Focusing error

[thechrisproject]

In some shots where I shoot wide open (f/5.6 on a 135 lens) I've noticed that the plane of focus on the negative/scanned image is consistently in front of where it was when I set the focus on the ground glass. Here is an example:

http://farm4.static.flickr.com/3443/3872411855_34a7b4a8b3_o.jpg

I'm fairly certain that when I composed and focused on the ground glass, the plane of focus encompassed the window frame and the mask on it. I wasn't really doing any adjustments if I remember correctly. When I scanned the shot, it certainly looks like the plane of focus is about 1-2 feet in front of where I wanted it to be. I have another shot where the subject was about 30 feet away and the plane of focus is a few feet in front of it.

What are my possible culprits and how do I find out for certain?

It's likely to be user error, but which kind? Bumping the camera forward? Loose rear standard? Or could it be something with my holders or the way I put film in, or a mechanical misalignment of some sort? Something with the lens?

Usually I stop down a little so this isn't much of a problem, but it's certainly annoying. If I have to do some rounds of lens chart tests and suchlike, I certainly will.

[Bob Salomon]

1 ground glass position
2 lens -they aren't made to be used wide open
3 film holder

So first have the gg checked, is in installed properly? Ground side down? Is it the proper position?
Then try shooting at the apertures the lens was made to perform optimally at.
Are you out of focus with all holders or just one holder?

[Vaughn]

It seems like the left side of the window frame is sharper than the right side. Perhaps a little front and/or rear swing was used?

Since the plane of focus shifts forward, the distance to the lens and the film is slightly longer than the distance between the lens and the GG.

So it is a matter of type of camera and film back you have. If the GG had a fresnel lens on the inside of the GG (rare to have them there) and the back was designed for it, then if the fresnel lens went missing, then the GG would end up just a little closer to the lens -- and would create the situation you are having.

I once had questions concerning my 8x10. I took off the back and measured down from the front side (towards the lens) to the GG. I then inserted a film holder, pulled the darkside and measured down from the front of the holder to the plane of the film in the holder. I did not use a ruler or any such fancy measuring devices -- just a piece of paper I marked the distance and then compared it. I was lucky, the distances were the same.

Vaughn

[Robert A. Zeichner]

If I had to bet on the cause, I would say your ground glass is not properly aligned. Of course it would help to know exactly what camera you are using. I think it safe to assume you are shooting 4x5. The film holder depth error spec for 4x5 is +/- .007". It is common for holders to exhibit errors within that range, but if you gg is for example, shallow of 0/0 by say .008" and your film holder is deep by .007" (within spec for the film holder), the total error will add to 0.015" and you will now be out of spec. If you are using a Linhof Technika, and the ground glass hasn't been messed with, it is likely to be just about perfect as the pads on which the glass sits are very precisely adjusted at the factory. On the other hand, if using a camera with a plastic gg as a friend of mine once had, where the gg actually warped, focus will always be off. Wooden cameras can be surprisingly well adjusted where gg alignment is concerned, but many are not. Tell us a little more about what you have and maybe this will better point to the source.

[Bob Salomon]

"If you are using a Linhof Technika, and the ground glass hasn't been messed with, it is likely to be just about perfect as the pads on which the glass sits are very precisely adjusted at the factory."

As long as there hadn't been DIY adjustments over the life of the camera. or they took the pads off (which we see quite a bit of).

[Robert A. Zeichner]

"If you are using a Linhof Technika, and the ground glass hasn't been messed with, it is likely to be just about perfect as the pads on which the glass sits are very precisely adjusted at the factory."

As long as there hadn't been DIY adjustments over the life of the camera. or they took the pads off (which we see quite a bit of).

Just like I said, if it hasn't been messed with.

[rdenney]

2 lens -they aren't made to be used wide open

Bob, this statement confuses me mightily and maybe you can clear it up. It seems to me that we focus the lens wide open, and then we stop it down and perhaps check focus. In all my experience, I've never seen a focus shift when stopping down, except when using just the rear cell of a convertible lens.

But if the guy is using the lens at the aperture at which he focused it, there can be no focus shift because there is no change in aperture. I just can't make your statement square in my mind.

Rick "who focuses lenses wide open" Denney

[Nathan Potter]

Well I notice what Vaughn has noticed - right side of window is sharper than the left side. I'd go through a critical focus check of the system.

