Remote follow focus

[Martin Amada]

Tom I think you are spot on with everything you wrote. Prototypes never make it to market in many industries because once the costs of tooling up for mass producton are calculated and compared to the market share available, the numbers don't add up. And, people realize how much time and effort (not to mention capital) before you actually start to make money. I've been there and done that many years ago. Not to mention the fact that we are camera ops and filmmakers and presumably most of us would like to spend our time making movies.

What further complicates this particular product is that everyone has different lenses and thus different gearing needs. But I  agree that a DIY kit in some form might make sense.

It would be great to hear from more people with M35 adaptors about what their lens rotational needs are. There may be some acceptable  trade-offs here.

For example if a servo with a 180 degree range can acheive good accuracy and I replace the existing gear on my servo with a large one (which I  am now doing) that gives me a roughly 1:1 ratio between the servo and lens gears. Now I have 180 degrees of barrel rotation. That may be enough for 70% of the shots I need. If my servo can go 300 degrees which is what my robotics guy is working on, that may be enough for 95% of the shots I need.

Martin

[Martin Amada]

Ever looked closely at the gears that Redrock sells with their follow focus - the ones that will adjust to various size lens circumferences? It doesn't look to me like they can travel more than a full revolution. Do they ever have to?

Martin

[Martin Amada]

Below is a pic of the big gear I had made to replace the small one on the servo. Here's how much the barrels of my lenses rotate from one extremity to the other.

Zeiss 28mm - 120 degrees
Nikkor 50mm - 120 degrees
Nikkor 85mm - 270 degrees

Looks to me like 180 degrees of servo rotation with a big gear will be enough for the vast majority of racks I will want (and be able to do) and that 300 degrees will cover them all. At least with these three lenses.

Here's another reason why I believe the demand for an affordable remote follow focus will exceed that for the manual one: I have a good Manfrotto tripod but its load capacity ain't that great. When  I grab hold of the knob of my manual ff, I have to be very careful not to ruin the shot by shaking the camera. I could sandbag the sticks and probably will in many cases, but the fact remains the big boys with their much heavier cameras and tripods don't have to worry as much about this. But replace the manual ff with a remote one - problem solved.

Martin

[David L. Holmes]

Hey Guys,

Good info here   180 degrees of movement is doable without modifying a servo.  Anything more then that will require the use of a Dremel tool to modify a servo for continual movement, but then you will lose your positioning pot (which is what a "Servo" is, if you remove the pot you have a motor with a gear box).

One thing about DIY projects, they look like DIY slap together projects until you have the time to refine the look after many different prototypes.  I like the simplicity of the giant gear, but it looks quite frankensteinish   Once I can get a decent little stepper motor, I can experiment with the idea of a smaller gear wheel turning at a 3:1 or more ratio to the input pot.  The only concern would be repeatability accuracy.  On a good plus note, the stepper would be a lot quieter then the servo.

How about starting a new thread with our combined efforts and ideas for one of these (or both) issues?

Dave

[Martin Amada]

By all means start a new thread Dave.

I think you are on the right track looking at using a stepper motor.

Have you tried a unmodified servo with 180 degrees and have you found it to be accurate with a standard R/C transmittor, and with no oscillation after it achieves position? If so it may not be necessary to go much further, at least not for my needs.
When I put this question to the tech people at Servocity this was the response:

"Basically all servos have 254 points of reference.  If you spread those points out over 180 degrees, the servo will have a point of reference every .7 degrees.  If the servo is left stock (90 degrees), it will be twice as precise.  On the 785 servo, [that's the one I'm using now] those 254 points are spread out over 3.5 rotations.  Even though it is a very low cost servo, the 425bb http://servocity.com/html/hs-425bb_super_sport_bb.html might work for your application.  Just make sure that it will easily turn your knob."

The servo he recommends comes 90 degrees stock, but can be modded at the factory for 180 degrees. Now run it with this off-the-shelf programmable controller (remote though not wireless) and if it's accurate that is a lot of bang for the buck.

http://servocity.com/html/servo_recorder_playback_contro.html

Martin

Martin

[David L. Holmes]

They probably mean 90 degrees CW and CCW, or 180 degrees total travel.  So far I have achieved 180 degrees of rotation with very little if any oscillation.  There are other people who have programed in redundant feedback error sampling that have improved on the oscillation problem.  Instead of just reading the pot once, they read the pot 3 times, then divide the results by 3 for an averaging effect.  I'll try to post a short demo video later on and see if this could meet your needs.

Dave

[Martin Amada]

Dave I assume they are doing this using the Arduino board, putting it between the servo receiver and the servo?

May I ask what servo you are using to get 180 degrees and little if any oscillation? What transmittor are you using?

