Linear Actuators and control method

Discussion in 'The Projects Forum' started by GtoRx7, Feb 19, 2011.

  1. GtoRx7

    Thread Starter New Member

    Jan 8, 2011
    10
    0
    Still working on a project to control intake length, but was thinking instead of a R/c servo, to use a linear actuator instead. I found the specs for the actuator I need (70mm/sec and 110N force @ 12v) but what method do they get controlled? I see you can get them with potentiometer feedback, does this mean they can operate similar to a servo? Such as x signal = 1in length, y signal = 2in length so on so fourth? Here are the ones I was looking at-

    http://www.warnerelectric.co.uk/Electrak_LA1-S___LA1-SP.pdf
     
  2. beenthere

    Retired Moderator

    Apr 20, 2004
    15,815
    282
    You may notice there is an electric motor involved in driving the actuator. I have used one in which I had to blacken teeth on the drive gear train and use an optointerrupter to count revolutions, which could accurately be related to actuator extension/retraction.

    If the company included a shaft encoder, then you get the same data, only with finer accuracy.

    Instead of driving to a balancing resistance, you count pulses and stop the motor when the count equals the desired distance.
     
  3. John P

    AAC Fanatic!

    Oct 14, 2008
    1,634
    224
    Yes, in principle you could use these units in the same way as a hobby servo, but bear in mind that the hobby servos are very small and inaccurate. When you scale everything up, you might notice the limitations. And of course, you would have to provide the electronics to link the feedback potentiometer to the motor drive.
     
  4. debe

    Well-Known Member

    Sep 21, 2010
    946
    184
    Have seen those 12v linear actuators with in built potentiometer used for an auto pilot on a boat. worked quite well the actuator is conected to the rudder by a small arm, the pot provided feed back to the autopilt as a servo system.
     
  5. GtoRx7

    Thread Starter New Member

    Jan 8, 2011
    10
    0
    So out of the box, a linear actuator just has a positive and negative wire? Feed it 12v to reach extension, reverse polarity to retract? And if that is the case, is it safe to assume the potentiometer versions simply have a pot connected to the ram tube and outputs the variable ohm reading?

    I like the idea of counting teeth, not sure how to implement it. For my project it definitely does not need to be extremely accurate. Going to use it on a variable intake length setup. for example at 5000rpm of the engine, the actuator would be at full extension aka 5 inches, 6000rpm 4 inches, 7000rpm 3 inches, 8000 rpm 2 inches so on so fourth.

    Maybe someone has a better device to use considering my control signal available? The programable computer in the car has a output, which is a duty cycle of 60hz or 90 hz. 0-100% duty. The signal is a ground or negative pulse. I can easily attach to a solid state relay and make this a positive 12v pulse, or 5v pulse instead. Any ideas to avoid complex systems?
     
  6. Bernard

    AAC Fanatic!

    Aug 7, 2008
    4,172
    397
    Some ramblings of senior citizen: Supose we call 5 in extension a 1 in opening. a motor driven threaded rod with 1 mm pitch would be about 127 mm, with sensor giving one output per revolution feeding a counter. A second counter with about one second gate time counts input pulses, ie 8k rpm = 133 pps; both counters feed a comparator like 7485. Count hi or lo give drive motor direction attempting to make thread count = PPS. Non linearity to be covered by others.
     
    Last edited: Feb 20, 2011
  7. Dyslexicbloke

    Active Member

    Sep 4, 2010
    420
    19
    If your not trying to be super accurate a feedback pot will work well, assuming its of reasonable quality. It has one advantage over pulse counting in that you always know where things are so you will not need to drive to a stop in ordert to calibrate and start counting after power up.

    It is relativly easy to build a voltage comparitor that drives a couple of PWM generators, based on the error, out of a few standard opamps.
    You would end up with a voltage controlled positioner ......

    I am about to build this circuit to run a modified RC servo but only the bridge componants would change for a bigger version.

    If your interested I can let you know how I get on.
    Al
     
  8. GetDeviceInfo

    Senior Member

    Jun 7, 2009
    1,571
    230
    I would be interested to hear what the mechanics are for adjusting your runner dimensions. This would also help in determining appropriate linkage and drive requirements. Feedback may be a function of manifold vacuum as opposed to engine speed, with provisions for heavy accel/decel.

    As a small group of wanna be racers, we played with air throughput, which included tuned runners, turbos, etc.
     
  9. GtoRx7

    Thread Starter New Member

    Jan 8, 2011
    10
    0
    I am able to use both manifold vacuum and rpm with the computers output, but most likely will be using only rpm. I am building Air horns for the individual throttle bodies which will slide in and out. 70mm/sec would be fast enough, and 5 inches of travel would be enough adjustment. Its on a race mazda rotary engine, and very much same principle as the 787b engine for the guru's out there.

    Anyway, after doing more searching I found linear actuators that move in and out based on voltage input. 1-5v controls position. This would be perfect, I would simply need a converter to change the negative duty signal from the car computer to a voltage output. Anyone have a circuit idea to accomplish this?

    Output from computer is 60hz with 0-100% duty cycle, negative ground pulse.
     
  10. Dyslexicbloke

    Active Member

    Sep 4, 2010
    420
    19
    I am about to try and build a sample and hold circuit that will generate a voltage proportional to the input pulse width .....
    My requirement is 40 to 60 Hz producing a 0-10v signal and I am going to try with opamps and analouge switches ...
    I will let you know how I get on if your interested.
    (I guess you dont need the PWM positioning circuit any more right?)
    Al
     
  11. GetDeviceInfo

    Senior Member

    Jun 7, 2009
    1,571
    230
    Being a bit of a mechanic, I'd consider a scissor actuator for each tube, with a common center link that connected to a bellcrank. This in turn is driven by a geared motor. I'd think for reaction time you'd need a motor capable of holding a position, as gearing wouldn't be high enough for self holing. So maybe a stepper, or frequency positioned servo.

    As far as your computer output being modulated, I'd buffer it, average it through a filter, then if required, scale it, to produce a reference position signal. Leave room for summing other signals (scaled if required) 'cause I still thinking vacuum.
     
  12. GtoRx7

    Thread Starter New Member

    Jan 8, 2011
    10
    0
    Sounds fairly similar to what I need. Keep me updated on what you come up with
     
  13. Dyslexicbloke

    Active Member

    Sep 4, 2010
    420
    19
    I have designed the attached circuit which discharges a capacitor for 1/2 cycle and samples the voltage before charging it again on the following half cycle.

    It seems to work ok in the sim but I am expecting it not to be stable enough when I actually build it.

    Its a good exersize though and significantly simpler than aranging a PIC to produce an analouge signal and programming it to set a voltage based on pulse width.

    (Its based on an NE556 dual timer and an MC14016BCP quad switch)

    Al
    [​IMG]
     
Loading...