It is currently Thu Dec 05, 2019 9:35 am




Post new topic Reply to topic  [ 68 posts ]  Go to page Previous  1, 2
 Motorizing a focus rail 
Author Message
User avatar

Joined: Wed Mar 18, 2009 3:07 am
Posts: 22
Post Re: Motorizing a focus rail
shutterdrone wrote:
Daniel wrote:
One thing I like about ladyada's Motor Shield, is that it has a release command. When you call this command, it cuts of the power the the motor. Handy for timelapse to save power between exposures and to keep the motor cool. I guess you could rig up a separate circuit to do this with the easydriver board as well if it was a requirement.


Some caution when doing this -- if you're micro-stepping a stepper, the only thing keeping it in its current position is the continued application of voltage. Once you remove the voltage, it will likely fall into its nearest full-step position (or even further! It can move several steps based on how much torque is applied to its gearing) - so, I'd be careful about doing that in TL, when it may cause it to cycle between the same position over and over, with a little bit of randomness. Mostly this will happen when you're making less than a full step of motion. Also remember that when you first apply power, the motor will lock into its nearest stable position - meaning you could have up to two un-controlled movements. (In fact, there's no guarantee that it will remain in a micro-stepped position other than equal to a full-step with power applied, but I haven't seen a lot of impact of this.)


While you could certainly use worm-gearing to prevent the weight of the camera from driving the motor into a new position when the power is removed (the gear can't turn the worm, in theory =), you can't control how the motor responds to fluctuating power.

I'd be interested in seeing how well this actually plays out in real life. Remember to base your opto-coupler not only on voltage, but also current rating. You want up to 750mA to pass through it safely =)


Yep, thanks for pointing this out Shutterdrone. I should have put this in more context. I was using ladyada's motor shield using full steps and running it in "double" mode, which means 2 coils are activated at once (for higher torque). I did a few tests just using the release function I was referring to. From the few tests I did, it seemed to stay in position every time. But I think this kind of thing could behave different in the field due to any number of variables such as fluctuating power as you mentioned. I guess some sort of physical optical encoder would be the safest approach?

Sounds like you are cooking up a crazy/cool timelapse recipe there Shutterdrone, can't wait to see how it all turns out. Especially the touch screen! Are you using processing for the graphics?


Mon Mar 23, 2009 3:35 pm
Profile
User avatar

Joined: Wed Jan 14, 2009 8:34 pm
Posts: 626
Post Re: Motorizing a focus rail
shutterdrone wrote:
I'd be interested in seeing how well this actually plays out in real life. Remember to base your opto-coupler not only on voltage, but also current rating. You want up to 750mA to pass through it safely =)

now if I could get past this stupid problem _I'm_ having *lol* (Think new chipsets with spotty support, secondary touch screen displays with their own craziness and spotty support, and the miriad issues of software serial interfaces on an MCU. *sigh*)


Do you know a part number for these high-current optocouplers? I've not encountered one but they sound useful. I would think the more common approach is to (use an optocoupler to) switch a power transistor, but having them both in one package would certainly be nice.

MILapse has rather cruelly been teasing us with the promise of your Great Machine. The touchscreen sounds interesting, sort of like CamBlock. I've been working with the 16-bit PIC 24F series, and have the Explorer16 development board and touchscreen LCD module. Microchip have a royalty free graphics library. I tried some of their demos and it was amazing; buttons, sliders, menus, bitmaps, animations, you name it. I've been using it in some projects - the potential is mind-blowing.

http://www.microchip.com/stellent/idcplg?IdcService=SS_GET_PAGE&nodeId=1406&dDocName=en024858
http://www.microchip.com/stellent/idcplg?IdcService=SS_GET_PAGE&nodeId=1406&dDocName=en532570

Anyway, back on topic. Being able to turn off the motors whenever possible means potentially HUGE energy savings. I didn't consider any method other than full-stepping. Instead of being a nuisance the detent torque will actually help hold the machine in place. As for wormgears, unless you get a really good wormgear the backlash and general wobbliness will probably be greater than your theoretical stepper precision. Motion timelapse is a tricky business indeed.

