Some years ago I was invited to visit the photo lab at National Geographic.
What a shock. I knew they worked almost totally 35mm. But what they had in
their lab at the time was a lineup of 12 10x10 Durst Laborator 184 enlargers,
all with colorheads. " Where do you do your 35mm work?" I asked. "Right here"
was the answer I got. I told them I thought they would be using Leitz Focomats
or some other dedicated 35mm enlargers. I got a very patient reply that laid
out their strategy. It went like this. " The bigger you enlarge, the more
damage is done by vibration and lack of perfect alignment. What is needed is a
massive enlarger that has been machined into alignment and cannot be misaligned
even by the most careless lab technicians. So the largest heaviest most rigid
machine available is far and way the best for 35mm. I couldn't find fault with
that argument in any manner or form especially when you see the quality what
National Geo turns out routinely. I compromised. I got the Durst 138s 5x7
enlarger for my 35mm. work.But I guess by now Nationall Geo is all digital.
(sigh)
Arthur Kramer
Visit my WW II B-26 website at:
http://www.coastcomp.com/artkramer
I'm not quite sure I buy this.
Small, light structures tend to be more rigid than large, heavy
ones, given simiar materials.
Also, their natural frequency tends to be higher making it easier to
damp and easier to prevent mechanical coupling with building
structures.
Of course, resistance to vibration depends on idividual design, but
I rather think its easier to make a low vibration structure which is
small than one which is large.
Also keep in mind that the strength of a material varies with its
cross section while its weight varies with volume. So, as a general
statement large structurs of the same dimentions as smaller ones are
weaker. Compare a spider to an elephant to get some idea of this
effect. The elephant is by far the weaker structure.
---
Richard Knoppow
Los Angeles, CA, USA.
dick...@ix.netcom.com
Richard, FWIW, my experience with machine tools says
heavier is better.
> Compare a spider to an elephant to get some idea of this
> effect. The elephant is by far the weaker structure.
This may be true but which one will be moved more by
a gust of wind?
--
Ken
--
I follow you reasoning Dick.But it is a case of comparing the rigidity of
let's say a Focomat with a Durst 184. In my experience the 184 wins every
time.I think we must make direct comparisons of one machine against another
rather than broad principles.
I just don't want to see a spider the size on an elephant :-)
______________________________________________________________________
Posted Via Uncensored-News.Com - Still Only $9.95 - http://www.uncensored-news.com
<><><><><><><> The Worlds Uncensored News Source <><><><><><><><>
artk...@aol.com (ArtKramr) wrote in message news:<20020717165349...@mb-fe.aol.com>...
I love my massive floor standing Durst L138S with motor driven vacuum
baseboard and Super Chromega E dichroic head. It is truly an Elephant. (grin)
> <snip> ... So the largest heaviest most rigid
> machine available is far and way the best for 35mm. I couldn't find fault with
> that argument in any manner or form especially when you see the quality what
> National Geo turns out routinely. I compromised. I got the Durst 138s 5x7
> enlarger for my 35mm. work.But I guess by now Nationall Geo is all digital.
> (sigh)
Apparently not.
Geo's current FAQ claims that nearly all their photographers use 35mm
transparency film, "such as Fuji Provia 100, Fuji Velvia 50, Kodachrome" and
that their cameras are usually "Canon and Nikon 35mm SLRs and the Leica M6
range finder." 300 to 400 rolls taken per story is average, with longer
stories sometimes consuming 1000 rolls. I would assume that they scan
the negatives to produce their rotogravure print drums.
My guess is that when you see Geo converting, you can assume digital has
equalled 35mm in quality. They are much more quality conscious than most
publications.
That's reassuring to hear. I know they switched from 10x10 Dursts to 10x10
Omega many years ago In fact I bought my 5x7 Super Chormega E color head from
them when they converted. It is now on my Durst L138s. Back then they used only
Kodak materials. Seems National Geo was heavily invested in Kodak stock. Since
the stock took a dive, I guess National Geo figured it was ok to go elsewhere.
But it is nice to hear that they are still using 10X10 monsters for their 35mm
work.
Depends on surface area vs: mass. Probably the elephant wins here.
Which one will survive falling off a ten story building?
