Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

VFDs and Slip Frequency. A question for the Gurus. 2

Status
Not open for further replies.

waross

Electrical
Jan 7, 2006
26,720
1
36
CA
I am wondering if the slip frequency in either RPM or Hz below synchronous speed has the same effect on motor performance regardless of the actual speed.

Put it another way: Starting with a speed/torque curve of an induction motor, If we label the speed axis as slip speed, either in RPM or Hz, (Not in percentage) and starting with zero at synchronous speed and increasing to 100% or 60 Hz at zero speed, can we then use the curve for any commanded speed?

eg: If full load current and torque occur at 40 RPM slip or 1.33 Hz, will we find full load torque at 40 RPM slip, rgardless of the output frequency of the VFD?

Is the slip frequency the primary cause of torque or are there other factors involved?

I think that this is so, but I would like confirmation.
My school days and my library predate VFDs by a few decades and I would like to be sure of my understanding.
On the other hand ask me something about Amplidynes. grin
Thanks

Bill
--------------------
"Why not the best?"
Jimmy Carter
 
Replies continue below

Recommended for you

Yes, the torque curve moves horizontally along the frequency (or RPM) axis. At least at the first approximation. So, having slip frequency 1.33 Hz produces the same torque at 10 Hz as it will at 40 Hz.

But there are things like windage and other losses that eat some of the torque at 40 Hz. So the actual torque may be somewhat lower at 40, or 60, Hz than what you can see at 10 Hz.

Then, of course, you don't get the same torque at 1.33 Hz slip when you get into the "field weakening" speed range (above 60 or 50 Hz, depending on where you are).

Also, at 10 Hz, don't use the high torque for prolonged times if you haven't the cooling needed.

You knew that already.

Gunnar Englund
--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.
 
Thank you Gunnar.
That was my understanding.
I understand the windage and field/weakening issues.
I wanted to be sure that there were no other significant issues.
Also, as I understand, at low speeds the motor will develop the same torque at the same slip frequency, but with less windage a little more torque will be available at the output shaft.
An issue of torque developed versus torque delivered.
image_hju3sp.png

Does anyone have a link to a similar graph that is marked in slip Hz or RPM?
Such a graph would be useful in undestanding and in explaining the performance of motors driven by VFDs.
The percentages don't work once you change the applied frequency.
Thanks again, from me and from others.


Bill
--------------------
"Why not the best?"
Jimmy Carter
 
I had forgotten that thread.
Thanks again Gunnar.
(And Jeff.)

Bill
--------------------
"Why not the best?"
Jimmy Carter
 
The amount of torque available at low speeds is also dependent upon the type of control algorithm used in the VFD. Simple Scalar Control (aka V/Hz) will not allow good torque production at low speeds because the voltage drop caused by the stator winding resistance becomes a larger and larger percentage of the total circuit as the speed slows. You can in most cases manually compensate for that using what’s called “torque boost”, but the point at which this becomes significant will change with loading on the motor. Vector Control in general solves this problem by using a closed loop feedback of the rotor position so that the mP can automatically tweak the V/Hz ratio when it sees the motor performance begin to deviate. Then you get into which TYPE of Vector Control you need as you begin to parse out exactly how much accuracy you need and how slow you want to go, right down to having full torque at zero speed (with the right motor design).


" We are all here on earth to help others; what on earth the others are here for I don't know." -- W. H. Auden
 
Thanks Jeff.
That's the kind of "gotcha" I was concerned about.
At about what speed V/Hz ratio become problematic?
For instance is the simple model good down to 20% speed?, 30% speed?

Bill
--------------------
"Why not the best?"
Jimmy Carter
 
I would say that 10% is OK. You still get full torque. But the slip shows.
BTW, there are sensorless (no encoder needed) VFD:s that go much lower. I have been promoting them for decades. But no one seems to believe it is possible. It is.

Gunnar Englund
--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.
 
