Nice RB67, but there's this one little thing

[yessammassey]

I picked up an RB67 after dealing with some older 6x6 cameras with horizontal cloth shutters. Frame exposure consistency issues were coming up a little to often, so I figured that I could do away with them by switching to a leaf shutter. The bump in format size was more than just a nice bonus, too.

I'm using a Pro-S, with 90mm f/3.8 C & 65mm f/4 K/L. I have been cropping out everything past the blue edges here -

But lately, in part because I'm so used to closely scrutinizing my scans for uneven exposure, I've started to notice something a little off in the color balance of the vertical edges of the negatives from the RB67.

Now, this effect can be subtle, but hopefully this image illustrates what I'm trying to get at. Check out the color of the red fire lane curb stripe on the left. In reality, it should be the same color as the rest of the red paint -

Here's another example. This time, it's on the right. Fire lane curbs do a good job of emphasizing the existence of whatever kind of problem this is. On other surfaces, like green grass, it's more subtle -

This effect isn't always apparent. Much of the time, it's basically invisible. And as you can see, the side that gets affected can change. That kind of makes me doubt that a light leak is the culprit, and I think there are fresh seals everywhere but around the mirror, but I guess there could be similar light leaks on both sides of the camera. That could cause the appearance of the aberration to occur on one side or the other, depending on the direction of the light source. Maybe light is getting in around the mirror?

It's not likely due to the lens shutters, I suppose?

Maybe an internal reflection of some sort? Flare? An optical issue with the lens elements?

I'm just not familiar with all the quirks of this camera yet, nor am I very good at diagnosing the causes of film aberrations.

One last question: does that blue crop line look correctly placed? I figure that the outer right and left sides show where the film curves away from the pressure plate, and therefore isn't perfectly flat, and should be cropped out.

 

[MattKing]

I think that the change in hue is just the angle of the light and the effect it has on irregular surfaces that are extremely off axis.
And I print a fair percentage of my RB67 negatives full frame and am usually happy with the sharpness and resolution right at the edge of the image. AFAIK, the pressure plate is larger than that image area.
Neither of your lenses is a macro lens, so there will most likely be some curvature of field.

 

[Sirius Glass]

I agree with Matt about the color change due to the light angle.

 

[flavio81]

If this is produced by the lens, which i doubt, it would be produced by lens group separation (lens cement going bad). This creates a spot on a lens group, which will have different color balance.

 

[shutterfinger]

A light leak of any kind or intensity will affect a larger area than a single color band that runs across the image such as the curb does in your examples. You will notice the density difference in the negative also.
If you still suspect your camera or lens then get some ISO 400 B&W film and shoot an evenly lit mid tone monotone wall with the camera square to the wall. It will show any flaws in exposure or development quickly. Be sure to meter the wall in a 6 or 9 cell grid of the area that will be captured on film to ensure the lighting is even.
Check your roll film holder for proper tensioning of the film by loading a scrap roll of film or the backing paper of a processed roll and advancing it through the holder verifying the tightness to the pressure plate on each frame. Low supply spool tension will allow the supply to unroll an extra amount per wind and be loose at the supply guide roller, too high tension will make winding harder. Check the pressure plate depth in a 9 cell grid to the holder shell's face, it should be the same at any point.
Always check the negative with a good loupe when trouble shooting problems. If its on the negative it should be in the print or scan, if its in the print or scan but not in the negative then its the enlarger, scanner, or printer at fault.

 

[BMbikerider]

I don't know if you process the film yourself but I had something like this with 120 film (both B&W and Colour) used in my Bronica ETRSi. and in past times, my Rollieflex 3.5f. Not every frame but on most. I do all my processing using a JOBO CPE2 and tried processing with the drum rotating at the slow speed and this eliminated the problem. It seems the problem was all down to the developer not reaching edges consistently with the faster speed. It was also there on the negative edges between frames. That I cannot explain, but by changing the processing as above, the problem went away.

 

[shutterfinger]

I use Jobo tanks and manual roller base. I found increasing the amount of chemicals by 10% eliminated rotation problems. The instructions say to use 270ml and I increase it to 300ml which solved edge unevenness. My manual rotation speed is 50 to 60 rpm.

 

[Dan Daniel]

Any chance it is the scanner, not the negative? Change the orientation of the negative in the scanner and see if the position of the flaw follows the film or stays in the same relative scanner frame position.

The same method can be used to check an enlarger for illumination issues, so hopefully the veboten d word inclusion will not lead to exile.

