Vector mode disadvantages?

[cokeguy]

Are there any disadvantages in using vector mode against using V/Hz mode in typical fan, blower or pump apps? We normally don´t enable vector mode in a VFD unless it is really required, but I don´t know if it is a safe or good practice to enable it always even if the application doesn´t demand it. Almost all of our drives now are Saftronics GP10 drives with that capability, and yesterday one of our technicians asked me that question, and I couldn´t answer it.

 

[jraef]

There are applications where you cannot use Vector mode, such as multiple motors driven from a single VFD, or operation over base motor rated frequency etc. But for 1 motor - 1 drive, base speed and under applications there really is no down side to using Vector mode. In the past it may not have been worth it for you to spend extra to get Vector control for a simple centrifugal pump, but now it is a standard feature on most new drives, so there is no reason not to use it other than to avoid having to tune it to the motor. With Autotune now, even that has become a simple task as well.

That said, I recently ran into a VFD that had trouble integrating its operation with its own internal PID controller because I had it in V/Hz mode and for some unknown reason it required SVC mode. I have no idea why that would make a difference, but it would not operate correctly in V/Hz mode under PID control. It kept ignoring the PID output speed command and going to full speed because it couldn't see enough loading on the motor model (this was determined after 4 hours on the phone with the manufacturer's tech support gurus). I think sometimes manufacturer's engineers take the tweaks too far now, this was unnecessary in my opinion.

http:/

http://www.jraef.com

Eng-Tips: Help for your job, not for your homework Read faq731-376

 

[ozmosis]

One disadvantage of using "vector mode" rather than linear or variable V/F control on fans/pumps/blower applications is that a common use of VFDs in this business is to connect more than one motor (multi-motor applications) to one drive. If you have a "vector mode" (either open-loop or closed loop) then the method of achieving this mode entails getting more motor data and (usually) dialing this into the VFD. If you have more than one motor connected, then this makes it rather difficult(impossible) for the VFD to know what each motor is doing.
Also, gaining the energy savings on centrifugal fans/pumps is typically achieved by setting a variable (or quadratic) V/F. I don't follow that setting a VFD in "vector mode" will allow this curve to be simulated and therefore you would get the same energy savings you would out of a (variable) V/F.

 

[DickDV]

As mentioned, the principal disadvantage of using vector mode on fans and pumps is that the motor ID has to be rerun if the motor is changed out.

Benefits, on the other hand, would include better "catch a spinning load" performance and, at least for some drives, slightly better efficiency and less motor heating at light loads due to active management of motor magnetization.

None of these are really big issues either way, in my opinion. I set up most fans and pumps scalar.

 

[jraef]

You make a good point on the "energy savings" feature sed2, I hadn't thought of that. Although that isn't true for every brand of drive (I just did an informal survey of manuals I have at hand), it appears to be the case in a majority of VFDs. In particular though, his GP10 is one of those that does not restrict the energy savings mode to V/Hz operation, but it does automatically disable it if in Torque Limit operation while in SVC mode.

http:/

http://www.jraef.com

Eng-Tips: Help for your job, not for your homework Read faq731-376

 

[CJCPE]

This would be a good question for a knowledgeable contact for your preferred supplier. It would be interesting to see what "inside" people for various manufacturers have to say. I believe that some manufacturers maintain quite a bit of SVC functionality when in the V/Hz mode.

 

[Drivesrock]

Hi all.

In Vector mode the drive is controlling independantly the motor flux current and the rotor current which produces the torque. This means that the drive should supply only the rotor current required to maintain the required speed. The motor should therefore be running at an optimal condition for the load and save energy.

Remember that in Vector mode the drive is monitoring the speed of the motor - either by direct or indirect method.

Some drives even reduce the motor flux automatically (parameter selectable) if the speed is below nominal and the load is light, thus giving even more savings. This I believe, is sort of akin to the square law volts/speed curve in straight fwd V/F (V/Hz) drives.
Many drives nowadays have auto-tuning (of the motor) required in V/F mode as well as Vector mode as mentioned above.
In some pump aplications such as for iron-ore slurry it is desirable to have a constant V/F curve or Vector control to be able to get the motor away from zero speed with a 'clogged up' pump - one that has been stopped for a while with the solid in suspension decanted. This is also true for 'fly-catching' fans especially if they are free-wheeling backwards.

My recomendation is to always try and use vector mode for optimal motor performance - no matter what the application.

Having said that....

Sometimes it is better to have the motor fully fluxed and the drive just outputting constant V/F. The rotor 'can't' runaway from the stator and the drive isn't trying to be clever!

The disadvantage of Vector mode is that the speed loop in the drive should be tuned. Unless the engineer or technician in front of the drive on site has a computer and the appropriate S/W (many drive manufacturers nowadays provide this free to display and monitor drive performance) and the knowledge/experience and even time to adjust the P & I gains of the speed loop this may not be possible. Also, a direct speed feedback device (encoder) maybe required to achieve optimal response from the speed loop for 'dificult' loads - which just isn't practical on a fan or pump. I have had applications where my pride wanted me to leave the drive in Vector control but pratically I couldn't achieve that. The motor control (speed and current) was unstable - and the drive tripped on overvolts or overcurrent.

Just as a footnote;
I have successfully run two 'identical' motors in parallel in Vector mode - this was a moving machine application and the difference in performance was remarkable.
It is possible to run in Vector mode above motor base speed. The drive must be a good one though - the motor control math above base speed in 'field weakening' range is rocket science stuff! (also needs very fast processing speeds). Some manufactures can really only do that with an encoder fitted but there is at least one that can do it encoderless (I've done that too).

