Collimation Suggestion

[Thor Wixom]

Here's a thought (read opportunity) for someone like Doug Underdahl...

Get an inexpensive prime and "Lock" it in at a known, fixed focus distance... say 10 feet.

Resell the lens as an absolute reference point for collimation. It would ONLY be used it for collimation, not for shooting, so the "quality" of the glass wouldn't really matter, as long as the mechanics of the lens were permanently set at one focus distance.

In fact, I'd be willing to purchase 2, maybe 3 lenses at various focal lenghths with "absolute" set focus distances. That way I could cross check them and make sure the collimation is accurate at various focal lengths.

Seems like it would take the guess work out of collimation, maybe even eliminate the chance of setting it wrong. It would be relatively inexpensive, and should be dead accurate every time if it was permanently set by a professional lens guy.

Would anyone else be interested in this?

-Thor Wixom

 

[Douglas Underdahl]

Shop standard lens

Shop standard lens

Yes, that makes sense, to have a "shop standard" lens of your very own. When I worked at various camera repair outfits we would have a lens like this, usually a 24mm focal length, that was old and beat up but was set correctly and lived in a drawer and never was used for anything other than setting ground glasses in film movie cameras. It was a cheap lens, maybe even scratched, but it was dead on, focus-wise.

If you want one of your own, here's what - buy a cheap prime in the mount of your choice, aka the mount that you have on your RED. Right now that's probably Arri PL but soon it could be Nikon or Leica or BNCR or Oct-19 (working on all of 'em).

Now here's how we made sure that the lens in question was correct.

First, you get a lens depth collar. This was a collar that was made of stainless steel that was of the correct "depth", or flange focal distance for the camera in question. For Arri that's always 52mm. For Nikon it's 46.50mm. Then you rest the collar on a granite standard, which is an incredibly smooth flat granite block that's certified and traceable to some such and such international standard. Then you put a dial gauge into the block, and zero it, so that the gage is reads zero when it's plunger is 52 or 46.50mm, etc.

Now you put that zeroed gauge into your Richter or Chrosziel collimator and use it to zero the focal plane micrometer. Then you put your shop standard lens on the collimator, and use it to make sure that when that lens is set to infinity, that the image is 52mm or 46.50mm or whatever from the film plane.

I remember that the Chrosziel was the best collimator, hands down, but you could do just as good work on the Richter. The Chroziel had a video camera which saved your eyes and helped to show customers what was going on. It was ten grand, whereas the Richter was four grand.

Anyhoo, now you've got a tested, shop standard lens, one that you can use on your RED to set the flange focal distance as you see me doing in the Youtube video. Search there on RED CAMERA LENS MOUNT.

It all goes back to that collar, which you buy from Arri or the other camera manufacturers. That's the ultimate standard that you set your other stuff up from, the starting point.

What the freak is this guy talking about? I don't get it.

Well, here's a shortcut. Get a brand new lens in the mount that you are using on the RED and have it checked by a tech that you trust. Or buy a brand new Nikon mount lens, and check it on a Nikon camera. Focus the lens on something at infinity, like the treetops in the distance (well, that's not infinity, but it's close. Only God can make a tree and God lives somewhere beyond infinity, which you can actually set your lens to, if you want). Now test that lens on the Nikon at 6 feet, four feet, etc. Compare tape and eye focus. If it's correct, use it to set your RED.

Put it in a drawer and only use it to set your RED, and now you have a shop standard lens.

Buy the cheapest prime lens that's about 24mm from B and H that you can get. Probably it's made by Tamron or Tokina or someone like that.

For Arri, I'd go with a 16mm format prime, like a 16mm Schnieder, then use an adapter to change it from Arri Standard to Arri PL. It's okay that it doesn't cover 35mm and you'll see vignetting like Porky Pig saying That's All, Folks, because you just need it for focusing, not the actual image.

