Home Made Lens

[Steve H]

All,
I already picked the brains @ APUG, so I will tap into your craniums here too :D . Im going to make an achromatic lens to try some portrait work with. I found a lens from surplus shed (FL=235mm, dia=60mm), and designed a housing for it. Working off of information from Alan Greene's 'Primitive Photography', this lens should have lots of coverage on the 4x5 format. I also followed his recommendations for the location of the aperture.
One issue that I do have, is that I need to make sure Im 'doing the right thing' here for the look I want. Just to clarify (see the end of the APUG thread), an achromatic doublet will give me the soft focus look, right ?
However, this is where Im stuck. Im not sure if Im going to go with a rotary wheel aperture, or the thimble caps sort of like what the Imagon has. I suppose waterhouse stops are an option too, but I see these as the least desirable.
I attached two basic drawings of the design, I appreciate any comments you may have.

FYI - The APUG discussion can be found here:
http://www.apug.org/forums/showthread.php?t=30154

[Steve H]

One design flaw that my drawing may/may not have is the way the lens faces. In looking throught the LVM, I only see two lenses that would be somewhat simular in construction - the Agfa Igenar f/8.8 105mm, and the Imagon. The agfa is setup how the lens in my drawing is; whereas the imagon has the thinner lens in the front.
I am thinking that I will have to wait and see what the lens profile actually looks like before I can make this determination; or trial and error. Unfortunately, all that I know is that the lens is 235mm in focal length, 60mm in diameter, and is a cemented achromat.

[Donald Qualls]

Achromats originally made for small telescopes or binoculars (almost certainly the source for this lens) might be oriented either way, though the majority of those using common crown and flint are installed with the biconvex element toward the world; the only examples I've seen with the concave element forward used a relatively exotic glass for one or the other. Though such a lens can give very good performance in a telescope, an eyepiece is picking up only a tiny piece of the central "sweet spot" and you can expect considerable off-axis aberrations in a photographic application; typically, you'll see a very sharp center and increasing amounts of coma (radial smearing) as you get farther from the lens axis. You're likely to also see increasing color fringing (which records in B&W as plain old blur), called "lateral color", with distance from center, and some longitudinal color error in a plain achromat of that type (by definition, an achromat focuses two colors to the same point from an infinity light source -- an apochromat focuses three colors to the same point, but generally requires three elements, usually one being an exotic glass). Spherical aberration is typically very well corrected at infinity in commercially made binocular and telescope achromats, but focus to portrait distances and you may see some of that as well, along with significant field curvature (again, not a problem with an eyepiece that uses a few millimeter circle in the center of the field).

So, bottom line, yes, on film you can expect a sort of soft focus effect that will get sharper as you stop down. In telescopes, I'm used to seeing the aperture mounted immediately in front of the glass (spacing of a few millimeters in this scale); mounting it further forward will result in vignetting or light fall off.

[Steve H]

Donald,
Thanks for the info. I spaced the aperture according to what Alan Greene recomended, and also judging by what I have seen in 'real world' examples. Somewhere around 1/6th the focal length away from the front of the lens.

[Steve H]

All,
The other day while browsing the local farmer's market, I came across an old brass projection lens. The price was right (under $10), the glass was crystal clear, and at first I thought I had a petzval (Judging by reflections). I wasn't too hopeful as it was only 5.25" focal length, but I took it anyway.
Taking the lens apart, I found it to have a cemented positive meniscus achromat in the front, with an uncemented negative meniscus achromat in the rear. The front element is rather large (64mm), and has a focal length of 230mm......So needless to say, I cancelled my order from the surplus shed.
Yesterday, I made a quick MDF lens board and mounted the lens using its barrel (all of the pieces of the barrel are threaded together). Thankfully, that still gave me 40mm of barrel to work with, as my aperture calcs placed it somewhere between 32-38mm away from the glass.
Taking the average of that number, I affixed the aperture 35mm away from the glass. I didn't see any vingetting. However, to be honest, I didn't see any when I pushed the aperture all the way against the glass either. Even with some big rise, there really wasn't much fall off until after 2" of rise or so. Could the image circle possibly be that large ? Did I goof somewhere ? The only time I saw some vingette was when I affixed the 'snoot' shown in the last photo; and even that didn't show w/o some rise.
I still wasn't sure what was the correct orientation for the lens to face, either ---> ((], or ---> [)). In looking throught the LVM, I only see two lenses that would be somewhat simular in construction - the Agfa Igenar f/8.8 105mm, and the Imagon. The agfa is setup [)); whereas the imagon has the thinner lens in the front.
So now that I have it mounted, I figured I'd give it a shot. This is what I found:

F=230mm
Wide open = f/3.2
Stopped Down = f/16

----> ((] Gives you a super sharp image stopped down in the center. The OOF areas aren't as 'hazy' as a typical meniscus would be, but rather everyhing swirled around the focused area. There is a pretty sharp cut-off as well between in and out of focus. Wide open its still really soft however.

