Any engineers or hobbyists recommend a good stepper/servo motor for film project?

[Matthew W. Phillips]

In my spare time, I have been experimenting with microcontrollers and controllable motors. Ive been using Atmel AVR controllers with Atmel Studio (not arduino). Lately, I learned how to use a ULN2003 IC to control a stepper with the AVR chip so that the induction of the motor doesnt come back and bite my uController. I used MOSFETS before with a DC motor but I prefer the fine tuning of the stepper.

 

Anyhow, I was able to programmically control a 5V 28BYJ-48 stepper do move at any speed I wanted but the max speed is still way too slow to control, say, a film projector or the motor in a film camera. My goal is to find a low cost stepper that I can put in my circuit so I can replace old film gear (Super 8, for example) with crystal sync'd stepper motors. My circuit already contains external crystal controller via 16 MHz crystal on XTAL1 and XTAL2 and 2 22 pF caps on the ground. So the timing is more or less rock solid.

 

So anyone have any ideas of a sufficiently fast and cheap stepper motor that is 12V or less and small enough to use in a film camera and/or projector unit?

[Phil Rhodes]

Sounds like an interesting one - I just finished some AVR work for an LED strobe light. Unfortunately my knowledge of steppers is fairly slight so I can only offer moral support, but do let us know how you get on.

 

P

[Gregg MacPherson]

Chris Millar from over on the F vs D thread was or is a proffessional mechatronics guy and may ooze helpfull ideas. When you are over there just ask him, invite him back here. He may not have noriced your thread.

 

Cheers,

Gregg.

[Chris Millar]

ello,

 

Doing a ME in mechatronics (by coursework) but its from one of the loftier ivory tower complexes so practical knowledge is next to zero (cybernetics anyone?). They teach us how to learn apparently, and we fill the gaps with our own curiosity - thing is there is no time (or money) to do that !

 

However,

 

I have played around with stepper controlled cameras and projectors - my early attempts long ago to build my own crystal/timelapse/PC-controlled Bolex resulted in the realisation that steppers are waaaaay to noisy/vibrational for any job that requires you to be in the vicinity. You've either got speed and vibration, or really good precision but next to zero rpm. Since then my personal build experience has been with gruntier servos (160v) running a CNC mill closed loop (linuxCNC) and those issues are gone, I do note however that a motion control rig I worked with occasionally 2 years ago used huge steppers for it's boom movements and base rotation (an ex-LOTR kiwi home-brew one I had the pleasure of learning from the builder - kuper controlled if that's interesting for you). I wish I paid more attention to how they were run drive wise as they didn't have the noise issues at all, but they did drive the machinery via a relatively softer timing belt ('soft' due to the distances involved) rather than direct mechanical metal vs. metal gearing or shaft couplings.

 

Of more recent experience has been working with Walking with Dinosaurs and How to Train Your Dragon Live - servo, servo, servo, hydraulic, hydraulic, pneumatic, hydraulic, servo, servo etc... No steppers involved that I spotted ;)

 

Good on you for going with AVR over arduino - it's a path to a more considered and optimised system, but then again the support for arduino is much (much) for approachable for beginners (which when it comes to that stuff we all are for years longer than we realise). I myself started down the the 16-bit PIC path as that is what Creature Technology worked with. Still rate myself at about level 1 on a scale of 1 to 10 in terms of knowing.

 

Anyhoo, wish I could be of more assistance - when I was playing around with cameras I just scoured the net for info and found it lacking or mired in a bog of over complicated techno speak (purposefully or ignorantly it had the same effect). I think the reality is that most of the motors in the scope/size you're looking at are very cheap, just go ebay a bunch of them... see what happens.

 

Oh yeh, scope out user: Glenn Anderson in these forums - he actually does things :o

 

http://www.synccine.com/nxtindex.htm

[Matthew W. Phillips]

thanks Chris for giving me points to ponder. The stepper I was working with was extremely quiet and could not possibly have been heard inside of a camera body or projector casing. However, when I did a test of mounting a takeup reel to it for Super 8, the max speed only moved at about 14.5 frames/second. There is a 12V version of the same motor and I am no properly trained EE so let me ask you, should the same motor (5V) being driven by 12V increase the speed or only the torque? Like I said, I dont know much theory at all.

 

I feel like I may have to go back to DC motors and using MOSFETS. However, this becomes more expensive and harder to get exact. I did write an interrupt routine in C that handles PWM for a DC motor but then there is a ton of math formulas I have to figure out for converting the movement of the motor to frames per second, etc. At least with steppers, the data sheet shows how many steps per rotation and I can program it accordingly.

 

Also, DC motors with MOSFETS seem to be harder for me to control the induction of. I did the MOSFET and Zener Diode but I still seem to burn out more MOSFETS than I do right so I must be missing something or doing something wrong.

 

Any advice you can give? Yes, I intend to hunt ebay for more low cost steppers though.

