I have seen this done several times on Nature on PBS and a few other programs. You have a really nice time lapse with a storm moving in, or at night with the stars crossing the sky and there is a slow pan or tilt with the camera. Obviously this is done with some kind of motorized unit because the movement is ocurring during the timelapse, but I have been unable to find anything online. Can someone enlighten me as to what type of device this is and where they are available? Has anyone done this that can speak of their experience?

Many thanks,
David

Do a web search under "Motion Control Rig" and you'll find what you're looking for

Thanks Paul! It seems that the best way to go is to do a series of digital still photographs stitched together into a Quicktime. Then you have high-def resolution as well.

Harbortronics do some low cost timelapse controllers for Nikon stills cameras as well as a motorised pan head.
http://www.harbortronics.com/

Thanks Alister. I also found these guys http://www.bmumford.com/photo/camctlr.html. Seems like a good option as well especially since the motorized head will ramp up and down making a more natural pan.

a cheap alterative is simply applying something to the extent of the ken burns effect to an HD clip.

David,
some years ago I made a pan timelapsed take in 35mm film. We made a graduated carboard ring and attached a piece of wire (as a pointer) on the head of the tripod. Then using a stopwatch we made the pan and the timing for each fotogram (we shoot in bulb). The result was awesome.
I know, is not a very advanced technique but for one or two takes will do if you have to wait some seconds between each fotogram.

Another approach might be to use a makeshift "hourglass" - cap a piece of pvc pipe on the bottom and drill a hole in the cap; size of hole determines rate of flow of sand. Place a piece of tape over the hole. Fill pipe with sand and set in a support that keeps it in place vertically. Place a pan under the pipe to catch sand for reuse. Place a weight in the top of the pipe. Attach cord to weight and run over pulley above pipe that redirects force to pan control arm on head. Start filming and remove the tape to allow the sand to flow.

A great idea, Jim!

motorized though is to put a little electric motor and gear the bejeepers out of it to get it to turn painfully slowly when the force is applied...attach motor and gears to a lazy susan.

I really like the sand timer idea though...that's really cool!

Martin and Jim,

Those are both great ideas. Jim, I take it you have actually made one of these? How do you determine the size of the hole for the sand? Sounds like trial and error to get it to move at a specific rate unless there is a formula (I hate math so algebra scares me). To elaborate on Martin's idea I think a protractor mounted to the tripod head would work well. Then you could pan by specific degrees. One way or another I am going to give this a shot in few months, I'll post some video then. Thanks for everyone's great ideas!

David

for the sand method...you could have a selection of end caps with differently sized holes for the bottom of the tube that screw on. You could test out the speed of each of the end caps by marking a fill line on the tube for the sand and timing the rate of the descent...or to get more accurate measure for these purposes, the angle of change for your lazy susan...I would recommend you put another pulley on the other side and a lighter weight to provide some tension on the string that goes to the lazy susan. Cut a groove in the outside of the lazy susan so the string won't jump off and line it with a large rubber binder for traction.

David, I haven't built one yet. Calculating the flow rate should be pretty simple. First you need to know the vertical distance the weight must travel and the time you need it to travel in, call those values H and T respectively. Next determine the volume in the tube for H. If D is the interior diameter of the tube and R is the radius D/2, then the volume V is pi * R * R * H cubic inches (substitute your favorite unit of measure). The flow rate needed is V/T. For testing just take a fraction of T and adjust flow rate until you get a corresponding fraction of V.

I originally came up with the idea to implement an extremely slow dolly tracking movement, so it seems to have a few possible uses.

[edit] Also, it occurred to me that the rate of travel can be made to change by varying the diameter of the tube. Some calculus might be involved for complex rates, but it's doable.

The sand idea sounds good - but my concern about it is I don't think the top of the sand will descend at a constant, smooth rate. With a hole in the center of the cap, the sand will 'funnel' toward the center of the pipe. This will probably result in periodic mini 'avalanches' and collapses of the sand that would cause the weight to jump downward. Multiple smaller holes (kind of like a salt shaker) might help, but I still think it would be hard to prevent the sand from periodically 'avalanching' if the flow wasn't perfectly even.

A clock motor/gearbox might also work. This would be an interesting project to try to solve over a weekend.

EDIT: Instead of sand, how about using water and a fishing float and sinker? With a valve you could vary the flow rate. This would also eliminate the avalanche issue.

The mechanical ideas sound like a lot of fun to try, but if you want to get it done clean and simple, just shoot the time lapse with a digital SLR using a wide angle (or even fisheye) lense, then do a pan & scan in post.

Nick.

Thanks for that one. It never entered my head to dream that up on my own.

I have my doubts that the sand or water device would work unless you used a pulley and pan head with an extremely low friction rating. I can see anything turning that slowly having big problems with 'slip-and-stick', IE, jerky motion. Come to think of it, that could be a major obstacle with any device trying to move a heavy camera that slowly. You'd have a lot of starting friction in the pan head to overcome. Any driving device you connected would tend to repeatedly build up tension before overcoming the starting friction, resulting in tiny bursts of movement rather than a smooth, flowing pan. At a normal panning speed this isn't a problem, since once you overcome the starting friction, the 'moving' friction comes in to play. At those extremely slow speeds, you would really never get the head into 'moving' friction. To overcome this, you would need some sort of zero-friction pan head, like a block of wood floating in water.

Mounting the camera on a large lazy susan-like disc and connecting a clock motor gearbox to the outside edge might work. Sounds like a fun project for next weekend!

I don't think pulley stiction should be much of an issue, you can easily find one with low mass and the tension to pull the arm can be quite high. Stiction of the head is definitely an issue, but I think the issue applies as well to the motorized drive - both solutions need to apply sufficient force to move the head at a constant rate and a decent head is assumed. It's possible to choose a weight sufficient to do the job with the dropping weight method.

I really like the water clock approach which could be highly adjustable. The motorized solution would of course be easier to use, a stepper motor and timer control is the usual approach in telescope tracking solutions. There is also a device called the "barn door drive" which is a poor man's astrophotography solution. It uses two pieces of wood hinged at one end and a threaded rod turned at a constant rate (e.g. quarter turn every 15 min) to adjust separation of the free ends (with hinge axis pointing at North Star).

Check out my solution... very inexpensive too: http://www.youtube.com/watch?v=Ni89TBOTCUA