Happy Vector or V/F control everyone!

Drivesrock.

 

[cokeguy]

Thanks everyone, only one question, what about running the VFD above base speed in vector mode? I didn´t see that restriction in the manual or elsewhere. I realize that it´s probably pointless to enable vector mode if you plan to work at or above base speed, but still, what happens or could happen if you do?

For example, we have a few GP10s in vector mode driving conveyor belts, with widely varying speeds sometimes going as high as 90-100 Hz and as low as 10-15 Hz.

 

[jraef]

My recomendation (sic) is to always try and use vector mode for optimal motor performance - no matter what the application.

I'm sorry, but that is not a good suggestion. I know of no vector drive that states it can operate in vector mode with multiple motors downstream, and some cannot allow operation above motor base frequency (see below).

Cokeguy,
You are correct in that the GP10 does not really tell you that you can't operate in Vector Torque Mode above base motor speed, but they also don't really tell you everything. What they are doing (and others like them who don't bother to explain) is that once you get over base motor frequency, you are in virtual V/Hz control anyway. Think about it; if what a Vector control algorithm is doing is to constantly update the torque and flux producing elements of the motor model vectors in order to maintain speed or torque, and torque is directly tied to voltage, what can it do when you have maximized voltage? It has no more to give. So at base motor rated frequency you will be at maximum voltage, therefore maximum torque anyway. Above that you are operating in a constant HP mode, meaning that your shaft torque is going down as speed goes up. Vector control essentially means little at that point other than to possibly LOWER torque if not necessary. Some vector drives lack the ability to do even that so they must warn you ahead of time not to take the motor over speed in Vector mode because they can become unstable as the algorithm attempts to correct for error without all of its tools.

To give them the benefit of the doubt, one reason not to explain all this up front is that they don't necessarily know what your motor base frequency is going to be. So if you theoretically buy a 120Hz motor, you can in fact have full vector control up to 120Hz. I say theoretically because I have never seen a 120Hz motor available off the shelf. I also picked 120Hz because that is the max. frequency that the GP10 goes to. Other drives that are capable of 400Hz output or more are usually the ones who make the statement up front about maximum frequency limits (usually 100-120Hz) in Vector mode because even if you have a 400Hz rated motor, they cannot maintain the Vector algorithm at those high frequencies anyway.

http:/

http://www.jraef.com

Eng-Tips: Help for your job, not for your homework Read faq731-376

 

[ozmosis]

drivesrock
I'm not so sure tuning a drive (for use in "vector mode") should be seen as a disadvantage. Any process of gaining more information on the motor you are connected to should be seen as useful. Motor data within the drive is not just used to improve dynamic performance, it is also used within the drive to provide better protection for the connected motor for IsqT function of overload protection etc. However, this really only works on a one-one basis which is why I'm surprised you've set up two motors in "vector mode". I would have thought there would be a real conflict of data being measured by the VFD with two motors as I cannot see electrically two motors with exactly the same characteristic at exactly the same speed/load etc. If that was the case, why not just use one bigger motor?

 

[cokeguy]

Thanks jraef, so should we DISABLE vector mode in the GP10s if running above base speed? I understand the fact that vector mode is meaningless above base speed, but like I mentioned with the conveyor application, if we disable vector mode we may have problems at lower speeds, and since we can´t disable/enable it on the fly, we´d have to change gearings to set 60Hz as top speed, etc...

 

[cokeguy]

Please disregard my previous post, I guess I didn´t read jraef´s post closely enough. It clearly states that there should be no problem in vector mode above base speed with GP10s, they simply work in virtual V/Hz mode. Thanks everyone....

 

[DickDV]

What is this about vector mode not being useful above base speed?!!!!

Exactly the opposite is true. If you want improved speed and torque control at any speed, sensorless vector or certainly flux vector is the correct choice. I depend on ABB's vector control for applications as high as 320Hz and the control is excellent. This is not a plug for ABB. I have done this with Magnetek (Yaskawa) and Allen-Bradley as well.

I your drive manufacturer tells you not to use their vector mode about motor base speed, I would respectfully suggest you seek another brand. I would suspect any brand that does not offer this capability.

 

[Drivesrock]

Hi all.

Sorry jraef, I should have been clearer in my thoughts and said I recommend Vector control ... for SINGLE motor applications.
There have been manufacturers in the past (I will not say who) that claimed to be able to run multiple motors in Vector mode but I believe this was confusing Voltage vector control with Flux vector control which is what we are really talking about here.
But I have run 2 'identical' motors in parallel in Flux vector control on a bridge crane type application. There was one motor on each side of the machine and one big one was not possible. The motor manufacturer guarenteed less than 5% diference between 'identical' motors and the solution worked very well in flux vector control. I was careful to ensure equal cable lengths, which were short anyway. Each motor had individual overload relays.

Sed2developer: The disadvantage I am refering to is the fact that in many drives (in vector mode) the SPEED loop has to be setup manually to achieve the appropriate speed response of the connected machine - motor+load. This is different to the auto-tuning features of most drives for the CURRENT loop which model the motor characteristics. Adjusting the speed loop cannot be done well by looking at the LCD/LED display of drive for the drive responses (eg. speed or Current F/B values) and adjusting the P&I gains. In the olden days 'scopes were used (thinking mainly DC drives now), nowadays - Laptops. If the drives specialist hasn't got one and/or doesn't know how to properly tune the speed loop the motor/load speed regulation can be poor and the drive can trip on overvoltages or overcurrents.