Servo 2 positions with single 555 - fast

Thread Starter

bio88

Joined Feb 17, 2011
34
Hello - been reading up on servo tester/controller circuits using the 555. Here is one I am considering. They use a pot and I assume that is used to operate the servo through a range.

Well, say you have a design that requires a servo to travel 90 degrees - and have 2 pts. zero and 90.

Can you first establish two points using the POT and then substitute with 2 resistors and a switch?

Also, if a given servo is rated to travel say, 60 degrees in .09 ms will the 555 servo controller circuits be able to run the servo that fast? Or putting it another way how does the 555 circuit above control and handle speed of a given servo?

Lastly, if you needed to incorporate a delay into the circuit how would that be done if it could be done? Example go from 0 to 90, pause for 5 seconds return to 0...
 

jpanhalt

Joined Jan 18, 2008
11,087
Hello - been reading up on servo tester/controller circuits using the 555. Here is one I am considering. They use a pot and I assume that is used to operate the servo through a range.
That pot is used to set the trigger level of the circuit. Note that it also uses an LDR to sense light levels. If all you want is to control movement of a servo with a resistor and switch, there are better or simpler circuits for that. What is it you are trying to do?

Well, say you have a design that requires a servo to travel 90 degrees - and have 2 pts. zero and 90.

Can you first establish two points using the POT and then substitute with 2 resistors and a switch?
Yes.

Also, if a given servo is rated to travel say, 60 degrees in .09 ms will the 555 servo controller circuits be able to run the servo that fast? Or putting it another way how does the 555 circuit above control and handle speed of a given servo?
The 555 does not control the speed of a hobby servo. Hobby servos move at their own design speed unless you use a more complicated circuit to control them. That is usually done with a microcontroller.

Lastly, if you needed to incorporate a delay into the circuit how would that be done if it could be done? Example go from 0 to 90, pause for 5 seconds return to 0...
That is quite doable. I sense you have very little experience with servos and think you should begin by describing what you are trying to do, then get the servo part working, then go on to adding delays.

John
 

Thread Starter

bio88

Joined Feb 17, 2011
34
If all you want is to control movement of a servo with a resistor and switch, there are better or simpler circuits for that. What is it you are trying to do?
Thanks John -

I need a servo to travel at its max specified speed from say, 0 to 90 degrees and back to 0.

I would like to use a momentary switch to start the cycle moving to 90 degrees.

Then a pause/delay at 90 degrees for some set time.

Then return to 0.
 
RC servos generally arn't that accurate, at least the cheap ones I've used arn't ....
Some will respond to a single pulse, others need a continual string of pulses usually with a low duty cycle. (Short pulses big gaps)

There are 2 things that will directly affect accuracy, the length and quality of your pulses and the quality of the pot in the servo.

Speed is usually a function of the servo its self as was previously mentioned.

Have you considdered a solanoid it sounds like it might be a better fit for your requierments. Connect one to a mechanically constreigned crank, possibly even two so you have positive drive in each direction.

Al
 

Thread Starter

bio88

Joined Feb 17, 2011
34
The accuracy is not important actually +/-5 degrees would be ok. It is for opening and closing a small 6" W by 20" L door

I have experimented using both AC and DC solenoids. DC had a very very small stroke for rated force. While the AC was rip roaring strong, fast and noisy just about pulling the little door off its hinges. Also it is AC and this is an outdoor application so I don't want to have to deal with UL or Nema 4 enclosures etc.

I have 2 prototypes running right now - both actuating as needed but on simple manual switches. One is the AC servo and the other is this one using a DC motor, pinion and spur gear with DIY slip clutch.

The DC motor is the most smooth but it is only turning a smaller 6x6 inch door/flap 90 degrees. What a waste of revolutions right? Only a 1/4 turn of the spur gear? So you can see that perhaps a servo would be best.

Any help appreciated - this is a cool invention and I hope it turns out well.
 

jpanhalt

Joined Jan 18, 2008
11,087
A servo will give you the accuracy and power over distance that you need. Here is a simple schematic for a servo driver taken from another thread on the forum. I have built it x3 and it works. See comments below.



Note, R3, R5, and R6 were chosen to give 1.1 MΩ. A 1 MΩ might work, as will two, 2.2 MΩ resistors. The pot (R2) and be substituted by a SPDT switch and two resistors, as you asked in your first post.

