Variable Speed Step Motor

Well it might help if you identified which stepping motor you've picked or are we supposed to be playing 20 questions for us to help you do your project. And why in the world would you use a stepping motor for a variable speed application, stepping motor are supposed to move in steps. Also for any motion control system to be "very precise" you should use some form of a closed loop system. Otherwise all that you will actually know is what speed or position you are requesting, not what you actually have.

I know, your actually looking for an excuse to fail here. Well I'll grant you that.

From what you've told about this complicated job, it cannot be done.

ok,very good points! i don't know very much about step motors let alone controllers (I'm just a home hobbyist) so now that i think about what you said its clear to me that i need to use a brushless DC motor with a digital pressure reading in the piston that runs to a microcontroller to control the rpm of the motor to maintain the right pressure. this is what you mean by closed loop right? kind of like a PID system for pressure. this way its much less electronics and i still have the accuracy i need.

Exactly my point, one should monitor directly the parameter of interest. This sounds like a much better design approach. But I do wonder why you and many people immediately jump to the use of a brushless DC motor. There are many very nice brushless motors with built in commutation control. But what is wrong with using a simple brushed servo motor? Now if you happen to have a brushless motor sitting on your desk, like I do, then by all means use this more complicated package. But I suspect you can be much more frugal using a brushed DC motor here.

Is this a good place to start:-

http://www.galilmc.com/learning/tutorials.php ?

They look reasonable. But being a manufacturer product site, you have to realize their focus will be for you to buy their product. Naturally they want you to succeed with their product. But I suspect that this site expects a lot of established fundamental knowledge, so it may not be a place to start. For stepping motor fundamentals Professor Jones maintains a great tutorial. For brushed DC motors, Microchip has a nice application note. But these will obviously pale in comparison to a formal class on electric motor fundamentals at your local university.

If he uses a pot to set the speed, the reference precision will be worst than 5% (unless he uses a 15 turns pot).

Then a simple op amp with a PI feedback on your target parameter will do the trick. Use a cheap DC motor found in almost any scrap machine.

What is the reaction time needed? For a small system, in slow reaction mode (>5 sec), a very simple PI feedback will be very precise for almost anything. If fast (<1 sec), it will be more difficult.

Given his know how, as simple as possible is is more likely to succeed.

Good luck.

A stepper controller driving a stepper motor is controlled by pulses.

If you use a 200m steps per revolution stepper you need to supply 200 pulses per min to achieve 1 RPM. For 200 RPM you will need 200x200 pulses per min.

Think about a 555 to do the pulses.

yea, it seems that all this stuff about steps and pulses is making this project more complected then it needs to be. and in alignment with what redfred said i am starting to think it may not be anywhere as effective as i need it to be. thanks for the great advice!

Sound like some type of dispensing equipment. A gear pump would work better. Direct connection to the motor, a known displacement for flow and a sensor can be incorporated to control pressure. A continuous controlled flow

A piston has to stop and retract to draw in and refill. Even if your piston pump draws and pumps in both directions there is hesitation at the ends of the stroke.

It sounds to me like you are looking for a "chemical injection pump", which are quite common, although expensive. One example of a manufacturer would be http://www.sidewinderpumps.com/

If you are not interested in purchasing a relatively expensive ready-made solution, you may want to have a look at how others accomplish this to get some good ideas. Google "chemical injection pumps" for 856,000 suggestions...

Stepper motors run on frequency (pulses/min). You want to have manual interface of potentio meter.

Thus

V to F (Potentiometer analog to frequency) converter to suitable pulses.

Then you have to confirm if the stepeer motor can accept 200 X 200 pulses per minute.

If with the 200 rpm and the gearing required to drive the load of piston and the hydraulic pressure, motor can drive the mechanism, it will work.

The pressure depends upon overall hydraulic resistance of your circuit.

If you get the system driven flow rate will be directly proportional to the rpm of the motor.

Computerized control as you want should be possible.

Without knowing what you're trying to do, I wonder if you have considered using pneumatics?

