Image stability & short ends

[Nicholas Kinsey]

I am a Canadian DOP and we have shot two 35mm feature films with an Arri 35BLII rebuilt by Visual Products. The mags are like new and were taken apart and the clutches were changed. The camera is in very good shape and I have had technicians check it out regularly.

 

On each feature we have had image stability problems. We shoot a total of about 65,000 feet of short ends, all Kodak stock, for each film. There are always a couple of scenes where the stability is unacceptable (probably less than 2% of the footage) The problem doesn't seem to come from a single mag (1000 and 400 footers). One day you are using a mag and everything's fine, then the next day the mag is producing jerky pictures. The pictures seem to shake from left to right, not so much vertically.

 

Perhaps someone out there is familiar with these older cameras and can explain to me and to my assistant how to avoid this kind of problem. Maybe it has to do with the loop which is too tight or something about using short ends. We have been obliged to scan these jerky images and to stabilize them digitally frame by frame, then return them to 35mm negative film. The result is very clean, but expensive and a pain in the ass.

[Dan Goulder]

Have you noticed more of a problem with certain stocks? I've occasionally noted this problem primarily with 5245 (50D) stock. The actual thickness of this stock may be slightly greater than that of faster stocks, and may interact with the gate differently. Another possibility might be inconsistency with the perforations. (I don't recall ever having this problem with any other stock.) It's hard to track down, especially when most of the footage comes out real solid. I don't know if putting an extra perf or two into the loop would help.

 

I've been experimenting with different post production remedies, so far with only partial success. Could you please share your post production remedy for this problem? Thanks.

[Bernie O'Doherty]

I do a lot of work on Russian 35mm cameras and find that the users generally run short ends to save money. I find that lateral instability can be eliminated by adjusting the spring tension on the side nubs.

Most times the tension is way off. When the camera body door closes it applies pressure to spring the spring loaded lateral pressure nubs. A few grams/sq.mm change in pressure can make an astonishing difference.

Do you think you might need to have your side/lateral distance on your aperture plate checked on the Arri?

I know the design is different from the Russian cameras, but there may be common ground with regard to exact demensionality.

I know you don't have factory steel-pitch film samples, but maybe you could simply take a few short-end stocks and overlap them over a foot or two and see if the line-up both ends.

Cheers, Bernie

[Stephen Williams]

I do a lot of work on Russian 35mm cameras and find that the users generally run short ends to save money. I find that lateral instability can be eliminated by adjusting the spring tension on the side nubs.

Most times the tension is way off. When the camera body door closes it applies pressure to spring the spring loaded lateral pressure nubs. A few grams/sq.mm change in pressure can make an astonishing difference.

Do you think you might need to have your side/lateral distance on your aperture plate checked on the Arri?

I know the design is different from the Russian cameras, but there may be common ground with regard to exact demensionality.

I know you don't have factory steel-pitch film samples, but maybe you could simply take a few short-end stocks and overlap them over a foot or two and see if the line-up both ends.

Cheers, Bernie

 

Hi,

 

I have noticed Fuji can be slightly wider than Kodak stock.

 

Stephen

[Nicholas Kinsey]

Have you noticed more of a problem with certain stocks? I've occasionally noted this problem primarily with 5245 (50D) stock. The actual thickness of this stock may be slightly greater than that of faster stocks, and may interact with the gate differently. Another possibility might be inconsistency with the perforations. (I don't recall ever having this problem with any other stock.) It's hard to track down, especially when most of the footage comes out real solid. I don't know if putting an extra perf or two into the loop would help.

 

I've been experimenting with different post production remedies, so far with only partial success. Could you please share your post production remedy for this problem? Thanks.

 

Thanks for the comment. We don't use 5245 stock, mainly 5217, 5218, 5279, 5277, etc. The shots that are giving me trouble in post are 5217 Kodak stock. So I am sure whether this explains anything.

