Page 5 of 5 FirstFirst 12345
Results 41 to 50 of 50

Thread: something different from galvo scanners?

  1. #41
    Join Date
    Jan 2006
    Location
    Charleston, SC
    Posts
    2,147,489,446

    Lightbulb

    Quote Originally Posted by dream View Post
    Hm, instead of going the long road and attempting to use forensic science in an online board to find evidence about potential use of multiple accounts, why not ask an admin? I'm sure you'll find the results interesting. And also save time for everyone involved.
    The admin has been involved, for over a year now. It's just never been made public. (And honestly, was never meant to be public, but oh well.)

    Anyway, back on topic...

    The Servo Recorder/Playback unit is a very interesting idea, but at $110 for 4 channels, that yields just two X/Y pairs. That's pretty close to what a sound card DAC would cost. Also, to "program" the unit, you would need to twiddle the knobs in real-time to record the movements. That could get difficult if you wanted circular motion. (Ever try to rotate two knobs in a sine-wave pattern while one is 90 degrees out of phase with the other?)

    Another suggestion was the Adafruit 16-channel servo driver. This unit is only $15, and if I read the spec sheet correctly, it allows for 16 separate PWM outputs. Also, it appears to be able to drive smaller RC hobby servos directly. (Not sure if the gear-driven ones would require more current than this thing could supply though - we'd need to look into that further.) The real issue though is that it receives all it's data over a 2-wire connection from a controller like an Arduino or a Raspberry Pi. So now you need to program all that I/O in the Arduino.

    I have never worked with the Arduino, so I'm not sure how simple this would be. I would think that for someone with a programming background it wouldn't be too difficult, but I'll reserve final judgement on that for someone with more experience.

    Still, if the Arduino (or the Raspberry Pi) can be programmed easily, then using the Adafruit board would seem to be the cheapest solution. You can get a pair of small RC servos for $10, and building a mount out of sheet metal (or even wood, in a pinch) can't be all that expensive.

    What are other people's thoughts? Has anyone here driven a servo pair (or several sets of servos) using an Arduino? If so, how hard was the programming?

    Adam

  2. #42
    Join Date
    Jan 2015
    Location
    Seattle Washington USA
    Posts
    52

    Default

    a hobby servo takes a pulse between 1 and 2 milliseconds, usually 50 times a second or a standard of expecting a pulse every 20 ms. 1.5 ms would be the center position. the pwm driver only is used to generate this control signal. there are hobby servos aimed at robotics that have digital inputs as well as the ability to do things like report their position and state.

    mechanical hysteresis will probably be worse than control resolution, but it seems like that will not be an issue.

    but seriously, $5 rc servos. for less money than getting a fancy beer, or two, you could play with such magic in your own home as well.

  3. #43
    Join Date
    Jan 2006
    Location
    Charleston, SC
    Posts
    2,147,489,446

    Default

    Quote Originally Posted by lightlinked View Post
    a hobby servo takes a pulse between 1 and 2 milliseconds, usually 50 times a second or a standard of expecting a pulse every 20 ms. 1.5 ms would be the center position. the pwm driver only is used to generate this control signal.
    That was my understanding as well, though I think there are other servos (so-called fast-acting) that expect a signal 90 times per second (or is it 120? I forget...)

    But yeah, even at 50 times a second that will be more than fast enough for this application. My concern is whether the PWM driver is enough to actuate the servo, or if you need an intermediary amplifier...

    there are hobby servos aimed at robotics that have digital inputs as well as the ability to do things like report their position and state.
    Probably more advanced (and more expensive) than needed for this project...

    but seriously, $5 rc servos. for less money than getting a fancy beer, or two, you could play with such magic in your own home as well.
    I know, right? I bought a couple of the DMX servo drivers from Lasershowparts a while back. For just a tiny bit of cash you can install a servo and one of those boards so you can drop a diffraction grating in front of the scan head. It's a very cool effect! Amazing how cheap some of this stuff has become...

    Adam

  4. #44
    mixedgas's Avatar
    mixedgas is offline Creaky Old Award Winning Bastard Technologist
    Infinitus Excellentia Ion Laser Dominatus
    Join Date
    May 2007
    Location
    A lab with some dripping water on the floor.
    Posts
    9,890

    Default

    Except what he is using in the second video is not a RC servo, its a stepper motor or brushless motor with feedback, and one who's driver exceeds the cost of most galvos. The clue is his using G code (CNC programming language) to control it.

    The PWM RC servos jitter 40-50 times per second when still. The Digital servos jitter 400 times per second when still.

    Steve
    Qui habet Christos, habet Vitam!
    I should have rented the space under my name for advertising.
    When I still could have...

  5. #45
    Join Date
    Oct 2012
    Location
    Germany
    Posts
    1,478

    Default

    ... RC-servos aren't meant for high accuracy positioning - you'll have to gear them down with maybe 50:1 or more, so they will be pretty slow compared to their normal speeds.

    In my (high resolution) servos from maxon is a gear-box with 200:1, what's limiting the max. possible speed with 100kHz stepping from the STEP+DIR-servodriver and a 500 lpi encoder on the motor shaft.