Your system could be a tank with pressurised air forcing the fluid (water) out the bottom, changing the air pressure via elec operated solenoids & pressure regulators would give different system pressures.

Small air compressors, solenoids & pressure regulators are quite cheap and easy to use.

Since the speed you require is 0-200rpm I would suggest a permanent magnet ac syncronous motor.

I have seen these in action and they are brilliant. The ones that I have seen can run of a single phase to three phase variable speed drive (VSD), they operate at full torque over the complete speed range, they dont require a cooling fan and even when stalled do not go over current.

By interfacing with a small Process Logic Controller (PLC) with Analogue I.O you could control the system in real time.

Since when has PLC - PROGRAMMABLE LOGIC CONTROLLER become PROCESS LC??

Woops, I wasnt thinking straight! but you get the idea.

good one! that sounds like just what i need. did some research on plc systems and it looks a lot like a solution to the problem of a main systems controller for the whole machine, as there is a good deal of solenoid valves and relays in my design.

thanks a lot! this is one of the most helpful posts so far! you have me started off on some ideas for a very functional design.

Keep in mind that as the name implies a synchronous motor changes speed with the frequency of the voltage applied to it.

I built a motor controller for a stepper motor using a 'full step' configuration and an HC11 microcontroller to control the motor speed. The idea of a stepper motor is the potential for accuracy. The HC11 can be interfaced to accept l/P (inputs) from any electronic device and be programmed to react in the way that you need to controller the motor speed. The microcontroller is very expensive when compared to a potentiometer though, but gives much flexability in the interfacing. The program for the HC11 is written in assembly language. The stepper motor I used used 12 VCD. Also I used 4 NPN transistors, 4 diodes, 8 resistors, and TTL IC to interface the HC11 to the motor control curuit. I have a schematic that worked in controlling the motor.

scotty_f

Check out Automation Direct www.AUTOMATIONDIRECT.com They have everything you need for your project at reasonable prices. Their catalog is very informative and they have very good tech support.

I am in no way affiliated with them. I have just been using there products for many years for many successful projects.

Their products are not necessarily the best on the market, but what I like is the choice of components and the fact that they are for the most part compatable with each other.

Good luck with your project. And remember, even if the outcome is not what you expect, you are going to learn a lot from it.

BenJ,

The cylinder with ball screw drive that you describe is a form of an intermittent positive displacement, PD, pump. I think that your liquid flow control requirements may be mutually exclusive. It is a problem to simultaneously control flow with a positive displacement pump and to also modify the same liquid flow with a needle valve. Think what happens when someone closes the valve! The required motor torque increases with total head pressure of your pump. I agree with the earlier reply by ozzb, scrap the ball screw / cylinder idea and substitute any PD pump or metering pump. Be sure to flood the pump inlet with vapor or gas free fluid. There are many varieties available, some with continuous analog adjustment of the displacement per revolution. Many pumps are compatible with water and even aggressive chemicals. You can still drive the pump with a step motor, but you should know that any pump has a lower speed where pump leakage may become dominant.

I suggest that you evaluate the energy required to pump the "water" into your system at the peak system pressure. If you size the motor drive for the PD pump to account for the maximum pressure and flow required with some extra power to account for motor and pumping losses, you should succeed. If you need digital precision, you can get it with an angle encoder on the pump shaft. Would 4096 pulses per revolution be sufficient? They are available off the shelf. For the ultimate in accuracy you should measure the liquid flow instead of inferring it from the volumetric displacement of the pump's rotation. Look at PD meters or digital flow meters such as turbine or vortex shedding for volumetric flow measurement. You could automate a blending operation by measuring the "wild" product flow, but we can only guess what you need, since you haven't told us about the overall process.

Luther M

actually the needle valve was only an example of a way to set an initial "dummy resistance" for test/experimental proposes, not a way to adjust the flow during operation. the actual medium that the water will flow through will be a particulate mass that will have an adjustable grain size and therefor causing a variable resistance according to the "pack" of the particles in said filtered reservoir.