_________________
Little Bramper website and on Timescapes
Link to Time-lapse FAQ


Mon Mar 23, 2009 5:02 pm
Profile
User avatar

Joined: Sat Mar 07, 2009 8:37 pm
Posts: 364
Location: Dubai
Post Re: Motorizing a focus rail
Not quite there yet. Still looks a little janky:



DELRIOUS

_________________
http://www.vimeo.com/timelapseinc
http://www.vimeo.com/delrious


Tue Mar 24, 2009 12:58 am
Profile
Post Re: Motorizing a focus rail
Daniel wrote:

Yep, thanks for pointing this out Shutterdrone. I should have put this in more context. I was using ladyada's motor shield using full steps and running it in "double" mode, which means 2 coils are activated at once (for higher torque). I did a few tests just using the release function I was referring to. From the few tests I did, it seemed to stay in position every time. But I think this kind of thing could behave different in the field due to any number of variables such as fluctuating power as you mentioned. I guess some sort of physical optical encoder would be the safest approach?


The more I think about it, the only problem would really be those axes supporting thrust loads instead of radial loads (i.e. tilt axis) - I did some tests last night on a _pure_ radial axis (pan/dolly/truck on a level plane) and found no appreciable movement breaking power after micro-stepping in 20:1 gearing w/ a 4lb load.

Daniel wrote:
Sounds like you are cooking up a crazy/cool timelapse recipe there Shutterdrone, can't wait to see how it all turns out. Especially the touch screen! Are you using processing for the graphics?


Actually, using "sub-processing" (lol) - I'm using the Touch Shield Slide from Liquidware -- http://www.liquidware.com/shop/show/TSL ... ield+Slide coupled w/ a Duemilinova w/ 328P processor, and their extender shield and LIon power pack. Haven't gotten much done on the graphical UI yet, but I'll explain in a second what I have done... =)

astronomerroyal wrote:
Do you know a part number for these high-current optocouplers? I've not encountered one but they sound useful. I would think the more common approach is to (use an optocoupler to) switch a power transistor, but having them both in one package would certainly be nice.


Actually, I was looking for some after posting that (I just saw opto-coupler and responded based on my knowledge of the common ones I use supporting very little current) - and I can't find any that seem to fit the bill. I think you are correct in the use of an opto-coupler + transistor. If you want them in one package, you'd go with an SSR (solid-state relay), which certain models combine both an opto-coupler and mosfet.

astronomerroyal wrote:
As for wormgears, unless you get a really good wormgear the backlash and general wobbliness will probably be greater than your theoretical stepper precision. Motion timelapse is a tricky business indeed.


Yes, backlash will be a highly probably issue - wobbliness will not *grin* The design I've been working on for a while, and will start machining in the next month or two will have a lot of structural strength in a small package. To be honest, in most cases one wouldn't even want to use the theoretical maximum resolution, at one shot per second, moving 0.004 degrees between shots (the minimum motion of the design I've worked out) would take 12.5 hours to achieve a 180 degree movement. That would be used in a minimum of cases, I'd imagine. I've been using 0.009 degree movements in a spur-gear setup with a LOT of "wobbliness" and backlash, and not have noticed any ill results in a radial axis (again, thrust would be different, and requires _no_ wobbliness). That is, errors occur (at a rate somewhere around 1 every 800 requests) but they are not visible in the output video, as the lack of minimal change in a single frame is not generally perceivable. Of course, if you increase that rate, or the size of the error, then that all changes...

astronomerroyal wrote:
MILapse has rather cruelly been teasing us with the promise of your Great Machine. The touchscreen sounds interesting, sort of like CamBlock. I've been working with the 16-bit PIC 24F series, and have the Explorer16 development board and touchscreen LCD module. Microchip have a royalty free graphics library. I tried some of their demos and it was amazing; buttons, sliders, menus, bitmaps, animations, you name it. I've been using it in some projects - the potential is mind-blowing.


Hehe, ok, we'll let the cat out of the bag (I'm not big on making announcements before stuff is ready to ship, but I'm also aware of the pain of teasing forever =). Part of the big delay is I spent a few months on that red herring of dynamic external exposure control (fun, but almost not worth investing the time and money to productize).

