Can a VFD's analog pins be used to read sensor values?

[belae1ka]

Hi all,

For my senior design project I am controlling a large three phase motor from the program Labview through MODBUS.

The sensors being used in the control system of my project are a TMP36 temperature sensor, a S-type load cell, and a rotary encoder to measure RPM.

Instead of using a USB Data Acquisition device (DAQ), I would like to know if it is possible to wire my sensors to the VFD and have MODBUS read the sensor values.

Also, Instead of using a rotary encoder to measure RPM can I read in the RPM of the motor using the VFD?

The VFD I plan on using is the DURApulse GS3 series AC drive (1 HP, 230V AC)
Model name: GS3-21P0

Load cell: https://www.amazon.com/Keli-Electri...coding=UTF8&psc=1&refRID=4A2W5VXCYSXYYG0WWN0H

Temp sensor: http://www.analog.com/media/en/technical-documentation/data-sheets/TMP35_36_37.pdf

Thank you.

 

[just the cowboy]

Yes and no

Depends on what the sensor for temperature is?
If they are both analog output decives yes I think the GS3 has two analog inputs.

As for RMP from drive it will only give you what the motor is doing, not the machine so if you have a broke coulpling the motor will run and tell you speed but machine will not be moving.

There is no logic in the VSD so if you bring in the analogs they are for speed control only, but everything can be read with modbus.

 

[the_msp]

While it may be possible I think using the DAQ would be a beter solution as you are already using LabView. USB DAQ's aren't that expensive, as far as NI hardware goes...

 

[Gene Bond]

I do it all the time with Control Techniques drives, but I have no idea if the AD drives are open architecture like the CT drives. With CT, you can read any variable/parameter in the drive through Modbus. For example, Analog Input 1 is 07.001, 0.0-100.0% (40700 in standard Modbus mode).

 

[belae1ka]

Depends on what the sensor for temperature is?
If they are both analog output decives yes I think the GS3 has two analog inputs.

As for RMP from drive it will only give you what the motor is doing, not the machine so if you have a broke coulpling the motor will run and tell you speed but machine will not be moving.

There is no logic in the VSD so if you bring in the analogs they are for speed control only, but everything can be read with modbus.

I am a bit confused on what you meant when you were talking about the RPM. I need feedback of the motor's RPM (what the motor is doing) so I can plot it with respect to force (from a load cell) as the RPM increases . For my project, the force on the load cell will increase as the motor goes faster and faster and acquiring the FORCE vs RPM graph is the goal of my project . My project is to build a shock dynonameter. The VFD states that one of the output terminals is a digital frequency output which is a square wave that pulses a train output. I assume that RPM can be acquired through MODBUS though. Let me know what you think. Thank you.

 

[ELake20]

https://www.automationdirect.com/static/manuals/gs3m/ch5.pdf

Yes you can do this.

So your labview app would only be starting/stopping the drive and reading the status?

 

[just the cowboy]

For what you are doing no problems modbus will let you read and write. Your true feedback then is the change in force, which will prove that the motor is running. What i was pointing out was since you are not using an encoder on the load you have no way of knowing the motor is changing speed or actualy turning the load.

 

[belae1ka]

https://www.automationdirect.com/static/manuals/gs3m/ch5.pdf

Yes you can do this.

So your labview app would only be starting/stopping the drive and reading the status?

Yes,

step 1 would be to click a button so that the motor runs until a desired temperature is met by a temperature sensor (this is the "warm up phase"). My project is to create the control system for a shock dynonameter (the temp sensor will be placed on the shock to measure its temp).

step 2 would be automatically start the motor after the warm up phase has been completed, this time plotting the motor rpm with respect to force exerted by the shock (force output from a load cell sensor. Once a desired RPM is reached, the motor will turn off. The final result of my project is to acquire the characteristics of the shock being tested (the force vs RPM graph).

Thank you very much for providing the link regarding durapulse MODBUS communication. That will definitely come in handy.

