Hey guys/gals,
I'm getting into FV control of induction motors...

http://en.wikipedia.org/wiki/Vector_control_(motor)

Does anyone know of any resources the find info on this stuff, it's pretty scarce and a decently high level of mathematics... as you can see below, the formula page on the wiki is too blurry to make out some of the subscripts.

I'm also opening this as a discussion and will be posting some of my findings. Feel free to contribute.

 

Flux Vector Control varies both the magnitude and direction of the magnetic field in the rotor of an induction motor, hence the name "Vector" control.

Without citing any mathematical models, vector control is implemented by using the input from a position sensor on the rotor shaft so the resultant magnetic field is always aligned tangent to the rotor.

This maximizes torque and provides quick response to a changing load on the motor. Vector control is very responsive to torque variations caused by driving an elastic load such as the winding machine for cable elevators.

By contrast, scalar control varies only the magnitude of the field without precise control of its direction. Scalar control is often called "Slip Control" which varies the angular velocity of the field in the stator in response to the angular velocity of the rotor as detected by a speed sensor on the rotor shaft.

 

Thanks for posting, from what I can tell we’re on the same frequency… I’m going to dive a little deeper into what I've been reading about. Please note that I’m not an expert on this subject… I’m just learning, so my information might not be %100 correct.

An AC motor has a voltage range, a frequency range and a rated fixed constant “flux”. The units for flux are Webers… A Weber is Volt*Sec or Volt*(1/f) or Volt/Hz. Therefor the ratio of voltage to frequency into the motor must be fixed to maintain constant flux... Now, unlike a DC motor the input current is not directly proportional the output torque, this this not immediately obvious and I will get more specific.


-Fig 1
Note that current it not directly proportional to torque…


-Fig 2
This is due to what is called the "magnetizing current" associated with the motor... Its vector is at a right angle to the Torque producing current (real) ... The resultant of these two vectors is the complex input current to the motor. The current at x = 0 on fig 1 is the magnetizing current, from what I can tell... this is responsible for fluxing up the motor, and is fixed through the normal operation. As you can see by the vector diagram (Fig 2) It's also completely imaginary... So if I have the proper thinking...

The ratio of Volts/Hz is kept constant... this means that the ratio of Amps/Hz is kept constant... The magnetizing current is Fixed but the Torque-Producing current is not fixed ... This is what gives the non-linear ratio between the input current and output torque. A standard system (non FV) will give far less superior control of torque.

So the idea behind flux vector is to change the magnitude and direction of the torque producing current (which is linear to the torque) without touching the magnetizing current's magnitude or angle wrt. the torque producing current.

I'm just going to leave it at that for now to sleep on... maybe other people will pick at it.

 

Flux Vector Control varies both the magnitude and direction of the magnetic field in the rotor of an induction motor, hence the name "Vector" control.

Without citing any mathematical models, vector control is implemented by using the input from a position sensor on the rotor shaft so the resultant magnetic field is always aligned tangent to the rotor.

When you say "magnitude and direction of the magnetic field" I'm thinking that this is directly related to my "Magnetizing Current" ... as a switching current in a highly inductive environment is what's responsible for producing a magnetic field...

 

As a practical matter, scalar (slip control only) just controls the strength (magnitude) of the resultant field in an induction motor.

The idea is to keep slip at the optimum value to maximize the magnitude of the field and keep the torque relatively constant.

However, a vector drive also controls the direction of the resultant field by adjusting the rotation of the field in the stator. Imagine that you had a steel bar on a pivot and you could make it rotate by holding a magnet next to it.

In order to keep the torque constant, you would have to keep moving the magnet so the force is always tangent to the end of the bar. That is a simple form of vector control. However, it gets a bit complicated when the field is induced in the rotor of an induction motor and the system has to adjust both slip and the direction.

 

I'm bringing the thread back from the dead. I have a few questions regarding FV control.

1. Two motors mechanically linked, the motors are ganged off one drive and there is an encoder on one of the motors? I'm told FV is not possible, no one has given me a clear answer as to why... My thinking it because the angle of the flux current and torque producing current are not always in phase (between the two motors) ... and this will mess everything up... essentially improper commutation and confusion of the drive.

2. If one was to get FV working on this system (For all intensive purposed lets assume it's one large motor with the same HP as the two existing motors added together) ... How would this react to a sharp change in load? Right now I'm using DC current injection for holding purposes ... it's working well for many cases, but the load isn't consistent and sometimes the motor will break away.

3. SV vs. FV - Sensorless vector vs. Flux vector ... how do these compare (Even more specifically on Allen Bradly PF753 drives), I'm confident that SV measures the output voltage and current, uses the phase shift to determine the inductive component and thus the mag. component of the current ... this seams like it should be able to get close to closed loop vector? Does having feedback really effect the approximation? I'm thinking this might relate back to my question on impact loading.

4. "Full torque at zero speed" - What the hell does this even mean... It's quite obvious what it means literally, but what does this actually mean? How is this possible? If you have full torque at zero speed will a full mechanical load... excellent full torque! We're happy! What if you have full torque at zero speed and no mechanical load... won't your motor accelerate? so that's what the feedback is for? it noticed a move then it drops the current down to zero to stop it from accelerating? so we're left with no torque at zero speed? - This is something else I haven't gotten a clear answer for. Some people have told me some things, some people tell me other things... "Flux up the motor" - that's a hot one

So I'm assuming the mag component of the current shoots up, not causing rotation... why can't this be done in SV?

5. If a motor is rated for "Full torque at zero speed" - what does this mean? Can I just ram FLA through it at 0hz and everything will be okay?

If anyone know the details, I would be very grateful

 

For starters I have used two identical motors on a dual axis lathe for spindles using a single WEG VFD, with good results.
The lathe part is held at each end and driven by dual motors on each spindle.
If the motor has an encoder then this would not be using flux vector for rotor position but encoder feedback?
With sensorless vector, the controller keeps tabs on what is happening regarding rotor position in order to attempt to keep the vector angle optimum, instead of encoder, it uses feedback via voltage current etc.
The majority of motors have maximum Continuous torque at zero rpm, this means they are capable of this continuous torque (current) at the rated load, if this is exceeded, the motor enters the peak torque area and can usually only sustain this for a very brief period, otherwise destruction can result.
IOW, the torque curve for most motors has a continuous torque curve, and a peak torque rating.
Max.

 

If the motor has an encoder then this would not be using flux vector for rotor position but encoder feedback?

Can you rephrase the question? currently the system is using v/hz, there is an encoder on the motor, it's currently not going back into the drive. If it was terminated into the drive and configured properly it believe it would be used by the drive for rotor position and as a tach. for speed feedback.

 

In your post #6 you said a motor had an encoder?
If you use it you require the encoder option on the VFD, but having never implemented encoder for dual motors, only a single, I would think you would need one on each motor?
Then no longer flux vector operation.
Max.