Jump to content
ODP

Car Mounted Steadicam

Recommended Posts

I am designing mounting system to adapt a steady cam to a car.

Thus far, we are thinking the "Crawler" transmission would be most useful, in that it is a slow 30:1 ratio, and would be able to navigate over most terrain.

We haven't done this yet, and are still open to suggestion.

Here is our call for help:

OPEN DOME PRODUCTIONS IS REQUESTING ASSISTANCE DESIGNING / SELECTING RC FILMING VEHICLE

We are in search of a larger RC Truck, similar to the Killer Krawler 1/6 kit, BUT are open to all suggestions that fit our criteria as follows:

Purpose of this vehicle is: To mount a professional HD Camera system and slowly traverse the woods and urban environments for stable fisheye videography.

REQUIREMENTS:

Smooth suspension is our top key component, so the filming comes out silky smooth when in motion. Should float like a luxury car on any terrain, handling woods or asphalt

Should carry a payload of up to 10 pounds with no problem, the more the merrier.

Electric and quiet

Low gear ratio for slow, controlled movement

Larger, soft tires to absorb some of the vibration.

We would like the motors, the radio (3 channel), 2 x performance batteries, charger, speed control included.

The plastic body is not important. We will remove it and mount a camera system. The stability and smoothness of ride is top priority.

OPTIONS:

We are also interested in attaching a gyroscopic stabilizer to the vehicle for ultimate smooth ride. Please advise if you know of an ideal toolset for that.

An interchangeable gear ratio may be nice, enabling us to go from ULTRA-SLOW to MEDIUM speed movement.

A controller with adjustable sensitivity (min/max) levels may be helpful in obtaining very constant movement.

A small ccd video feedback camera is an option, but not a requirement.

CONTACT US:

Purchasing Timeframe: ASAP! October 2009

BUDGET: Under $1,000 - Negotiable

Contact Person: Dave Walen anamoglam-AT-gmail.com

Please respond with any details, prices, suggestions, questions, etc... We're looking for any help we can find.

post-5729-12568228555_thumb.jpg

Share this post


Link to post
Share on other sites

I doubt highly, that you will be able to build such a vehicle for less than $1,000. I did quite a bit of set work and EM model work for ILM some years ago. What you are asking, is a complex piece of machinery. First, it will have to have some weight to it, to reduce 'bouncing' and allow stabilizers to react more smoothly. You would also need to use a completely belt driven primary reduction drive, to reduce noise and vibration. Use belts and appropriate stacked pulley ratios to achieve the top end speed and torque that you require. Use a brushless out-runner motor to drive it. Encase the motor and gearbox in a sound insulated casing and rubber isolator mount the drive to the chassis to prevent any vibrations from causing vibration to travel to the steady rest assembly. The suspension should be completely independent, not a solid axle type like your picture shows. If you want all terrain capability, independent suspension is the only way to go. Otherwise, if one wheel goes down, one will go up. You will upset the balance and attitude of the entire chassis. You would wind up with three wheels on the ground instead of four, disrupting the balance and thus stability of the mount. Of course, four wheel drive. In your four wheel drive system, use belts where you can. Where you cannot, use helical gears, not spur gears. This will run smoother and with less tooth impact, causing less noise and vibration not to mention heat. We are talking serious money here.

Honestly, using an RC vehicle will pose you many problems in filming. There will be no real escape from bumping around, no matter how well you design and build the vehicle. Stabilization electronics development and components alone, will run you several thousand dollars. Especially for use in unusual terrain. If filming can be done in shorter takes, you should use a rail system. A simple mono-rail or dual rail made from pipe and flanged bearings would hold a platform that would ride the rails with far greater stability. In fact, the filming would be as smooth as riding on a smooth floor. With a properly designed aluminum trestle or saw horse type mount, the rails could be easily transported from point to point, to film as you go, along a very long path. Another idea would be to use several tripods, with adjustable legs, to hold a single, round rail. Pivot the camera mount plate sideways and mount the rail mounting bracket to it. You could then make a vertical camera mount post, that is counter weighted to hold the camera in a fully vertical position, as the bearings ride along the rail. You could easily attach the rail to the tripod mount plate with the use of a simple bolt, that would hold the rail (horizontally, in reference to the mount plate on the tripod) a few inches from the tripod mount plate. This way, the camera could be powered by a belt driven drive that is much smaller, less expensive and far more stable. The camera drive mechanism could consist of the three bearings to ride the rail (rubber to reduce noise), and a rubber hobby aircraft wheel or similar, driven by your belt driven gear reduction. The rubber drive wheel would simply ride the rail. Use a rubber pinch wheel (spring loaded) that is free running, to help traction of the drive wheel on the rail. If you use longer lengths, of smaller diameter plastic tubing, you could bend around turns much easier than using aluminum or steel tubing. You could couple tubing ends with internal couplers that simply plug together for easy assembly, disassembly and transport. This will enable you to film your takes in various locations, with arcing turns, dips, etc., to follow terrain.

