RED and zeiss high-speed colimation issue??

[Felipe Perez-Burchard]

Hi everyone, its been a while since I last posted here...

 

A quick search didn't yield any results for this issue, but maybe I didn't dig deep enough, please let me know if you've seen this or read about it:

 

I've used the RED quite a bit since its introduction, I've been out with it in the desert in high budget and low budget productions of all kinds, but in the last film I did an issue presented itself that I had not encountered before regarding a shift in back-focus or collimation I think.

 

Let me mention, in the past I've put Ultraprimes, master primes, cooke S4s and Nikon mount lenses but particularly with the RED I hadn't used the Zeiss High-Speeds; pure circumstance of the projects, this was the first time.

 

So the first time we saw this issue we were shooting in some damp tunnels in Guanajuato and half-way during the shoot the 1st ACs marks didn't work, even on an 18mm (in fact it was most present in the wide lenses) something that was about 30' away he had to focus to about 10' in order to get focus (they were working fine at the beginning of the day) and on the long lenses close to infinity wouldn't seem sharp.

I know the rental company very well and they get the backfocus adjusted at Birns and Sawyer in LA and I trust them, but another body was ordered and the next day (we were on night shoots) it arrived and we swapped it and tested it and it had the very same issue... so it wasn't the body... my guess was that given the humidity of the place and old lenses there had been a shift in the flange distance somehow.

When we were back in the city with the rental house resources we went to check them again... they were fine, everything worked properly, but we switched the lenses anyway just to be sure (for another high-speed set).

A week later shooting in a park (yes it had been raining at it was damp) half-way through the shoot the same thing happened... all the more confirmation for my theory I thought -- it had been raining.

That evening we took the gear to the rental company and while the lenses were still wacky, we tried a set of ultra primes and everything fell in its place, totally perfect... it had to be the lenses!!

Well, that was close to the last day of shooting of the 2nd Unit (of which all this was on) and the production decided to stay with the lenses and they were mostly fine the rest of the shoot (the weather was cooperating).

Since we wrapped the camera went to work as the B camera on the 1st Unit with another crew and another set of zeiss high-speeds; with perfect weather in the last couple of days the same thing happened to them, and in fact they stayed funky... and are still giving the 1st AC problems.

 

Is this a known issue with high-speeds?

has anyone seen this before?

Im sure its collimation of the lenses, but it happened on three different sets, and the main camera in 1st unit, I hear doesn't have the issue (havent seen myself)...

This is very puzzling and mysterious, if anyone can shed some light on the matter I'd be very grateful.

 

Thanks for your time and responses.

 

Best,

 

-felipe.

[John Sprung]

This is a strange one. What I can bring to the discussion is to run the numbers thru the optics formulas.

 

Running an 18 mm lens focused at 10 ft and 30 ft, the difference in the position of the lens comes to 0.0028 in. = 0.071 mm. By FFD standards, that's a big error.

 

Given that the 10 ft. mark on the lens worked for an object that was actually at 30 ft, what happened is that the lens moved closer to the film by that amount. Whatever it was, it caused the lens to shrink. The AC had to turn the ring to move the lens farther away from the chip.

 

If it's internal to the lenses, it should cause the same problem on film cameras, and be visible in their reflex finders.

 

 

 

 

-- J.S.

[Thomas Dobbie]

Hi,

 

I've occasionally had a similar issue,when the camera gets very hot,the back focus seems to go out,sometimes by quite a bit.

I've seen it suggested that the large PL turret on the Red,also acts as part of the heat management system,if true it would also suggest

that some of that heat is transferred to the lens.

It doesn't sound like this was the issue in your case,as what you describe doesn't suggest overheating.

Interested to now if anybody else has experience of this.

 

Tom.

[Oli Soravia]

Is this a known issue with high-speeds?

has anyone seen this before?

Im sure its collimation of the lenses, but it happened on three different sets, and the main camera in 1st unit, I hear doesn't have the issue (havent seen myself)...

This is very puzzling and mysterious, if anyone can shed some light on the matter I'd be very grateful.

 

 

 

This sounds really strange. I`ve shot an entire picture only with the HS in all weather conditions, but never had any problems with them. Although the ultra primes are better lenses, I prefere the look of the HS because they are not as crisp as the newer lenses, but still give you enough speed for the low sensitivity of the RED. And, the look doesn`t become too "HD-sharp". Let us know what comes out of it, could be good to be aware of those problems. Best. OLI.

[John Sprung]

It doesn't sound like this was the issue in your case,as what you describe doesn't suggest overheating.

 

Indeed, it's the wrong direction. Heat causes metal and glass to expand. What happened here was contraction. Assuming aluminum and a whole inch of material, to shrink by the observed amount, the lens would have to be colder than the ambient temperature by 215 degrees F = 120 degrees C. So, we can absolutely rule out temperature as a factor.

 

 

 

 

 

 

 

-- J.S.