 

[film_man]

I'd start with the above suggestion, scan it a few times with different orientations and see if it follows the film or if it is random. Then you'd know if it is the film/camera or scanner. If it *is* the film and camera then you could have a light leak or it could be film processing. The effect is very light and kind of uneven, in the first sample I can see it on the wall but not on the road. On the second one if I look carefully I can see it on the curb but that's about it.

What film is this? Is it expired by any chance?

 

[tedr1]

Lens performance limitations would be my first choice of explanation. It is the nature of a lens that the contrast and resolution performance is inferior at the edges compared to the center, this is quantifiable and is shown on charts of lens performance. Reduced contrast at the edges also reduces color saturation.

 

[wombat2go]

Rotate the neg 180 degrees and do another positive to compare.
That should show if it is in the neg or not.

 

[yessammassey]

I'd start with the above suggestion, scan it a few times with different orientations and see if it follows the film or if it is random. Then you'd know if it is the film/camera or scanner. If it *is* the film and camera then you could have a light leak or it could be film processing. The effect is very light and kind of uneven, in the first sample I can see it on the wall but not on the road. On the second one if I look carefully I can see it on the curb but that's about it.

What film is this? Is it expired by any chance?

For sure, scanning issues are the first thing I check when I find something amiss with any film images. I checked these on a flatbed and on a light pad + FF DSLR + macro lens. The hue shift is there on the negatives. And the negatives are Portra 400, fresh out of the fridge.

I agree that the effect is more apparent on the image of the store wall than it is on the image of the air pump station. Here, by really abusing the scan of the pump station, I can try to show if the aberration is covering any more of the frame:

... I've highlighted in yellow where I think you can see some hue shifts. The top right corner could be something else, but there's also a little bit of hue shift towards the bottom right, appearing on the highway wall.

Same camera, lens, and film. Here it's on the left again. The JPEG compression artifacts really make it obvious, I think. This time appearing darker:

And on the right. Grass shifting towards cyan..

All told, the examples I've posted came from three different rolls.

I wonder about light leaks, because if I fire the 'shutter' (which really just actuates the mirror/baffle assembly when there is no lens attached), then half-cock the return level it so that the mirror still sits flush against the bottom of the viewfinder, like so:

(I've heard that it's supposed to rest unevenly, maybe not that much though) .. but that the light baffle goes on back down, like so:

, then take it into the darkroom and shine a very bright LED into the lens throat or the viewfinder, I can see just a barely perceptible amount of light coming in through one side of the baffle to the film back. It's really very, very subtle illumination, but it's there.

Maybe it's just wishful thinking that there would be a light leak causing this issue. After all, the leak looks like it is on just one side, while the aberration can appear on either side. I'd much rather be able to fix the issue by upgrading to a Pro-SD body than replacing the expensive 65mm lens that I ordered from Japan.

I think that the change in hue is just the angle of the light and the effect it has on irregular surfaces that are extremely off axis.
And I print a fair percentage of my RB67 negatives full frame and am usually happy with the sharpness and resolution right at the edge of the image. AFAIK, the pressure plate is larger than that image area.
Neither of your lenses is a macro lens, so there will most likely be some curvature of field.

This is an interesting possibility. And good to know about cropping to the edges. I was wondering.

Even though I have the 90mm C, I've been using the 65mm K/L pretty exclusively since I got it. I think all of the examples I've posted that show this problem were taken with the 65mm. It does have a manually-adjustable floating element. The user has to read the focal distance off the side of the bellows and adjust the element to match. Maybe the hue shift toward the edges is a result of an incorrectly-ranged element.. I have forgotten about it sometimes. I wish I could remember if I had forgotten about it when I took these pictures or not!

If this is produced by the lens, which i doubt, it would be produced by lens group separation (lens cement going bad). This creates a spot on a lens group, which will have different color balance.

I do think that my 65mm may be slightly decentered, as unhappy as I am to consider it. The bottom left corner seems a little softer, although the difference is subtle. I don't see any basalm separation or haze in the glass when I shine a light through the lens, but I suppose separation could still be occurring. Would a misaligned floating element potentially cause the same kind of aberration?

I don't know if you process the film yourself but I had something like this with 120 film (both B&W and Colour) used in my Bronica ETRSi. and in past times, my Rollieflex 3.5f. Not every frame but on most. I do all my processing using a JOBO CPE2 and tried processing with the drum rotating at the slow speed and this eliminated the problem. It seems the problem was all down to the developer not reaching edges consistently with the faster speed. It was also there on the negative edges between frames. That I cannot explain, but by changing the processing as above, the problem went away.

