I am extremely new to robotics being 58 years old, but was hoping
someone here might be able to give me some guidance;
I'm working on an idea that works much like a seesaw, is there a
sensor that would send a signal to a servo or other device that would
work 180 degrees from the signal? Meaning if a seesaw was up this
would send a signal to offset that motion in direct proportion to it
wioth the intent to always keep it in a level state? I'm so archaic I
was thinking that a rheostat might be able to do it but I do hope we
are way beyond that device for this problem.
Any and all help is greatly appreciated.
There are a range of options to do what you want. But the most
useful for hobby-type work are sonar and infrared (IR).
An IR distance sensor can output a pulse or voltage level that's
proportional to the distance between the sensor and some reflecting surface
(such as the ground). They work up to 10 meters (usually at
the expense of looking like a pair of binoculars) and usually give you
a continuous output.
Simple sonar modules can do the same thing. But they are periodic
and normally require a little bit more of control circuitry to set them off and
interpret the output. Again, they typically generate an output pulse with the
width proportional to distance, but many can also generate a voltage level,
or even send a packet of bits that encode the distance as a small number.
There are dozens of sonar modules around. They can work upto 10 m or so.
Both options are low-cost i.e. <20 USD many places.
Next you'll probably need something in the way of a microcontroller to
read the output from your sensor and control some effector like a
motor or servo. Typical of how many would approach this work are the
numerous boards based on AVR8's. Want to learn some assembler programming
and binary programming? You may need to.
On the effector side, you need a servo controller (cheap servos usually
need a pulse of a certain width to move to a particular position in
their range of -- say -- 0 degrees up to 100 degrees).
Some AVR8 boards may have control circuitry for a few servos, or you
may need an extra little board -- usually about 10mm x 10mm for 1
or 2 servos -- to do the work via some kind of serial interface to
the AVR8 board.
Not to give too much of a plug to anyone, but I've worked with Pololu
parts for years and they are pretty cheap and have enough doc for
most old codgers with tinkering-in-the-back-shed genes to make head or
What profiteth a man if he gain the whole world but lose sight of the
basic results of multivariate decision theory?
There are a number of solutions. The one that will probably carry
you the furthest forward is to start learning how to use
microcontrollers. The microcontroller that seems to have the
most "buzz" these days is the Arduino. If you live near a
metropolitan area, I would recommend trying to find out if
there is a local robotics club. That is an excellent way to
get going. Failing, that join some of the robotics club mailing
lists -- HBRC (Home Brew Robotics Club), SRS (Seattle Robotics
Society), DPRG (Dallas Personal Robotics Group) and start asking
questions. Failing that, visiting your local library and seeing
what they have in starter books is worth doing. Many of got going
via Gordon McComb's Robot_Builder's_Bonanza. You should seriously
consider getting a subscription to Servo Magazine as well.
This list used to be much more active, but it got seriously nuked
a spammer and most people have left.
P.S. Full disclosure, I'm currently president of HBRC; hence,
my bias towards robotics clubs.
Yes a rheostat would do the job, but the modern terminology would be a
potentiometer for something at low power. If you were to take apart
the servo motor in a radio control car, you would find that a
potentiometer is used to measure the position of the output arm.
Modern hobby servomotors are available in a great range of strengths
and speeds. You might even take one apart, and remove the
potentiometer, or simply cut its wires. Then, measure the resistance
of the original potentiometer, and find another with a duplicate
value. You would mount this new potentiometer so that it connects to
your seesaw. Then to make the whole system worm, you would need to
feed the servo motor the correctly pulsed signal to stay centered.
There are pre-made devices for this which were designed for the radio
control hobbyist to test their servo motors.
I don't know your application, and the ones Mr. Gramlich mentioned the
ultrasonic or infrared sensor might be a better choice since they
don't wear out like the potentiometer would. Also, the servo motor I
mentioned would constantly be drawing power if the seesaw experienced
a constantly unbalanced load.
An optical encoder (as used by pre-"laser" computer mice) might also
Depending on your needs, simple limit switches may suffice (they're
also cheap and generally reliable -- see the buttons on your mouse).
Limit switches may also simplify the circuitry if you do use a rotary
encoder (otherwise you have to track the velocity). Tilt right until
one switch is hit, then start tilting left.
Assuming the seesaw should hit both limits, the rotary encoder only
adds a proportional control.
To avoid microcontrollers, a couple DPDT relays (or transistors) could
implement the required switching. If your rotary encoder has a
variable-voltage or variable-resistance output, then it could drive
the gain of a transistor or other power amplifier.
P.S. Here's a nice internal view of a microswitch.
You probably just haven't told us enough, but based on your description,
the best solution is to use glue. That is, glue the seesaw so it can't
move. That automatically adjusts the force needed to keep it level. :) :)
(everyone else already provided you with "good" ideas).
Polytechforum.com is a website by engineers for engineers. It is not affiliated with any of manufacturers or vendors discussed here.
All logos and trade names are the property of their respective owners.