    You can calculate possible speeds and accuracies with stepper motors too - a common 1.8°-stepper has 200 steps per revolution (so a 0.9°-stepper will have 400) -- with microstepping (1/16 or 1/32 should be good enough) you'll get higher angular resolution and lower mechanical vibrations ... I'm driving steppers with up to 1/256 microstepping, where the motor is pretty silent, compared to the loud 'humming' of coarser stepped driving ;-)

    If this stepping resolution isn't fine enough, you can use a gear-box with an external spring-load to prevent resonancies ...

    Viktor

  6. #46
    Join Date
    Jan 2015
    Location
    Seattle Washington USA
    Posts
    52

    Default

    these cheap analog servos have quite abit of jitter when moving slowly/little load. (shaky movement) their gearbox/motors ARE capable of smooth silky motion, its just the servo electronics are junk, scratchy pots instead of optical/magnetic encoders, acceleration in the servo loop software would help too. there is the openservo http://openservo.com/ conversion which might make it better.

    personally I've been looking into building a really precise remote bounce mirror with geared steppers and servo control but this thing would be like hundreds of dollars in parts. (for new media art installation)

    Click image for larger version. 

Name:	JSH_2577.jpg 
Views:	11 
Size:	746.0 KB 
ID:	46889Click image for larger version. 

Name:	JSH_2578.jpg 
Views:	15 
Size:	954.9 KB 
ID:	46890

    the thing with the big bearings are for spinning some artsy junk on stage

  7. #47
    Join Date
    Feb 2011
    Location
    New Hampshire
    Posts
    3,513

    Default

    the thing with the big bearings are for spinning some artsy junk on stage
    Or flipping pancakes.

    You might look at this:
    http://www.trossenrobotics.com/robot...oboturret.aspx

  8. #48
    Join Date
    Jan 2006
    Location
    Charleston, SC
    Posts
    2,147,489,446

    Smile This is starting to come together, I think!

    Based on the animation example posted above, I'm not sure that both X and Y movement is required. If we only need single-axis movement, this just got a whole lot easier...

    Zorn - please confirm that you need both X and Y movement.

    As for servo accuracy, what do you guys think? If this is for an aerial display, is accuracy really an issue? I mean, yes, RC servos can jitter a little bit, but if you are sweeping the beam around slowly, I don't think you'll notice any jitter, and if your accuracy is off by a half a degree, will it really matter? Clearly audience safety isn't a factor if the beams are pointed into the sky. (I'd still install an aperture mask to prevent the beams from dipping below the horizon, but that's not a big deal.) Aircraft interference is obviously an issue, but again, accuracy shouldn't matter there.

    One other thing I was thinking about is ditching the mirrors entirely. Hear me out...

    If we assume the plan is to use a high power blue diode (445 nm), then why not mount the diode in a big heat sink (say, an Aixyz module installed in a Z-bolt heat sink) and just use the servos to pan and tilt the whole heat sink? Mount it on a yoke and use one servo to pan the whole assembly left and right while the other servo pans the mount up and down in the yoke...

    It would give you a very wide range of motion without any worry of the mirrors touching like you'd have in a traditional X/Y scanning setup. I'm interested to hear what others have to think about this, but personally I think it would be an ideal solution.

    Regarding control, my understanding now is that he would like to use Blender 3D to develop the animations and then export the data to an Arduino board, which in turn would talk to one of the Adafruit servo boards linked to above. Initially, I thought this was needlessly complex, but it turns out that other people have already done this, so it might be fairly easy to borrow their code to make this project happen. For sure if it works it would make the process of creating new effects damned near effortless.

    One question I have concerning RC servos: there are 3 wires. I'm assuming they are hot, ground, and PWM signal, and that the hot and ground are connected all the time. If that is the case, then you could wire all 16 servos to a robust DC power supply (hot and ground), and then just use the output from the Adafruit board to provide the control signal on the 3rd wire to each servo, right? Sound reasonable?

    If so, that's good, because it means that by using the Adafruit board you can control 16 servos for less than $1 each, plus the cost of the main DC power supply (which, conveniently, can also power everything else). Add in the cost of the servos ($5 each) and you've got a very cheap solution! OK, you also need the Arduino board, but they're like $30. So yeah, this is really coming in way under budget!

    Comments?

    Adam

  9. #49
    Join Date
    Feb 2008
    Location
    East Sussex, England
    Posts
    5,248

    Default

    I don't think I can get servos to rotate a module because of the weight.
    You'd undoubtedly be setting sensible balance points, but even if you weren't, even budget servos have the capability to move a decent amount of weight.
    Think large flight surface at large deflection on fast moving model aircraft and the associated forces pushing back, usually via a linkage that increases the pressure.
    Frikkin Lasers
    http://www.frikkinlasers.co.uk

    You are using Bonetti's defense against me, ah?

    I thought it fitting, considering the rocky terrain.

  10. #50
    Join Date
    Jan 2006
    Location
    Charleston, SC
    Posts
    2,147,489,446

    Default

    OK, so positional accuracy doesn't seem to be a problem, although we definitely need two servos for each beam because we do need full X/Y movement.

    Regarding the idea of a moving yoke vs mirrors, I agree with Norty in that the servos should be able to handle the load, but if you plan to gang several diodes together and either knife-edge or PBS them together, then the resulting module will probably be too large and cumbersome to fit in a yoke, so we're back to mirrors again. And that's not a huge problem - mirrors will work just fine - but the final range of motion may be reduced.

    I'll leave the Python scripting and linking to the Arduino to someone with more experience. But the mechanics seem to be pretty much there. Let us know what you find out from the Arduino forum.

    Adam

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •