vfd control

[jcole]

I installed a powerflex 70 drive recently for a progressive cavity pump. The manufactor said to intall constant torque drive. The powerflex 70 was suggested by salesman for this application.

Most of our applications (waste water) are fan and pump type (variable torque).

When setting parameters for this new applicaiton, I was not sure what type of control to use: v/hz, sensorless vector, vector.

Isnt v/hz used for variable torque loads? So I assume sensorless vector and vector would be for constant torque?

Why would the manufactor require constant torque? Its a positive displacement type of pump but the pressure never gets above 30 psi. So starting torque would not be that high.

Thanks for any replies.

 

[Jraef]

Yes, a PD pump is a constant torque application; the torque requirement does not vary with speed, which is the definition. Regardless of the speed you run it at, the HP to Speed ratio remains the same because the energy necessary to move the fluid varies in a linear manner with speed. That means that at 50% speed, you need 50% HP, which means you still need 100% torque.

With a centrifugal pump*, as the speed drops the pump moves less water, but it does so at the cube of the speed so as speed drops, it is doing LESS work at the load in terms of HP to Speed. So at 50% speed, the HP required by the load drops to .50 x .50 x .50 or 12.5% HP! So much less HP means a lot less torque, which means a lot less current which means the devices inside the VFD will not be strained as much and can be smaller / cheaper. That's the only thing that VT vs CT ratings of VFDs mean.

I would go with sensorless vector control (SVC) on a PD pump, most of them are speed controlled for metering purposes so SVC will give you accurate repeatable speed performance even if the commanded speed is low. V/hz will also probably work fine as long as your commanded speed is never less than maybe 4:1, i.e. 15Hz.

Full on Vector control would require a speed sensor (encoder) feedback from the motor shaft, totally unnecessary on a PD pump.

 

[jcole]

Thanks again for the reply Jraef.

These progressive cavity pumps are a type of positive displacement pump according to manufactor. I am not a pump or drive expert obviously. What I dont understand is why the amps will increase with speed(more flow) like a centrifugal pump would. If it was a constant torque load wouldnt the amps stay the same no matter the speed?

I know the hydraulic (high pressure) piston type positive displacement pumps I use to work with drew about the same amount of current no matter what the flow was. I guess what I am trying to say is that this pump acts like a centrifugal pump.

The pump works fine with either control (v/hz, sensorless vector).

What is different about a centrifugal pump design that cause amps to be directly related to flow (varible torque) than a positive displacement pump where amps stays the same no matter the flow (constant torque). A lot of positive displacement pumps are variable volume but the amps dont change when you change the volume.

Just trying to understand the difference I guess.

 

[Jraef]

... What I dont understand is why the amps will increase with speed(more flow) like a centrifugal pump would. If it was a constant torque load wouldnt the amps stay the same no matter the speed?

You are misunderstanding something here. Amps follow torque so if the torque is constant the amps are constant. With variable frequency you are changing the HP of the motor with speed because you are lowering the voltage at the same time as the frequency and maintaining a constant V/Hz relationship, so the motor's torque remains the same. So for example, if you start with a 10HP 460V motor drawing 14A at full load and you put a VFD on it to slow it down to 30Hz (1/2 speed), the motor is now a 5HP 230V motor. 5HP 230V is still around 15A full load (there are other minor issues involved in the current / voltage relationship) so your current is staying relatively constant. But any motor will only draw as much power as it needs to do the work it is connected to. So if, in the case of a centrifugal load, the WORK at 1/2 speed drops to 1/6th of full load, then the motor is only USING power for roughly a 3/4HP motor at that point and the current for a 3/4HP 230V motor is somewhere around 3A. So it's not that amps INCREASE with speed in a constant torque application, it's that amps rapidly DECREASE with speed in a variable torque application.

I know the hydraulic (high pressure) piston type positive displacement pumps I use to work with drew about the same amount of current no matter what the flow was. I guess what I am trying to say is that this pump acts like a centrifugal pump.

How so? Are you seeing the amps dropping with speed?

The pump works fine with either control (v/hz, sensorless vector).

Thought it would, but if you are testing, try running it at 10Hz or 5Hz in V/Hz mode with a load on it and you may see a difference. If you will never run that slow, don't worry about it.

What is different about a centrifugal pump design that cause amps to be directly related to flow (varible torque) than a positive displacement pump where amps stays the same no matter the flow (constant torque). A lot of positive displacement pumps are variable volume but the amps dont change when you change the volume.

It's complicated, but do a search on "affinity laws" and you will see some good papers on what it's all about. In a nut shell though, think of a centrifugal pump as being like a machine with a centrifugal clutch, like the kind you used to see on go-carts as a kid. The faster it goes, the harder it grips the stuff it is trying to move and the more work it performs. So a pump motor is sized for the BPH (Brake Horse Power) required at the MAXIMUM flow, head and pressure it must deliver. But when you reduce flow, you are "slipping" more of the load past that "clutch" and the motor is not required to work as hard for the lesser amount it is pumping.

Just trying to understand the difference I guess.

I'm an EE but when I got into the VFD business years ago, I had a steep learning curve on mechanical engineering principals. I had to take ME classes in school, but I never thought I would have to pay that much attention... Turned out I was wrong.

 

[TxEngr]

J-

I sure wish I'd payed a lot more attention in that thermodynamics class. I just had to learn it all over again when I got to the 'real world'. Let that be a lesson to those of you still in school!