First make sure the front standard is exactly parallel to the rear standard. You need to use vernier calipers or any solid length rod or bar about the focal length of the lens you're using and rack the bellows in until the rod just touches one of the corners of the front and rear standards. Check the other four corners by noting that the rod just touches those four corners with less than a hairs width gap. Note that the rod needs to be exactly orthogonal to the standards. If you've got swing and tilt engravings on the standards note the positions when you've adjusted for planarity.

Next make sure your lens board sits flat against the front standard and is not warped at the lens mount opening.

Next set up a finely divided yard stick centered about 6 ft in front of the camera and disposed at an angle off the optical axis so that you can see the divisions in the GG. Position it near the center of the view and focus with the lens at maximum aperture on 15 inch mark using a high quality loupe. LOCK down the standards and recheck focus critically. I generally use a 210mm lens on 4X5 for this so the magnification is roughly 1:10.

Load your best film holder with film and with great care. DON'T disturb standards, camera or tripod. Take the image with lens wide open and develop the film. Polaroid is nice but do you want to check the 545 back or your best film holder? Develop film. After you've exposed and removed the holder check the focus again to confirm that you didn't move the setup.

Inspect film using a good loupe and determine if the sharpest spot is at the same 15 inch location. If not note the new location in, hopefully some fraction of an inch difference and record it.
Now you have a delta in the DOFld and using the appropriate formulas this can be converted to an equivalent DOF error probably in a few to several mils.

Also before you start this test and as mentioned above by others make sure that the ground side of the GG faces the lens and rests unobstructed on the flat surface of the rear standard, or as Bob points out small pads or flats found on some standards such as Linhof. If using Quick or Readyloads or 545 backs make sure the pressure plate is adequately spring loaded and sits exactly in the proper plane. It would also be wise to check that the film sits relatively flat against the backplane (plenum) of the holder once the dark slide is removed. By all means waste a sheet to confirm this with all your holders.

All this is a PITA but its like a detective story so have fun. ;) ;)

Nate Potter, Austin TX.

[thechrisproject]

Lots of great responses, thanks all.

The camera is a Chamonix 045n-1. The lens is a Nikkor 135 f/5.6. The ground glass is as it was installed at the factory, which I'm fairly certain is the correct orientation.

I don't focus with a particularly powerful loupe. I don't particularly want to go through too much nitpicking if the problem is just user error and poor technique. Although it's probably a good idea to go through the nitpicking anyway, for the experience.

If you're familiar with the Chamonix you'll know it's pretty easy to have the standards not totally parallel. It's pretty free and loose. I do have vernier calipers though, so I could check it when doing testing. If that's the cause I can deal with it.

[ic-racer]

Shim your ground glass back (away from the lens). You can use trial and error, or a quick calculation. Did you focus on the window and the frame on the wall is in focus? If so your error can be estimated by the 1/p + 1/x = 1/focal length equation.
So if you focused at 5000mm and the film showed focus at 4500mm from the camera, the ground glass is off by around 1.3 mm (if you are using a 150mm lens)

[Robert A. Zeichner]

Lots of great responses, thanks all.

The camera is a Chamonix 045n-1. The lens is a Nikkor 135 f/5.6. The ground glass is as it was installed at the factory, which I'm fairly certain is the correct orientation.

I don't focus with a particularly powerful loupe. I don't particularly want to go through too much nitpicking if the problem is just user error and poor technique. Although it's probably a good idea to go through the nitpicking anyway, for the experience.

If you're familiar with the Chamonix you'll know it's pretty easy to have the standards not totally parallel. It's pretty free and loose. I do have vernier calipers though, so I could check it when doing testing. If that's the cause I can deal with it.

The friend to which I referred in my first response never used a loupe for focusing. I asked him why not and he said he felt his visual acuity was good enough without magnification. I think he was in denial as age does have its effects. I suggested he set up his camera and focus by eye alone and then do so with the loupe. There was a revelation! Another photographer I know who, like me is in his 60's also focuses without any loupe at all and never mis-focuses. He's blessed with incredible vision. We're all different in this regard.

That said, I think to arrive at any meaningful conclusion about your ground glass/film plane coincidence, you need to do a more controlled experiment. First, focus critically, using a loupe, on a two-dimentional subject, making sure that every inch of the ground glass is in as perfect focus as you can achieve. That will rule out things like parallelism because you will have to adjust the movements to get lens plane and film plane perfectly parallel with the subject plane. Once you have done this, make some exposures and process the film. If you can leave the camera set up while you do this, you will be able to go back and compare negative and gg under magnification. Forget trying to measure depths with a vernier caliper as making this measurement is trickier than you might think. Without a surface plate, a test fixture that would permit testing the depth at 1 inch intervals across the 4x5 film area and a dial depth gauge calibrated in increments an order of magnitude smaller than .001", you would really not be able to arrive at any conclusive results. All you are trying to confirm is that what you see on the gg is exactly what you are getting on the film. You just need to compare the two side by side to see any difference. If after comparing two or three negatives made with different holders to the gg, you see a consistent difference in sharpness, than it's fairly safe to say the gg is mis-aligned and you need to have it corrected. A lot of folks claim that it doesn't matter all that much because of all the human error involved in setting up a camera and focusing. It is exactly because of human error and the depth variations in film holders that the gg should be precisely in the right position.