So, yes, it's funny you mention camblock, as I had drawn up a system design that was similar back before I started on the dynamic external exposure control setup. It was funny (not in a "haha" way!) to see that stuff once I was already in the process of procuring machining equipment and designing hardware! At any rate, it's not such a big deal as his system when/if it comes to market will certainly be more capable than what I'm working on it. It will also be much more expensive, but as they say - if you need all the bells and whistles, then you can probably afford them too... =)

So, here's the deal: I'm working on an open-source hardware and software design. The first thing to come out will be the operating system, calling it the "OpenMoCo OS" - it will consist of two software components at first, an engine component that does all the control work, and a graphical UI running on a 320x240 touch-screen interface. The whole thing can be completely implemented using a one of the gadget packs from liquidware -- a duemilinova w/ 328p chipset, a touch shield slide, an extender board, and a lithium power pack (if desired). The extra components fit right on the extender shield, basically 6-8 diodes and a dual opto-coupler. In the next couple of months I will be putting out the first motor prototypes and enclosures and such - again, all open-source using open-source components. It's going a little slow as I'm doing all the machining in-house (literally, in my house =) to avoid expensive set-up and machining fees that would push the cost beyond what we could all afford.

I'm getting close to releasing the engine component, and we'll have a a website up soon for it, where everyone can join in and add their contributions if they like. The UI is just beginning, so that will take some time. Unfortunately, there aren't too many pre-built libs for the touch shields yet, but I'll help to produce some of those.

Here're the features of the engine:

* Shoot-move-shoot style control
* 4 axis individual stepper control, with the stepper drivers being external (i.e. easydriver, probotix, gecko, etc.)
* Motor ramp-up/down, either linear or smooth (see Daniel's post about smoothstepping on his blog)
* Camera intervalometer with pre-focus tap. Exposure times up to several hours, interval times from 1 second to 18 hours. (Camera exposure and intervalometer are non-blocking, other things can occur during exposure -- see below)
* Pre-and post-shot delays
* Intervalometer is "optimistic" in that in will always try to achieve as close to your actual interval time even if some other action or activity you've programmed would normally take too long, without disturbing those activities
* Alt IN/OUT port - trigger flash before camera, trigger second camera, trigger program start by external signal, etc.
* Action scripting -- generate actions that can modify motor distance moved, motor ramp settings, turn on/off motors, change interval time, change exposure time, etc.
* Keyframing -- trigger scripted actions on time, camera, and motor keyframes (trigger action #1 after 30 minutes of program run time, or trigger action #2 after motor #0 has moved 15 degrees, or trigger action #1 after 300 shots, etc.)
* Dual serial control interface. All settings and capabilities can be configured via hardware or software serial (can use both at the same time!) during program execution using a simple serial protocol. (Ok, so not that simple, but I'll get to that part =) This is because the touch-screen shields for the arduino use a software serial communication. This way you can both interact from the UI and script things from your computer or laptop at the same time. Just hook it up via a USB cable.

I'm finishing up QA on the engine and writing a perl library to wrap up the serial protocol first -- so it can be used directly with a computer, until I get a decent UI package for it out. At first I'm chasing the touchshield route, and then I'll work on cheaper interfaces using other common shields. I expect to release the engine in the next week or so, barring any major issues.

I wanted to go this route, as not even mentioning the people on here that work on timelapse DIY, there are about 4-8 people a month going on the arduino forum looking for advice, and there are a lot of blogs out there working on similar stuff. I wanted to make a simple to use engine that could be combined with all sorts of hardware, and didn't require a lot of work for the beginner. (The current prototype for the controller is just using off-the-shelf stuff from liquidware and a couple common components as I stated earlier, no soldering needed for a simple setup.)

I'm also working on production, as well, to be able to sell complete 2-4 axis setups (no assembly required) for those of the non-diy persuasion. They will include the ability to daisy chain motors 2-deep (two cables from the controller run 4 axis), again using inexpensive cabling that you can find at any computer store, custom-machined to be ready to pop right on a tripod for pan/tilt, with all the necessary bracketry, etc. The pricing will be below any similar system, due to the fact that the electronics are all off-the-shelf, only physical machining needed.

