What is the 5-link movement in Arri (and Moviecam) cameras?

Gergo Gosztolya · October 28

I know that Moviecam initially introduced a 5-link movement system, which was advantageous due to its quietness compared to other movement systems (I'm not sure, but maybe they're four bar linkages?), and that Arri implemented this exact system in their Arriflex 535, 435, 235, and Arricam ST and LT cameras, but does anyone have information on what it actually is?

Because I couldn't find any documentation on it other than every Arri user manual including it as a buzzword. So how is this system different from the other systems in Mitchell and Panavision cameras, and how does it achieve a quieter operation? Is there even a single photo showing the mechanism? (best I found was https://www.youtube.com/watch?v=DT9_2Zp0Euc, timestamp 1:31)

Thanks in advance for the help!

Tyler Purcell · October 28

The multi-link movement has been around for a long time, Arri developed it before Bauer (Moviecam) but what Bauer figured out was something Eclair figured out WAY before.

The original Arri multi-link pulldown claw goes into the perf, then goes down, then goes out of the perf.

Bauer's big invention was the pulldown claw going into the perf AND DOWN at the same time. At the end of the stroke, it's moving and pulling out at the same time. This helps to reduce noise considerably.

Gergo Gosztolya · October 28

Just sharing this for future reference:

I've done some patent digging, and thought I'd share what I found to save anyone else some time if they want to understand how these kinds of movement systems work.
Here are the relevant patents I found (in chronological order):

US4402581A (1980, Fritz Bauer)
this was published before the release of the Moviecam Super/SuperAmerica which have curious "Patents pending" inscriptions on their movement modules.

US5225860A (1989, Otto Blaschek, ARRI)
if you look at the YouTube video I linked earlier, FIG 3 looks pretty similar to it (but don't quote me on that).

US20050206845A1 (2003, Walter Trauninger, ARRI)

Using these resources I think you can get a pretty good idea of how such a mechanism works.
If I missed something, or got something wrong, feel free to let me know!

Tyler Purcell · October 30

Here is a better explanation, mind my mistakes, I correct myself as I'm off the cuff talking.

https://www.dropbox.com/scl/fi/to86uh73xwiw9q17rkk42/Multilink-discussion-movement.mov?rlkey=0dlaze54jtof25hpuelnbz4g8&dl=0

Gergo Gosztolya · October 30

Thank you for the demonstration! Just to make sure I understand correctly, the point is that the claw tip goes almost fully into the sprocket before starting the downwards movement, and only comes out of the sprocket when the movement is finished.

This seems to align fairly well (though not fully) with the second patent I mentioned (US5225860A), which mentions "the tips of said claw penetrating shallowly yet sufficiently into the film perforations and, at the end of the stroke, being pulled in the shortest possible time nearly perpendicularly out of the film perforations."

So I'm curious about the specific mechanism in your video. What camera model was that from? I'm trying to piece together a timeline: did Arri initially believe deeper claw penetration was preferable and later move toward shallower engagement, or did the camera in your demonstration predate the patent, with later models adopting shallower penetration as the optimal design?

Also, you mentioned having worked on Arricams, so I'm also interested in your experience with them: did they use a similar deep initial penetration to what you demonstrated, or a shallower one closer to the patent's description?

Thanks again for the help.

Tyler Purcell · October 30

5 hours ago, Gergo Gosztolya said:

This seems to align fairly well (though not fully) with the second patent I mentioned (US5225860A), which mentions "the tips of said claw penetrating shallowly yet sufficiently into the film perforations and, at the end of the stroke, being pulled in the shortest possible time nearly perpendicularly out of the film perforations."

You have to think about what makes the noise when the pulldown is happening.

1) The film sliding on the gate (even though there is a gap) makes noise.
2) The pull down claw going into the perforation hole makes noise.
3) The registration pins going into the perforations make noise.

So what are the things you can do to help change those things.

1) You can have as little components touching the film as possible. In the case of the commercial quiet cameras, they generally have a small single spring loaded section of the pressure plate, which is the only thing holding the film against the gate, only in the area where the film is being exposed (aperture region). The film is then sorta held in place by the sides, but with a very small gap (float) which is adjusted via shims in the later model movements like Moviecam and Arricam.

