go away spammer

Author Topic: 20-25 amp servo system.  (Read 3413 times)

0 Members and 1 Guest are viewing this topic.

Offline TreetpflyerTopic starter

  • Jr. Member
  • **
  • Posts: 7
  • Helpful? 0
20-25 amp servo system.
« on: November 21, 2008, 02:15:30 PM »
Hi,

I pretty new at this, but I have a some questions about a simple system I'd like to make.  I would like to make a stand alone position servo system that is like an R/C servo system with a manual pot control,  just built for more current.  The system requirements are basically: pot to control 1-2ms pulses, servo gyro like one used in R/C helicopters, and a 12V, 20-25 amp servo motor to control a hydraulic actuator.  This would be for tilt control to balance one degree of freedom or axis.  It seems pretty simple but, since I'm new to this I'm sure there is a very simple(inexpensive) way to accomplish this.  I figure the servo will have a separate linear pot on the actuating arm of the hydraulic actuator.

Thanks for any info.

Offline ArcMan

  • Supreme Robot
  • *****
  • Posts: 519
  • Helpful? 4
  • Mmmm... Plasma
Re: 20-25 amp servo system.
« Reply #1 on: November 22, 2008, 09:41:51 PM »
Define "inexpensive".  If you mean about $250, then you're right.

Offline szhang

  • Robot Overlord
  • ****
  • Posts: 140
  • Helpful? 1
    • szhang.net
Re: 20-25 amp servo system.
« Reply #2 on: November 22, 2008, 09:50:31 PM »
A) Why do you need a gyro.
B) Are you sure you need 12V @ 25A (300W)?
C) Simple? This thing has enough power to melt 16 gauge wire http://www.powerstream.com/Wire_Size.htm
D) Inexpensive... I guess ArcMan already covered this...

Offline Ro-Bot-X

  • Contest Winner
  • Supreme Robot
  • ****
  • Posts: 1,431
  • Helpful? 25
  • Store: RoBotXDesigns.ca
    • Ro-Bot-X Designs
Re: 20-25 amp servo system.
« Reply #3 on: November 22, 2008, 10:29:25 PM »
You can build your own closed loop servo system with a microcontroller, potentiometer (connected to an A2D pin) or a shaft encoder, and a powerfull motor driver (PWM controlled). You can also add the Gyro to the control loop. All you need to do is find a powerfull enough motor driver, the rest is easy.
Check out the uBotino robot controller!

Offline TreetpflyerTopic starter

  • Jr. Member
  • **
  • Posts: 7
  • Helpful? 0
Re: 20-25 amp servo system.
« Reply #4 on: November 24, 2008, 12:14:19 PM »
Ok.  $250 is what I would consider as cheap.  I need it to run on 12 volt(car battery).  Any idea as to where I could find these parts?  What type of controller would I need?  Would this require alot of programming?  Thanks.

Offline ArcMan

  • Supreme Robot
  • *****
  • Posts: 519
  • Helpful? 4
  • Mmmm... Plasma
Re: 20-25 amp servo system.
« Reply #5 on: November 24, 2008, 12:27:13 PM »
OK.  I would use a Dimension Engineering Sabertooth drive to run the motor, a high-quality pot for position feedback and a microcontroller for position control.  Wire your position setpoint pot into the micro as well.  I would tie the position feedback pot to the motor output shaft with a toothed belt.  Then implement a PID control loop in the micro.  Viola.

Offline TrickyNekro

  • Contest Winner
  • Supreme Robot
  • ****
  • Posts: 1,208
  • Helpful? 15
  • Hardware and Firmware Designer
    • The Hellinic Robots Portal
Re: 20-25 amp servo system.
« Reply #6 on: November 24, 2008, 01:08:08 PM »
I wouldn't rely on pots... They tend to loose accuracy pretty fast... The best method is with an optoelectronic IR sensor pair...
Like QRD1114 or emitter - receiver tower style... I don't know how they are called sorry for my English...

But optoelectronic are better, and even better are hall effect sensors but I have never used them to help you....
For whom the interrupts toll...

Offline ArcMan

  • Supreme Robot
  • *****
  • Posts: 519
  • Helpful? 4
  • Mmmm... Plasma
Re: 20-25 amp servo system.
« Reply #7 on: November 24, 2008, 01:59:05 PM »
I believe you're thinking of optical incremental encoders.  Not a good choice for a servo positioning system.  They don't know their absolute position - so if you power off your system, it would require re-homing.  Absolute encoders, which always know their absolute position, are important for servo positioning.  Now you can get an absolute optical encoder or a resolver, but they are pricey.  That's why a pot is the best choice for a budget servo positioner.

Offline TreetpflyerTopic starter

  • Jr. Member
  • **
  • Posts: 7
  • Helpful? 0
Re: 20-25 amp servo system.
« Reply #8 on: November 24, 2008, 04:10:52 PM »
Ok, I see what your getting at.  I was looking into that particular drive.  Then a more powerfull version of this..

http://www.geocities.com/bourbonstreet/3220/bigservdrv.html 

On the servo driver, I was wondering where the pot signal wire would go.  Then somehow add an R/C gyro between the position pot and the controller to mimic the R/C setup.

Or buy a microcontroller that I can somehow add the gyro to.

