load cell vs motor current

L

Thread Starter

lynn vanderbloemen

I am developing an application to remove solid waste from a fluid vessel. The greater the motor load, the more waste to be removed, the longer the time the motor runs. Motor current is a continuous signal that can be used. Would a load cell give a more direct and stable feedback to the control? I do not have experience on the type of signal a strain gauge would generate. My limited knowledge of this would be with static measurements at a scale or not very accurate measurements on a moving scale. I would appreciate information on this type of signal.
 
K

Kirk S. Hegwood

Lynn,
I would and have used a load cell. This would give you a much more accurate signal of what is happening with the motor. With current, any
change in voltage (Ga. Power guarantees +- 10%) will throw your signal off of any parameters.

Kirk S. Hegwood
Hegwood Electric Service, Inc.
 
D

David Leese Dresser Valve Div., Hallibur

I would also agree that the load cell is better. Motor current will vary with voltage and with motor temperature. Even if you could hold V and t constant, for AC motors (you don't specify your motor) the current will never vary linearly with load, particularly if you operate the motor near it's upper torque limits.

I have a product with an electric motor and an option for a current-sensing torque limiter. This isn't a very popular option because the variations described above limit the effective
resolution. The general agreement here is that it would be better to develop a shaft-speed-sensing torque controller, but that's just an idea that we haven't pursued.

David L.
 
Responding to query:

Presuming load is proportional to motor torque, and as long as you have ability to monitor current and voltage, then I suggest you install an
input power monitor. Then, motor output power, hence load can be determined directly, without inference.

Output of power monitor can be any measurement you need.

Regards,
Phil Corso, PE
Trip-A-Larm Corp
 
R

R A Peterson

Instead of current feedback try horsepower feedback. Several companies make them (Load Controls comes to mind). Current does not change much with load but HP used does.
 
hello lvander:
you can use load cell for in motion weighing with proper mechaical equipment.
we use them for in motion coal weighing application on a conveyor belt.
regards

s.sathiraj
 
G

Guy H. Looney

Lynn,

It really depends on how much resolution you need and the type of motor & drive you are using. Also, you asked for stability. Is the motor switching directions, changing speeds, or stopping and starting? If the answer is yes to any or all of those questions, you've made it harder to use the motor's torque as a reliable feedback for force. You'll have to know how much torque is used to accelerate the load's inertial mass and how much is required to keep it moving. Further the torque will be constantly changing during the acceleration period to compensate for the PID loop.

Many digital AC servos allow you to monitor torque directly from the amplifier. However, the resolution is typically limited to 8 bit (1 bit for sign, 7 for magnitude) and the signal is typically 0-3VDC (3VDC equals peak torque). Let's say you had a mechanical system that could generate 200 pounds of force. If 200 pounds of force equates to peak torque, the best resolution you could hope for would be 1.5625 pounds. If you need more resolution than that you would have to go with a load cell. To compound the problem, you need to factor in how much of that 200 pounds is due to inefficiencies & friction. From experience, I would tell a customer this example would provide a repeatability of about 5 pounds (however, the mechanics would dictate the reality of this answer).

If you don't have a servo of this type, I would not recommend measuring Horse Power. Horse Power is not a good measure of torque due to the fact that it doesn't account for losses and it requires that speed must be constant at all times. If speed isn't held constant, then you'll have to plot the torque vs. the speed as it changes.
HP = [Torque (oz*in) * Velocity (rev/sec)] / 16,800

Monitoring the current is a bit better, but still doesn't account for the losses in the system.

You asked for stability. Is the motor switching directions, changing speeds, or stopping and starting? If the answer is yes to any or all of those questions, you've made it harder to use the motor's torque as a reliable feedback for force. You'll have to know how much torque is used to accelerate the load's inertial mass. Further the torque will be constantly changing during the acceleration period to compensate for the PID loop.

The bottom line is it depends on how good you need the feedback to be. A load cell will definitely provide you with the most reliable information. Think of it like a linear encoder vs. a motor's rotary encoder. If you have a servo driving a ball screw table, the linear encoder is going to tell where the load is while the rotary encoder is going to tell you where the motor is. Which one is better? Obviously the linear encoder is more reliable, but the question is do you need it. Often times the answer is no.

If you do go w/ a load cell make sure you pay attention to the resolution it has. You don't want to spend a bunch of money on a device with infinite resolution and find out that your control system can only handle 12 bit.

Thanks,
Guy

Guy H. Looney
Sales Engineer

Regan Controls, Inc.
475 Metroplex Dr.
Suite 212
Nashville, TN 37211
phone: (615) 333-1940, ext. 322
fax: (615) 333-1941
[email protected]
www.regancontrols.com
 
G

Guy H. Looney

Current is proportional to torque:
Torque = Kt + I
Where Kt is the torque constant of the motor & I is the current being used.
Current always changes with load.

HP = [Torque(oz*in) * Velocity (rev/sec)] / 16,800

HP will change based on either a change in velocity or torque.

Guy H. Looney
Sales Engineer

Regan Controls, Inc.
475 Metroplex Dr.
Suite 212
Nashville, TN 37211
phone: (615) 333-1940, ext. 322
fax: (615) 333-1941
[email protected]
www.regancontrols.com
 
Responding To Guy Looney's comment:

Current also changes inversely w/ voltage. Therefore, the more Technically Correct parameter to use is Power, P=SQRT(3)xVxIxPF.

Regards,
Phil Corso, PE
Trip-A-Larm Corp
 
Top