!c


Tue Mar 24, 2009 7:58 am
User avatar

Joined: Wed Mar 18, 2009 3:07 am
Posts: 22
Post Re: Motorizing a focus rail
Yeahhhh! Bring it on!!

That all sounds way cool Shutterdrone! If you pull this off you will be at rock star status in my book. I've been wresting with the idea of affordable motion control for years. Ever since I found out that the rig I was designing an interface for costs around 100 grand to buy! Hopefully you can shave a few dollars off that :)

I am planning at some stage to design an animation timeline gui (similar but not as sophisticated as the camblock gui) for my own needs. Features would include:
- A scalable timeline
- set/delete keyframe buttons
- playback and next keyfame buttons
- fcurves with adjustable tangents for keyframes (very much like camblock gui)

As of yet I haven't decided to do this with flash and action scripts or do it more low level in processing. Maybe we can start a new thread on this or just talk privately. But once your site is up an running, maybe I can contribute to the UI part of your project. Also if you are wanting to reach another audience (low budget VFX houses/artists/cinematographers) I would love to share my experiences with motion control from a motion picture pov which might help your software/hardware design reach a wider audience.

Anyway, Exciting times! Thanks for sharing.

Dan.


Tue Mar 24, 2009 4:57 pm
Profile
User avatar

Joined: Wed Jan 14, 2009 8:34 pm
Posts: 626
Post Re: Motorizing a focus rail
Wow, good man. A massive philanthropic gesture. I won't even bother trying to describe how impressive that sounds. I have lots of Qs but I'll wait until you start releasing things. Hardware+firmware+software, that's the complete package.

By the way, I decided to try one of those TAOS chips, despite the fundamental problem of metering mostly reflected vs incident light. Really interesting to be able to see inside the chip ...

As for Delrious' focus rail. The progress being made is prodigious, gripping.

_________________
Little Bramper website and on Timescapes
Link to Time-lapse FAQ


Tue Mar 24, 2009 5:38 pm
Profile
User avatar

Joined: Tue Aug 26, 2008 3:30 am
Posts: 824
Location: Sydney, Australia
Post Re: Motorizing a focus rail
Regarding the concerns of mechanical backlash in particular axes, a programming tip is to pre postion your axis. eg Start the pan a little before your actual start position.
For example if your pan went from 0 degrees to 90 degrees - start the move at say minus 5 degrees (theoretical 355 degrees position of your 360 degree panning circle)

This gives the mechanics time to "settle" and your stepper drive "pulse" train/sequence to figure itself out relative to mechanical tolerances of your gear.


Tue Mar 24, 2009 6:15 pm
Profile
User avatar

Joined: Fri Jun 13, 2008 5:54 am
Posts: 611
Location: Oslo, Norway
Post Re: Motorizing a focus rail
Church, you're my hero too!!

_________________
Canon 400D, 50D, 5D Mk II. Canon L 16-35/f2.8, Sigma 10-20. Adobe Creative Suite 4.
website:
http://www.magictimelapse.ch/en


Wed Mar 25, 2009 3:16 am
Profile
Post Re: Motorizing a focus rail
Oy - I'm so sorry for hi-jacking your thread Delrious! (I'll start my own thread before too long here.)

I'll continue only momentarily *grin* I'm happy to have everyone's input, I would like to see it as a community project (open source) more so than a commercial product. Anything sold is just to help people solve their problems, and I'd like everyone to own the code/designs and be contributors. I think that's the only way for something like this to survive in the long-run, and plus - well, that is the whole point. =)

Delrious - I think you're making great progress. I'd love to see some photos of the setup, as I'd like to see the relative size of the linear stage you've gotten. Please keep us updated as you go on!

!c


Wed Mar 25, 2009 5:38 am
User avatar

Joined: Sat Mar 07, 2009 8:37 pm
Posts: 364
Location: Dubai
Post Re: Motorizing a focus rail
More test video. I don't have the shutter wired in yet, as I'm waiting for my opto's to arrive in the mail. I think that may be contributing to the shakiness. Also, the camera is just resting on the stage at this point (no tripod mount built yet).