I hope these ideas have helped you in some way. Honestly though, an RC vehicle will not give you a stable filming platform, at any speed, for so little money, in all terrain conditions. You really should use a cable, or rail system to avoid the terrain altogether.

Share this post


Link to post
Share on other sites

Indeed. If you want professional level results, you should start by adding at least a zero to your budget, and a few months of development...

Share this post


Link to post
Share on other sites

Indeed. If you want professional level results, you should start by adding at least a zero to your budget, and a few months of development...

I think your point about belt drive, and trying to silence the transmission of the crawler/car/truck is great.

as to the cost, well, we are trying. It looks like just the car will run about 1400. and paying someone to adapt the frame, and make mounts for a steadi cam could be were the cost cutting is going to be done... ie, it may be ourselves tooling away the day.

The idea is to place the vehicle on site, and be able to remotely control it, not to catch running scenes, over rough terrain.

it could be on smooth grass, or over gravel. or a forest floor, littered with the occasional branch. In fact, the vehicle may only be making a move once every 4 seconds, to create time lapses...

Share this post


Link to post
Share on other sites

This does clear things up quite a bit regarding the use and environment of the vehicle. It does make engineering it much easier as well. I do believe that machining your own parts would be the most economical way to do things. Prototype machining or one-off design and machining costs a fortune, if you have to pay someone else to do it. In fact, I have my own machine shop for this very reason. I design and prototype various devices. If I had to pay another company to do it, it would be far beyond my budget. Paying a company to do it is only practical when such prototyping and development leads to a product line where thousands are expected to be manufactured and sold.

Certainly use belts instead of gears. You can accomplish reductions such that you would need and have an almost whisper quiet drive system. Coils in brushless motors can whine, so be sure to insulate the motor noise, or suppress the coil noise with a feedback audio circuit (noise canceling)which is basically just a microphone circuit, amp and speaker that picks up the motor noise and plays it right back at the motor, 180 degrees out of phase. It will help to cancel out much of the audio being emitted by the motor itself, by a relatively high percentage.

Back in the day, a belt mounted steady cam mechanism allowed us to run down a forest path with little image instability. So, with a slow crawler such as you need, for intermittent use and a steady cam mechanism, you could keep it pretty simple and cost effective. Provided you do the work yourselves through friends or associates. Certainly in any case, go with a fully independent suspension. Instead of using bevel gears to transfer power from a drive shaft (fore and aft) from the transfer case, use a belt instead. So not only can your reduction be belt driven, but so can your transfer to each wheel. Straight toothed bevel gear boxes are expensive and noisy. Not to mention the cost of helical gears!

Better still, but more expensive in some respects, is to use independent drives. One motor and belt reduction per wheel. An encoder, be it magnetic or optical, could send feedback to the driver circuit that controls all four motors. This way, you don't have to deal with differentials or posi-traction drives. A fully independent suspension, with independent motors per wheel will give you the most stability and control. But then again, at a higher level of complexity and cost. The gain from this approach would be a superior machine however. The cost would not be so much more however, provided you have a robotics enthusiast on your team to help with the motor driver circuit and programming. To keep costs way down, get two old computer mice. Each has two optical encoders in them. Use them! Just shield the diode and receiver assembly from ambient light.

I personally, would use as many aluminum extrusions as possible, to build the chassis. This will help keep costs and initial machining down to a minimum, yet offer structural integrity. I would then use one brushless out-runner motor, per wheel. An out-runner by nature of its design, is a reduction on its own. A low Kv out-runner, can then be coupled to the wheel using a timing belt. Use the smaller timing pulley on the motor and the larger on the wheel for further reduction. This will put the drives right on top of the wheels, allowing you to have optimal chassis clearance possibilities and a truly independent suspension and drive system. You could even cut costs even further, by using four, surplus stepper motors. With a microstepping drive circuit per motor, you can achieve precise movement of the vehicle. Both motors and driver circuits are readily available, either new or surplus. Coupled with a closed loop optical encoder and a robot board, you could have a very effective motorized camera platform. With stepper motors, you would not even need a belt reduction. Stepper motors can run very slowly and quietly and are fully sealed.

I would use optical encoders (shielded) to read rpm from each drive. This would then be routed to IO ports of a robot control board, which is readily available for a reasonable price. The robot board could control each motor at the same time, using the encoder inputs to know if one wheel is slipping, etc in respect to the other wheels. With an encoder on the steering and some clever programming, you could make each wheel act as though it were driven by a differential to account for differences in speed between an inside versus outside wheel in a turn, while at the same time, acting like a posi-traction drive, where none of the wheels would slip, as they unfortunately would with a common differential drive. Complete traction control. More complex electronically and a little more money, but well worth it if you are looking for something very practical for filming use in various environments/conditions. But, you will need someone to help you out with the programming and setting up the electronics of the robot board, such as interfacing it to the vehicle.