Martin

[Peter Chung]

I made a follow focus using a 555 timer and can get a little more than 180 degrees. I posted a short video on YouTube a while back if you want to take a look: http://www.youtube.com/watch?v=PSlcbi0GnVw&fmt=18

You can trim the range for shorter range lenses as well as switch rotation direction.

As for 254 reference points, I have no idea where the tech got that information.

[David L. Holmes]

The Arduino has servo control capability built in (called a servo library).  Therefore all you have to do is have an analog input (the potentiometer) and then a digital output to the servo.  Then in the program you map the input to the servo output, turn it on and your good to go!  

Simple, but maybe not as refined as we would like.  Only further experimenting will tell.  I have to go to work now, but later on I will try to demo what I'm talking about.

My servo is an unmodified Hitec hobby servo.  

So far this is a wired solution, not a wireless.  More on that to come...

Dave

[Martin Amada]

Peter, that looks very good from what I can see in the video. Can you control the speed? Would you like to share a bit more information? Like what servo you used and what a 555 timer is and does? Have you considered adding programmable presets?

Are you planning to use it in one of your own rigs, or are your aspirations strictly commercial?

Martin

[Peter Chung]

Thanks, Martin. Most servos are "speed controllable" meaning that the slower you turn the control knob, the slower the servo turns.

You can use just about any servo. The one in the demo is one of the most inexpensive Hitec servos I got from the hobby store. A 555 timer is a general purpose timer IC that is used for lots of projects. There are programmable presets via potentiometers.

Hope that helps,
Peter

[Peter Hoare]

Hi,

My name is Peter Hoare and I am the designer of the Hocus Focus wireless unit.

This is a full featured wireless focus unit, aimed at 35mm adapter and DSLR users, who use 35mm SLR lenses.

The motor mount is milled from aluminium, and includes a locking Hirose connector to connect to the receiver unit via the included Hirose cable. The motor also features a quick change gear hub, so you can change between gear pitches for cine and broadcast lenses in a matter of seconds.

RX/TX units are CNC waterjet cut, and laser engraved. The transmitter has an ARRI sized follow focus wheel, with a dry wipe marking ring and pointer. There are two programmable stops on the transmitter, designed to limit motor movement so you cannot over turn and damage the lens or the motor. This can also be used to correctly use newer Nikon lenses that have an infinitely rotating focus ring.

The transmitter unit has 10 user changeable channels, and it has a belt clip and a lanyard point and runs on a 9v battery for approx 10 hours and the RX unit can be powered off a Sony NP (or EX style battery). It also accepts a 12v 4pin XLR power input from the sled if available. Canon, Panasonic and JVC plates are also available if preferred.

The range is around 200 meters (650 feet).

The unit is priced for stabiliser owners and builders. We have not announced final pricing yet, but it is very reasonable, and includes Transmitter, Receiver, motor, lens gear, battery plate and motor cable, all in a case with custom cut foam.

We will be announcing more details soon, but for updates please see the website here www.hocusproducts.com and register on the mailing list. Its safe, we dont share or release your details, it is purely for sending out Hocus Focus updates every now and again.

Any questions, drop me an email, pete<at>hocusproducts<dot>com.

[David L. Holmes]

Thanks Pete,

Nice looking product you have there.  Since this is a DIY forum, how about telling a little about the insides of this gadget?  What kind of motor are you using on the camera?  Servo or Stepper?  How about the wireless?  What systems did you test this unit out on?  Do you have any pics of the finished system mounted on a camera?

Thanks,
Dave

[Martin Amada]

Pete, I second what Dave wrote. This looks like a very well thought out product, but it would be helpful to have more information.

Also, if your price is going to be "very reasonable" I think it would be in your interest to let us know just how reasonable. I can't speak for folks like Dave and Peter who seem to have the electronics skills needed to go as far as they want with their projects, but if I could buy your system or a DIY version of it for what I consider to be a "reasonable price" I'd probably jump at it.

That's because sometimes we build to save money too. Redrock was smart. I bought their DIY kit for $100 and built an adaptor that I don't think differs much from their production model costing more than 10 times that amount. Now I'm going to buy their $400 achromat because I can't make that myself. They realized that only a very small percentage of their customer base would choose the DIY route, and that providing the DIY option would help build their customer base. The proof is that they continue to offer the DIY kit as an alternative to buying their adaptor. You may want to think about doing something similar. People on this site can rig their own enclosures, markable dials, and rail mounting systems.

You also may want to think about offering your product with and without lens gears. Some people have them already.  I had my own custom made so that they would fit my lenses perfectly, and they weren't that expensive.

Martin

[David L. Holmes]

Ok, so here is my thread that is going to start this Arduino based Follow Focus going.  Let's see what we can do 

Dave

http://hbsboard.com/index.php/topic,3882.0.html