>
>"[SNIP]> Compare a spider to an elephant to get some idea of this
>> effect. The elephant is by far the weaker structure.
>> ---
>> Richard Knoppow
>> Los Angeles, CA, USA.
>> dick...@ix.netcom.com
>
>I just don't want to see a spider the size on an elephant :-)
>
>
>
And you won't:-) It would be crushed by it's own weight!
I agree with this. Many enlargers are not very well designed from
the standpoint of rigidity. Very large format enlargers are really
industrial tools and tend to show better engineering, or at least more
expensive construction. Actually, they wouldn't work at all if poorly
made.
Probably the big Saltzman enlargers were the ultimate. They look
like Brigport vertical milling machines.
However, I stand by my statements of physical principle, essentially
smaller structures are stronger and can be made more rigid than large
ones. A small enlarger would be much easier to make stable than a
large one.
Isolation from vibration is another matter. At one time it was
thought that mass was the key. It was then discovered that the proper
design of a structure as a filter was the key. This is the combination
of masses and compliances along with damping.
For instance: in the design of a loudspeaker enclosure it is
important that the enclosure does not transmit vibration. Ideally, it
should be infinitely rigid. While at one time very massive structures
were recommended (hollow and filled with sand) it turns out that such
heavy structures are not necessarily rigid. The mass simply lowered
the natural frequency, sometimes to a point of being a very efficient
radiator at the lowest frequencies. The best structures turn out to be
the lightest and most rigid possible. The natural frequency is raised
allowing much more efficient damping. A laminated structure of, for
instance, plywood and buliding paper or rubberized cork, forms a
relatively light structure which is very well damped. If it is broken
up into small areas by braces it becomes very rigid. The low
fequencies which the enclosure is supposed to contain are well below
the natural resonance of the panels in a region where the radiation
efficiency is very low. The result is a very effective enclosure which
is much lighter than one made of very thick plywood or, worse yet,
particle board.
While this is removed from enlargers it illustrates physical
principles which apply to both.
Similar considerations are found in the design of (especially)
aircraft structures and automobiles.
A proper enlarger design would attempt to minimize the weight of
supported parts to raise the resonance frequency of the enlarger head
on its support column (a pendulum) to well above expected vibrational
frequencies, and provide damping to prevent the system from becoming
excited into a self sustaining vibration.
Large, massive, structures have frequencies which can be at the
exciting vibration found in buildings. If it can be placed below them
there will again be a loss of efficiency in coupling, which is
desirable. However, there is no direct relation between mass and
stiffness, a more massive structure is not necessarily more rigid,
although it may seem to be.
Enough of the very off topic stuff.
Depends. If the spider falls first and then the elephant you will have a
squashed spider under the (also squashed) elephant. Reverse the order and the
spider may survive.
;-)
Bert
My reason for wanting an enlarger a size or two up from 35mm is to get even
illumination at the baseboard. With my D2 + cold light using 35mm or 2-1/4,
illumination at the easel is even within 0.1 stop. When I go to 4x5, I get a
0.6 stop dropoff at the corners. I have to burn corners like crazy with 4x5
negatives.
> My reason for wanting an enlarger a size or two up from 35mm is to get even
> illumination at the baseboard. With my D2 + cold light using 35mm or 2-1/4,
> illumination at the easel is even within 0.1 stop. When I go to 4x5, I get a
> 0.6 stop dropoff at the corners. I have to burn corners like crazy with 4x5
> negatives.
How do you measure dropoff? Even a state-of-the-art Apo-Rodagon-N 4/150
has a dropoff of 0.6 f-stops (measured at f8) at the corners, according
to the Rodenstock literature on Paul Butzi's site. So, a enlarger
with a perfect light source has this dropoff in the corners.
AFAIK a few enlargers (notably the Leitz Focomat V35) have a uneven light
source (brighter at the corners) to correct this falloff. The V35 is
adjusted to his Focotar 2.8/40.
Martin (who never saw any effects of light dropoff in his 4x5-enlargements,
but probably has taking lenses with falloff, and the lighter density
in the corners help...)
You are not using a 135mm enlarging lens instead of a 150 are you?
What focal length lens are you using for 4x5. If its 135mm a good
deal of the fall off is the lens.