Offline ArcMan

  • Supreme Robot
  • *****
  • Posts: 519
  • Helpful? 4
  • Mmmm... Plasma
Re: 20-25 amp servo system.
« Reply #9 on: November 25, 2008, 09:37:57 AM »
I don't mean to sound like some kind of techno-elite here, but 555 timers are soooo 1970's.
You can get much better accuracy, stability and circuit simplicity with a microcontroller.

Offline TreetpflyerTopic starter

  • Jr. Member
  • **
  • Posts: 7
  • Helpful? 0
Re: 20-25 amp servo system.
« Reply #10 on: November 25, 2008, 11:07:44 AM »
Thanks for the help.  As for 555 timers, very old technology I know.  That's why I'm here, to get advice on my project.  I've been looking at the Stamp microprocessor for the project.  I'm not a programmer at all, but I am learning quite a bit.  Originally I was looking at simply making a "tough" positional servo system just like the current RC systems.  It would be a stand alone system that would simply correct for yaw.  Just like that of an RC helicopter with a heading lock gyro and servo setup without the reciver.  It would just be more powerful.  Think of it as a system for a full sized car as compared to an RC car. As for the gyro, if the rear wheels brake loose and slide to the left or right, the steering would turn into the direction of the slide, just as you would while driving a real car.  Or maybe full size go cart.  I'm starting to understand that this is alot more involved than I originally though.

Offline TreetpflyerTopic starter

  • Jr. Member
  • **
  • Posts: 7
  • Helpful? 0
Re: 20-25 amp servo system.
« Reply #11 on: December 14, 2008, 05:08:29 PM »
ok,

So far I have a Arduino Duelmilanove micro controller set up for a hobby servo(to start working on the programming).  My plans are for the driver mentioned above.  The gyro is set between the servo and the controller for now.  It works ok.  I would like to install a start/stop switch for the program next.  Also the drive does not have the signal wire like the hobby servo so I need more info on how/where to put the signal return to the controller.  As for PID, it sounds great but I have no idea as to how or where to put it into the code.  Any help would be great!

Thanks,

Offline Soeren

  • Supreme Robot
  • *****
  • Posts: 4,672
  • Helpful? 227
  • Mind Reading: 0.0
Re: 20-25 amp servo system.
« Reply #12 on: December 14, 2008, 06:05:53 PM »
Hi,

Originally I was looking at simply making a "tough" positional servo system just like the current RC systems.
[...]
I'm starting to understand that this is alot more involved than I originally though.
A KISS approach would be to use the chip from a servo and just add power MOSFET's to drive the larger motor (wouldn't get anywhere near the $250 mentioned, especially with free MOSFET's :))

However, I have to agree that a potentiometer should not be  used for anything that could potentially harm people if it failed, since it _will_ wear down sooner or later (I guess sooner), so a digital feedback would be proper (and some second line of defense, like a fail safe kicking in if certain error conditions are met).

How many degrees will the hydraulic control arm move (or how far will the point of control travel)?
How fine a resolution do you need?


If you just need open and close, perhaps with a sort of mechanical start-slow-then-speed-up device, it would be even simpler.


And don't be afraid to use the 555 if that's the best/simplest solution. I _am_ techno-elite and I will, at any time, prefer the well tried and proven stuff, which is rid of infant symptoms and micro-code bugs, not to mention your own code, which will show off the bug, that you didn't catch when debugging, at the most inappropriate instant - at least when dealing with stuff that could maim or kill if loosing control.

Real elite stuff isn't about using some particularly new technology (hey, the concept of microcontrollers aren't new either, I build my first PC-like computer back in the mid-seventies, when most of the users here were just a naughty thought in a teenagers mind ;)), it's about the results you get (even with a string and a paperclip if that's what you have handy).

If you know that you're an ace on micros, go ahead and have a blast, but if not - stay off them for potentiel lethal devices - you don't drink and drive either... I hope.
Regards,
Søren

A rather fast and fairly heavy robot with quite large wheels needs what? A lot of power?
Please remember...
Engineering is based on numbers - not adjectives

Offline TreetpflyerTopic starter

  • Jr. Member
  • **
  • Posts: 7
  • Helpful? 0
Re: 20-25 amp servo system.
« Reply #13 on: December 14, 2008, 06:38:39 PM »
Thanks for the help,

As simple as an off the shelf system can be.  I've got my Micro controller working well so far for just the hobby servo.  I was thinking of just adding a mechanical switch to the signal wire to the servo.  This would give me an on/off function to the system, yet keep the controller going so when I flip the switch it will become active again.  The actuation would be through around 90-120 degrees.  It's an idea that I've been playing with for a camera platform on an aircraft.  For taking video that will stay horizontal, so that aircraft bank will not be seen in the video.  It's also could be used for stabilizing other camera types keeping it vertical during ground mapping.  The use of a heading lock gyro between the controller and servo should give me what I want.  Just need it to handle around 100lbs.  The reason for hydraulics is that I want to install a solenoid valve between the two cylinder supply lines so that I can flip a switch to allow it to be "rolled" back and forth for installation and removal for different cameras quickly.  I'm starting out as simple as I can and "grow" the system until it's ready.  I'm starting out with one axis, then moving on to more once I get things perfected.  I know that gyro stabilized cameras mounts are commercially available, but they are far too expensive for my project.  And anytime you add the word "aircraft", insurance companies start freaking out!

 


Get Your Ad Here

data_list