Re: [Frameworks] need help with dual 16mm projection

From: David Tetzlaff (email suppressed)
Date: Sun Sep 19 2010 - 08:38:58 PDT


How you could achieve this depends in part on how long you want it to
run, and whether or not there would be an attendant. If it's s loop,
I'm guessing you want it to run unattended for a long time. In which
case, I would only trust some sort of mechanical linkage.

I don't know what kind of motors projectors have, but there are two
kinds of AC motors: a standard one has it's speed determined by the
input voltage. These are subject to a good bit of variation. A
'hysterisis synchronous' motor has it's speed governed by the
frequency of the AC current. A 'crystal control' doesn't adjust
voltage, but rather the frequency clock. There are other kinds of
circuit that regulate and stabilize voltage, but I doubt that voltage
accounts for any inherent speed differences between two projectors.

If two projectors use standard AC motors, are plugged into the same
source, and one runs inherently faster than another - the difference
would be due to how efficiently each total system uses the available
voltage. In which case, the two could be more roughly equalized by
dropping the voltage on the faster one with a rheostat, but never
locked tightly together.

If the projectors use synchronous motors, they are trying to run at
the exact same speed already as the are both governed by the same AC
clock from the mains, and any variation would be due to the different
amount of mechanical drag the transport puts on the motor, or belt
slippage or some other mechanical factor.

If you had two projectors that were very close in speed, or could be
made very close in speed by voltage controlling one of them, you
should be able to find a way for an attendant to keep them running.
Basically, you'd feed the loop through the faster projector first, and
the slower one second, starting with just a little slack between the
two. As the loop ran the slack between the two would increase. The
attendant would have to watch the growing slack, and put some kind of
mechanical drag on the fast projector to slow it down momentarily
until the slack got taken up. How exactly you would do this would
depend on the projector, and might require some kind of knob to be
bolted onto one of the drive sprockets or the projector case opened so
drag could be applied to the main driveshaft. Basically, this would be
a film projection version of how Steve Reich made his early phase
shift pieces: momentarily slowing the tape decks just by putting some
drag on the capstan with his thumb.

Alternatively, if the projector speeds do respond to voltage changes,
you could put two rheostats in line with the faster one. The first
would be set to get the two projectors as close to exact speed as
possible. The second would be set to slow down the faster one just a
bit, and this would be wired with a switch that would add it or remove
it from the circuit. Thus the attendant would only deal with this
switch: in one position the slack between the projectors would grow,
in the other position it would shrink.

Assuming this sort of set-up would work with the projectors at hand,
it could theoretically be automated to remove the necessity of the
attendant. In this case you would put the slower projector first,
start with a certain amount of slack, and have the film run through a
set of rollers that would move up and down as the slack increased/
decreased. At the point of tightest allowable slack, the rising roller
would trip a switch that engages the second rheostat on the fast
projector, making it slow down so the slack increases. I think this
would work just as such, but the switch might me switching on and off
enough to make the speed change visible and annoying. If so, perhaps a
timer could be rigged in the circuit, so each time the switch is
tripped the slow-down rheostat is engaged for a certain length of
time, and then goes off. Again, the whole premise here is that your
projectors would respond to small changes in put voltage with changes
in speed, which is not necessarily the case. Also, a crude system like
this attached to the supply lines of the projectors might (or might
not) affect lamp brightness in an unacceptable way.

Mariah's original post did not specify what sort of relationship the
two projected images are meant to have in physical space:
superimposed, overlapped, side by side, on different walls (side or
opposite). Such physical requirements would determine the feasability
and difficulty of creating a mechanical hack to lock the two
projectors together.

One way or the other, I think it would be a doable hand for a someone
handy with tools and electrical controls, would take a little bit of
money for parts, and quite a lot of time and effort in experimentation
to get something that would work. But such is the nature of innovation
in the arts!

> There are two ways of doing this: electrical or mechanical. Either
> you use a common drive shaft, as was used for Dan Graham's Body
> Press film, or you have a crystal controlled voltage supply that
> locks both projectors to the same voltage. Such a system existed for
> three Elf projectors. The Arts Council of England had some and
> William Raban made a three projector time-lapse film, Thames Barrier
> (1977), using it. There's also an English filmmaker called Elizabeth
> McAlpine who has made some Super 8 films with six or so projectors
> that are mechanically connected via a series of chain drives.
>
> Nicky Hamlyn.
>
>
> On 18 Sep 2010, at 22:45, Steve Polta wrote:
>
>> This issue has been discussed here before and as far as I know
>> nothing very conclusive has come from it, i.e. no real solution has
>> ever really materialized.
>>
>> Quite a while ago I was interested in this and tried it out. I was
>> using Super-8, for the record. The problem is obviously that your
>> projector will not run at exactly the same speed, the result being
>> that, if the second projector is faster, the slack distance between
>> gets shorter and shorter until the film snaps; if the first
>> projector is faster, the slack builds up until you have a pile of
>> film between the projectors. The trick is indeed to use projectors
>> with rheostats (i.e. speed control knobs). This is part of the
>> reason I was using Super-8: the Elmo projectors I possess have such
>> knobs, which subtly increase the speed in the range of 1-2 fps. My
>> biggest problem was confusing the projectors—i.e. turning one up
>> when it should have gone down, etc.—but this seems easy to correct
>> with practice.
>>
>> Ultimately I just abandoned this work so I have nothing to show for
>> it. I think it could be potentially very interesting so I encourage
>> others to pursue this. I can't say I've seen such knobs on portable
>> 16mm projectors (but they may exist) but modification is always a
>> possibility. As precedence, I've heard of a Sharits work which may
>> have done this; also some early Luis Recorder pieces ran the same
>> film through the same projector twice (looped over itself) and were
>> very interesting.
>>
>> Good luck!
>>
>> Steve Polta
>>
>> --- On Fri, 9/17/10, D Dawson <email suppressed> wrote:
>>
>> From: D Dawson <email suppressed>
>> Subject: Re: [Frameworks] need help with dual 16mm projection
>> To: "Experimental Film Discussion List" <email suppressed
>> >
>> Date: Friday, September 17, 2010, 3:14 PM
>>
>> Depending on which projectors you are using, there may be a
>> rheostat in one that will let you slightly adjust the FPS up or
>> down... You can adjust and set this while they are running and it
>> would help you tweak the slack if they were running slightly
>> differently from one another.
>>
>> Deco
>>
>> On Fri, Sep 17, 2010 at 5:56 PM, mariah garnett <email suppressed
>> > wrote:
>> I am trying to rig up an installation involving one 16mm film loop
>> that runs through two projectors simultaneously. Does anyone have
>> any advice on how to synch the 2 projectors so that the slack
>> remains constant on both sides? I think I either need to rig up a
>> slave motor or slave one projector to the other.
>> Any advice?
>> Thanks!
>> Mariah.

_______________________________________________
FrameWorks mailing list
email suppressed
http://mailman-mail5.webfaction.com/listinfo/frameworks