DELRIOUS

_________________
http://www.vimeo.com/timelapseinc
http://www.vimeo.com/delrious


Wed Mar 25, 2009 7:07 am
Profile
Post Re: Motorizing a focus rail
Delrious wrote:
More test video.


Looks pretty damn good! So this is using that pre-manufactured motion control table or a retrofit focus rail?


Wed Mar 25, 2009 8:55 am
User avatar

Joined: Sat Mar 07, 2009 8:37 pm
Posts: 364
Location: Dubai
Post Re: Motorizing a focus rail
That's using the pre-manufactured linear stage that I linked above with the camera just resting on the stage. I want to make sure I can make this thing work before I build the brackets to put it on my tripod. It's still not as smooth as I would like (I'm still tweaking the voltage/delays). The video above was at 1:1 and it still looks a bit shaky. If that were using my MP-E at 5:1, it wouldn't look nearly as smooth. I'll take some pictures of my full setup this weekend (just a Breadboard that looks almost just like Dan's howto setup + my linear stage) as I'm out of town atm.

DELRIOUS

_________________
http://www.vimeo.com/timelapseinc
http://www.vimeo.com/delrious


Wed Mar 25, 2009 5:35 pm
Profile
User avatar

Joined: Wed Jan 14, 2009 8:34 pm
Posts: 626
Post Re: Motorizing a focus rail
Are you using some sort of small weight and pulley on the far end? Sometimes helps to have things under the constant influence of a force, rather than floating. Having a force pulling the table down the track will also mitigate possible problems associated with static friction, which might otherwise cause jankiness. Mr Timescapes uses a weight on his dolly, and I imagine it's common practice.

_________________
Little Bramper website and on Timescapes
Link to Time-lapse FAQ


Wed Mar 25, 2009 6:03 pm
Profile
User avatar

Joined: Sat Mar 07, 2009 8:37 pm
Posts: 364
Location: Dubai
Post Re: Motorizing a focus rail
Time lapse of my whole setup:



DELRIOUS

_________________
http://www.vimeo.com/timelapseinc
http://www.vimeo.com/delrious


Sat Mar 28, 2009 6:24 pm
Profile
User avatar

Joined: Sat Mar 07, 2009 8:37 pm
Posts: 364
Location: Dubai
Post Re: Motorizing a focus rail
-Got rid of the rest of the jankyness by just slapping on a bigger power supply and readjusting the pot

-Used two small blocks of wood for the tripod->linear stage->focus rail coupling - works great (thanks for the tip shutterdrone).

-Got a cheap laptop to serve as my slave computer. Allows me to monitor remotely as well as have another stage in my processing pipeline.

Right now as I type this, I am doing my first official tripod mounted shot of my subject matter (which in case anyone hasn't figured out by now, is live coral polyps under magnification).

Left todo:

-Interleave intervalometer with stepper motor (waiting on my optoisolator from sparkfun).
-Package everything in a nice box with break aways.
-Figure out the hall-effect sensors (I briefly tried everything suggested in this thread, and none of them have worked, but honestly, I haven't been spending time on this one, I'll get back to this at the end).
-Rotary Encoder: Looking at my motor I don't think that this stage was built mounted with one (they are an option for this model)
-Write user frandly software. Right now I just edit the parameters of the c program for each move. I would like to create a basic scripting mechanism that will let me write out a shot description to a text file and feed that into the program as the instructions. Also, allow a fast forward, rewind to start mode to test out shots.

DELRIOUS

_________________
http://www.vimeo.com/timelapseinc
http://www.vimeo.com/delrious


Sun Mar 29, 2009 9:32 pm
Profile
User avatar

Joined: Sat Mar 07, 2009 8:37 pm
Posts: 364
Location: Dubai
Post Re: Motorizing a focus rail
And alas a shot of my whole setup in action as I wait for this shot to finish. This is probably the worst photograph I have ever taken (it's very dark and I'm far away).