2) You can adjust the pull down so instead of it going straight in and down, it comes in at a slight angle and starts to move down as it goes in. This means, you aren't sliding on the perforation as the film is standing still. This is where the "flick" noise comes from, the act of the pulldown claw going into the hole and the film flexing as a consequence because even the most precise perforation cutters and emulsion, still has a slight deviance that the motion picture camera needs to adjust for. The measurement from the top of the stroke to the bottom of the stroke is called "pitch" and that helps combat slight shrinkage and perforation issues. On 35mm cameras this is a more critical adjustment due to the system using 2 pull-down claws on these cameras. They also changed the shape of the pulldown claw pretty radically, it's rounded off at the tip, so when it first engages with the perforation, it doesn't quite fill it right away, which further helps to reduce noise. Some cameras like the Aaton 35III and Aaton Penelope, both use a duck beak pulldown claw, it's a single flat surface which comes up and into the perforation hole at an angle. It relies on the film being "thrown" down by the movement in order to align the next perforation hole for the pulldown since those cameras don't have registration pins. It's why a camera like the Penelope can be so quiet and so small, it is the smallest quiet 35mm camera ever made. Arri never adopted that pulldown system because they were all about registration and of course, they wanted to own (patent wise) what they sold.

3) Registration pins and overlap are both very interesting. So as the pulldown pulls the film down, the registration pins recess into the movement as depicted in the video. The overlap is where the pulldown claw and the registration pins are both engaged, for a very short period of time. This is like a "handoff" to prevent the film from shifting its position when the pulldown engages. Also remember, the shutter hasn't fully closed when the pulldown claws are starting to engage, so any shift in the film would be a problem. Enter the rounded off registration pins, one filling the perforation vertically and one horizontally. This helps greatly with the sound level on this hand off, especially during the end of the pulldown stroke where the pins are going into place. On the older movement, the pins can really get stuck on the film, but on the newer designs, that's not the case at all. The film has a bit more wiggle room until the pins are more extended. This was a simple fix, but it does reduce the sound considerably.

5 hours ago, Gergo Gosztolya said:

So I'm curious about the specific mechanism in your video. What camera model was that from? I'm trying to piece together a timeline: did Arri initially believe deeper claw penetration was preferable and later move toward shallower engagement, or did the camera in your demonstration predate the patent, with later models adopting shallower penetration as the optimal design?

They're both BL's, first one is Arri's older design from the BL1 - 4 (1970s) and the Multilink is from the BL4S (1989). I have lots of videos with the later generation Arricam and Moviecam movements, but I can't see to find them. I must have put them in a folder on my old computer and named them wrong or something. I will dig and post them if I can find them. I've rebuilt many Moviecam movements and they are a real thing of beauty, just a lovely design.

Not sure about the patents, I just fix and use cameras.

5 hours ago, Gergo Gosztolya said:

Also, you mentioned having worked on Arricams, so I'm also interested in your experience with them: did they use a similar deep initial penetration to what you demonstrated, or a shallower one closer to the patent's description?

The Arricam's movement is very similar in function to the BL4S one I demonstrated (the 2nd multilink one).

There are some minor tweaks to the function tho; more of an angle of the pulldown going into the perf being the main one. I'm honestly not sure if the penetration is less, that's a good question. If the cameras come in and they have issues with noise, it's generally the movement not being square with the gate, so we re-shim the pressure plate on the movement to compensate. This quiets them down right away and gets people back on the shoots. So I haven't really measured the penetration because it's not a necessary thing to really work on unless you dismantle the movement for some reason. Generally owners of Arricam's are rental houses, so the cameras are lubricated all the time, so they're usually in good working order. I've worked on a few ceased Moviecam's tho, mostly Super America's so I have taken apart that movement a few times, but they go back together again as they came apart, there really isn't much of an adjustment available for penetration so I haven't needed to measure it. I actually haven't rebuilt a BL movement, most of the BL work is lubrication, belts, electronics, optics stuff like that. I have way more experience in movement rebuilds with older Arri cameras like the 2C and 3, which always seem to need work and of course the Aaton's which are way easier to rebuild.