives exhibit a weird problem I have never seen before and am at a loss to explain. Throughout the roll and with different image locations, there is a faint inverted image of the frame that was being exposed superimposed on the main image. The faint image appears to be rotated 180 degrees from the main image rather than being a reflected mirror image. (See the attached print test strips. The image with the star has been manipulated so that rectangular part shows better on the scan.) She was using an older Canon AE-1 on "B" and bracketing exposures from 1 second out to 2 minutes at f/5.6. The foam padding around the viewfinder where the mirror seats when a image is being recorded on the film has deteriorated and is essentially missing. All I can think of is some faint image is somehow being reflected back and forth off the mirror or pentaprism surfaces because of that gap and ending up on the film upside-down.

Any other ideas?

She is bracketing exposures as you mentioned, but turning the camera round between exposures and the camera is double exposing (it's a Canon).

I had some ghosting/double images when scanning some of my high contrast slides. After initially blaming my lenses I found the film did not have the ghost image. It was coming from my scanner and internal reflections. Not nearly as bad as what you show though and not inverted either. That really is weird.

Poltergeist?

Lamar said:

I had some ghosting/double images when scanning some of my high contrast slides. After initially blaming my lenses I found the film did not have the ghost image. It was coming from my scanner and internal reflections. Not nearly as bad as what you show though and not inverted either. That really is weird.

Click to expand...

It's definitely on the negatives and throughout the roll. I also played around with it a bit in Photoshop and confirmed the secondary inverted image is the same size as the normal image, although it does not align perfectly when rotated 180 degrees. The angular relationship is the same but the secondary image is offset a little bit when rotated.

smieglitz said:

It's definitely on the negatives and throughout the roll.

Click to expand...

So can you rule out cliveh's explanation?

pentaxuser

I had this problem once in a similar situation - high contrast night shots (of St. Peter's Basilica and the Vatican, if anyone's interested).

All I did to rectify this was make sure that I used the viewfinder cover before I took the shot - my only guess as to what caused this was internal reflections from somewhere within the pentaprism. The camera I was using was an AE-1 too, with a less-than-perfect mirror - maybe some light leaked through the old (possibly worn) coating on the mirror?

Don't even know if this is possible, but I can't think of anything else that would cease to be an issue by simply covering the VF.

jimbop said:

I had this problem once - similar situation, high contrast night shots (of St. Peter's Basilica and the Vatican, if anyone's interested!!). It turned out to be internal reflections from within the camera - presumably stray light reflecting around inside the pentaprism.

All I did to rectify was make shire that I used the eyepiece cover - I guess that was enough to reduce the reflections from the viewfinder eyepiece.

Click to expand...

I'll have her try this again while covering the eyepiece.

But still, if the mirror flips up and blocks the imaging light from hitting the viewfinder/pentaprism during normal exposure, why would blocking the viewfinder from the rear affect an internal reflection, especially one originating from the front of the camera)? I can understand how it might affect the meter reading depending on where the cell was located, but the image? The secondary image does not appear smeared or distorted in any way which I would assume would happen if the mirror was somehow refracting the image. Also, why would a night exposure present this problem when daylight exposures do not exhibit it? I would again assume a daylight image would have a general fog or show the brightest areas in a scene (specular reflections, etc.,) also copied from the viewfinder/pentaprism if that was the source.

I have seen pressure plate dimpled patterns reflected and imaged over an exposure (especially with HIE infrared and Olympus OM cameras) but if that was the source, the secondary image would not be inverted relative to the normal one. Ditto if the mirror was acting as a pellicle mirror. (Which BTW was originally introduced into 35mm cameras by Canon and only followed/copied later by Nikon after an 11-year lapse.)

The problem remains (if you will excuse the pun) baffling.

If it were a reflection, wouldn't the image be reversed, not simply inverted?
A prism may not reverse the image, but I'd think it would be difficult to reflect an whole image in this manner while the mirror is moving - I'd expect more of a light-leak from the area, not a full ghost image, unless there are a number of reflections occurring to produce this. Just guessing.

