VFD machine tool spindle acceleration pattern and stall prevention 1

[fastline12]

Still setting up a Mits drive on a machine tool spindle and have a couple a general questions.

1. Acceleration pattern - Mits is recommending an "S" shaped pattern instead of linear and indicate a reduced accel time. Curious if an "S" pattern is typical of spindle apps?

2. Stall prevention - Mits indicates that after the base frequency (usually the rated motor frequency), the stall prevention boost reduces with increased frequency. Motor base is 60hz and max freq is 250hz but very concerned about reduced stall prevention at higher frequencies. They indicate current is reduced but voltage is not. Is this typical? Base speed is well below most typical operation speeds.

 

[ScottyUK]

Are you maintaining a constant voltage/frequency ratio above 60Hz? If the output voltage tops out at 60Hz then there is an inverse relationship with available torque as the frequency rises. At 250Hz you'll have less than 1/4th of the torque you had at 60Hz.


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[fastline12]

I will have to confirm setup from the previous drive but I think how it was setup was max voltage at 60hz and carry the voltage to the higher frequencies.

I guess I was wondering if others here have setup machine tool spindles and how they are typically setup?

 

[LionelHutz]

It's fairly common to have a specially designed motor which will carry the V/Hz ratio to the expected operating speed. In other words, the motor could be a 230V/250Hz or 480V/250Hz motor.

 

[fastline12]

I know this motor is an off the shelf motor. Nothing special at all. 4 pole, 60hz, 230/460V, stock bearings, no extra balancing, being pushed to 250hz. Has done that for 15yrs.


From what I gather from you, I should indeed be using the rated frequency of the motor for the peak voltage target in the drive?

 

[cswilson]

On the "S" acceleration: "S-curve" acceleration -- so called because the velocity-vs-time plot is S-shaped -- is usually done for reasons of smoothness. It gradually ramps up the acceleration, and therefore torque, to a maximum, then gradually reduces it to zero as you hit maximum speed. (You do this instinctively when you drive a car -- if you're older than 17...)

For the same peak acceleration/torque value, it actually takes longer than constant acceleration, which instantly steps the accel on and off. But many people value the smoothness more.

Curt Wilson
Delta Tau Data Systems

 

[fastline12]

Obviously if we set base speed at 250hz, the voltage will peak at 250hz BUT that seems it would make the motor weak everywhere? There seems to be some debate whether the sensorless vector system will actually improve anything above base speed with no more volts to give?

The main reason for the vector system is to improve every aspect of the spindle. Accel time, decel time, and power in a cut. I am starting to question where we will end up here.

Could the S curve be more tailored to larger spindles with big power? This is a 10HP spindle and even the ones a few years newer with vector systems accelerate about 10x faster and use the exact same motor. I am hoping I am on the right path here.

 

[LionelHutz]

What is your line voltage?

 

[cswilson]

In a vector control system, most commonly the base speed is set the same as the line frequency the motor is designed to work from. Below this speed, the motor is current-limited, so you have a constant amount of torque available from zero to base speed Power available and voltage used increase linearly with speed).

Above base speed, the field is weakened inversely with speed to keep the back EMF below the available voltage so there is still voltage "headroom" to drive torque-producing current through the phases. However, the torque available falls off inversely with speed in this range. Power available and voltage used are constant in this range.

As you surmise, if you were to set your base speed/frequency to 250 Hz instead of 60 Hz, you would be limiting your torque at low speeds to less than 1/4 of what would be available with a more standard base speed setting of 60 Hz.

Curt Wilson
Delta Tau Data Systems

 

[fastline12]

Line voltage here is 240V. Everything in the drive will be set at 230V. I guess there is some discussion as to HOW you get a performance improvement above base speed when switching from v/hz mode to SVC or encoder feedback? It seems that after base speed, full voltage is delivered thus the only thing that can really change is current and wave form? Or maybe the drive is better at determining actual slip thus compensating the output frequency better to try and maintain speed?

I was told that simply by tuning the slip ratio in v/hz, you can achieve better acceleration and constant rpm. We are going to have to reinstall the old v/hz drive for now due to our time line and would like to tune it better to achieve better accel/decel, and maintain speed better under load.

I have run identical machines with the same spindle motors and only difference being a vector drive system and speed regulation all the way to 150% spindle load was very impressive. Load goes up but spindle does not bog down. Able to make much heavier cuts. I am curious to learn "what" exactly is getting that extra performance.

 

[LionelHutz]

You will set the drive to the motor parameters of 230V and 60Hz then.

If you had 480VAC available then you'd set it to 460V and 120Hz and keep the motor connected for 230V. This way, you'd get full capabilities up to 2x the rated speed of the motor.

 

[cswilson]

I'll try to provide an intuitive explanation of vector control. Your guesses are pretty close.

Traditional V/Hz control is open loop. It outputs an electrical signal of a frequency corresponding to the instantaneous command value and a proportional voltage as determined by the preset V/Hz ratio. (Note that this is not technically a "slip ratio", although it effectively reduces to one in the steady state at constant velocity.) It does this with no knowledge of what the motor is actually doing. The slip in this case is simply the physical reality of the difference between the electrical frequency of drive output frequency and the rotor mechanical frequency (i.e. speed).

The preset V/Hz ratio allows you to trade off torque for speed. A higher ratio can give you more torque at the cost of a lower base speed.

Vector control, by contrast, is a closed-loop technique. The term "sensorless vector control" is somewhat of a misnomer, as it does use voltage and current sensors inside of the drive. Technical papers usually call it "shaft-sensorless" control. Anyway, it is continually measuring/estimating the physical rotor speed and outputs an electrical frequency that creates the slip frequency needed to get the torque it wants to maintain the desired velocity.

In vector control, there is a V/Hz ratio (often implicit), but it is related to the slip frequency rather than the electrical frequency itself, as in open-loop control. Often this ratio derives from the "magnetization current" setting of the drive. If this setting is left constant over the range of operation of the drive, you have the same speed range as an open-loop V/Hz range (but with improved operation within that range due to the controlled slip).

However, a vector drive, because it has knowledge of the actual rotor speed, can vary this setting dynamically as a function of the rotor speed, something an open-loop drive cannot do. So it can be set high for high torque at lower speeds, then reduced at higher speeds to keep the needed voltage within range. This reduces the available torque (there's no free lunch...) but it does permit some torque to be generated.

This "field weakening" technique sets the instantaneous V/Hz ratio for the slip frequency inversely proportional to the rotor speed when it is above base speed. As you note, this has the effect of outputting a full voltage signal above base speed, but varying the characteristics of the signal.

Curt Wilson
Delta Tau Data Systems