---> [)) Gives you that typical meniscus hazy/dreamy look. Still pretty sharp when stopped down, but it slowly bleeds into hazyness.


Im going to go with the first one, as I think it would work best for me. If I were doing studio work however, I would probably go with the second one, as for tighter shots I think it would work best.

When I put that snoot on (f/4.5), it makes some real nice images on the GG.

In any event, here's the photos:
http://www.flickr.com/photos/veedubrocco/sets/72157594220442744/

[Uli Mayer]

Steve,
as far as I know the most common practice with those simple lenses ( whether singlets or doublets ) was to place the lens behind the stop, its less strongly curved surface, respectively its hollow side being oriented towards the light. I found two - almost coinciding - recommendations about the aperture's best distance: It should be placed within 1/8 - 1/10 of the lens' focal length in front of the first surface in order to minimize distortion ( this is what Ludwig David said in " Photographisches Praktikum", 1931). Hans Harting (designer of the Heliar) wrote in "Photographische Optik", 1948, aperture distance should be no less than 8 % from the lens surface. But he wrote this while dealing with the "Frontar", a simple Goerz doublet that - surprisingly - had the stop behind the lens.
Uli

[Steve H]

Uli,
Interesting, thanks again for the info. I will try to take some photos today with the lens in both orientations; but its really hard to do so in bright sunlight w/o a shutter, or any ND filters :D ....But I'll see what I can do.
As I said earlier, I had the aperture placed @ 35mm away from the front glass surface. This would come out to be about 15%. I honestly saw no difference at all in having the aperture 5mm away from the glass either (2%). I suppose this is due to the large image circle ? How could I test this properly (besides purchasing an 8x10 !) ?

Thanks again,

[Uli Mayer]

Steve,
If it's just distortion that you want to test (and not chromatic aberrations) : for extending exposure times you can fix and stack all types of filters in front of the stop . By varying the aperture's distance while carefully watching the ground glass edges - straight building walls are a good test object - you should be able to see at which position lines start bending inward rsp. outward. I would then fix the stop right in the middle between both positions.
Uli

[Steve H]

Uli,
Thanks. But as I said earlier, even with the aperture resting against the glass, I don't see any bending in a building. Maybe if I shift the rear standard all the way to one side of the image circle and try it again I may ?

[Uli Mayer]

Steve,
this may be a good idea. As far as I understand it (now) you'll have zero distortion (an orthoscopic image) if the aperture is placed directly in front of the lens. By increasing its distance one will introduce barrel shape distortion. So why not place the stop right in front of the lens? - Because, as the scriptures tell, this is not the position minimizing coma. For more reliable information than I could provide with my half-knowledge in optics, you may try to get hold of a copy of Eugene Hecht " Optics". Fig.6.25 and 6.34 in Chapter 6 "Aberrations" show various effects of a too close or a too distant aperture position. The 1/8 - 1/10 f.l. recommendation, I referred to above, may be the "best" compromize in tackling more than one lens fault.

Uli

[Steve H]

Thanks again Uli - I really appreciate your help.
Im going to do some more testing this week, but I do hope to have the barrel cut over this coming weekend. We'll see how it goes.

[Steve H]