What lens is the student using? Filters, etc? Could they be photographing through a window? Very wierd.

Can she lock up the mirror before making the exposure? Might not solve it, but it might tell you something.

"There is a fifth dimension beyond that which is known to man. It is a dimension as vast as space and as timeless as infinity. It is the middle ground between light and shadow, between science and superstition. You're moving into a land of both shadow and substance, of things and ideas. You've just crossed over into the Twilight Zone. "

get her to replace the foam

Xmas said:

get her to replace the foam

Click to expand...

Stranger things have happened, but I's start with the foam. It is not a "bumper" as it's commonly called but a light trap to prevent light
coming through the eyepiece and fogging the film.
Your student is making images at night. Why couldn't the effect be caused by light through the VF?

Probably all sorts of logical reasons, but the easiest to eliminate is the foam and VF.

I don't think so but would a filter be the cause of the reflection?

Check the back of the mirror for a reflective surface.

John Koehrer said:

Stranger things have happened, but I's start with the foam. It is not a "bumper" as it's commonly called but a light trap to prevent light
coming through the eyepiece and fogging the film.
Your student is making images at night. Why couldn't the effect be caused by light through the VF?

Probably all sorts of logical reasons, but the easiest to eliminate is the foam and VF.

I don't think so but would a filter be the cause of the reflection?

Click to expand...

confirmed if you take photos at night the only filters that are 'safe' to use are the non planar ones that Pentax used to market circa 1965. The light can reflect off the film on to the filter and back to the film where it will be in focus but upside down. The Pentax ones merely spread the hot spots so I've not bothered.

Dont ask about digital or plate cameras.

But for example F2s don't need foam on their back doors but have foam all the way round the mirror box as well as a light trap all the way round the mirror they act together just like a dark room light trap.

So you need to clone the as new foam style and only use domed filters if you must as even plane multi coated are a risk.

I have a ping pong bat with black velvet both sides for night shots.

But, if the film is so reflective and there is something (filter, back of mirror, etc.,) reflecting the film reflection back to the film inverted, why wouldn't this be obvious in daylight exposures as well? The path of incoming light rays would remain the same both at night and during the day. The absolute intensity would be greater during the day so I would expect the effect to be intensified then, but that does not seem to be the case. Also, If one of these explanations did fit, why is the inverted image sharp and the same size as the normal exposure? I would think all those reflections would be causing the light to travel a different focal length and thus be out-of-focus.

The day vs night comparison really complicates any explanation.

Some other thoughts: the light rays coming into the camera are refracted and travel a path the either ends up with a focal point on the groundglass screen, or when the mirror is out of the way, on the film. Those paths have identical focal lengths. The image coming through the pentaprism is not being focused at the same focal length as the lens. Every part of the screen image is at the same plane and the diopter in the viewfinder has a separate focal length that brings that planar screen image into focus. And if the viewfinder groundglass image was being reflected, refracted, or transmitted, the mirror would have to be in the down position and the shutter closed. If the mirror was up, there wouldn't be any image reflected to the screen as all the light would be traveling through the camera directly to the film.



The image on the focusing screen is reversed laterally when viewing without a pentaprism to correct it but it appears upright (ala TLR as above) while the image at the film plane will be inverted and laterally reversed on the emulsion surface (ala view camera groundglass image or taking lens of a TLR). But, the focal-plane shutter is closed when the mirror of a 35mm slr is in the light path so the film couldn't receive an image at that time. The mirror must be out of the way to image at the film plane. But if the mirror is out of the way, there is no image on the groundglass screen of the viewfinder, and nothing is reversing the image coming from the lens that would put it upside down at the film plane at that focal distance. And even if something were below that screen, any image from it would have to be reflecting from something facing the opposite direction of the mirror when down for normal viewing. If there was a reflective something at 90 degrees to that resting mirror orientation and the shutter was open, the reflection would be inverted and laterally reversed at the emulsion. But there is nothing there in that position and angle when the shutter curtain is open (or closed for that matter). But if there was, the secondary image on the emulsion could be recorded as it appears on the now-developed film.

