Arriflex 16SR 11pin external speed control signal level?

[Vincent Wolfram]

Hello,
I'm tinkering on adding a small external lcd with footage counter and selectable speed control for my SR I, inspired by the P+S Evolution upgrade side display.
Note that these additional speeds wont have to run at perfect sync speeds for my particular use case.

If the frequencies are generated by a microcontroller like an arduino and a signal generator like the AD9833 they wont be in perfect sync. 

Online, i found the following document (PDF) that talks about a 3200 PPF signal on pin 8. I presume PPF stands for pulses per frame, which results in a 16khz to 240khz range. (5-75FPS)

Does anyone have any documentation on exactly what kind of signal the camera expect? just a square wave? what voltage level? duty cycle?


attached below is the same PDF document, hopefully preserving it in case the website shuts down in the future

Film camera accessory connectors.pdf

[Aapo Lettinen]

Would you happen to be interested in my Aaton footage counter, it is going to have Arri SR compatibility with 20 speed crystal generator and oled display with footage counters? I don't have possibility to test it with SR here but because you are in Europe you could ship it free for software upgrades if it needs some tweaking with the SR. The other customers use it with Aaton so would love to get some arri user feedback too 😊

[Steve Switaj]

Well, oddly, I'm actually pretty familiar with that particular document, because I wrote it.

It's my website. Granted, not a spectacular website by any means. I do a lot of motion control work and put up some technical information about 15 years ago so that when people called me on a weekend with a technical question I could say "go look at document XXX". I don't think I've updated it in a decade. Really should get on that.

Anyhow, I believe I got that connector information from either Rodger Reddy at Arri or from Clive Tobin. 

I was always a little suspicious of that 3200 PPF number. It seems a bit high to me, since most cameras use 200 PPF (pulses per frame), which correlates to the old 4.8KHz PilotTone frequency.

Over the years I've had a lot of experience with the 11 pin connector on 35mm cameras, and they are in fact 200 PPF, but I never had a chance to examine the signals on an SR/SRII. 

3200PPF seems like a pretty high frequency, but it is 16x the normal rate, and divide-by-16 circuits are trivially easy, so maybe there was some magic factor in the way the SR accessories generated the signal or maybe the SR's had 400 cycle encoders on their motors.

As far as I recall, all the signals worked with normal 5V TTL  levels, though some cameras will have a resistor inline with the output driver to limit short circuit current, and if you're trying to drive a circuit with a low-level input impedance this can lead to inadequate signal level.

[Steve Switaj]

Also, you mention the AD9833 so I'm going to assume you have some familiarity with digital electronics.

You might want to check out the Microchip family processors with an NCO peripheral.

It's an adder/accumulator system like the AD9833, but designed to generate a digital output instead of a sine wave. Since it's an actual peripheral inside a reasonably capable microcontroller that can do some math it makes for a really easy frequency source when you'd like all those 'odd' frequencies, like 23.976, etc.

It's mostly available in their 8-bit PIC family chips, but I think they also use it in some of the Atmel AVR family parts

appnotes here: https://www.microchip.com/en-us/products/microcontrollers/8-bit-mcus/peripherals/waveform-control/nco

 

 

[Aapo Lettinen]

Yes should be 200ppr , 5V logic level 50% duty cycle.

I have no idea where the 3200ppr comes from, it must be a misunderstanding of some sort. These cameras have some kind of "check the internal crystal frequency" output which is 70-something kHz and maybe the oddball figure comes from that. 

Using a 3200ppr signal on camera meant for 200ppr would likely make it speed up uncontrollably to close to 200fps and absolutely explode the mechanism so not a good idea to use tens-of-kHz signal as first test! 

Normally these would be tested with a function generator which has easily adjustable output. You would connect the frame pulse output of the camera to a frequency counter or oscilloscope with readouts and start with something like 500Hz signal in this case. And when getting it turning, ramp up the speed to suitable low level like 5 to 10fps and then confirming what exact fps you get from the shutter pulse output. 

Divide the frequency generator output Hz with real measured fps and you get the ppr value. Please report here what kind of figure it is but it is surely the 200ppr because practically every camera has the same value than the encoder slot count per frame shot.

If Vincent wants to sponsor my youtube tutorials I could make a video of how to get exact crystal frequencies directly from the MCU without separate generator. It is lots of work to make the tutorial so would otherwise do the video in 2026 when having more time