T&L Publications

Hello,

I am new to the board, and just getting used to how to find things on here so forgive me if I ask about things that have already been posted... at least until I get my sea-legs

I am trying to set up a very simple circuit for my new servos to be controlled by a DPDT switch to act as a forward and reverse control.
In my case I am going to run a Hitek HS-55 with a 6-volt battery pack to be controlled by an SW12-ND DPDT MOM/OFF/MOM rocker switch.
Servo Wires:
- Signal/Control = yellow/blue
- Positive = red
- Negative = black

I know how to jumper the DPDT switch to act as a forward-reverse control, but I am worried about frying my servos.
How do I connect the servos (2) to the 2 leads of the switch without causing smoke?

Thanks in advance!

Phillip

Hi there,


Welcome to the forum.

You mean your servo legs? (little pun on how some servos vibrate back and forth a little)

The servo i have requires a different pulse width to turn the shaft left or right.

LOL, Servo legs

Trying to get the wiring down for this sort of thing...
http://www.youtube.com/watch?v=iS2yRKO-5qk

I know he is using only 2 servos (One on each side of the visor at temple level)
And there is no controller board. He has somehow wired the servos direct to a DPDT switch... not unlike a power window in a car.
This type of switch-servo interaction is what I am trying to capture.

Hi again,


Yeah, he he, servo legs, as if you were standing on one of those servos that was vibrating

I wouldnt mind hearing more about this DPDT method either. I have a servo i would like to run
that way too without having to use a controller chip. If i use a controller chip i can do it
any way i want to so i dont need a DPDT switch (two momentary push buttons) but would
be nice to be able to do it without a controller chip.

If there is room for the servo, then there is room to hide a dual 555.
First 555 oscillates at pulse rate. Other, triggered by first puts out pulses for one position of servo. A simple on-off switch parallels a second capacitor with timing cap of second 555; this changes pulse width, and servo moves to other position.

For intermediate positions with mom-off-mom switch, I don't see anything less than a 8 pin uP to "remember" last toggle of switch, and put out a stream of pulses with width for current position. Smaller than a dual 555. IO pins left over for other functions.

A SOIC package could be glued upsidedown to servo and wired dead bug style.

Cheers,

Most R/C servos expect to see a 50 Hz pulse on the signal line, with a positive pulse width close to 4.8V and variable from 0.5 to 1.5 milliseconds. The pulse duration controls the servo position.

The red wire must be battery (+), usually around 4.8V. The black wire must be wired as (-) ground. The control pulse is sent on the third wire.

The 555 idea is good. The switch can be set up to change the resistance and pulse width of an astable multivibrator. Or one side of a 556 dual multivibrator can be set up as a 50 Hz oscillator and the second stage as a monostable to vary the pulse width.

I have never used a 555, so that is a far as I can help, but "It's all been done before". Have a look at the servo drivers on this page:

http://www.sentex.ca/~mec1995/gadgets/gadgets.htm

You may also check the links found on this page:

http://www.sentex.ca/~mec1995/index.htm

It's probably a modified servo. I have (had?) a few servos that had the internal electronics disconnected and the rotation limiter removedso they were basically small gearmotors in R/C servo packages. Once that's done, it's trivial to wire up a DPDT switch to get direction control.

MrAl wrote:You mean your servo legs? (little pun on how some servos vibrate back and forth a little)

I had a talk with a PHD in Engineering. That vibration or "hunting" is instability and has only two causes, (if you don't count the 50 Hz RC signal vibration):

1) Excessive loop gain
2) A "dead band" ie: gear lash. The correction signal will cause the mechanism to continuously overshoot the target position.

Good R/C servos have less gear lash than the el cheapo's.

Thanks all... I think I have the answer as one of you touched on...
The only way to accomplish the task without using a 555 or other type of controller is to lobotimize the servo. This renders it a dumb DC motor with a nice gear reduction package. The motion will be controlled manually via the DPDT switch.

Traditional RC servos have 3 leads: black for ground, red for +6 VDC (5 will work), and orange or yellow for a 50-60Hz pulse train. The width of the pulses determines where the servo wants to be. If you don't apply pulses, the servo position doesn't update. The range of pulse width is from about 0.5 milliseconds to about 1.5 milliseconds.

If the width of the pulses sent to the servo goes outside of those boundaries, it is possible to strip the gears. (Many servos have plastic gears.) Some higher end servos are not fussy about the pulse rate. The pulses can be sent at rates such as 300 Hz. Other servos will act strange if the pulse rate is outside the range of 50-60 Hz.

I learned all this years ago when I made a servo "stimulator" for testing servos installed in an RC plane I built. The "stimulator" circuit consisted of a PIC16c71, a crystal, a bypass cap, a 5k ohm potentiometer, and 4 AA batteries. The PIC16c71 contains an ADC, so a simple program can read the pot position and generate the pulses for the servo. The servo position follows the pot position.

I'm new to this forum, so I'm not familiar with how such info is shared here. If there's interest, I can post a schematic and source code for the stimulator.

-Mark

When I say that servos will "act strange" if the pulse rate is outside of 50-60 Hz:

If my servo stimulator ran with a pulse rate of around 300 Hz, the Futaba servos I had worked just fine, but the HiTec servos would "hunt" and oscillate back and forth around the commanded position. Who knows why, but I just modified the program to send pulses at about 60 Hz and all was fine.

not sure if this helps or not...i am sure there is a tester here in the states available also..
http://www1.conrad.de/scripts/wgate/zco ... results=20

WOW ! long address line..lol

anyway this one that i use.. i have 3 of the kits for some weird reason..
http://www1.conrad.de/scripts/wgate/zco ... results=20

this uses the HEF4001BP IC and a 50K pot, a few other discreete components. its useful for positive or negetive pulsed servos. will work with any servo it claims. as far as model use goes.
ok good luck..