My brain hurts...

Xmas said:

confirmed if you take photos at night the only filters that are 'safe' to use are the non planar ones that Pentax used to market circa 1965. The light can reflect off the film on to the filter and back to the film where it will be in focus but upside down.

Click to expand...

Confirmed. The test to prove the point is so simple you will have to laugh...

Take any SLR with a filter indoors and turn on a light that you can focus on.

Loosen the filter and move it around. You will see the ghosts on the ground glass - upside down and right reading. They will move about when you move the filter so you will know that's where they come from.

Ghosts are not perfectly in-focus but close enough if you were to stop down. Notice also how the ghost (with lens wide open) looks about the same intensity as the direct light source when you stop down to about f/16

So obviously you don't have as significant a problem during daytime, the effect is not as noticeable during the day as it would be at night because at night you often include lights that are so much brighter than the background.

But next time I hear it said that "using a daylight or UV filter" does not significantly impact image quality... I will remember this simple test and what I just saw.

Yep, filter. Here is a perfectly clear example of a UV filter at night...




-X

I have seen that before. I didn't realize it came from a reflection off the UV filter though. I just thought it was a lens anomaly.

Fast14riot said:

Yep, filter. Here is a perfectly clear example of a UV filter at night...

-X

Click to expand...

OK. Thanks. I will find out if she had a filter on the lens. If I'm understanding this correctly. only a flat ("planar") filter will cause this effect with a strong light source present in a high contrast lighting situation. If no filter, no secondary inverted image, and nonplanar filters or rounded lens elements will not cause the effect.

This sounds like the same principle used in an old fun house trick where a person or ghost appears to be along side you when you look in a mirror. What you're really seeing is a reflection on a piece of glass that is in front of the mirror, but not parallel to it. The object being reflected is out of your field of view.

The Regent is a great subject as well as venue. We were really happy when the city of Allegan restored the marquee. Old theaters are great.

smieglitz said:

OK. Thanks. I will find out if she had a filter on the lens. If I'm understanding this correctly. only a flat ("planar") filter will cause this effect with a strong light source present in a high contrast lighting situation. If no filter, no secondary inverted image, and nonplanar filters or rounded lens elements will not cause the effect.

Click to expand...

Correct. If she had the lens pressed up to a window, the same might happen.

E. von Hoegh said:

Correct. If she had the lens pressed up to a window, the same might happen.

Click to expand...

A window or uncoated filter would real bad.
A single coated filter less so
A multi coated better
But no filter best

I've only ever seen the Pentax non planar filter advertisements once in 65 or so getting one in your size probably mission impossible.

The ghost image will be pin sharp if the primary image is pin sharp but the ghost will soften faster than the real image Id need to think more about the latter.

The rule is remove filters for night shots.

Two other things are important the reflection off the film on to filter and back to film does occur on the lens air to glass surfaces but the curved surfaces spread the light and can create spurious distorted images.

The iris similarly.

If you have two different lenses take two shots.

This does not compromise using a multi coated filter during the day as the ghost image will be normally be below the fog level and not detectable.

But note normally does not cover shooting with a low sun in frame or just outside frame or using a centre filter with a Hologon on a sunny day.

An M8 with an IR filter and low sun near to frame similarly needs the filter removed.

I normally use a single coated lens and a single coated filter cause I like the signature but I don't take low sun shots.

I keep clear or uv filters on all my lenses, as well as caps when not actually being used (I dislike cleaning lenses, it's better to keep them clean). But, whenever difficult situations arise or I want the Nth degree of what the lens is capable of, the filter comes off. At this point the camera is probably on a tripod, so it isn't the hassle some like to make it out to be. My preference is for muticoated filters, even (especially!) on uncoated lenses, single coated are second, I don't use uncoated unless I have to.