[Brian Ellis]

Lots of great responses, thanks all.

The camera is a Chamonix 045n-1. The lens is a Nikkor 135 f/5.6. The ground glass is as it was installed at the factory, which I'm fairly certain is the correct orientation.

I don't focus with a particularly powerful loupe. I don't particularly want to go through too much nitpicking if the problem is just user error and poor technique. Although it's probably a good idea to go through the nitpicking anyway, for the experience.

If you're familiar with the Chamonix you'll know it's pretty easy to have the standards not totally parallel. It's pretty free and loose. I do have vernier calipers though, so I could check it when doing testing. If that's the cause I can deal with it.

I don't know what you mean by "particularly powerful" loupe. What loupe do you use? A 4X loupe is common, some use stronger, some use a little less but you certainly should be using a 4x or thereabouts loupe (and "more powerful" isn't necessarily better, it actually can be harder to focus with an 8x than a 4x especially with a Fresnel lens). At f5.6 and a moderately long (in absolute terms) focal length lens such as 135mm, which is going to put you up pretty close to the subject for a photograph like the one you've displayed, you have very little depth of field because the three factors affecting depth of field all are working against you - relatively long lens, close to subject, wide aperture. So your focus has to be spot on and it's probably not going to be if you're not using a decent loupe.

I owned a Chamonix for a while and really liked the camera but I thought the Fresnel left something to be desired compared to the Maxwell viewing screens I had on my previous cameras. Fresnels are inherently difficult to focus but the one on the Chamonix seemed a little more difficult than others I've used. If you're open to spending about $250 I think you'd find a Maxwell screen a delight to use. Or you could go with a plain ground glass, which won't give you as bright a screen as you get with a Fresnel but which is easier to focus.

I didn't find it difficult to keep the front standard aligned on my Chamonix. I just lined it up with the front edge of the camera bed by eye and that seemed to work fine. But calipers certainly wouldn't hurt.

[BennehBoy]

Chris,

I thought I had focus pretty nailed when I wasn't using a loupe, the moment I used one I was rather alarmed to see how far off the mark I was.

You can snag them pretty cheaply off the net.

[boris]

i used a 7x loupe for a long time with good results, suddenly i realized that 30% of my negativs weren't perfect. ..to shorten up the story my glasses were to blame
(-1.5 diopter on both eyes, i am 43 years old).
now for focussing i take my glasses off and my negativs are fine again. :)
boris

[Leonard Evens]

Let me give you some quantitative information, which may or may not be helpful. Using a 4X loupe, the typical depth of focus at f/5.6 is about 0.56 mm. That means, if you focused twice or more on the same target, the positions of the lens might be that far apart. Using a 2 X loupe, it would be twice that or about 1.12 mm.

Such errors are inevitable in that you just can't see well enough to distinguish one focusing position from another as long as they are within the depth of focus. Be that as it may, such errors would be random in the sense that they could vary over the entire possible range from undershooting to overshooting. If you repeated such measurement many times and took the average, you would end up essentially right on.

On the other hand, if we suppose you are focusing at about 2 meters or 2000 mm, with an error of about 300 mm (just less than 1 foot), the error in the position of the lens is about
(135/2000)^2 x 300 which comes out to about 1.4 mm. If it is consistently that large, it would seem unlikely it could occur just due to random focusing errors, and there is some systematic error. As others have noted, the most likely culprit is the position of the gg relative to the position of the film.

[shadowleaves]

Lots of great responses, thanks all.

The camera is a Chamonix 045n-1. The lens is a Nikkor 135 f/5.6. The ground glass is as it was installed at the factory, which I'm fairly certain is the correct orientation.

I don't focus with a particularly powerful loupe. I don't particularly want to go through too much nitpicking if the problem is just user error and poor technique. Although it's probably a good idea to go through the nitpicking anyway, for the experience.

If you're familiar with the Chamonix you'll know it's pretty easy to have the standards not totally parallel. It's pretty free and loose. I do have vernier calipers though, so I could check it when doing testing. If that's the cause I can deal with it.