Image

DELRIOUS

_________________
http://www.vimeo.com/timelapseinc
http://www.vimeo.com/delrious


Sun Mar 29, 2009 10:56 pm
Profile
Post Re: Motorizing a focus rail
Delrious - looks great! You've made a lot of progress in a very short time! Are you sure you haven't done this before? =)

!c


Mon Mar 30, 2009 8:05 am
User avatar

Joined: Sat Mar 07, 2009 8:37 pm
Posts: 364
Location: Dubai
Post Re: Motorizing a focus rail
Honestly, this was way easier than I thought, I just needed a few people to tell me so and point me in the right direction :P

For my day job, I do programming on embedded systems (focused on parallel programming for the IBM cell processor for a certain video game console)...the programming/pipelining/workflow aspects of this project are easy for me, as they are the same sorts of problems I have had to solve over and over again throughout my career, and getting instant feedback is quite rewarding. The "hard" part was going to home depot to buy the right machine screws, connecting the wires and drilling the holes (this just seems wrong).

DELRIOUS

_________________
http://www.vimeo.com/timelapseinc
http://www.vimeo.com/delrious


Mon Mar 30, 2009 9:50 am
Profile
User avatar

Joined: Sat Mar 07, 2009 8:37 pm
Posts: 364
Location: Dubai
Post Re: Motorizing a focus rail
I got the shutter circuit running last night and did another test. The test failed when I managed to burn up my easydriver about half way through (it doesn't work anymore...going to order some more).

Looking at the shots that came off, there is some resonance going on in the rail, it seems to speed up and slow down over ~60 frame intervals. I think I need to spend some more time messing around with powering this thing, as clearly I'm doing something incorrectly if things are melting.

DELRIOUS

_________________
http://www.vimeo.com/timelapseinc
http://www.vimeo.com/delrious


Wed Apr 01, 2009 8:28 am
Profile
User avatar

Joined: Wed Jan 14, 2009 8:34 pm
Posts: 626
Post Re: Motorizing a focus rail
If I recall correctly the Easydriver sources <1Amp. Since you're also microstepping perhaps there's a stiff spot in the mechanics that's causing periodic stalling/erring of motor - I don't know what your motor's rated at, but driving it quite hard, using full steps, might be the safest approach.

Even though I had to solve the issue above - and now my motors run smoothly - I still get so-called periodic errors in my worm drives simply because the worm wheel and gear aren't set perfectly. Mechanically that's been the biggest bug I've had to deal with.

Perhaps things might be clearer if you work out how many stepper steps correspond to this periodic pulsing. Also, is there any discernible side-to-side wobble of the screw/table?

_________________
Little Bramper website and on Timescapes
Link to Time-lapse FAQ


Wed Apr 01, 2009 11:53 am
Profile
User avatar

Joined: Sat Mar 07, 2009 8:37 pm
Posts: 364
Location: Dubai
Post Re: Motorizing a focus rail
I'm bored of messing with the voltage on this thing - it's just too time consuming to make fine adjustments and then have to make test videos to verify if the resonance is gone. The good news is - I just found the native driver for this motor for $10 on ebay...$5 cheaper than an easy driver.

DELRIOUS

_________________
http://www.vimeo.com/timelapseinc
http://www.vimeo.com/delrious


Wed Apr 01, 2009 4:13 pm
Profile
User avatar

Joined: Sat Mar 07, 2009 8:37 pm
Posts: 364
Location: Dubai
Post Re: Motorizing a focus rail
If anyone is curious as to what warrants a truck shot on this scale, here are some stills from my movie. These are straight from camera to imageshack (no editting or cropping).

Image
Image
Image

DELRIOUS

_________________
http://www.vimeo.com/timelapseinc
http://www.vimeo.com/delrious


Wed Apr 01, 2009 11:34 pm
Profile
User avatar

Joined: Wed Mar 18, 2009 3:07 am
Posts: 22
Post Re: Motorizing a focus rail
Stunning! Really nice pics. Can't wait to see them in motion!