[Satsuki Murashige]

I have flange issues all the time with the Red. I mainly use them with Super Speeds (rotating between about 4 different sets and five different bodies). Part of the problem in my case is that only one of the rental houses or private owners whose gear I work with get their lenses collimated on a regular basis, but I always take note of how far off they are in prep and none of them are off by more than 6", most by less than 3".

 

Even so, at some point in the day, sometimes twice a day, the flange will go way out. So I'm curious whether the original poster tried resetting the flange in the field before sending the bodies back. I just expect to have to do it now, but now I'm wondering if other AC's have the same issue on bigger shows. That would be a total nightmare...

[Felipe Perez-Burchard]

I have flange issues all the time with the Red. I mainly use them with Super Speeds (rotating between about 4 different sets and five different bodies). Part of the problem in my case is that only one of the rental houses or private owners whose gear I work with get their lenses collimated on a regular basis, but I always take note of how far off they are in prep and none of them are off by more than 6", most by less than 3".

 

Even so, at some point in the day, sometimes twice a day, the flange will go way out. So I'm curious whether the original poster tried resetting the flange in the field before sending the bodies back. I just expect to have to do it now, but now I'm wondering if other AC's have the same issue on bigger shows. That would be a total nightmare...

 

Thank you all for your replies...

 

I'm still puzzled by this though. I agree with John that it seems heat is definitely out -- can humidity contract objects (even though that is probably not it either as I mentioned)?

 

Regarding Satsuki's question; well, we adjusted the back-focus on the body the first time as opposed to the mount on each lens, but it ended up being ridiculous because we had around 6 different back-focus marks for our 10 or so lenses... you see it didn't seem to be off by the same distance (although always in the same direction) on the 18mm it might be off 20' and on the 85mm it might be off only 5' -- which makes sense I think for the distance the barrel was moving.

 

I'm going to try and get a hold of the body and lenses when 1st unit wraps before they go out again and do some more thorough testing, but I'm also not exactly sure what I should try other than see if it happens on other bodies or with other lenses in the exact same situations (which I did already but it was a bit rushed).

 

On another note, I agree with Satsuki that high-speeds or super speeds are useful when shooting HD to get rid of that hard edge (if you are going back to film then I think you will gain the softness on the print, but for TV definitely); I've done this with the F35 and never had these kind of issues (had other issues but purely camera design related).

 

Thank you all for your time and thoughts on the matter...

[monday sunnlinn]

this will definitely show how little I know about the internals of a good lens, but is it possible that the change in humidity of the air between the elements are actually moving the individual elements by changing the air pressure?

Edited September 13, 2009 by monday sunnlinn

[John Sprung]

-- can humidity contract objects (even though that is probably not it either as I mentioned)?

 

... you see it didn't seem to be off by the same distance (although always in the same direction) on the 18mm it might be off 20' and on the 85mm it might be off only 5' -- which makes sense I think for the distance the barrel was moving.

 

I'm going to try and get a hold of the body and lenses when 1st unit wraps before they go out again and do some more thorough testing, but I'm also not exactly sure what I should try ....

 

Humidity doesn't affect the materials we might be concerned with -- aluminum, glass, steel, etc. It has a big effect on things like wood, which is why you always store the material for a hardwood floor in the room you're doing for a few days before you install it. Generally, things expand with higher humidity -- the water from the humidity fills up pores and pushes outward.

 

Right, a constant flange error would show up as bigger distance errors on wide lenses, and smaller errors on long lenses. On simple lenses, you can see that the long ones move farther when you turn the focus ring.

 

What I'd do is call Clairmont and see if they'd undertake to check everything out for you, and put it right. Theirs is probably the best optical shop in town.

 

A random error of about three thousandths sounds to me like something is just a tiny bit loose somewhere. You might try just pushing and pulling on the front of the lens while it's mounted on the camera. You'd probably not be able to feel a movement that small with your fingers, but if you push in and test, then pull out and test, you might find that the focus error changes. Try a wide lens first, as it'll reveal the difference better.

 

 

 

-- J.S.

[Satsuki Murashige]

Update: I just finished ACing a short with Super Speeds and the 18-85 Red zoom. No flange issues at all, I never even had to set it at prep - it was spot on and remained that way throughout the shoot. Heavenly!

 

Now I think it could be a mount build quality issue on the camera side. Some of the mounts I've worked with are awful. If you've opened one up, you'll see that the threads inside are plastic and can easily skip a tooth. When you tighten the bolts on the flange adjustment, the flange consistently shifts - without a collimator it is impossible to tell by how much. Someone should really make an aftermarket PL mount with a better build quality.

 

The other thing I would check is to see how clean the flanges on the mount and the lenses are. If there's dirt build-up, that could be throwing the FFD off.

[Tim Carroll]

Update: I just finished ACing a short with Super Speeds and the 18-85 Red zoom. No flange issues at all, I never even had to set it at prep - it was spot on and remained that way throughout the shoot. Heavenly!