Ok, here comes THE answer....

This a design fault that exists on EVERY chamonix camera that uses a Fresnel, namely the 045n1, 057n1, and a few other models. YES, your 045n1 does have this focus shift.

I should have posted this in a separate post. I posted a while ago about this in a Chinese forum, that I did a thorough investigation into the focusing error of Chamonix 045n-1 cameras. My conclusion is that clearly the factory alignment calibration is done without fresnel on the ground glass (which was confirmed during my chat with Mr Xiang Yu, the owner of Chamonix and designer of all Chamonix cameras), yet their cameras are shipped with a fresnel placed between the ground glass and the lens.

This apparently causes the focus shift, usually about 1/3 of the thickness of the fresnel.

The problem can be easily illustrated by this pic. I drilled a hole in the center of the fresnel of 045n1, and capture the focusing results using my D50 camera and a macro lens. Clearly, the fresnel is the only cause of the focusing shift. If I focus through the hole that I drilled, or without fresnel completely, the focus will be perfectly accurate.

This in my opinion is the biggest fault in 045n1's design. I communicated with them about this, but unfortunately they think the users are going to use below f/32 for landscape anyway so it won't be a big practical problem in their opinion. I'll leave that up to you guys to decide whether you want to drill a hole in the fresnel or to remove it completely or to live with the focus error.

[Bob Salomon]

"fallacy" that is a false belief. What you mean is a fault not a fallacy.

[shadowleaves]

"fallacy" that is a false belief. What you mean is a fault not a fallacy.

thanks for pointing it out Bob!

[Robert A. Zeichner]

My conclusion is that clearly the factory alignment calibration is done without fresnel on the ground glass (which was confirmed during my chat with Mr Xiang Yu, the owner of Chamonix and designer of all Chamonix cameras), yet their cameras are shipped with a fresnel placed between the ground glass and the lens.

This just about guarantees focus error!


This apparently causes the focus shift, usually about 1/3 of the thickness of the fresnel.

Depending on how they mount the Fresnel, the focus shift could be worse than that. If they sandwich the Fresnel between the mounting pads and the gg, you can add the total thickness of the Fresnel to the optical shift as well.


This in my opinion is the biggest fault in 045n1's design. I communicated with them about this, but unfortunately they think the users are going to use below f/32 for landscape anyway so it won't be a big practical problem in their opinion. I'll leave that up to you guys to decide whether you want to drill a hole in the fresnel or to remove it completely or to live with the focus error.

If by below f32 they mean f45 or f 64, then it would seem they expect users to live with diffraction artifacts right out of the gate. That's unacceptable. Many LF lenses designed for 4x5 achieve their best performance at f22 or wider.

With longer focal lengths, maybe under certain circumstances, the use of a smaller f stop would mask the problem, but I seriously doubt an error that large would not be noticed with a short focal length of say something in the range of 75mm - 110mm.

[shadowleaves]

This just about guarantees focus error!



Depending on how they mount the Fresnel, the focus shift could be worse than that. If they sandwich the Fresnel between the mounting pads and the gg, you can add the total thickness of the Fresnel to the optical shift as well.



If by below f32 they mean f45 or f 64, then it would seem they expect users to live with diffraction artifacts right out of the gate. That's unacceptable. Many LF lenses designed for 4x5 achieve their best performance at f22 or wider.

With longer focal lengths, maybe under certain circumstances, the use of a smaller f stop would mask the problem, but I seriously doubt an error that large would not be noticed with a short focal length of say something in the range of 75mm - 110mm.

Chamonix arose as a ULF camera maker a few years ago. In all their cameras larger than 8x10, there is no Fresnel whatsoever. I think that's the reason why they calibrate their cameras without ground glass - they've just got used to it. When they decide to make the 045n1, apparently the calibration procedures are simply inherited from what they used to do with ULF cameras.

In the 045n1 camera, the fresnel is mounted with its grooves facing the lens, and the smooth side of the fresnel is touching the coarse side of the ground glass. I believe this will cause a shift around 1/3 thickness of the fresnel for lense longer than 90mm. For shorter lens, the smaller entering angle will cause more complicated shift that varies across the entire frame, and it's not very easy to calculate.

From my conversation with the Chamonix people, it seems that they expect the users to use 045n1 as a landscape camera mostly. I agree that beginning at f/32, the shift will become not so obvious. But it's clearly visible at f/22, and unacceptable at f/16. I'm actually surprised that the OP is the first poster I've seen complaining about the focus shift of 045n1 cameras which is due to a problematic design. Where are the other users of 045n1?

