Flowrate to Drive DC Motor?

You mention the frequency output of a flowmeter to control flowrate but the more common output is current. If this is the case, a very common application is to wire the 4 to 20 madc output signal (represents 0 to 100% of the flowrate) into a setpoint controller which allows you to pick a control point from 0 to 100%. Take the 4-20 madc output and feed it into the motors speed input. If the motor does not accept 4-20 madc as an input but accepts a potentiometer, select a controller with a resistance output. Omega Inc. is a good place to start.

You didn't mention what type of flowmeter you have. If I have to guess, it's something like a Signet flow sensor. These are paddle wheel types that put out a series of pulses whose frequency is proportional to the flowrate.

Okay, if your DC motor speed control uses a potentiometer with all three terminals connected, like so:

you're going to need a frequency-to-voltage converter. The voltage should be equal to V+.

If your speed control has a voltage of 10Vdc, you're in luck since you can get a standard 0 to 10Vdc frequency-to-voltage converter. If not, you'll need to reprogram your speed control. If your V+ is, for example, +24Vdc, and your converter is 0 to 10V, you reprogram your speed so that your maximum speed occurs at 10V instead of 24V. Of course, whether you can do this or not depends on your speed controller's capabilities.

The type of flowmeter that I will try out first is a paddle wheel type and yes the freq out is proportional to the flowrate. I would like to convert the frequency out to either a current or voltage using a signal conditioner. Then use that output to drive a motor. It seems that in the example you gave, the voltage would vary to change the motor speed. I think there would be a torque issue with that, that is why I want to use PWM, so that my voltage is always max. Most PWM motor speed boards I find use a pot for the adjustment, I need to find a way to use my current or voltage output to vary the PWM unit. I'm looking for a way to interface the 2 circuits. I havn't found anything just yet, so any info is appreciated. Thank you.

Most PWM motor speed boards I find use a pot for the adjustment

That's what I meant. The drawing I gave above is the pot that is being used for adjusting the motor speed. Remove the potentiometer in the drawing I gave above and put the voltage output of your converter into Vin (referenced to COM). It will work (I've done this before).

Here's a more complete drawing, if you didn't get what I meant.

You could also look at Servo controllers, where you give it a specific signal, and the flow-meter is the feedback signal, the servo controller will ramp up/down to keep the flow rate in proportion to the required input signal.

Another approach is to use a motor that runs a constant speed, and then a "Eddy-clutch" and possibly a TECO eddy current controller, then your using the same as I have written above, where the motor (prime mover) drives a constant speed, and the eddy clutch varies the output speed to bring the feedback signal to equal the referance signal.

Thanks, I think this would be the simplest way to go about it. The motor that the flowmeter signal would be driving is a peristaltic and would provide good accuracy for me. My main goal is to give a steady dilution of the chemical into the resovoir that is being filled at varying flowrates. Again, thank you.

I overlooked something important.

You're supposed to have a controller in between the F/V converter and the input to the speed control. That's so you can set the flow-rate that you need. If you hook it up as I drew it, you won't be able to control the flow.

Alternatively, you could reverse the configuration of the converter so that a high flow will produce a low voltage and a low flow will produce a higher voltage. That way, the speed control will settle at the half-way point. Then you adjust the speed control 0-100% setting to get the flow-rate that you want.

In my current setup, I want to just make it self adjusting and it seems that in your first diagram would work for me. The long term though is to process the incoming frequency through a controller, which then would signal the motor speed based on user programmable presets. If the user wants a dilution of 1oz./gal, the flowmeter would send out the current flowrate to the controller which would in turn process the signal and send out the required voltage to the pump to deliver such a rate (the pump would be calibrated for rate internally as well). As the flow changes, the processor compensates the motor voltage to match, which theoretically should work. But for now I just need to show that it can be done manually in a sense until the design goes to the software guys. Thanks for all the info.

I don't think anyone has mentioned this: another possibility would be to use a digital potentiometer, controlled by a microcontroller (a Basic Stamp, for instance). The advantage of this approach is that if you have an existing drive for the pump (controlled by a pot) and if you are happy with that drive, then the controller is simply "setting the knob" on the drive (you are just replacing the drive's pot with the digital pot).

Other than what I have just said, I know nothing about these things -- I have a stamp and a digital pot I plan to play around with, because I may have an application where using one would work well.

Those digital pots are good for creating set points, I have used a few.

But if your making a "Dosing pump" like what the Guest is suggesting, then the digital pot will make it cumbersome, why add that into the design if your adding also a microcontroller?

The "No Micro" approach would be a "up/down" press buttons to control the D-pot then the output going to the speed controller.

The digital approach, (micro) skip the D-pot, put in 2 thumbwheels for setting from 00-99% then run the PWM from the flowmeter into the uC, do some calculations then run the PWM output from the uC into the PWM circuit of the speed controller.

Personally a uC is an overkill for a simple dosing pump, if your looking to buy something, I'm not sure if this company is still around, "Acromet" they sell dosing pumps for companies.

The company I work for actually makes the dosing pumps. The reason for using the controller is because the setpoints have to be user adjustable. The controller needs to vary the dosing pump to match the changing flowrates on the different resevoirs. With a single manifoild feeding 8 resevoirs, the flowrates can vary significantly depending on how many solenoids are actuated to allow for fill. As the flow changes, the dosing pump will have to match it to deliver the proper diltution. The controller will take the freq out of the flowmeter, process it and adjust the correlating dosing pump for that resovoir. Right now I just need to spec out the parts we want to use, validate the theory of operation, then I can hand it off to the software guys to bring it all together. The current motor controller, it turns out is not PWM, it just varies the motor voltage via a pot. Ideally, I think PWM is the way to go, as the motor will have full power when on. The system, in the end, has to be "smart". We do not want to have manual adjustments to have to be made, it must be controlled by a controller and software. Thank for the input, keep it coming.

Guest:

You might consider a retriggerable monostable (one-shot) multivibrator coupled directly to the motor by a driver circuit and eliminate the conversion to DC. At low flow rates (frequency) the duty cycle of the driving pulse will be low and it will increase linearly with frequency, to 100% at max flow.

DickL