 

Generally our short ends are all fresh Kodak stock that have been in a freezer for while. We never have problems with this material. Luckily with short ends is that you are constantly changing mags so there may be problems that you are not going to see because you are only printing good takes. I remember we shot some 10 takes one time of a 3 minute scene (270 feet). We ended up throwing out the first 8 takes (we didn't even develop the stuff) and only printing 3 takes. As I said, we are talking about maybe 2-3% of our material. We used to think it was a mag problem, but we seem to get the same problems with different mags.

 

Basically, in a feature you have the neg scanned at places like Cinebyte in Toronto. They scan it 2k and then use a stabilizer software program which stabilizes the whole sequence and then outputs back to 35mm negative. 2k is fine for medium shots and close-ups, but not good enough for a wide shot with no movement. And 4k isn't cheap. The only other way is the old optical printer.

[Nicholas Kinsey]

I do a lot of work on Russian 35mm cameras and find that the users generally run short ends to save money. I find that lateral instability can be eliminated by adjusting the spring tension on the side nubs.

Most times the tension is way off. When the camera body door closes it applies pressure to spring the spring loaded lateral pressure nubs. A few grams/sq.mm change in pressure can make an astonishing difference.

Do you think you might need to have your side/lateral distance on your aperture plate checked on the Arri?

I know the design is different from the Russian cameras, but there may be common ground with regard to exact demensionality.

I know you don't have factory steel-pitch film samples, but maybe you could simply take a few short-end stocks and overlap them over a foot or two and see if the line-up both ends.

Cheers, Bernie

 

Thanks Bernie. A good comment. A large number of American independent filmmakers shoot on short ends and the Kodak material is excellent. You can stock it in a freezer for years and take it out and shoot with it, and you won't notice the difference with fresh stock. It is nothing less than amazing how well Kodak stock keeps in a freezer. But most of our material is fresh and coming off American TV series in the spring. We shoot in the summer so we get pretty fresh stock. I don't think perf size on short ends can be the problem.

 

Of course on a low budget feature in 35mm shooting on short ends, you are going to wind up with some 250-300 rolls of film instead of 70-100 rolls if you were shooting 1000 foot loads. This means more buildup of film shavings in the gate and in the mags, because you are changing film more often. Maybe this is the problem, dirt accumulating in the gate. In an Arri 35BL you don't have a side/lateral pressure plate. You have a channel, a pressure plate, and a set of two registration pins. The registration pins are supposed to hold the film steady.

 

I remember on our last shoot towards the end (33 days) when we had shot some 50,000 feet, the first assistant wasn't cleaning the gate all that much and there was an accumulation of fine dust. The film rides on this on its edges and then maybe pushes against the pressure plate from time to time causing registration instability.

[Bernie O'Doherty]

[The perfect cheapo short-end camera would have interchangable vari-width aperture plates, adjustable pin and lateral vacuum channels !!

[Leo Anthony Vale]

Have you noticed more of a problem with certain stocks? I've occasionally noted this problem primarily with 5245 (50D) stock. The actual thickness of this stock may be slightly greater than that of faster stocks, and may interact with the gate differently.

 

With B/W, it's usually the slower the stock, the thinner the emulsion. Smaller grains packed in tighter.

It seemed to be the same with Color reversal, so probably the same with color negative.

It's also one of the reasons slow stocks are sharper.

 

So perhaps a tad more wiggle room in the gate with the slower stocks

[Jason Eitelbach]

It probably is the loop. I?ve shot three shorts with out schools Arri BL-4 and have had that happen twice.

 

The first time was on a film test. One of the rolls we projected had registration problems I called everyone, Camera people said it was the printer, lab said it was the camera. Shot another registration test the day before production, no problem and no problem with any footage in the film.

 

Flash forward 18 months the camera has just come back from Arriflex factory service, we shot a registration test no problem.

 

We get the footage back and we have problems from about half way through our shoot with just one mag.

 

There is a way to set the loop too tight at the top. You must follow the guides even a half frame too tight will compromise the registration.

 

You?ll notice that the camera will be slightly louder, a little ?clickey?. I should have stopped shooting with it, my instinct told me that something was wrong, but I kept hearing that camera tech telling me ?The BL is one of the best registered cameras?.? And I thought that if it was running surely it would be registering properly, not so.