[NewBearings]

http://www.largeformatphotography.info/forum/showthread.php?t=49328

[newmoon2night]

I owned a Chamonix for a while and really liked the camera but I thought the Fresnel left something to be desired compared to the Maxwell viewing screens I had on my previous cameras. Fresnels are inherently difficult to focus but the one on the Chamonix seemed a little more difficult than others I've used. If you're open to spending about $250 I think you'd find a Maxwell screen a delight to use. Or you could go with a plain ground glass, which won't give you as bright a screen as you get with a Fresnel but which is easier to focus.



I have a Chamonix 45N-1 and also have a Maxwell screen, but haven't paired the two yet, though I did have an initial look.
On the Chamonix the fesnel slot is slightly shorter slot than the glass slot (and the fresnel screen is a different length from the ground glass). The Maxwell viewing screen is the same length as the Chamonix ground glass, so when fitted it doesn't sit proud at the back, and this results in the four retaining screws not being at 90 degrees, and the plastic or carbon fibre glass protector doesn't fit.
When I asked Bill Maxwell he emailed back 'Your original Chamonix screen was a ground glass/fresnel sandwich. The new screen is a modern "one piece" focus screen. You don't put my screen where the original fresnel was (which sits lower than the focus plane). My screen sits at the same level as a simple ground glass would.'
Has anyone come up with a solution for neatly pairing the Chamonix 45N-1 and Maxwell screen?

[Michael Rosenberg]

This was discussed a couple of months ago, and there were several people who saw this problem. See: http://www.largeformatphotography.info/forum/showthread.php?t=49328&highlight=Chamonix+fresnel

Also what I found with careful measurements, which support what shadowleaves reports above:


Re: Focusing problems with the fresnel groundglass on the Chamonix 45N-1?
I was a little baffled with this thread. Two people with the same camera having identical problems with a particular focal length lens. I would have thought that the Chamonix being a new camera the position of the gg/fresnel combination would have been tested. So I thought I would test my own Chamonix, and using mm ruler verify any small differences if they existed.

For a target I used a window mat on a mat board that would provide a bright line (reflecting the overhead light), and a dark line that would be a shadow from the overhead light. Both lines would be parallel to each other, and the target would also test flatness of the gg (i.e. there should be no change in focus of the parallel lines).

I checked three lenses, my 58mm Schneider XL, 90mm Linhof Scneider Angulon f5.6, and my Apo Sirornar S 135mm. All were tested with either the fresnel in the proper position, with it on top of gg, and no fresnel - just the Chamonix gg. I used a Scneider 6X loupe to check for focus.

With the 58mm lens the focal shift +0.5 mm with the fresnel in the proper place. With the 135mm lens the focal shift was +1 mm with the fresnel in the proper place. This should not make a big difference in depth of field unless shooting wide open I would think. However, with the 90mm lens the focal shift was +3.5mm with the fresnel in the proper place. There was no focal shift when the fresnel was placed on top of the gg. I have tried 2 other gg that I have, and made the same observation.

One possible explanation is the focal length of the fresnel is affecting the focus of 90mm lenses. I do not have another fresnel to test. But I think hereafter I will keep my fresnel on top.

Mike

[Michael Rosenberg]

I have a Chamonix 45N-1 and also have a Maxwell screen, but haven't paired the two yet, though I did have an initial look.
On the Chamonix the fesnel slot is slightly shorter slot than the glass slot (and the fresnel screen is a different length from the ground glass). The Maxwell viewing screen is the same length as the Chamonix ground glass, so when fitted it doesn't sit proud at the back, and this results in the four retaining screws not being at 90 degrees, and the plastic or carbon fibre glass protector doesn't fit.
When I asked Bill Maxwell he emailed back 'Your original Chamonix screen was a ground glass/fresnel sandwich. The new screen is a modern "one piece" focus screen. You don't put my screen where the original fresnel was (which sits lower than the focus plane). My screen sits at the same level as a simple ground glass would.'
Has anyone come up with a solution for neatly pairing the Chamonix 45N-1 and Maxwell screen?

Did Bill mention which side is facing the lens? I assume the ground side.... The only way to make the screen protector work is to shim the posts so that they are flat. The Maxwell screen is much better than the supplied fresnel, but getting it to fit may be not an insignificant issue.

Mike

[Bob Salomon]

"When I asked Bill Maxwell he emailed back 'Your original Chamonix screen was a ground glass/fresnel sandwich. The new screen is a modern "one piece" focus screen."