All,
Yesterday, I took out my rigged up lens for a test - just to be sure it gave me the look I wanted on film before I committed the cutter to brass. So I took my cardboard f/stops and went down to a local wildlife/land preserve on the Brandywine River.
Being a nice evening, I enticed my girlfriend to come along; with the promise of a relaxing evening bbq/picnic. Unknown to her at the time, I was also looking for my first portrait .
After the BBQ and some idle chitchat, I retrieved the camera. It was about an hour before sundown, and the sun was touching the trees. I needed this sort of light, because my 'Galli' shutter isn't honed yet . So with 25asa film (I don't have any 77mm ND filters, which would be *just* too small anyway), I was able to get close to repeatable accuracy.
I wanted to shoot a variety of textures, as these old lenses seem to display different characteristics when shooting different patterns. So I found some woodland scenes with a lot of foliage, some shots of just trees in a row, and a sneak portrait of my girlfriend relaxing.
Upon developing the negs today, I am VERY pleased. The shots that have the canopy in them give a sunburst effect, with a center sharpness that can be adjusted via f/stop. Wide open (f/3.3 or so), the focused area is nice and soft, slowly sunbursting out. At about f/5.6 the center starts to sharpen up, and stand out of the background. By f/11, individual textures on blades of grass can be seen, but yet the OOF area swirls away. At f/22 (the smallest one I made), nearly most of the neg is sharp, with a slow softness at the edges.
It doesn't give the concentric distortion in the background that a petzval gives, but rather a 'warp speed' sort of look. This seems to come and go dependant on the subject, and the area of focus...Im beginning to think that the burst-like look is dependant upon bellow extension. At infinity, etc there is more glass used to make the image on the neg, so the light refracts off of the curved surfaces more. Whereas when the bellows are extended to focus on something closer than infinity, these abbreviations are cast to either side (into the bellows), and only the center portion of the image is projected on the neg. It isn't like the look when a lens doesn't cover the format - its different (plus, at infinity, I have like...2.5" of coverage still !). It would be interesting to test this theory on an 8x10.
The portrait is what impressed me the most. Its smooth and sharp without being too sharp. There isn't any of that sunburst distortion either. We'll see what it prints like, but judging by the negative, this is turning out to be a worth-while project !!!
As I said earlier, my Galli shutter isn't honed, so I have shots that are mostly over exposed by a stop or so. There is one photo that is exposed well, and it shows some real nice contrast - MUCH better than the old Helostar I was messing with earlier.
Hopefully I can cut the barrel this weekend. I still don't have a packard to mess with, but at least I have the electronics figured out. So slowly but surely Im getting there...

I'll try to get some scans later this evening.

[Steve H]

All,
Here is my final design that I am going to import into the lathe shortly. I have made provisions for a 67mm filter, and I adjusted the aperture for its proper placement (thanks again Uli). Finally, I changed the profile a bit to decrease weight.

[Patrik Roseen]

Steve H, I'm impressed and thrilled by your 'project'. I really enjoy these kind of threads!

Would be interesting to see that 'warp speed-kind of look'.

[Steve H]

Thanks Patrik. My scanner isn't playing well with Linux...So unfortunately I won't have scans until tomorrow.

[Steve H]

Here's the negs. I promised scans....so here they are, in a 'kinda sorta' way :D
My scanner doesn't play well with Linux, and I have yet to bother to purchase one that does (I had ordered a 4870...But after getting one that was DOA, I discovered that Epson isn't the best company in the customer service dept) .
So....In order to keep my promise to you, I had to place the negs on my light table, photograph them with my digicam (ca. 1999), and invert them. Please don't judge such things as exposure and contrast, as its impossible to see. Also, be take notice of relative sharpness, not how sharp it looks on screen. One of these years I'll get a scanner and do it right.

f/3.8
http://static.flickr.com/92/216077434_02a61ea75d_b.jpg

f/5.6
http://static.flickr.com/69/216077431_f33b6b0729_o.jpg

f/11
http://static.flickr.com/40/216077428_cb4811f3bb_o.jpg




In the other thread, I commented about how the distortion seems to be dependant upon bellow draw just as much as it does on the aperture. You can see what I mean by looking at the f/5.6 shot (closer than infinity), and then looking at the f/11 shot (infinity).

Thoughts ?

-Steve

[Steve H]

f/22
http://static.flickr.com/59/216077429_462bcd270c_o.jpg

[Patrik Roseen]

I like it SteveH...now this lens has got 'bokeh' doesn't it?
The effect from using f3.5-f11 is what I like the best and I understand why you said warp-speed kind of look. You could probably photograph any kind of subject and people would call the result 'Art'. Wow!
Edit: Looking at the f22 it moves me too ;-)

[Steve H]

Thanks Patrik. Im curious now to see what an acual print would be like...As contrast and shadow/highlights are all out of wack due to the uneven illumination of the light table.

[Joseph O'Neil]

I love the images - they look like a lot of fun. Take some pictures at a Civil War re-enactment, do some sepia toned prints, man, you could have a lot of fun. I am thinking those old portraits where mostly just the eyes and the nose was in focus.

joe

[Steve H]

I have no idea why the 'Edit' button isn't showing up.....So I apologize for dead links above.

I adjusted the levels and messed with GIMP a bit. Here's the results:

f/3.8
http://static.flickr.com/82/216317797_119eaedcd3_b.jpg

f/5.6
http://static.flickr.com/59/216317796_45ee7d9dbf_o.jpg

f/11
http://static.flickr.com/91/216317792_cd87535c9a_o.jpg

[Steve H]

And, lastly

f/22
http://static.flickr.com/83/216317795_8cf13b9bd7_o.jpg

If the mods catch this, can you please PM me how to edit the posts ?? I read the FAQ, but I just don't see the edit button.