[Chris Millar]

Here I am about to show how little I know (I'm not an EE (yet), I actually got into the ME via a convoluted pathway that involved a partially completed architecture degree :o )

 

Speed, maybe only very indirectly and minimally (and not worth mentioning, but I'm just pre-empting a smarty pants reply from a lurker) - 'only the torque', well, yes, but a stepper is designed for an appropriate input voltage only, you can play about with it sure, but by a factor of more than double, you're asking for funny smells.

 

Try looking into Phase Locked Loops - its what Clive Tobin uses/used in his crystal motors for Bolex cameras - he used to post here also. Rather than having to do weird stuff with numbers to try to convert a PWM wave into an RPM, have an encoder and a feedback loop, so you can let the system tune itself to the rate you're after. An encoder in part gives you that same 'step' control that your refer to being easier to program (there is a linear relationship between encoder and drive output).

 

Advice is head to the electronics forums - there are some wizards there who take on well written requests for info like a home-work assignment and over night you'll find a personalised schematic turn up in your inbox. Give me maybe two years and I'll be able to help with a little authority, but for now - sorry ! :mellow:

 

Glenn Anderson really does know steppers by the way - he built his own moco rig with them.

 

and yes, I agree steppers are quiet, but wait until you mount them on a echoey frame... the nature of the digital stop/stop really becomes apparent, yes, there are frequencies you can drive them where the step speed means there is no idle time and they go very quiet indeed, but as for the rest, think coffee grinder! :o

[Matthew W. Phillips]

I see no need to create a phase locked loop. Seems like overkill, me thinks. At least if you use a microcontroller. My problem with DC motors is I cant seem to get a MOSFET to work right. It seems silly but these things really are my bane. I purchased a logic level MOSFET tonight so you would think it would be easy to use...not for me it isnt. I tried hooking this thing up every which way possible and nothing. Either the MOSFET comes on and wont switch off or it is off and wont switch on. Im done with MOSFETS for now so I think Ill just have to try to find a good stepper/servo motor that I can use ULN2003 for. Honestly, a stepper would be better anyway since its brushless and should have a longer life than a DC brushed motor...most likely quieter too.

[Jean-Louis Seguin]

In a similar vein, I was always wondering how you could modernize a JK optical printer by using a different approach to the motors and sequencer with the main goal of making the whole thing much quieter and minimizing vibration. This printer was designed in the 1960s using the technology of the period. I'm sure there must be vast improvements since then in motors and control technology.

 

Jean-Louis

[Chris Millar]

I see no need to create a phase locked loop. Seems like overkill, me thinks. At least if you use a microcontroller.

In terms of redundancy a microcontroller is overkill compared to a PLL

 

I tried hooking this thing up every which way possible and nothing.

Best way to learn if you're allergic to spec sheets (as I am) is to buy an off the shelf driver and reverse engineer it.

 

Honestly, a stepper would be better anyway since its brushless and should have a longer life than a DC brushed motor...most likely quieter too.

Not quieter at all - and that is from experience - think about the nature of how they are driven, 'forward' is actually composed of gazillions of stop/start/stop/start motions each of which has an associated inertial force that is translated into the structure that it is mounted to. Yes, servos stop and start also, but not at the frequencies that a stepper does. The issue of longer life in steppers vs. brushed is a consideration for long duty applications, not intermittent use that a camera or projector. There are always brushless DC and AV servos to keep in mind also (they are what I use in my CNC mill).

 

As for a product I know is used in the film industry (as I was involved in building a custom hot-head with them) are:

 

http://www.animatics.com/products/smartmotor.html

 

and these guys I've worked with also (many many motors) - brushless, used in constant duty display animatronics:

 

http://www.maxonmotor.com.au/maxon/view/content/index

[Matthew W. Phillips]

Chris, a microcontroller is much easier to use for someone like myself who has a programming background and prefers code over hard engineering. The concept of I/O ports and interrupts is much easier for me to understand compared to cap/resistor values and feedback loops.

 

We can argue forever about DC motor noise vs Brushless motor noise but the fact of the matter is right now I CAN make a stepper work and I CANT get a DC motor to work with any degree of control. Kindof makes my decision for me, doesnt it? UNLESS I can find a way to make DC work, its a moot point.

[Chris Millar]

k,

 

If you're keen on steppers go for it, if they turn out noisy then try belts (like you see in scanners, vending machines and so on) instead of direct coupling

 

Have you sent Glenn A a PM ?

 

For the information you're after he is the best bet that I know of around here.

 

Dom Jaeger, a name you might recognise from here - works with him on occasion.

[Matthew W. Phillips]

I will PM if I can find them

[Chris Millar]

search function - top right of the frame:

 

http://www.cinematog...?showuser=35197

 

google works too

 

As does the link to his personal website I posted above ;)

Edited December 9, 2012 by Chris Millar

[Matthew W. Phillips]

Searched some AVR boards and discovered that I just need to get a high torque 12v stepper. This way I can get enough speed out of it before the torque becomes too low.

 

Not to start an argument but where did you hear that brushed motors were quieter? Every site I have looked at says steppers are far quieter and that only servos are more quiet. Just curious where you saw that so I can review it.

