Powerflex 753 and High Inertia Loads

Hi everyone. I've run into a perplexing issue and before I travel too far down the "we have problems" path I'm hoping someone might have some extra insight. Here is the short rundown:

I have a 150hp motor running a centrifuge that is being controlled by a Powerflex 753 drive. Ethernet communications but we are using the VFD simply for ramping functions. Maximum speed of 60hz and using a preset speed of 60hz, no dynamic control. We have had this machine set up with a 700 second accel time and 350 second deccel time since installation. Very recently it has become apparent that it is not taking anywhere near 700 sec to reach reference, in fact it is only taking 231 seconds to reach reference. After verifying every parameter that I am personally comfortable with, I decided to double my accel time and clocked in at 356 seconds to reference. With the longer ramp time I did notice individual amp spikes throughout the curve with each spike corresponding to a sharp bump in my output speed.

With all of this said, I have started down the line of thought that heavy interia inherent with this machine is helping me outrun my ramp time. Could this be the case or am I missing something here? To make matters worse I drew comparisons with like equipment elsewhere and when trended in our historian I'm seeing the same behavior with them as well. I'm relatively green in my position and it seems like anytime I dive into an issue these cans of worms come out and bite me.

If you want a predictable and repeatable acceleration time, you need to do this in SVC (or Sensorless FVC) control mode. In basic V/Hz (Scalar) mode, the drive has no idea what the motor is doing, it just puts out a ramp rate that corresponds to the slope of an angle from minimum to maximum speed. The output HP of the motor will track that slope fairly consistently. If, at any time on that curve, the output HP of the motor exceeds the HP requirement of the load profile, that may result in the motor accelerating faster than anticipated. This is the same thing you will experience with a Soft Starter, which is, in essence, what you are doing with this VFD by the way.

But with a Closed Loop control option implemented in the VFD, the VFD knows what is happening in the motor and responds accordingly. Even though we use the term "Sensorless" in SVC or with regard to FVC (in the case of PowerFlex 755s), that doesn't mean it is open loop. The "sensors" are inside of the VFD, not attached to the motor, but the VFD does know how fast the motor is spinning / accelerating / decelerating.

Of course, you didn't specifically SAY that you were using V/Hz mode, I am assuming so. If in fact you ARE using SVC or FVC, then the issue may be that you have enabled the "Bus Regulation" feature, which may be affecting your acceleration profile. If you don't need that, turn it off.

jraef said:

If you want a predictable and repeatable acceleration time, you need to do this in SVC (or Sensorless FVC) control mode. In basic V/Hz (Scalar) mode, the drive has no idea what the motor is doing, it just puts out a ramp rate that corresponds to the slope of an angle from minimum to maximum speed. The output HP of the motor will track that slope fairly consistently. If, at any time on that curve, the output HP of the motor exceeds the HP requirement of the load profile, that may result in the motor accelerating faster than anticipated. This is the same thing you will experience with a Soft Starter, which is, in essence, what you are doing with this VFD by the way.

But with a Closed Loop control option implemented in the VFD, the VFD knows what is happening in the motor and responds accordingly. Even though we use the term "Sensorless" in SVC or with regard to FVC (in the case of PowerFlex 755s), that doesn't mean it is open loop. The "sensors" are inside of the VFD, not attached to the motor, but the VFD does know how fast the motor is spinning / accelerating / decelerating.

Of course, you didn't specifically SAY that you were using V/Hz mode, I am assuming so. If in fact you ARE using SVC or FVC, then the issue may be that you have enabled the "Bus Regulation" feature, which may be affecting your acceleration profile. If you don't need that, turn it off.

Click to expand...

Thank you very much for the reply. We are definitely using V/Hz mode as you figured. Mostly I believe because no one took the time to explore other options. Also we have moved off soft starts towards VFD due to some major malfunctions before my time and so far VFDs have proven to be more robust from what I understand. From what I've looked into thus far is the autotuning functions something worthwhile to explore or a waste of my time?

jraef, I'm a bit surprised by your comparison of V/Hz accel ramp control to softstarter ramp control. There shouldn't be any comparison because the softstarter is always 60hz and the ramp is a torque limited ramp. Under that system, if the motor unloads, it will accelerate faster. On the other hand, even in VFD V/Hz mode, the motor is forced to follow the accel frequency ramp up to max speed whether loaded or unloaded. The motor speed may vary around its slip speed limits but no further as it follows the ramp up to speed. Of course, in Sensorless Vector or Flux Vector control the motor is forced to follow the ramp even closer as the ability to slip under varying load is greatly reduced.

I'm sure you know this. You must have just misstated what you meant.

I guess I have to admit not meaning to imply they were the same, but in re-reading it after your comment, I realize that I did. You are right, not the same thing.

But I have seen this issue with V/Hz control before. Yes, the motor SHOULD follow the frequency more closely, but sometimes it doesn't and because the V/Hz mode doesn't even know if the motor is connected, it has no idea if it is accelerating faster either. Yes, as it goes beyond the slip speed it SHOULD start to lose torque and slow back down to the commanded speed. But as drives became more capable, they have had "features" added to improve performance that sometimes have odd side effects. Bus voltage regulation, added to help prevent OV tripping on high inertia loads during speed changes, is one of those and can cause unwanted side effects if not turned off when not needed. In the 750 series drives, it might be enabled by default (I'll have to check later), but the manual tells you to do a motor Autotune even if you are using V/Hz mode, or else turn off the Bus Regulation. Hardly anyone reads the manual however...

Yes, DC Bus Regulation is shipped as Enabled, with the action set to "Adjust Frequency". So under the right circumstances, it will override the ramp rate of rise in order to accommodate an increase in bus voltage, resulting in a faster accel time.

So either do an Autotune and use SVC by setting P035 to a 1 for “Induction SV” and P070 to a 2 for "Static Tune" (meaning it will not need to rotate the motor), or turn off the DC Bus Regulation by setting parameter P372 to "Disable". If however you change a speed reference too fast, you may get an OV trip, but being that you are essentially using the VFD as a glorified Soft Starter, you will not likely be doing that.

I'm trying to picture how DC Bus regulation would come into play when accelerating a centrifuge. I've seen it used to avoid regen trips when decelerating but how is it going to be activated on accel? Maybe if the load tries to accelerate faster that the programmed rate (not typical of a centrifuge) so the motor is holding back as a brake to enforce the longer accel ramp. Any other situations that come to mind? I don't see a Low Voltage trip on unexpectedly difficult acceleration---that would tend to generate an Overcurrent fault.