New 20x24 Camera

greenspun.com : LUSENET : Large format photography : One Thread

I currently use a 20x24 box type camera fixed at infinity focus with a Goerz 30" Artar which I love but the near focus distance is 20' at f128 so I've been limited to shots off bridges or overlooks etc. I wish to build a second box camera using this lens which will give me a focus range of say 6' to 300'. I've attempted to run the calculations and figure if I move the lens out 4" to a total 34" from the film plane and tilt it downward 10 degrees i.e. the Scheimpflug Effect, that this will give me the desired DOF. Can anyone confirm these calcs or steer me in the right direction?

-- Mark Stevenson (marks@cwo.com), March 21, 2001

Hi Mark –

The formula relating image distance, object distance and focal length for a *simple* lens is

1//focal length = 1/image distance + 1/object distance.

Plugging in 30” for the focal length and 34” for the image distance (between lens and film) gets us an object distance of about 250”, or 20’. Unless, of course, my brain isn’t hitting on all cylinders this early in the morning.

So, this tells us that an object at 20’ is exactly in focus, and says nothing about depth of field. I hate to admit it, but DOF field calculations always seem like voodoo, mostly because you have to make some assumptions about acceptable circle of confusion. Can anyone educate me?

A caveat is that your Goertz Atar is not a single element simple lens, so this calculation is no doubt off a little. Having said that, I’d be willing to bet we’re pretty close.

Post a picture of that camera somewhere. I want to see it!

- Kevin

-- Kevin Bourque (skygzr@aol.com), March 22, 2001.

As Kevin said, how much extension you will need depends on what circle of confusion you can live with. It will also depend on the tilt angle, which in turn depends on the height of your camera above whatever is at your 6' near-limit.

One piece of good news: although Artars aren't 'thin' lenses in the technical sense, they are simple and symmetric enough that you can get away with measuring all distances from the iris position.

If you really can get away with only a 4" extension, you can probably get away with making a sliding extension tube out of two PVC pipes and putting that on the front of the camera. A more sophisitcated version would be to make a lensboard with a graphlok 'front' on it and mount the lens on a cheapie graphic 4x5 which you then use for focussing and tilts. Either method will eventually vignette if you use too much extension.

-- Struan Gray (struan.gray@sljus.lu.se), March 22, 2001.

Mark, what are doing with the film? I assume not enlarging it? If not, you can start by using standard DOF formula using a cc of .3, which is an acceptable print resolution. If you are familiar with tilt, the basic rule is J = fl/tilt angle, where J = the distance in ft. below the lens the plane of sharp focus will intersect. So in your case 300mm/10 = 30 ft below the lens, the plane of sharp focus will intersect. As you change the focus distance while using tilt, you you change the angle of sharp focus, but do not change where it intersects below the lens. I suggest looking at Merklingers quick time movies on his page to clearly see this.

Now my question is, what type of film do you shoot in a camera that big? Any color film made that size? As mentioned above, anyway to sneak a peak at this monster?

-- Bill Glickman (bglick@pclv.com), March 23, 2001.

I'm using HP5 20x24 inch sheets. It's available in boxes of 25 sheets for about \$350. Then I contact print. I created a spreadsheet which does all the calcs for HFD, NFD, FFD, DOF, and lens to film plane distance. The following data is pluged in: focal length, circle of confusion, object distance, and f-stop. The formula I'm using for lens to film plane distance is from Ansel Adams' The Camera and is 1/focal lenght minus 1/circle of confusion = lens to film plane distance. I would like to post a photo of the camera I've been using but I'm not sure how to do it. It's just a simple box made of 1/2 inch birch plywood. The whole top comes off so that I can place the film inside. No moving parts except for the aperature. I show that without the lens tilt and the lens to film plane at 34 inches my near focus distance is 11 feet and far distance is 378 feet. I'm guestimating that with a lens tilt of 10 degrees that the near focus distance will come down to about 6 feet. Am I guessing right?

-- Mark Stevenson (marks@cwo.com), March 24, 2001.