These were all processed by hand using either a plastic or steel tank / reels and the agitation was by inversion. I do screw up on occasion. Were you seeing something like this on the short or long edges of your frames? (or in the case of the Rollei, on the top or bottom edges, or on the left & right sides?) When I have density or hue changes on the long edges (most likely due to insufficient quantity of chems in the tank, or insufficient agitation) or irregularly placed streaks (probably due to bad temp control or expiring chems), that's when I am most likely to blame processing.

The one roll of Velvia I shot with this setup, I did have lab processed. I can't see any major hue shifts along the short edges of those frames, but the subjects were all more irregular than these examples I've posted, so it's hard to say for sure if they were completely unaffected.

But I'd much rather the issue I'm asking about be down to bad processing than any kind of equipment failure.

A scan, eh? I'm not going to come on here and be a self-authoritative and smarmy ass to point you to the dpug.

Ah yes, the DPUG Medium Format film camera subform. How could I have missed that...

I'm probably not going to waste time and money to print negatives made with faulty equipment or process, just to double check that what I can already see with a loupe, scanner, or macro lens is really going to show up on an enlargement. But if I do, can I mail you an 8x10 for analysis?

 

[wombat2go]

I just checked my Pro S body in the "half cocked" position.
On the first click, the angled door descends, but not completely into the light baffles.
It takes one or two more clicks until it is fully closed.
I did the darkroom flashlight test and after 2 minutes, I could not see any light around the door when looking in from the back side

 

[yessammassey]

OK, that's one tick I can make in the 'probably light leak' column. (The other columns are 'probably the glass' and 'probably the process')

Thank you for checking. As an RB user, I suppose you've never seen any of the sorts of problems I'm describing. Do you have the 65 K/L?

 

[wombat2go]

I have the C 90, the C 50 and the SF C 150

I have not seen the problems you show, except with uneven development that you mention.
Now I do not use the rotation stick, I do slow 360 degree inversions for 15 seconds then one slow inversion each 30 seconds thereafter
I question how it could be a lens problem, because "each dye-cloud on the plane is formed by rays from each element on the area of the lens"

If there vis a slight leak on your angled baffle, then light could get in via the viewfinder.

Hope you track it down soon.

 

[Dan Daniel]

I have the C 90, the C 50 and the SF C 150

If there vis a slight leak on your angled baffle, then light could get in via the viewfinder.

Hope you track it down soon.

This possibility, the viewfinder providing the light sneaking past a poorly seated seal, is worth exploring. I had a leak on a Rollei TLR that was driving me crazy until I realized that it depended on light coming through the focus hood/screen.

Just to be certain because I may have missed this- the area outside of the frame shows no signs of any light leaks? This is one of the indicators of a leak being from in front of or behind the shutter/light baffle. Leaks outside of the frame are usually from the back. Also put the negatives into the back as if being exposed and look at the film in front of behind the actual frame. Sometimes leaks happen when the film is coming off of or going onto spools, not while in the film gate itself, but they show up inside a frame.

 

[shutterfinger]

There should be no light entering the image path from the viewfinder opening with the mirror in the up position.

 

[wombat2go]

There should be no light entering the image path from the viewfinder opening with the mirror in the up position.

The mirror does not seal, there is a ~6mm gap at the bottom when it is down. All the sealing is done by the angled door at rear of mirror. (ref Pro S here)

Edit : You are correct that the mirror should seal against the foam when it is up.
But if the darkslide is out while composing, the only seal is the angled door.

 

[shutterfinger]

No light should reach the film except when the mirror is up and the leaf shutter in the lens is open. Light from any other source is unacceptable.
http://rb67.helluin.org/files/2010/02/Mamiya-RB67-Pro-S-Repair-Manual.pdf
pdf page 25

 

[paul ron]

For sure, scanning issues are the first thing I check when I find something amiss with any film images. I checked these on a flatbed and on a light pad + FF DSLR + macro lens. The hue shift is there on the negatives. And the negatives are Portra 400, fresh out of the fridge.

... I've highlighted in yellow where I think you can see some hue shifts. The top right corner could be something else, but there's also a little bit of hue shift towards the bottom right, appearing on the highway wall.


