Mixing motor types.

[perfectbite]

Thank you for the very interesting responses on the sylsens motors. Cost and availability put a system like that entirely of my range but I am looking for some kind of 'sympathetic' motor control like that.<p>Miguel made the suggestion that a stepper motor can position another stepper motor. Because this project of mine has substantial inertial loads a 'companion' stepper motor driver would not work. <p>However, could stepper motors (used in place of the resistors in a home built joystick) be used to position a worm drive, carbon brushed, 12 volt DC, open ended (limit switched at each end of 180 deg. travel) servo motor? How would I go about doing that? Would it even be possible? Smooth operation over the joystick's range with no signal to move = no movement would be perfect. <p>Discrete pulses from the re-positioned stepper motor(s) being the P of PWM?<p>Thanks.

[zotdoc]

I think that pwm stands for pulse width modulation, a way of digitizing DC current to allow for motor speed control - voltage stays constant but the motor only sees current with each pulse, the longer and more frequent the pulse, the faster the motor works.

[perfectbite]

Zotdoc you are right. PWM stands for Pulse Width Modulation. (Digital). <p>I have just recently come to understand the difference between digital and analog. <p>Edge triggered or ramp triggered. Is that essentially correct?<p>Right now I am looking at a Forrest Mims' schematic of a Power Pulse Generator using a 555, a 2N222 and a 350T. What if I don't use the 555 and substitute the joystick stepper pulses, fast/slow or none to the 2N222s which in turn will I/0 the 350Ts gates and then, from the 'switching' 350Ts, on to a MOSFET 'H' bridge? What kind of signal would the stepper motor have to produce to switch the 2N222 (or something like it) on and off? Hall effect transistors linked to the stepper motor's poles outputs? Ka-ching. A Hall efect switch at every stepper motor's pole? Be reasonable. <p>I ain't no brain but it seems to me that it could work and you lot would have to tell me if this is a digital or analog circuit. My problem is that I don't know how to get there from here.

[rshayes]

If you have a light mechanical load, you might consider a motor driven potentiometer. All Electronics (www.allelectronics.com)has three types of these as surplus items with 10K and 20K linear elements. The pot would be used as a position sensor to generate a feedback signal which would be compared to a voltage signal from a standard pot. The difference between the two voltages would form and error signal which can be amplified and used to drive the motor. When the system reaches balance, with zero errror signal, the motorized pot will duplicate the position of the other pot.

[perfectbite]

Stephen, The load isn't a light load by any stretch of the imagination it takes a lot of effort working against the device's moment to lift it back to center. The 12VDC windshield wiper motors I am using have sufficient power. Getting them to move when and how I want is the task before me. Robot type control won't work and I really don't want to go the way of PICs or servo control.<p>Quite a while ago there was a post here that said that steppers are decent generators. What kind of signal gets generated when a stepper gets driven? Sawtooth, square wave, sinewave? I don't have an oscilloscope. Does it make a difference if the stepper is bipolar or unipolar? Is there a minimum speed requirement or does each step generate a discrete signal/pulse?

[rshayes]

If you need the torque of a windshield wiper motor, then you should probably stick with that for supplying the driving force. Selsyns and stepper motors are not very good for delivering torque.<p>As a position pickup, stepper motors have two problems. First, the permanent magnet type (the most common type) have preferred positions even when unenergized due to the magnetic field. It requires a fair amount of force to move halfway to the next position, and then the motor snaps to the new rest position. Try turning a stepper motor shaft with your fingers and you will see what I mean. This detent action puts a variable load on the actuating motor and may make it difficult to accurately position the shaft. Second, stepper motors have no absolute position reference. In effect, they are an incremental device. The zero position will have to be established by other means, at least at power on and probably more frequently.<p>When the stepping motor is moved from one position to the next, a pulse is generated in the various windings of the motor. The height of this pulse will depend on the rotation rate. The area under the pulse should be constant. At low speeds, there will be a wide low voltage pulse. This will get narrower and higher at higher speeds.<p>A potentiometer geared to the shaft will give you and absolute position sensor, which can be used with an error amplifier to control the DC motor. The main disadvantage of a potentiometer is mechanical wear. It may last ten's of thousnads of cycles, but it won't last forever.<p>A moving mask that shadows two solar cells illuminated by LEDs can be used as a position pickoff.If the illumination is uniform, the output of a solar cell is proportional to the illuminated area. If the mask is arranged to add illuminated area to one cell while it subtracts illumination from the other cell, the position can be obtained by subtracting the two signals and then dividing by their sum. If the sum of the signals is kept constant by controlling the LEDs with a feedback loop, the no division is necessary. This is more complicated than a potentiometer, but it won't wear out.<p>Optical shaft encoders are available, but usually generate a digital code.<p>[ June 16, 2004: Message edited by: stephen ]</p>

[perfectbite]

Stephen, Thanks again but I think I'll go for low/old tech. Steppers sound wonderful but, as you have pointed out are very limited.<p>My idea now is to build the joystick control using one of those very tough blue squash balls filled with bicarb and superglued solid (per Edd Whatley) so that it has weight with a stalk sticking out of it and to rest that ball on two freely turning cross axis shafts, like a computer mouse (probably 1/2" dowels with double sided sticky tape for traction) and on each shaft use either home built discs with a edge of hot glued small bar magnets or a bicycle chain spur gear that 'trips' a Hall effect transistor as they rotate past. A rotation sensor attached to each shaft will provide direction to the H bridge. That way, depending on the circumferential ratios, I can generate a given number of discrete full amperage, even slightly over voltage 'go' signals to the cross axes motors at the speed and direction the joystick is moved. When the joystick stops, the motors stop. I may not have an oscilliscope but I do have a handy-dandy 6" Craftsman lathe in my workshop. <p>The eventual movement of the limit switch restrained controlled device will be relayed to the joystick operator via a baby room monitor so the system is not entirely without feed back.<p>BTW. Originally I bought the 2.4 GHz baby room monitor as a survelliance device to place on a pole so that I could look at my roof without climbing up on it. I put a stop to it two years ago but before that, when I was at work, my next door neighbour thought nothing of regularly sending her yard man, on a ladder from her side of the fence, up onto my roof to trim her side of my trees without my knowledge or consent! Unfortunately she is quite paranoid and, because I had called the police on her 3 days after the fact, (I knew she wouldn't listen to what I had to say in the matter), she had that whole side of her house equipped with sentry lights and video cameras and is convinced that at night I climb up onto her deck and scratch it with pins or knitting needles. (?!) The other neighbours tell me that she refers to me as "The Prime Suspect." <p>Because I didn't want to aggravate her paranoia or start a survellaince camera 'war' I never used this camera for its intended use and now have this baby room camera which I can to put to a lighthearted use. <p>(It truly is sad. My neighbour used to have her locks changed about every 18 months or so, now it is down to weekly/monthly locksmith's visits. I hope she doesn't start throwing lighted rags onto my roof to exorcise me)<p>Thank you all for all of your help and ideas and stories, as usual, any comments will be very welcome.

[jwax]

Train a laser pointer at her camera!