Mark, I ran the following numbers, 34" lens to film distance, cc = .3mm which is equal to 3 lpmm (about what the human eye can resolve, good target for contact printing), lens 30" or 762 mm, f128. With no lens tilt, I come up with the following,

Point of exact focus from lens 21 ft. Near DOF 15 ft Far Dof 37 ft.

So for starters, I am struggling how you came up with those numbers, I am using standard view camera calculations. I am reasonably confident they are accurate, but would hate to ever disagree with Ansel. :-)

Your goal of 6 ft to 300 ft (lets assume no tilt), assuming f 128, you must focus at 7 ft, which would give you a near DOF of 6 ft, but your far DOF will be 8ft. Bottom line, it's not possible unless you go to f 550 and focus around 12 ft. Even if f 550 was possible difraction would eat up an resolution and completely defeat using such a humungous camera.

Now, as for tilt.... I may stand to be corrected, but after reading Merklingers books and doing some experimentations with scaled drawings, adding tilt will not change the DOF at the point of exact focus! I know that sounds strange, but it really is true once you do the math. At very extreme tilt angles, this can change since it is possible to tilt so far downwards the entire ground below you will represent the plane of sharp focus... There is very few, if any, lenses that would allow for such huge tilt angles with out running out of image circle.

So if you don't exceed f128, using that lens and not drop your standards on cc, (which would be sensless), then I don't see how your goal can be obtained?

Correction to my last post... I wrote, J = fl/tilt angle, I meant to write, J = (fl/5)/ tilt angle. Sorry...

Hope this helps....

-- Bill Glickman (bglick@pclv.com), March 24, 2001.

Bill...I've been using a cc of .75 on the existing camera and my depth of field calcs have seemed to be right on. At f128 I get a HFD of 20'. There seems to be no degredation that I can see at these high f-stops. I have one photo taken at f64 which contains a 1 foot sign with six lines of small lettering about 50' away. I can't read the small lettering of the sign on the contact print with my naked eye. With a 6x loupe the lettering is very clear. I believe this lens is from the 1920's or even before. It's certainly not a red dot or anything. I saw somewhere to use a cc of .75 but this was assigned to the format. They also said to use .3 for 8x10 and .15 for 4x5. Say it is .75 for 20x24 with my 30" lens at 30" from the film plane. If I move the lens out to 34" what would my cc change to? I know I'm simplifying here because some of this stuff can't be measured. Anyways, it will cost about \$15 plus a days labor to make this new camera so I guess I'll give it a shot

-- Mark Stevenson (marks@cwo.com), March 24, 2001.

Now my question is, what type of film do you shoot in a camera that big? Any color film made that size?

Yes, Kodak Portra 400 (color neg) is available in 20 x 24. The only drawback: It's \$67 per sheet! See Wisner's site for more info.

-- Stewart Ethier (s_ethier@parkcity.net), March 24, 2001.

Mark, to convert from cc to lpmm, the converstion is 1/cc = lpmm. The best human eyes can resolve about 6 lpmm, while average eyes resolve about 3-4 lpmm. If you use .75cc, that equals 1.3 lpmm. This means your nears and fars, not the point of exact focus, will be reduced to this value, a bit less than what humans can resolve. I would consdider this a low standard for contact printing such large film. I would suggest at min. 3 lpmm. So in your example, if your sign was in the middle of your scene, i.e. close to the point of exact focus, than you should have no problems resolving small items...but not true towards the extremes, which is where the cc is at its largest, i.e. the value used in the calculation which is the least resolvable area of the entire scene.

Your question of ... If I move the lens out to 34" what would my cc change to? Mark, you are thinking of this wrong, your cc do not change. By moving to 34" you are changing the point of exact focus. This changes your near and far DOF distances based on the f stop and the largest cc you are willing to tolerate. I gave the DOF values in my last post. Remember, when you pick 34", what you are really doing is changing the point of focus. Mide Davis has an excellent DOF spreasheet that takes all the guess work out of these calcs. It is my understanding he sent this to you via email. His spreadsheet takes all variables into consideration.

