This is topic How accurate should the "optical" path be? in forum Film Handlers' Forum at Film-Tech Forum ARCHIVE.


To visit this topic, use this URL:
https://ft-forum.com/ft/cgi-bin/ubb/ultimatebb.cgi?ubb=get_topic;f=1;t=000787

Posted by John Walsh (Member # 168) on 03-22-2000, 02:12 PM:
 
You know how the center of a lens should line up with the (image) center of the film, and the reflector and lamp. How accurate should all of that be? Of course, the closer the better, but can anyone tell me exactly what the limit is, in the form of, " +/- .00X ."

Also, how would you go about lining everything up? I'm thinking of an, "order of precedent."

For example, the intermittent sprocket can't (usually) be laterally adjusted. So, maybe the first step would be to adjust the trap lateral guides to be directly in line with where the intermittent sprocket puts the film. Then, you might use that as a reference to align the lens; use the lens as a reference for something else, etc. This might be different for one projector to another.

I can see how the lens could be lined up to the reflector (using test film) but how would you insure that everything else was lined up (lamp, film, seperate swing-away anamorphic adapter.)
 


Posted by Gordon McLeod (Member # 33) on 03-22-2000, 09:10 PM:
 
Some of the nut and bolt intermittent sprockets have lateral movement
You need a alignemnt guide that LaVezzie makes for lining up the rails, guide roller and sprocket
 
Posted by John Eickhof (Member # 323) on 03-22-2000, 11:05 PM:
 
The standard guage available is the AG-1
now distributed through equipment dealers
by Edw. H. Wolk, Co., Inc. List price used to be around $40 (they have limited stock on hand, LaVezzi has discontinued this item)
This guage is ideal for flat & curved gates,
it is designed for use with CS (cinemascope
perf) and VKF type intermittent sprockets.
Check with you supply dealer for availabilty
and pricing. Or contact john at email above.
 
Posted by Gordon McLeod (Member # 33) on 03-23-2000, 09:47 AM:
 
I ended up haveing to machine my own since they don't have one for positrol sprockets
Also stainless steel film is very handy as well
 
Posted by Christopher Seo (Member # 7) on 03-23-2000, 10:46 AM:
 
What about screens in which the vertical center is changed due to 'guillotine' type masking? I gather that (at least with turrets) the lenses themselves are positioned to accomodate this. Does that mean that at least one lens is off-center to the film frame?
 
Posted by Gordon McLeod (Member # 33) on 03-23-2000, 11:10 AM:
 
yes most turrets allow for lens centering
 
Posted by Pat Moore (Member # 380) on 03-24-2000, 05:06 AM:
 
Hi Guys;
For what we call the "optical A-Chain", the lens is the key for the whole system. Projector manufacturers take great care to be sure the lens holder(s) are properly aligned to the film trap and gate. Before you change those relative positions be sure you have a really good reason to do so. The alignment of the reflector in that A-Chain ia also a key, perhaps most important. It's easiest to use the lens as a starting point, using a laser alignment tool. OK, not everyone has one of those, but that's the best way.
The relfector keys are Working Distance (its relation to the film plane) and its axial alignment to the lens and film centerlines. The reflector's optical axis must be the same as the lens and film. If it's not, you'll have trouble with light output, distribution, and image quality on screen.
Lots more to talk about, but it's true that the image projectio system has an "A-Chain" just like a sound system. If it's not right, what you get in the B-Chain (screen and auditorium) will never be optimized.

Pat
 


Posted by Pat Moore (Member # 380) on 03-24-2000, 05:20 AM:
 
Some thoughts on film trap alignment. This is really tough to do without the proper tools and suggested procedures vary among machines and gate/trap types. The film path through the trap to the intermittent sprocket is the most important part or the film starts doing some really strange things.
The film needs to be absolutely perpendicular to the intermittent sprocket. The AG-1 guage mentioned before is the best way to do this in the field, though the gage was originally designed with FLAT Traps in mind. Steel film is a good alternative.
Most film traps now are curved but the same thought applies. The AG-1 will show if the lateral guide rollers are at least close to where they should be. Rememeber the outside (soundtrack side)guide is fixed, so it's your reference point for the film plane.
Generally, you want to adjust the sprocket to the film path and not the other way around. Looking at the curved straps on the trap, make sure they are parallel and that each side has about the same spring tension.
You shouldn't really need to do this unless you rotate the sprocket or there is some real obvious indication that alignment is off. It's easy to check, but be sure before you start changing things without proper wherewithall.

Pat
 


Posted by Gordon McLeod (Member # 33) on 03-24-2000, 10:46 AM:
 
The ligth loss and distribution problems are far less if you recenter the scope lens holder in a theatre that has movable top masking rather than just cut an offset plate.
The lens mounts were designed to be the correction point for centreing from the advent of sound ( the original suppers had a swingable lens adjustment) to mag 35scope vs optical
Also all dual guage machines have lateral lens recentering to correct for the centreline of the frame differences
One of the first papers on the use of lens centering to correct image shift was during the boomof 3d when it was determioned that a better 3d effect (sharpness and convergence) would be obtained if the two machines were aligned parralel to each other and then only the lens mounts be shimmed over to converge the images
To this day the Imax 3d system relies on this technique in both the twin machine installation and the dual rotor design
Also I have found the best laser alignment mounts the laser in the back of the reflector and shoots forward onto a centre pinhole apperture rather than from the lens mount back
 
Posted by Pat Moore (Member # 380) on 03-24-2000, 12:37 PM:
 
I agree totally on which lens to reposition if required. The Scope lens is always a longer focal length and is therefore much more forgiving if it gets off-axis. The flat lens and format is far more critical and should just about NEVER be moved from dead-centerline.
As for the laser use, either way works if the tool is used correctly by a person that knows how to use it.
 