[Chris Millar]

No argument required (?)

 

answer: with my ears - but yeh sure, maybe my ears have been in odd places

 

Many servos are brushed motors.

[Dom Jaeger]

.. when I did a test of mounting a takeup reel to it for Super 8, the max speed only moved at about 14.5 frames/second. There is a 12V version of the same motor and I am no properly trained EE so let me ask you, should the same motor (5V) being driven by 12V increase the speed or only the torque?

 

Being a tinker in mechanics and optics I'm very far from an expert on steppers, but my understanding is that you can increase the speed of a stepper motor by upping the voltage, as long as the current is regulated to within the winding current rating. A higher voltage gets the coils to their current limit sooner, which allows the motor to run faster. Modern stepper drivers tend to be current limited, and the input voltage can sometimes be many times the 'rated' voltage of the motor. Depending on the winding inductance the torque will drop off at higher speeds/voltages.

 

From conversations with Glenn Anderson (who knows waay more about this stuff than I do) using a H-Bridge to reverse the voltage in the preceding coil (rather than simply switching the voltage off) can overcome residual current that slows the motor down (something which may also be incorporated in certain drivers).

 

Glenn doesn't visit these forums much, but if you wanted to ask him a question you could contact him through his website synccine.com. He's a pretty helpful guy.

[Chris Millar]

 

 

Being a tinker in mechanics and optics I'm very far from an expert on steppers, but my understanding is that you can increase the speed of a stepper motor by upping the voltage, as long as the current is regulated to within the winding current rating. A higher voltage gets the coils to their current limit sooner, which allows the motor to run faster. Modern stepper drivers tend to be current limited, and the input voltage can sometimes be many times the 'rated' voltage of the motor.

 

Interesting stuff! shows how little I know them

[Jeremy Cavanagh]

The above posts are interesting stuff. I would like to build some crystal sync circuits for motor driven 8 mm cams (either standard 8 or Super 8) but need to understand how motor speed is controlled most 8 mm cameras so if anyone can point me to a source of info that would be great. The crystal sync circuits are failry easy, the difficult part is interfacing them with existing systems such as those found in super 8 cameras (plus I am very rusty on my DC motor systems) so I have one particular question that someone may be able to answer straight off i.e. the 'flash' connection found on the side of cameras such as Canon's 814 and 1014 series does that fire the camera off frame by frame or does it send a pulse out of the camera to fire a flash lamp? If ist the former then I guess it would be fairly easy to control camera speed externally but I guess it can't be that easy.

 

Look forward to any answers, thanks.

[Jean-Louis Seguin]

I have one particular question that someone may be able to answer straight off i.e. the 'flash' connection found on the side of cameras such as Canon's 814 and 1014 series does that fire the camera off frame by frame or does it send a pulse out of the camera to fire a flash lamp? If ist the former then I guess it would be fairly easy to control camera speed externally but I guess it can't be that easy.

 

Look forward to any answers, thanks.

 

 

The "flash" connection or PC connector (for Prontor contact) is simply connected to a switch that closes for every frame exposed. This was the basis of many amateur sync sound systems in the 70s and was often referred to as 1/F sync. Crystal control devices for super8 cameras were marketed by many companies such as Super8Sound (now Pro8mm). Today, I believe only The Film Group still offers some of these devices. Check out http://www.webtfg.com/ and click on Film Group Products.

 

Jean-Louis

[Zac Fettig]

As for a product I know is used in the film industry (as I was involved in building a custom hot-head with them) are:

 

http://www.animatics...smartmotor.html

 

I've used one of these in my day job, and I can tell you, it was a breeze to use! Stepper and controller in one. Real quiet. Price however... not so great. They start at about $1200. Great for lab work. Also, a little too big for a Super 8 camera. Probably better for a B&H Filmo/Eymo or something.

 

If you're looking for cheap steppers, check out the surplus houses:

http://herbach.com/

http://www.sciplus.com/

http://www.allelectronics.com/

[George Odell]

Did someone mention The Film Group? Yes, in my fomer life that was where I spent many a late night developing crystal control for those, like me, who could not afford crystal control.

 

Still do it.

 

To answer a question posed, yes the basic "time clock" section of the circuit was more or less identical for the forty or so cameras we used to offer the conversion for. It was the "back end" that connected into the partiular camera's drive system that got complicated. Every camera model was different. Somes used machinical governors like the older Nizo's but most used some form of sevo system where by camera speed was controlled electronically rather than by spinning weights (yes that's how the Nizo's worked!). None offered anything close to a precise speed but the later offerings from the few camera companies that survived into the 1980's did get better.

 

Today crystal sync is not as necessary as when film was edited as film. Non-linear makes adjusting sound speed (clip length) an easy matter and unless you shot with a dying battery you more than likely can pull/slip the A-V sync back in.

 

www.webtfg.com and click on the left side link for Film Group Products. Visit the Q&A section. There is a ton of information there, as well.

Edited January 15, 2013 by George Odell