As for diffraction, just to get in the ballpark, take 1500/ fstop, so 1500 / 128 = 11.7 lpmm. So in your case, shooting for under 5 lpmm to film, diffraction is simply not an issue. To increase DOF, you can shoot at f 512, which would equal 3 lpmm...this would give you the DOF you desire without sacrificing any diffraction degredation....Hard to beleive you would get no degredation at that f stop! With your lens, this would equate to a 1.5 mm diameter lens opening. You are nearing pin hole photography. BTW, I saw some pin hole shots made on 8x10 and was shocked at how incredible the DOF was and how well resolved eveverything was...., of course no enlargement potential is the compromise.

I have considered these ELF (Extra Large Format) cameras myself in the past. After evaluating them and doing many calculations, I have come to this conclusioin about the ELF's. In the days of old, when film resolved very poorly, and had very limited enlargement potential, the format size was critical to get larger prints. As films resolving powers multiplied through the years, and when enlargement technology became mainstream, it superceeded the benefit of these ELF cameras. The reason is diffraction. Each jump up in image size requires you to further stop down, which continually reduces the lpmm the film can resolve. Therefore your final print sizes are restricted by diffraction, vs. smaller formats with are not bothered by diffraction, they can fully exploit modern films resolving potential. The enlargement process can provide the additional size required in the final print with less overall loss in quality. Something not possible many years ago, hence the reason these ELF died out.

The ONLY exception to this conclusion is when shooting subjects that have a very tiny amount of DOF, like shooting a wall, or shooting at infinity. Once you remove the DOF variable and the effects of diffraction than nothing can beat the huge ELF film size. In my case, I did not want to be limited to shooting walls or at infinity only to endeavor into such huge cameras. So the moral of the story is, you need to know your goals up front and then use the format and lenses that can acheive them. Another interesting point of lens tilt on ELF... The longer the fl lens, the greater the tilt angle needs to be to to obtain the same value of J. As you tilt, you loose image circle, it shoots up over the film...so in the case of ELF, you are very limited in the benefits of tilt since no lenses have 2000mm image circles. In addition, as you tilt, you begin putting the edges of the image circle in the film area, and this poor resolving area supercedes the cc values you are anticpating getting. Remember, the desired cc you use in DOF calculations makes the assumptions the lens can deliver this this value. With lens abberations, diffraction, MTF's and using image circle edges, the lens quite often can not deliver what you anticipate.

In practical purposes, IMO, I have concluded that 8x10 is the largest usable format today. And even in 8x10, you often are limited by diffraction, tilt angles & movements (do to limited image circles of lenses). So even 8x10 needs to be used wisely. After learning all this, and shooting 8x10 and 4x5 for awhile, I can clearly see why 4x5 is the ultimate LF size today. It offers the best mix of everything, and compared to a 20x24" camera, it will seem like a 35mm P&S.

I truly admire your tennacity to pursue these ELF cameras! What motivates you to do this vs. 4x5 and enlarging 5x?

-- Bill Glickman (bglick@pclv.com), March 24, 2001.

Bill...What you've said makes perfect sense. However, I have been shooting 4x5 and comparing my 20x24 contact prints with a 19x24 enlargement from my 4x5 is roughly equal to comparing an 11x14 enlargement from 4x5 to an 11x14 enlargement from 35mm as a general observation, basically no comparison. But that's my own opinion. People who veiw the 20x24 contact prints are fascinated by the detail, it really feels like you are there. Those viewing the 19x24 and 16x20 enlargements from 4x5 really like the prints but little mention is made of the detail. Hence my efforts to make a new 20x24 camera which would be more mobile and DOF friendly at the near end.

-- Mark Stevenson (marks@cwo.com), March 24, 2001.

Wow Mark, I am fascinated! Do you show your work anywhere in the New York area? I would love to see what you're doing with that monster! Got a million questions but I'll only ask two: Does this camera have a ground glass or do you just point and shoot? Can you reload out in the field or is it a one shot outing?

I tell ya, you really got the wheels turning in this man's mind!

Thanks, DL

-- Dennis Lee (Captdennislee@earthlink.net), March 25, 2001.