I have both condenser and cold light heads for my Omega D-2V. I found
that after carefully aligning the enlarger the condenser head gave
more uniform illumination especially at 4x5. The condenser is also
brighter for small negatives although its printing speed for 4x5 is
very short.
If you are using a 135mm lens about all I can suggest is going to a
150mm and some method of getting longer throw for the enlarger.
There are a number of 5x7 enlargers on the used market although they
are far from as common as 4x5. The most often seen is the Elwood. The
late models (after 1946) are good enalargers. The earlier ones will
need to have the lamphouse replaced with a cold light lamp. They had
silvered reflectors, nearly always tarnished beyond salvage. Late ones
used vacuum deposited aluminum which lasts forever if not physically
damaged.
The Omega E-3 is also a good enlarger but pretty rare.
5x7 stuff is large compared to 4x5, just on the margin of what you
can use practically in a kitchen darkroom.
You might also try maing an equalizer for the condenser head and
shorter lens. This is simply a negative exposed under the enlarger and
developed to a rather low contrast. Its placed in the filter drawer of
the enlarger or under the bottom condenser.
>My guess is that when you see Geo converting, you can assume digital has
>equalled 35mm in quality.
I'm not to certain they'll bother until storage gets to be
a good deal easier to carry. Just figure how many images one of
their photographers takes each day. Now imagine the number of
cards they would have to use to store all of those images in RAW
format.
Regards
John S. Douglas, Photographer
http://www.darkroompro.com
Cold light heads, especially round ones made to replace condensor heads are
notorious for not really covering the largest negative they are made to
cover. Using a longer lens with this head will help a lot, replacing the
diffusion material with a thicker piece of plexiglass will also help. You
may be able to adapt a larger cold light head to fit on your enlarger (I did
this for an 8x10 elwood enlarger I once used, it was very even, but not the
greatest enlarger in the world). The best solution is probably to buy a
color head to replace the cold light. It will be much more even and give
you easy filter adjustments for multigrade printing. With so many labs
switching to digital printers, I would expect you can find a used 4x5 color
head or enlarger with color head easier and cheaper than a larger enlarger.
Since I got a decent 8x10 enlarger (De Vere), I've been using it for
everything except 35mm. I don't find it to have more even illumination than
my 4x5 Besler with color head, but the enlarger is much more ridgid, easier
to align, stays aligned better, etc. I might use it for 35mm if I had all
the right pieces (negative carrier inserts and recessed lensboard), but
aligning the negative strips correctly when both ends are completely covered
by the 8x10 negative carrier is a bit of a pain.
John Sparks
>the right pieces (negative carrier inserts and recessed lensboard), but
>aligning the negative strips correctly when both ends are completely covered
>by the 8x10 negative carrier is a bit of a pain.
>
>John Sparks
I suggest you go to Carlwen rapid shift electron microscope carriers to solve
that problem. I have a full set for rmy Durst 138s Laborator
Where do you get them and do they hold the negatives flat? I typically use
either glass carriers or a half glass carrier (glass on top, metal mask on
the bottom).
John Sparks
>Where do you get them and do they hold the negatives flat? I typically use
>either glass carriers or a half glass carrier (glass on top, metal mask on
>the bottom).
>
>John Sparks
Carlwen Industries Inc., 19219 County Rd. 10, Hamel, MN
55340, phone 612/420-9400.
I have not used the big 8x10 enlargers but I have used Omega D5, Beseler
45MX and Devere 504 and none of them are as good as either of my two Leitz
enlargers: V35 and IIc. I am making big (16X and 24X) enlargements and by
far the biggest factor in corner to corner sharpness of the prints is in
staging the negative. Leitz enlargers are the only ones I have used the put
the negative in exactly the same spot and same plane every time. When
making big enlargements, the lens is very close to the negative and even the
most minute failure in positioning of the negative will cause some falloff
of sharpness of grain in the corners of the print. In my experience, only
the Leitz enlargers do this well.
The question was what enlarger is best for 35mm. My answer is it's the V35.
Ron McKinney
Truly. dr bob.
--
Manny Bhuta
Randolph, NJ USA
"Rhondal McKinney" <rmck...@ilstu.edu> wrote in message
news:B963FED4.104E%rmck...@ilstu.edu...
<snip>