 

Luckily we were shooting a lot of short ends, as well, so by the time this problem arose we were down to our shortest rolls and changing mags quite frequently so not all the shots were ruined.

 

What I?ve found is that I love coaxial mags on 16mm. They seem to be reliable, but on the 35BL they require a lot more attention. We?ve been taking them apart about every 2000-3000 feet to clean them out. Too much emulsion and gunk builds up in that loopty loop inside the mag throat, especially if you have inexperienced AC?s who sometimes reverse the film when loading and then they don?t work smoothly and they need to thread buttery smooth to work right.

 

Maybe someone with more experience can help flesh out my hypothesis further.

 

If you can keep on top of the mags I?ve found the 35BL to be a very reliable workhorse of a camera and I think they?re great for low budget features since they?re pretty cheap to rent. I wonder if anyone has had similar problems with the 535.

 

Thanks,

je

[Dan Goulder]

With B/W, it's usually the slower the stock, the thinner the emulsion. Smaller grains packed in tighter.

It seemed to be the same with Color reversal, so probably the same with color negative.

It's also one of the reasons slow stocks are sharper.

 

So perhaps a tad more wiggle room in the gate with the slower stocks

I don't have specific numbers, but having loaded a lot a mags with both 5245 and 500T (now 5218), I can say that the (50D) stock has the textural feel of extra thickness in the emulsion. You can certainly feel the difference in the two stocks just from hand-threading them through the mags. It also appears to be a softer emulsion, as well, and tends to have a greater potential to "foul the gate" than a stock such as 5218. Since I love the look of 5245, I've continued to use it while it's still been available (although I've got a few rolls of 5201, which I'm itching to try).

[Nicholas Kinsey]

It probably is the loop. I?ve shot three shorts with out schools Arri BL-4 and have had that happen twice.

 

The first time was on a film test. One of the rolls we projected had registration problems I called everyone, Camera people said it was the printer, lab said it was the camera. Shot another registration test the day before production, no problem and no problem with any footage in the film.

 

Flash forward 18 months the camera has just come back from Arriflex factory service, we shot a registration test no problem.

 

We get the footage back and we have problems from about half way through our shoot with just one mag.

 

There is a way to set the loop too tight at the top. You must follow the guides even a half frame too tight will compromise the registration.

 

You?ll notice that the camera will be slightly louder, a little ?clickey?. I should have stopped shooting with it, my instinct told me that something was wrong, but I kept hearing that camera tech telling me ?The BL is one of the best registered cameras?.? And I thought that if it was running surely it would be registering properly, not so.

 

Luckily we were shooting a lot of short ends, as well, so by the time this problem arose we were down to our shortest rolls and changing mags quite frequently so not all the shots were ruined.

 

What I?ve found is that I love coaxial mags on 16mm. They seem to be reliable, but on the 35BL they require a lot more attention. We?ve been taking them apart about every 2000-3000 feet to clean them out. Too much emulsion and gunk builds up in that loopty loop inside the mag throat, especially if you have inexperienced AC?s who sometimes reverse the film when loading and then they don?t work smoothly and they need to thread buttery smooth to work right.

 

Maybe someone with more experience can help flesh out my hypothesis further.

 

If you can keep on top of the mags I?ve found the 35BL to be a very reliable workhorse of a camera and I think they?re great for low budget features since they?re pretty cheap to rent. I wonder if anyone has had similar problems with the 535.

 

Thanks,

je

 

Dear Jason,

 

Thanks for the comments. I have looked at the gate and pressure plate of my Arri BL again and really can't figure out how lateral instability can really occur. You have two claws feeding the film forward and two registration pins locking the film in place before exposure. You have a relatively deep channel on both sides for the claw to operate in, so my idea of a film shavings buildup lifting the side of the film is really not possible. It could be possible in 16mm with only one claw moving the film forward but not in a 35BL. On the 35BL the film is stretched across two registration pins and then is a lot of room for shavings in the channels of the pressure plate (not flat, but with vertical channels). You could have a real dust pile up and it wouldn't affect stability.

 

Your idea of the top loop seems to be quite convincing. I will run some tests over the next few months and and check it out.

 

Thanks for your input.

Nicholas