Posted by John Walsh (Member # 168) on 03-26-2000, 05:06 PM:
 
Okay, but still; does anyone know for sure how close the optical center should be?

What do you do with a Cinemecannica V8, where neither the intermittent sprocket or the trap can be changed? I'm guessing a V8 is supossed to be machined correctly at the factory.

One reason I'm asking is because I find that many of those "lens bushings" (to adapt a regular lens to a 4" holder) are way off. One from Century was out-of round by 0.015" which seems alot.

On the Simplex bushings, they have two plastic screw things to hold the lens in, but also seem to hold the lens off center.

Also, the lens holders in some Simplexes we have (PR1050) seem off.

So, anyway, I was just wondering how close it should be. Maybe I'm getting too picky-picky, and 0.015" or so is still OK.
 


Posted by Pat Moore (Member # 380) on 03-27-2000, 07:29 AM:
 
Ideally, of course, you want zero tolerance for the lens and film centerlines but that's really unnecessary. 0.015" is pretty close. Some lens adapters are acually made eccentric to allow for adjustment. PR1050 Lens Holders are adjustable to allow for screen alignment -- we get them very, VERY close at the factory on turret machines.
Most important: the lens must be perpendicular to the film plane to allow even focus across the screen.
 
Posted by John Walsh (Member # 168) on 03-27-2000, 09:52 AM:
 
OK, if +/-0.015" is acceptable I'll go with that. I've been trying to get it to +/-0.005", and not very sucessfuly, either.

I'll ask two other related questions:

It is strongly implied (by projector mfg.s) that a 35/70mm projector can be changed from one format to another on-the-fly by just changing parts and settings, like the lens "mag-optical" setting. Yet, I never saw where the image was still centered on the screen after going from/to 35/70mm. The pedstal always had to be pushed left/right a little. Has anyone been able to from/to 35 or 70mm without moving the pedstal?

I worked with a tech guy (at UA) who would remove and throw away the studio guides on every projector. Is that good or bad?

Pat wrote:
"Projector manufacturers take great care to be sure the lens holder(s) are properly aligned..."

Not the Centurys I worked with. Admittedly, it was 15 years ago, but then they were totally off. Actually, it was the gate/trap; Thread film and close the gate with a brand-new-out-of-the-box Century, and it would pinch the film so hard it would strip the sprocket, (good thing it was not polyester!) You could see the gate was on an angle- they just screwed the four screws in and shipped it.

It's from working with those that I've become annoyingly suspicious of every optical path.
 


Posted by Pat Moore (Member # 380) on 03-27-2000, 01:09 PM:
 
Hi John;
I think your experience a few years ago with Century is one reason we bought them a few years ago. We've had our challenges but but I know the machine works pretty well "out of the box now.
Most 35/70's had some sort of a lens offset device to accommodate the change in film centerline. In the originall JJ it was the upper lens guide that was an eccentric bushing. The first Simplexes did something similar to that. The big Norelco AAII had an eccentric lens adapter, as I remember.
The real challenges came with the Turret projectors -- the Century addressed that the best, actually equipping one lens holder with an offset for the two formats. They always required a little fiddling around, though, since the actual required offset depended a bit on the lens and throw. You couldn't just chnage the lens centerline 0.055" and have it be right.
Studio guides. Heard that one, too. Each to their own but I don't agree with it. If they are adjusted correctly, they should stay because they really help lateral weave, especially on the film's reference side. The problem comes if they are set up too tight, or not properly lined up with the intermittent sprocket and/or the lateral guide of the film trap.

Pat
 


Posted by John Walsh (Member # 168) on 03-27-2000, 03:40 PM:
 
Sadly, the first Centurys I ever used were the same ones that made me dislike Centurys in general- which is probably unfair today. I tend to gauge all Centurys on these two machines. We had them rebulit in 1982, and had endless problems with them. After talking with several people, I got to talk to a guy named "Nick" in the drafting dept. who was very helpful. But then he quit. Then I talked to another guy who was good, but then he quit, too. They were JJ-2's with the eccentric bushing for the lens.

One time I completely wigged out on someone there; I don't remember who. One of the many things I was trying to get fixed/resolved was a heat filter. The JJ's had a holder for a round infra-red filter, just behind the trap. I kept ordering it, saying, "They are round shaped." and kept getting square-shaped ones. This went on about 6-7 times within a year. I finally said something like, "Dope! Don't you know the difference between square and round!" There were several problems that went on and on like this. I had another exciting go-around with a repair/exchange intermittent that would'nt work, but that another story...

I must say that I liked the design of Centurys, just not the quality of the manufacturing or some of the support people that were working there in 1982-87. I'm glad you guys bought them, because I think it's a good simple design if made correctly.

I got that little wedge-shaped thing to align the trap to the intermittent sprocket- but it didn't help with the gate at all.

If you feel like looking up my projectors, I think the ser no.s were 152 and 153, or 182 and 183- I can't remember which.
 






Powered by Infopop Corporation
UBB.classicTM 6.3.1.2