For the camera mount, I would use a 'balance' plate that is controlled by three servos. One would control tilt left/right and one would control tilt fore/aft. A gyro would detect vehicle tilt and automatically compensate by tilting the camera mount plate to accommodate changes in terrain angle, keeping the camera level. The third servo would control pan. The vehicle could potentially encounter a situation where the chassis will pivot right or left in respect to direction of travel, due to ground surface conditions, like a car sliding sideways on a snow covered road, slippery leaves or roots, sand etc. A gyro could control the pan servo, keeping the camera facing the intended direction of travel, even if the vehicle chassis has pivoted. Since each wheel is independently controlled, you could have the vehicle pivot itself out of such a rotation by controlling the direction of rotation of each wheel, much like turning a tank. This way you would not have to steer, while going forward, to correct for deviations in path. You could literally turn or rotate while in place, before moving forward again.

If you are feeling especially creative, you could use a common, hobby GPS unit, interfaced to the robot board, to coordinate stop and go travel of the vehicle. Travel to X coordinates and stop, etc. This would add a certain degree of automation to the vehicle, if such a thing is desired.

There are many, many options to such a camera mount. It all depends on your imagination, budget and utilization of as many readily available components/resources and having some clever people on your team. I am working on several products right now, so I don't have the time to help you out much more than offering ideas. Although, I do have to say your project interests me from a technical challenge point of view. I sincerely hope some of the ideas can help you out.

Share this post


Link to post
Share on other sites

In fact, the vehicle may only be making a move once every 4 seconds, to create time lapses...

Now that is a whole other thing... Do you want photo/timelapse, or full motion video? It's a VERY important question.

With your budget and the resources you seem to have, photo/timelapse is no problem. Take a shot, move, take another one, if the motion inbetween isn't all that smooth it doesn't matter as long as the vehicle moves correctly where it's supposed to go, then it's a "non-issue".

For full motion video you go a step further in complexity. Just imagine the treads you need on the tires if you want to be able to crawl over a branch, it will have to be deep enough so that it causes up-down motion when driving, already requiring serious dampening. A crawler also doesn't walk over obstacles smoothly, it will eventually pass them, but might slip and fall from the height of the obstacle a couple of times before it grabs it correctly to pass it. The dampening involved and mechanical presision needed to get something decent is scary.

From what I've seen so far in the motion imagery world (and which is also valid for many other things as soon as you get out of the consumer stuff), is that the function between "quality/possibilities" and "complexity and price" is all but linear, but can be modelled very well with 10^(quality level) magnitude orders, so with quality directly related to the number of zeroes on the price...

The common example of a camera gimbal:

Quality level 0: Attach camera, fixed: cost = $1 range (0.5-5), just a screw.

1: Manually moveable camera: Tripod head, cost = $10 range (5-50).

2: Simple motorised pan-tilt mount, using RC servos and cheap plastic parts/homemade metal ones: $100 range.

3: Basic stabilisation using R/C gyros or equivalent, giving you some degree of stabilisation that is better than nothing, but not particularly satisfying: $1000 range

4: Decent stabilisation for wide angle shots, with custom-made precision mechanics needed to get rid of slop and lack of rigidity at a level that is hard to reach with homemade stuff, along with custom drive and electronics: $10k range

5: Professional level system capable of very good results in most conditions, but not yet "perfect" (Wescam-style): $100k range

6: Excellent cinema-level, extremely smooth system, even at long focal lengths, that even manages to make "average" shots look good (Cineflex-like): $1M range.

Related hints:

- Levels 0-3 can be easily achieved with off-the-shelf consumer products and some minor hacking. Levels 4 and up almost definitely will require some custom-made parts, with engineering and proper tools. Bent/drilled/fastened alu/CF bars and tubes have their limit.

- Don't expect a $2500 system to perform heaps better than a $1000 one - even if the price difference is substantial.

- Financially going for the quality level below the aimed one thinking one can improve it oneself is usually just a false idea. If you count the development on your side - time spent, the equipment you buy and end ditching because it doesn't work as predicted, etc - you usually end up with the same bill as if you had brought out the needed cash at the start, but you maybe don't even have your system working how you wanted it, and lost ages in the process.

Analyzing your first post, It sounds like you're looking for level 5 results (from the "silky smooth" and "floating" requirements) with a level 3 budget (for under $1000 you basically can just get an R/C car to strap your camera to, without many improvements), hence the first responses. Maybe trade a bit on your expectations, talk a bit with your financial dept, and settle for level 4. Unfortunately it's the one that lacks commercial offers the most (but the one that would be the best compromise for a high number of users - go figure, the main reason being that the needs in this category are often very specialised and would make it hard to design something that satisfies enough users), so find someone with experience in building such things who can help you with design/construction (not talking of a specialised company, but someone like thorn3, who seems to know what he's talking about, have experience, and also have resources to put his ideas into practice). Yes that will cost something, he won't be working for free, but it will be way less than a proper engineering bureau. And you now have the budget to give him something he can happy with, and you will have much higher chances of ending with something that fits your needs and (revised) expectations, rather then ending ditching the whole stuff and losing money and a bunch of time because you didn't match your resources and expectations and didn't manage to make something yourself that matches your idea of it (which often happens...)

But that's just my own advice :)

Edited by Kilrah

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×