This might make it sound like a new technology. It isn't. Back in the 80s we introduced the Super Screen which was a product made by Fresnel-Optics. The Super Screen was a one piece screen with one side being a fresnel and the opposite side a frosted surface that served as the ground glass.
Fresnel-Optics was purchased by Beattie and the whole company is now Reflixlite. Beattie also used the one piece design on at least some of their screens. The problem with this technology, as it existed back then, was the screen material could warp if all sides were not supported in the camera back. This was more apt to happen in dry, hot conditions. Fresnel-Optics kept changing the properties of the plastic to be less prone to warping but never completely solved the problem by the time we discontinued it. A second problem was scratching the fresnel grooves when placing a loupe against it and deforming the screen if a loupe was pressed against it with too much force while focusing. To eliminate those problems a glass overlay screen of either clear glass or a grid patterned glass was sandwiched on top of the Super Screen.

The main reason why the Super Screen was discontinued was that the Linhof ground glass and fresnel lens became much brighter then their older ones in the early 90s eliminating the brightness advantage of the Super Screen.

So a one piece combination is not new. In fact the Boss Screen was a one piece sandwich as are some Wista screens.

[newmoon2night]

Did Bill mention which side is facing the lens? I assume the ground side.... The only way to make the screen protector work is to shim the posts so that they are flat. The Maxwell screen is much better than the supplied fresnel, but getting it to fit may be not an insignificant issue.Mike

This is the rest of Bill's email:

DO NOT alter your new screen or you won't be able to use it afterward. Please read the instructions where I refer to the situation where the original screen that is being replaced is a ground glass/fresnel sandwich. I strongly suggest professional installation by someone who is familiar with modern view cameras and modern focusing screens. Please show that person the caution that we put on the outside of the package warning about not using a microfiber cloth to clean the screen. Also show him the instructions for the screen installation.

If you wish to do this yourself, please read all the instructions. You will remove all parts of the old screen. That means the original ground ("frosted") glass, the original clear fresnel, and any shims that may have been under the ground glass (sitting on the ledges that the ground glasss rested on). The ledges where the original ground glass used to rest should be the place that my screen should rest (with it's "frosted" surface facing the lens of the camera). I am aware that the Chamonix cameras have had some issues with proper focus. The "frosted" focusing surface should face the lens of the camera and should be located 4.88mm from the front of the focusing frame (the front of the focusing frame is the side that would be touching the back of a film holder when the film holder is inserted into the camera). This is called the "adjusted" T distance (which is the distance from the front of a film holder to the emulsion side of a piece of film in that film holder). That should probably be measured by a professional. If the "adjusted" T distance is not correct, then the camera focusing frame may not have been manufactured correctly ... You could simply install the screen per my instructions, run some film tests at moderate apertures and see if the camera's focus is ok for your trip.

[Bob Salomon]

"If the "adjusted" T distance is not correct, then the camera focusing frame may not have been manufactured correctly"

That is an interesting statement. By whose criteria? Well before the Maxwell screen was conceived view cameras were made and were focusing correctly from manufacturers worldwide. So if a Linhof Technika from 195x or a Sinar Norma or a Deardorff or a Graflex have a problem with a Maxwell, and that camera did not have a problem with its' former screen setup, why would someone decide that the back was not made properly? And what would one do if they decide that they want to re-install the original screen or another screen from another company?

[GPS]

"If the "adjusted" T distance is not correct, then the camera focusing frame may not have been manufactured correctly"

That is an interesting statement. By whose criteria? Well before the Maxwell screen was conceived view cameras were made and were focusing correctly from manufacturers worldwide. So if a Linhof Technika from 195x or a Sinar Norma or a Deardorff or a Graflex have a problem with a Maxwell, and that camera did not have a problem with its' former screen setup, why would someone decide that the back was not made properly? And what would one do if they decide that they want to re-install the original screen or another screen from another company?

Well Bob, they probably and surely incorrectly suppose that all the cameras have just one and the same T distance... Which is in the same league as the Chamonix explanation for their focusing error. Just horror... :)

[Peter K]

Well Bob, they probably and surely incorrectly suppose that all the cameras have just one and the same T distance... Which is in the same league as the Chamonix explanation for their focusing error. Just horror... :)
If all this camera's backs are made for the "International large camera film holders" (http://home.earthlink.net/~eahoo/page8/filmhold.html) the T distance must be the same: 0.197 +/- 0.007" = 5 mm for 4x5" films. Horror or just the standard? :)

Peter

P.S. But be carefully with this site, not all numbers are correct. E. g. the thickness of 8x10" holders isn't 0.594" but 0,570" = 14,5mm.