 

Now I think it could be a mount build quality issue on the camera side. Some of the mounts I've worked with are awful. If you've opened one up, you'll see that the threads inside are plastic and can easily skip a tooth. When you tighten the bolts on the flange adjustment, the flange consistently shifts - without a collimator it is impossible to tell by how much. Someone should really make an aftermarket PL mount with a better build quality.

 

The other thing I would check is to see how clean the flanges on the mount and the lenses are. If there's dirt build-up, that could be throwing the FFD off.

 

I have not worked much with the RED camera but have looked them over at a rental house here in town where I help out with their lenses from time to time. I would not suspect the coolness and humidity was changing anything in the lenses, I would suspect that the coolness and humidity is changing things in the RED PL mount.

 

As previously mentioned, an expansion or contraction of a few thousandths of an inch can cause the issues you were seeing. With all the different materials used in not only the RED PL lens mount, but the materials holding the sensor in place and everything that supports the sensor in relationship to the mount, I could see them easily shrinking or expanding by a few thousandths of an inch with temperature and humidity changes. Many of the plastics, especially black delrin, expand and contract considerably with temperature change and humidity change.

 

The lenses in question use mainly stainless steel, aluminum, and brass to hold the glass elements in relation to the stainless steel PL mount. None of those materials is that sensitive to temperature and humidity changes when compared to the materials in the RED.

 

Best,

-Tim

[Chris Keth]

The other thing I would check is to see how clean the flanges on the mount and the lenses are. If there's dirt build-up, that could be throwing the FFD off.

 

This is a good point and reminds me of an episode I had a while back. Some trickster put a square of scotch tape on the lens mount when the flanges on the lens sit. It threw off the FFD noticeably and was invisible to a quick or casual glance. If I ever find out who put that tape there, I'll shake their hand while I wring their neck! :lol:

[Felipe Perez-Burchard]

This is a good point and reminds me of an episode I had a while back. Some trickster put a square of scotch tape on the lens mount when the flanges on the lens sit. It threw off the FFD noticeably and was invisible to a quick or casual glance. If I ever find out who put that tape there, I'll shake their hand while I wring their neck! :lol:

 

Thanks for the info/ advice... I just hope I get my hands on the camera before it goes out again; Clairmont is a bit far off and since the camera belongs to one party and the lenses to another party, it might be hard to get them in synch again.

 

But you guys just triggered a flashback to my very first RED shoot, when it first came out and there were only 4 of them ; At some point in the day we were going to put a net on the back of the lens and it threw off the focus even though it was perfectly and neatly put on so we went without it.

That flange tolerance is just on the margin...

 

Thanks again.

[Guest Robert Skates]

While we are on the subject of the RED PL mount. The RED PL mount has issues with some Century made PL lens mounts. It has to do with the 90 degree angle where the flange portion of the PL mount (the four wings that the camera's PL mount locks around) meets the part that extends into the camera's PL mount (body). On Zeiss PL mounts that 90 degree angle has a sharp corner. On Century made PL lens mounts that corner has a chamfer. ARRI PL camera mounts have a sizable chamfer. The RED PL mount has less of a chamfer. This causes many Century PL mounts to not seat properly on some RED mounts. It depends how tight the locking ring is. If the lens does lock in it will still be approx. 1-2mm further away from the focal plane than it should be. Very bad!

 

Has anyone else encountered this. This happens with some long Century converted primes. Century 1.4 & 2X extenders.

[Nathan Armstong]

So the first time we saw this issue we were shooting in some damp tunnels in Guanajuato and half-way during the shoot the 1st ACs marks didn't work, even on an 18mm (in fact it was most present in the wide lenses) something that was about 30' away he had to focus to about 10' in order to get focus (they were working fine at the beginning of the day) and on the long lenses close to infinity wouldn't seem sharp.

 

Is this a known issue with high-speeds?

has anyone seen this before?

Im sure its collimation of the lenses, but it happened on three different sets, and the main camera in 1st unit, I hear doesn't have the issue (havent seen myself)...

This is very puzzling and mysterious, if anyone can shed some light on the matter I'd be very grateful.

 

I have had this same problem with two different sets of Zeiss high-speeds on the red. The 18mm always wants a different back-focus than the tighter lenses.

 

I read somewhere on reduser that the protective glass in front of the sensor adds a slight bit of refraction, which particularly affects the wide lenses because they have a shallow back-focus. Does this add to up to everyone else? This is the theory I am currently going with. There is no protective glass in front of the gate on normal 35mm Cameras, so I am assuming that is why few people have considered this issue.

[Stephen Williams]

I have had this same problem with two different sets of Zeiss high-speeds on the red. The 18mm always wants a different back-focus than the tighter lenses.

 

I read somewhere on reduser that the protective glass in front of the sensor adds a slight bit of refraction, which particularly affects the wide lenses because they have a shallow back-focus. Does this add to up to everyone else? This is the theory I am currently going with. There is no protective glass in front of the gate on normal 35mm Cameras, so I am assuming that is why few people have considered this issue.

 

 

Hi,

 

The wideest lenses are the most critical, thats why you see errors with an 18 & not others in the set. The mount, the lens or both could be off.

 

Stephen