My suggestion then and now is to get this part working, then work on the delay. You will need more parts to make the delay and a clearer definition of exactly what you want to see as well as what you mean by or intend to use as a switch.

What is your level of experience building circuits?

Can you post the circuits for what you have running right now?

John
 

Thread Starter

bio88

Joined Feb 17, 2011
34
Thank you. I've not built beyond breadboarding. And even then it was LED circuits etc. I think I can read simple schematics and then put together and debug a circuit. I'm more build to print, plug and chug.

So I would likely take your schematic and try to maybe add this one to it to achieve a momentary to latching switch/
 

jpanhalt

Joined Jan 18, 2008
11,087
I suspect that you do not have an oscilloscope. That is why I suggested a step-by-step approach. It is easy to design a nice circuit on paper, but quite another story to get it to work with just a voltmeter.

As for your delays, I suspect a simple RC timer or one-shot (555-based, if you wish) and a relay will be the easiest to get going. You can substitute mosfet switches to avoid the relay, and it is possible with a little mechanical engineering to get the open and close action with a single resistor that is put in parallel with the resting state resistor...

In the circuit I provided, the servo should move CW (looking from the top) as the resistance is increased (pot turned CW). Putting a resistor in parallel will always reduce the net resistance. Hopefully, that will help with designing the geometry. Servos can be electronically reversed easily, and at some places, you can buy reversed servos, if that geometry doesn't work for you.

John
 

Thread Starter

bio88

Joined Feb 17, 2011
34
Interesting. Perhaps I can achieve even more simply with a DC motor, momentary sw's and 2 one-shots acting independently.

Assuming motor has clutch and/or hard stops so it doesn't burn out, using a momentary SW and one shot on motor turns CW for 1 second. End of stroke momentary makes contact and one shot to run motor CCW for 1 second.

Forgot the a delay in between motor CW stroke and motor ccw stroke.
 

jpanhalt

Joined Jan 18, 2008
11,087
You will never get even close, repetitive positioning with a typical (i.e., non-stepper) DC motor unless you add feedback. Then, you will have re-invented the servo. You could run the motor until it hits a stop (e.g., microswitch), then reverses to the next stop, etc. That is basically how your garage door operates.

Again, I am not sure I understand what you want to do. Is it a garage door?

John

PS: A one-shot typically produces just a signal. You will need to do some designing, like adding relays, to get enough current from it to run your motor. That gets you back to the garage door.
 
Last edited:

Thread Starter

bio88

Joined Feb 17, 2011
34
Here is a perfect example. Let's say it is an automatic dog/pet door that is rigid and opens to the outside. Let's also assume the dog can only go out, thus it is one way door, it has hard stops at 6 o'clock (home) and 9 o'clock (fully open). Cool?

Now dog walks up and some thing like this infrared prox sensor opens the door via latching relay and DC motor and gearbox.

Door is held in 9 o'clock postion by rare earth magnet. Dog exits. 5 second delay so that door doesn't close and trap dog on its way out.

5 seconds have passed, dog is well clear so door held at 9 o'clock hard stop now returns via motor power to home/6 o'clock. (I know gravity would return the door to home but we need motor power)

So lastly, as door returns to home hard stop there also needs to be a signal telling motor to stop, etc.

I appreciate all the help - I've proven the concept with me flipping switches manually but it really needs a nice circuit or means of doing this automatically.
 
Last edited:

jpanhalt

Joined Jan 18, 2008
11,087
If you get the mechanical part done, the electrical with microswitch stops and a delay is easy.

I once built a garage-type door that was 50' wide by 18' high and it was entirely controlled by limit switches. The part you might have trouble with is holding it open. In my case, a worm-gear winch was used, so it had good holding power when the motor was not energized. For your smaller door, I would also recommend that you use a worm gear drive for holding power. You will have to slow the motor down anyway.

For the delay, you might consider a simple mechanical timer off the shelf. I suspect electronic ones are also available.

John
 

Thread Starter

bio88

Joined Feb 17, 2011
34
Exactly John - I'm determined to achieve this application like the simple useless machine that another person made with mechanical switches and without using a 555 timer. So mine is like a useless machine - with a use ;)
 
Top