(I've heard that it's supposed to rest unevenly, maybe not that much though) .. but that the light baffle goes on back down, like so:

, then take it into the darkroom and shine a very bright LED into the lens throat or the viewfinder, I can see just a

I'm probably not going to waste time and money to print negatives made with faulty equipment or process, just to double check that what I can already see with a loupe, scanner, or macro lens is really going to show up on an enlargement. But if I do, can I mail you an 8x10 for analysis?

see that crumbly crap on the rim here in this picture? it suggests rotten foam on the seals that mates with this edge.

I dont think your porblem is light leaks but more of a scanner thing from what I see here. Do you see this on the prints or just scans? Try turning the neg to another orientation on the scanner n see if it changes sides?

The light inside your camera is the best test for light leaks. Expand the bellows all the way out as well when checking n be sure to examine the corners.. most likely place for leaks. If the mirror seals are suspected.. put your eye against the back n fromt as you shine the light form teh opposite end n see if you get anything leaking. Sometimes the VF allows light to leak on the film side as well.

BUT waht you are showing is so minor, I doubt its the equipment. Look at the prints, not scans!

 

[yessammassey]

Fortunately I have a backup pro-s body. Its mirror cocking lever grinds a bit, but the action is snappy and the baffle and mirror seals are all good. It's what I would start using for now while I worked on the leaky body... if I really thought that light leaks were the definite cause. But for now I'm not using any of it.

paul ron - based on what I've read while searching through this forum looking for answers to my question, you are maybe the top RB67 expert .. thanks for your input here. I can say with a pretty high level of confidence that the changes in color shown in my posted examples are not an artifact of digital reproduction. Repeated flatbed scanning, high resolution digital macrophotography and three different types of illumination have shown me that the relative color shift is retained in the negative itself. The discolored red parts of the affect images are even just perceptible when viewing the negative straight through a loupe.

Here's something I hadn't noticed before when inspecting the 65mm lens - just found this:

The leaf shutter 'star' isn't very symmetrical as it closes.

And there's a slight bend on one of the shutter blades, facing the front..

.. and the rear.

Could this have an effect on the final image?

Edit: The third image link isn't working, but it shows that there is another shutter blade bent in the same way, facing the rear of the lens.

 

[wombat2go]

The "bent" leaf is standard. I have some here and I recall it has been mentioned on an apug thread somewhere.

 

[paul ron]

its not easy isolating a problem if your equipment isnt up to par to begin with. you will fix one thing n still have tge same problem n then fix something else n something else till you totally rehabed the entire camera. so one thing at a time?

one blade has a bent edge, thats normal.

the irregular closing of the blades may be effecting your images but im certain that isnt normal. having a cla will fix it, then see if that takes care of the problem? meanwhile....

shoot with another lens to eliminate the lens as the cause?

you say you have another body? try that?

its just a pita process of elimination piece by piece.


do the light leak tests or save yourself some aggrivation n do all the seals as a prophelactic measure. it can only help when selling this body later anyway. seals are a diy weekend project n cheap enough... no brainer or rocket science.

then after all that... you'll find out it happens no matter what you fixed because it was the scanner all along. try scaning using different orientations n see if it moves consistantly with the negative anyway? i know you say its in the neg.. but humor me?

scaners do weird things.

oh and development can also be the cause, not the camera at all.

 

[shutterfinger]

The tips of the shutter blades bent outward is to prevent them from hitting during operation and damaging the blades or mechanism. It is factory correct, not a defect.
Shutter speed timing on a leaf shutter is from the instant it opens until its fully closed.
The shutter blades are mounted on a circular controller that should move with the weight of a 1/4 inch down feather touching it. Cocking sets the delay gear train and tensions the main spring aka the cocking spring. The controller has a weak spring that will move it to the end of its travel aka full open. When the shutter release lever is pressed all locks and other mechanisms are released from the blade controller and its spring moves the controller and blades to full open in .0001 to .0003 seconds then the delay gear train blocks the return mechanism from moving for the set time then releases the return mechanism and the main spring pulls the controller closed as allowed by the delay gear train. Speeds above 1/100-1/125 are achieved by a booster spring forcing the return mechanism closed quickly. Mamiya uses Seiko shutters. They are not overly complex. It is also normal for the shutter blades to close a millimeter from full open once the delay gear train takes over, hold position then close slowly, depending on selected speed, as the delay gear train runs down.

 

[Alan Klein]

i shoot with an RB67 both negative and positive film. Contact print your negatives or try using a chrome film to test the camera. Or you can look at chrome film directly and see exactly how the camera is performing. Running negative film thru a scanner leaves too much up to other issues. Good luck..