[Robert A. Zeichner]

If all this camera's backs are made for the "International large camera film holders" (http://home.earthlink.net/~eahoo/page8/filmhold.html) the T distance must be the same: 0.197 +/- 0.007" = 5 mm for 4x5" films. Horror or just the standard? :)

Peter

P.S. But be carefully with this site, not all numbers are correct. E. g. the thickness of 8x10" holders isn't 0.594" but 0,570" = 14,5mm.

The 0.197 +/- 0.007" tolerance for film holders really can't be applied to the ground glass position. Holders are made of plastic and thin aluminum and can be only so precise in their manufacture. The ground glass really needs to be close to perfect in its position so that the widest range of film holders will give satisfactory results. Suppose the gg is 0.007" shallow and your film holder is 0.007" deep. Right out of the gate, you are at double the tolerable error. If the gg were at 0.0, then holders that were shallow or deep by up to 0.007" should give acceptable results.

[GPS]

If all this camera's backs are made for the "International large camera film holders" (http://home.earthlink.net/~eahoo/page8/filmhold.html) the T distance must be the same: 0.197 +/- 0.007" = 5 mm for 4x5" films. Horror or just the standard? :)

Peter

P.S. But be carefully with this site, not all numbers are correct. E. g. the thickness of 8x10" holders isn't 0.594" but 0,570" = 14,5mm.

"If"!
"If there were fish in your backside we wouldn't need lakes" - my grandfather. :)

[pocketfulladoubles]

I'm actually surprised that the OP is the first poster I've seen complaining about the focus shift of 045n1 cameras which is due to a problematic design. Where are the other users of 045n1?

The focal shift is definitely there. Of course you wouldn't see it for most landscape shots, but it bit me on some architectural shots. I know I focused it right in the field, but the prints told me "Sorry, nice try". So, off with the stock screen and on with the Maxwell. I keep the stock screen for a backup.


My question is, why wouldn't someone just use digital calipers to check their film holders (with a piece of film loaded - I'm assuming), find the average depth, and toss out all holders not within 0.007" (the correct tolerance?) of the average value, and then adjust the ground glass plane to match either by machining or using shims?

[shadowleaves]

The focal shift is definitely there. Of course you wouldn't see it for most landscape shots, but it bit me on some architectural shots. I know I focused it right in the field, but the prints told me "Sorry, nice try". So, off with the stock screen and on with the Maxwell. I keep the stock screen for a backup.


My question is, why wouldn't someone just use digital calipers to check their film holders (with a piece of film loaded - I'm assuming), find the average depth, and toss out all holders not within 0.007" (the correct tolerance?) of the average value, and then adjust the ground glass plane to match either by machining or using shims?

This has nothing to do with users' film holders. In fact, Chamonix's ground glass position is perfectly accurate if you use them without the fresnel. The shift that we all see is caused by

1) the fresnel that is placed between the GG and lense which destines to cause a shift
AND
2) the fact that their ground glass position is factory-calibrated without the fresnel.

Of course you can use machining or shims to adjust the ground glass, but with super-wide lense such as Super Angulon 75/5.6 or anything wider (as an architecture photographer you probably use these a lot), the focus shift caused by the fresnel will be more complicated than shims or machining would solve. You will observe that the focus shift varies across the entire field, being small in the center and large in the corners; and what's even better, that depends on shift, tilt, and swing as well!

There're only three solutions that'd give you accurate focus without regarding to the lense or movement you use:

1) drilling a hole in the center (or anywhere that you'd focus often) of the fresnel
2) put the fresnel between the ground glass and the user - this includes Maxwell screens which are simply integrated pieces of fresnel and ground "glass", with the fresnel side facing the user
3) remove the fresnel.

[GPS]

One could suppose that Chamonix should have at least enough of the common sens and honesty to inform each user about this focus shift. Nope, they don't...

[pocketfulladoubles]

One could suppose that Chamonix should have at least enough of the common sens and honesty to inform each user about this focus shift. Nope, they don't...

Please. I'd love to spout the 'you get what you pay for' line, but I spend a lot of time in a lab with extremely high price and well documented equipment that has gone through countless revisions, and guess what? They are all full of gotchas and undocumented errors and bugs that the user unfortunately has to figure out the hard way.