I see VFD mfrs make statements about Scalar V/Hz being good for a 6:1 turn down ratio, so 10Hz on a 60Hz motor. I think that is very dependent on the nature of the load / machine. Case in point:

3+ decades ago I got involved in a project at Boeing wherein they "discovered" the magic of VFDs for their machine tools to avoid having their machinists changing belts to change speeds when working with different materials and cuts, i.e. titanium today, aluminum tomorrow on the same machine. Changing belts on sheaves was not only costing them time, the machinists were getting injured: think thousands of machines and machinists making airplane parts, only a 0.5% injury rate per month on the number of speed changes x the number of machines meant dozens of injuries every month for them. So they immediately embarked on a massive retrofit project, over 4,000 spindles all at once. One smart cookie engineer there however put the brakes on the project and commissioned a trial run, which is the contract I got, using "state of the art" drives (in the days prior to Sensorless Vector Control being released). It turned out that at spindle speeds below around 15Hz there were significant tool marks showing up, meaning the tool bit was "jittering" as it cut due to torque instability, despite the claims by the mfr that the drives were good for a 6:1 turn down ratio. We spent a week with their field service guy trying to manually adjust the drive settings to compensate, we never got rid of it to their satisfaction and the entire project was cancelled. Years later when SVC became commonplace we were putting VFDs on spindles right and left with no problems, but that experience led me to me temper my assessment of Scalar V/Hz control to say it is good for a 4:1 turn down ratio at best. That puts me at odds with manufacturers, even my own employers over the years, but experience is fact for me. If it were a conveyor or something where "jitter" was irrelevant, sure. But you can't make a blanket statement about where problems will arise without knowing the application.

Just to be clear though, that is only applicable to Scalar control. Since SVC has become the norm, I rarely see problems with torque stability now. To me there is no good reason to keep using Scalar control. Until recently the one argument was that Scalar control didn't require having to "tune" the drive to the motor. So if it was a centrifugal pump or fan and the VFD went out in the middle of the night, a rookie electrician could replace it with any VFD off the shelf without having to know how to perform an "auto-tune" procedure. But the latest generation of IIGBT transistors have a much MUCH faster turn-on time than the previous generation, as in 10x as fast, and that can cause issues in the motor circuit even in Scalar mode. So I now recommend that all drives be set up with an Auto-tune, even for Scalar control, effectively eliminating the only reason to use it. The only reason that option is still there in VFDs is because IF you have a situation where you have one VFD running multiple motors, you can't use SVC or any form of Vector control, it ONLY works using Scalar.


" We are all here on earth to help others; what on earth the others are here for I don't know." -- W. H. Auden
 
Thanks again Jeff and Gunnar, from me and I am sure from others following this thread who have gained a better understanding of the advantages and limitations of VFDs.

Is this an accurate sum up?: (If not suggest edits and I will change it)

1> Always use auto-tune and Sensorless Vector Control.
---Exception; multiple motors on one drive.
2> 4:1 Turn-down ratio means there are generally no problems down to 25% of rated speed.
3> Consider external cooling at slower speeds.
4> The common speed, torque, current curve may be used with modifications.
---The speed designations across the bottom of the curve must be changed.
Percent won't work. The labels must be changed to RPM slip and or Hz slip.
Zero% slip will still be zero Hz slip or zero RPM slip.
At 30% slip the slip RPM will be 540 RPM slip for an 1800 RPM based motor.
At 30% slip the slip Hz will be 18 Hz slip for an 1800 RPM based motor.
The point originally labelled as 0% slip and 100% Ns will now become the VFD output frequency and the synchronous speed corresponding to that frequency.
The graph is now usable down to about 10% of the motor's rated speed.

Bill
--------------------
"Why not the best?"
Jimmy Carter
 
I wouldn't say that it is as bad as Jeff says. That example was from an extremely demanding application. For most loads, you can use the 10 Hz limit without any problems. For fans and some pumps - you can go slower.

The FAG factory in Schweinfurth, Germany (now re-named) had similar problems when grinding the race-ways in bearings. But that wasn't a question of turn-down. It was about running the drive at certain frequencies where carrier frequency was an integer multiple of motor frequency and the intermodulation caused tiny speed changes that showed as a wavy race-way. The operators and management were not happy about that. "Now with built-in vibration!" wasn't a very successful slogan...

Gunnar Englund
--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.
 
Muthu: The motor is accelerating when starting. As the speed is changing, small speed variations are swamped out.
And precision grinding is generally done when the motor is at the desired speed, not while the motor is accelerating.

Bill
--------------------
"Why not the best?"
Jimmy Carter
 
Gunnar is right, in the vast majority of applications we don’t really care much about torque accuracy or stability at low speeds, I was giving an example of where it DID matter. But according to a report I read (1990 US census, which seems like ancient history now), 70% of all AC induction motors were used in pump and fan applications, so assuming that probably 90+ percent of those would be centrifugal and those are the ones most likely to have VFDs applied, a clear major majority of the applications for VFDs don’t need low speed torque accuracy beyond what Scalar control can offer. I just put it out there as a cautionary tale against making generalizations.