Wed Apr 01, 2009 11:59 pm
Profile
User avatar

Joined: Sat Mar 07, 2009 8:37 pm
Posts: 364
Location: Dubai
Post Re: Motorizing a focus rail
Any thoughts on how I should regulate power to this? I've got the Arduino board, the stepper motor (needs 24V) and an LCD (5V). I was thinking about adding a little plug to the box that will also let me route my camera power through the same box. I currently use a 30 VDC supply which is probably how I blew the easydriver out (on a side note, the stock driver has a huge heat sink).

I'm making stuff nice and working on the software while I wait for the new driver to arrive:

Image

DELRIOUS

_________________
http://www.vimeo.com/timelapseinc
http://www.vimeo.com/delrious


Mon Apr 06, 2009 6:01 pm
Profile
User avatar

Joined: Sat Mar 07, 2009 8:37 pm
Posts: 364
Location: Dubai
Post Re: Motorizing a focus rail
One month ago I came onto these forums and asked for some help on motorizing a focus rail.

Now I can do this:



(if it looks choppy watch it in low res, not sure why the high res is coming out so choppy on vimeo)

This is beyond the quality I had hoped to achieve - the native driver really hits this rails sweet spot, has some other nice features that easydriver does not and I have a huge amount of resolution to work with (which is nice for doing the little ramps at the start/end).

Just wanted to say thanks to all those who helped in this thread, especially Shutterdrone and Dan Thompson, both of who's work I referenced for my code/hardware and who have helped me out here immensely.

Going to post my code up here once it is in a fully workable form...it's too much of a work in progress still.

Truck shots are going to be nice, but motorized focus stacking + time lapse are going to be incredible for macro stuff.

Now to get working on my motor driven sun :)

DELRIOUS

_________________
http://www.vimeo.com/timelapseinc
http://www.vimeo.com/delrious


Thu Apr 09, 2009 12:16 am
Profile
Post Re: Motorizing a focus rail
Wow! Great work!

Glad it came together so fast for you =)

As for power? I'd just use the appropriate linear regulators for each one. If your camera is much like mine, you can base it off the design I did for mine here: http://roamingdrone.wordpress.com/2008/ ... ntax-k10d/

I used an LM338 as the LM317 did not support the current draw (even at a spike capacity of about 2A) for my camera when shooting. You can just use the appropriate R1/R2 combination for your voltage output.

It WILL generate a lot of heat coming down from 30V though! I think you can buy a straight-up 12V linear IC regulator for the arduino (or, might as well go straight to 5V, and hook up to the 5V pin instead of wasting energy using the 5V linear regulator on the arduino board).

What stepper driver did you end up using, btw? I'm looking for something a little bit better than the ED for my new housing design, as the ED doesn't have any appropriate mounting holes.

!c


Thu Apr 09, 2009 11:11 am
User avatar

Joined: Sat Mar 07, 2009 8:37 pm
Posts: 364
Location: Dubai
Post Re: Motorizing a focus rail
I used the CSD2112-T vexta driver, which is the driver that is specifically built for this motor. I found one for 10 bucks on ebay.

It's a bit beefy (I need to find a new enclosure to hold it). It has a fairly large heat sink that makes it about 2 inches cubed.

DELRIOUS

_________________
http://www.vimeo.com/timelapseinc
http://www.vimeo.com/delrious


Thu Apr 09, 2009 12:08 pm
Profile
User avatar

Joined: Sat Mar 07, 2009 8:37 pm
Posts: 364
Location: Dubai
Post Re: Motorizing a focus rail
I think the "right way" for this device/workflow to work involves me figuring out how to use a rotary encoder with this thing.

I want to be able set key frames when manually moving the stage for start/stop and then base my shots off that, which requires me to know the current position. Looking into this now.

DELRIOUS

_________________
http://www.vimeo.com/timelapseinc
http://www.vimeo.com/delrious


Tue Apr 14, 2009 8:39 am
Profile
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 68 posts ]  Go to page Previous  1, 2


Who is online

Users browsing this forum: Google [Bot] and 3 guests


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
Powered by phpBB © phpBB Group.
Designed by Vjacheslav Trushkin for Free Forums/DivisionCore. pozycjonowanie