Remember this - your RED came from the factory correctly set up for 52mm Arriflex depth. They set that excellent adjustable mount with, no doubt, a shop standard lens, probably a really expensive brand new Cooke S4 (just a guess). So when you get your camera, it's ALREADY SET FOR 52MM. So before you alter your lens mount, that is, turn that ring left or right after loosening those cap screws, make a mark on it, then a corresponding mark on the body of the camera. If it were me, I'd use a carbide scribe and scratch these marks permanently into the camera and mount. I'm gonna do this soon as I think my RED is coming this week! (happy happy joy joy)

So no matter how you might move your lens mount back and forth, you can always reset it to those original factory marks when you put that Arri PL mount back on the camera that it came with in the first place.

With a film camera, you could put your dial gauge right into the lens mount and use a flat gauge block in the gate to actually, physically measure the flange focal distance. Can't do that with the RED, there's a filter in the way and even if there wasn't you wouldn't want to put the tip of a chrome plated steel plunger onto your Mysterium because it would likely permanently damage it (but when they check you for interocular pressure, they do put something on your eyeball, when you go to the eye doctor. Boy that's weird. Hmm.)

So you need a shop standard lens to check that distance. In a film camera, you can clear the shutter (move it out of the way) and put the gauge in the camera to test the depth. But you can't use a gauge to check the ground glass/fiber screen, because the image bounces off the mirror shutter and there is no gauge that makes a 90 degree turn, as far as I know.

Whew. Hope this makes sense.

 

[Steve Gal]

to me it does.

 

[Harry Clark]

It's a good idea Thor, and an excellent summary Doug!
I typically bring my film cameras and lenses to CSC in New York, and my HD cameras to Abel in New York. It's a good reference to have a professional shop check the gear if you have any doubts; after all, we're (mostly) cinematographers, not technicians. And like going to the doctor, a second opinion is a good thing...
Besides, for me, it's a good excuse to socialize with techs whose company I enjoy!
But for quick checks, a cheap, "known" lens would be handy...
Cheers,
Harry

 

[chuck colburn]

The Richter lens standards for what ever lens mount were always spot on. I owned two of his autocollimators and used at least five other ones over the years, and they all read the same. The only thing I ever checked was the autocollimator optical unit itself if I hadn't used it before. I corrected one for an error of about .0005". The qualification test itself is fairly simple. So what I'm saying is if you have a Richter unit and you check it out you don't need a test lens unless you don't have a lens on the camera.
But I'm assuming you have the whole test bench not just the autocollimator.

http://www.chamblesscineequip.com/catalog/Richter R2 Autocollimator System.htm

 

[Mitch Gross]

A cheap known lens is good in concept, but lousy in execution. On most still lenses, if you breathe near the damn focus ring it will drift a few microns. Touch it (not rotate, just touch) and it could pop out by 50 microns, which is enough to put things seriously out of focus.

I've got a secret for you -- know how we set old Zeiss SuperSpeeds? On the collimator we quickly rotate them from minimum focus to infinity and back again three times to see where they naturally want to land. Because they can easily be way, way off if you just dial them in to infinity normally. And these are excellent lenses that have natural mechanical wear over time.

Perhaps it is possible to get some sort of simple lens that at a precise distance would only focus to infinity and do so sharply. But I wonder what part of the color spectrum it would focus, as a simple lens probably would have little to no coatings and intense chromatic abberation and frequency stretching. Red would focus at one distance and blue another. Hard to judge setting a lens mount with that.

 

[chuck colburn]

It would probally be best for most people to establish a working relationship with a "good" lens tech even if you plan to own your own collimator. A day or so with him/her would be worth your while. Oh yeah, and make sure you pay him a goodly sum for that time espcially if he's in the living hell of a rental house. lo

 

[Mitch Gross]

Oh yeah, and make sure you pay him a goodly sum for that time espcially if he's in the living hell of a rental house. lo

Wait a sec, I resemble that remark!

 

[laguun]

Wait a sec, I resemble that remark!

agreed.

 

[Thor Wixom]

Doug, you rock!

Mitch, I wouldn't have thought of the different wavelengths of light each having a different focus. That's why you're the tech!

Chuck, thanks for the link on the Richter.

By the way, Red has a lens they did this idea with. That's actually where I got it, but they used a high quality lens (don't remember which one, I was too busy staring at the Ducatti they set up for Red day).

-Thor