" We are all here on earth to help others; what on earth the others are here for I don't know." -- W. H. Auden
 
Yes. And you can have full torque at zero as well. A ride in a modern elevator shows that. Empty car or full, it accelerates from zero to full speed up/down, decelerates to zero and stops there when door opens. OK, mechanical brakes applied at zero speed. But that is not the case in container cranes, the whole cycle is run without brakes activated.

A "factlet" Elevators are balanced to "want" to go up when car is empty. But that doesn't change the picture much. Full torque is still needed when full. But design kW is lower than without the counterweight.

Gunnar Englund
--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.
 
Thanks again friends.
You have supported my understanding that the standard torque curve may be easily used for the overwhelming majority of VFD applications if the scale is changed from percentage to Slip Hz and Slip RPM.
Next question:
how high can you push the voltage and frequency on a standard induction motor before you encounter problems?
Is the peak voltage seen by the insulation the limit or is there some other unanticipated factor?

Bill
--------------------
"Why not the best?"
Jimmy Carter
 
I wouldn't say unanticipated. When your V/Hz ratio can´t go any further, then you are in field weakening. That is known and hardly unanticipated. The torque goes down and you are running constant available power instead of constant available torque.

If you wire a Y motor in D, then you can bring voltage and frequency up sqrt(3) higher nameplate voltage and frequency and get more HP out of it at the higher speed.

Or did you mean overvoltage due to reflections and ringing? The latter is entirely a question of what kind of insulation and insulation class the motor has.

Not quite sure what your question is about.

Gunnar Englund
--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.
 
waross said:
Next question:
how high can you push the voltage and frequency on a standard induction motor before you encounter problems?
Is the peak voltage seen by the insulation the limit or is there some other unanticipated factor?

Voltage is limited by design of the winding insulation, however most motor mfrs don't use a different insulation for 230, 460 or 600V, they just use 600V as the basis of selecting the magnet wire and use it for everything. But the motor will be designed around a V/Hz ratio, above which you start to enter saturation and below which you lose torque at the square of the delta from the design ratio. So for example if you have a 460V supply and a motor rated for 460V 60Hz, it has a V/Hz ratio of 7.67:1 and as long as you maintain that, you can get full torque from it. But once you get to 60Hz and 460V, if you continue to increase the frequency, you are dropping the V/Hz ratio and losing torque. HP (mechanical kW) remains constant at that point, for whatever that's worth, but the loss of torque begins to affect the amount of actual work you do in most cases.
VFD-Volts-Hertz-Curves_evm4aw.jpg



But because of how this works you can "play games" with voltages and frequencies. For example if you wanted to attain a constant torque capability over a much wider range of speeds, you can start with a dual voltage motor and connect it for the lower voltage, then use a VFD sized for the current at the lower voltage, but at the higher voltage supply. Then you tell the VFD to reach full base speed (i.e. 60Hz) at the lower voltage, that way you can continue increasing the frequency AND voltage at the same ratio and maintain full torque throughout. There are not a lot of applications that require this, but I've run into a few. Example: 1/2HP 230/460V motor, 480V supply, 480V VFD sized for the motor FLA at 230V 60Hz. Program the VFD to reach 60Hz at 230V, then run it to 120Hz, at which point it is 460V and the motor is fine with it, putting out full torque at 2x speed, which is incidentally going to be 1HP.


" We are all here on earth to help others; what on earth the others are here for I don't know." -- W. H. Auden
 
Forgive my poor wording, Gunnar.
Jeff said:
Program the VFD to reach 60Hz at 230V, then run it to 120Hz, at which point it is 460V and the motor is fine with it, putting out full torque at 2x speed, which is incidentally going to be 1HP.
This is purpose of my question.
I have heard of compressor skid manufacturers using a 50 HP, 230 Volt motor with a VFD at 480 Volts and 120 Hz to get 100 HP out of a 50 HP motor.
Keeping the V/Hz ratio constant, how far is it safe to push the voltage and frequency?
Will a suitable filter allow a higher voltage and frequency?

Bill
--------------------
"Why not the best?"
Jimmy Carter
 
OK, understand now.

Simple answer: Yes.

A more realistic answer: It depends.

If windings are designed for that voltage. If bearing life will be OK at that speed. If the motor/coupling are well balanced and vibration levels do not get too high. And many such things. Yes. It is being done routinely to keep weight down. And for aircrafts, the power frequency used to be 400 Hz. To keep weight and dimensions down. I think it still is.

Some regard that as close to overunity and cheating. I call it good engineering.

Gunnar Englund
--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.
 
Status
Not open for further replies.
Back
Top