If you call the manufacturer about it, it is not uncommon that they knew of the problem and just didn't disclose it because of the assumption, poor as it may be, that users would probably not be subject to the error. I recall this happened with Motorola's FPU once regarding a decimal precision error in Apple computers a long long time ago. They replaced the units for those in the scientific field only, stating that the typical user would not be subject to the flaw. It upset quite a few folks. And this was for a high-tech, engineered and documented product.

My point? It appears that this happened with the Chamonix fresnel. Now, to many, this is a gross oversight and unacceptable mistake. However, in reality, it should just be added to the laundry list of bugs that the manufacturer deemed not detrimental to the majority of users. If I'm not mistaken, they are correcting this on the next revision. And, you can always change the screen.

[GPS]

...
And, you can always change the screen.

You can - if you know about it. That's why I said it should be common sens and honesty to disclose it to each user. No need to defend lousy thinking...

[GPS]

...
I recall this happened with Motorola's FPU once regarding a decimal precision error in Apple computers a long long time ago. They replaced the units for those in the scientific field only, stating that the typical user would not be subject to the flaw. It upset quite a few folks. And this was for a high-tech, engineered and documented product.

...

It happened to PCs in the 90 something. The public outcry was such that they needed to remade the chip and correct the CPUs! No need to defend lousy thinking...

[Michael Graves]

"fallacy" that is a false belief.

You mean I've been using that word wrong all these years? It really DOESN'T mean when my wife....

Never mind.

[Robert Hughes]

That would be "phallacy"... or something like that...

[pocketfulladoubles]

It happened to PCs in the 90 something. The public outcry was such that they needed to remade the chip and correct the CPUs! No need to defend lousy thinking...

Yep you're right. Intel, not Motorola...

[sun of sand]

I think its sharpest left side -minute swing- and focused only a few inches in front of the window trim
the photo on wall looks sharp but its the nearest to window edge that is sharpest and I don't think its as sharp as the nearest to window edge of the mirror
the mirror is maybe 2.5 inches in front of the trim
trying to shoot cobwebs probably off by 4in

I think you just missed focus at 5.6 ..and have some swing somewhere

[MattTrask]

Ok, here comes THE answer....

<snip>

I'll leave that up to you guys to decide whether you want to drill a hole in the fresnel or to remove it completely or to live with the focus error.

Seems to me that there is a fairly simple solution to this, easier than drilling a hole thru the fresnel. A rectangular shim that is the same thickness as the fresnel could be attached to filmholder/roll-film-back/digital back/whatever to compensate for this apparent problem. Or just inserted between the film holder and the back of the camera.

My Horseman VH-R uses a similar technique with its Polaroid back to compensate for a similar spacing issue. The spacer is a standard Horseman SKU that ships with the Polaroid back.

[Bernard Kaye]

The more I look at your window picture, the more skewed it looks to my octogenarian eyes. I think the name given this by LF people is yawing or maybe perspective error.
I think you are too careful to be using defective equipment but since you are shooting wide-open, try the following: get your hands on a good 35mm. or 120 rangefinder camera; stand where you think you will be sighting head-on dead center to the window frame with the film plane parallel to the window frame-glass: see if the range finder keeps EXACT FOCUS-DISTANCE setting if you shift from right side of window frame to left side of window frame, back & forth. The old Contax 35mm. RF cameras would register a change in distance focus from one side of an electric switch plate to the other side (80mm.) because the wall is not so flat that both sides are equidistant from the camera lens and/or I was not holding camera to result in parallel film plane to switch plate. Only by stopping down could I get enough depth of field to have entire switch plate parallel to film plane in the eyes of the lens; now magnify that to an entire window frame, perhaps an old window frame that is not flatly parallel to your film plane no matter what. Can you be certain that you are in dead center to the window plane? Stop down: if you were doing landscape work, you could make sure not to rack lens all the way back to infinity which could result in a corner or side being minutely closer to film plane than infinity; you do not have this leeway in close-up LF with lens wide open. There are reasons that Linhof takes such pain to have camera front standard parallel to film plane.
Bob & are are not related but arrangements can be made.
Bernie

[jamie c]

how big of a difference is the brightness on the ground glass, comparing a chamonix with fresnel and chamonix without?

[europanorama]

Shim your ground glass back (away from the lens). You can use trial and error, or a quick calculation. Did you focus on the window and the frame on the wall is in focus? If so your error can be estimated by the 1/p + 1/x = 1/focal length equation.
So if you focused at 5000mm and the film showed focus at 4500mm from the camera, the ground glass is off by around 1.3 mm (if you are using a 150mm lens)
what is p and x?
Have the very same FL and Zone Z1 is too short. have all data only want to prove with this formula.