Best,

-Kevin

 

[belae1ka]

For what you are doing no problems modbus will let you read and write. Your true feedback then is the change in force, which will prove that the motor is running. What i was pointing out was since you are not using an encoder on the load you have no way of knowing the motor is changing speed or actualy turning the load.

If the shaft of the load is thoroughly connected to the shaft of the motor wouldn't the RPM of the motor=RPM of the load? I need to know the true RPM of what the motor is doing so couldn't I just read the RPM feedback? I need to plot the RPM of the motor with respect the force exerted on a load cell.

For example, let's say I want to see how much force was exerted on the load cell 5 seconds into the test...I would need to know what force was applied on the load cell at exactly 5 seconds in and I would need to know the true RPM of the motor exactly 5 seconds in. Does that make sense? I am a bit confused by your statement when you say you will have no way of knowing the motor is changing speed. If I plotted the RPM with respect to time, wouldn't I know that the motor is changing speed?

 

[DickDV]

The drive's analog inputs are configured for specific signals, ie: 0-10v, 4-20ma, etc. Usually, these can be scaled and offset if the incoming signals are slightly different, as in 0-20ma instead of 4-20ma.

That defines the signal the drive can read. It cares nothing about what is sending those signals. It is up to the person commissioning the drive to make the VFD respond in a particular way given the signal that is received.

So, the answer to your original question is, yes, the drive can respond to any device (including another VFD) that has a signal that conforms to the VFD input.

 

[robertmee]

Yes the RPM of the load would be the RPM of the motor, assuming the coupling is in tact and it is a 1:1 ratio (direct couple, no gearing). What the other poster was referring to was that assumption that the coupling is in tact. If it is not, then the motor could be spinning, but not the load. Depending on the size of the load, you can monitor the torque/current of the motor to see if the load is connected. Most motors uncoupled run a few tenths of an amp, but coupled may be an amp or more depending on the size of the motor/load. You can use that differential to know if the load is actually coupled to the motor.

 

[redbarron55]

Best to monitor the power since the amps on a VFD can be confusing.
Even an across the line motor draws about 30% of full load current due to the poor unloaded power factor.
Most drives will measure and can give you a power reading.
If you are using a force vector drive the RPM would be pretty close.
If you are want the use the analog inputs and read them from the modbus that is doable as well.
Many people just don't consider the calculated power when working with motors and that number can be pretty handy.
Pumps that are pumping take more power, oddly enough.
If you want a dynamometer then the power from the drive takes you right there.
Many drives will give you a % torque as well.

Good luck with your studies.
Back in the dark ages I worked with a guy on shocks with a similar setup and now the is known as Dr Who in NASCAR circles.

 

[thingstodo]

The Duraplus GS3 is a sensorless vector drive, so it will give you decent feedback on the speed of the motor because it is sensing/calculating the rotor position.

The amount of torque output by the motor is variable based on what is 'required' at the time. You can set a speed setpoint and have the drive figure out what it needs to put out for current, which adjusts the output voltage and phase angles. So you may need to monitor output voltage and output current to gather the data that you want/need.

As the torque rises and falls, the amount of slip (difference between frequency output by the drive and the speed the motor turns) does change. I don't have experience with this drive so I don't know how well it tracks.

I assume that you are trying to minimize your sources of error since this is submitted as a project. IMHO, the encoder is the best way - feedback is most accurate. But it costs time and money

 

[redbarron55]

If you want to know power the drive knows.
Ask it how much power it is putting out.
The volts and amps on a VFD are relatively meaningless.
Power is pretty accurate.
You probably want to know the SPM at the shock and the temp.
the drive analog inputs can accept the temp and modbus can read it back to your equipment.
If you want to know rpm more accurately and you have a gearbox for reduction to the final drive element you could put some sort of sensor to an DI and count that.
If you do have a reduction scheme then you need to know the power losses in that as well under load.

 

[rootboy]

Nowhere in the Modbus chapter for the drive did I see where you can read your AI values. So I would say, no, you can't.

Now that doesn't mean that there isn't an undocumented parameter lurking around in the drive that you could use. It would be best to call the tech support at AutomationDirect to find out for sure.