Sign up Calendar Latest Topics Donate
 
 
 


Reply
  Author   Comment  
Noknroll

Avatar / Picture

Sr. Member / Supporter
Registered:
Posts: 837
Reply with quote  #1 
Has anyone else noticed the Variable Line Frequency amplitude increase as a motor runs slower or is this just a coincidence with my particular motor? Just for the record, I'm not concerned about it. Just an observation that got me curios. pmp013-12 (2).png 
seanddd

Member
Registered:
Posts: 14
Reply with quote  #2 
Your frequency shows 73.33 would your 2xLF be constant and the harmonics of 1x vary or am I thinking of something else? The VFD's in our facility show dominant 4X with older Rockwell drives and a dominant 6x with the newer ABB drives we have.  The electrical guys here have said its how the drive is configured.  I am interested in learning more if anyone else can shed some light.  Also the 4x and 6x show up more in Acceleration Envelope and PeakVue.  
Big Al

Sr. Member
Registered:
Posts: 102
Reply with quote  #3 
Nok,

I've just checked a handful of my inverter driven motors and I can't see any obvious pattern. To be honest, I've got some inverter driven motors that are intentionally run at a single speed, and the line frequency peaks vary in amplitude even then.

Sean,

In my experience it's never exactly 4x RPM, it's always 4 and a tiny bit and equates to 2x Lf. The dominant 4x appears on 4 pole motors in the acceleration demod data and dominant 2x and 6x on 2 and 6 pole motors. The same peaks are often seen as side bands around RBPF in the velocity data.
Alex

Avatar / Picture

Sr. Member
Registered:
Posts: 190
Reply with quote  #4 
It could be due to some motor characteristics like torque or power at different speed. We all know motors are usually more sweating at lower speed. Often related with temperature increase.
John from PA

Sr. Member
Registered:
Posts: 940
Reply with quote  #5 
Is there a gearbox involved?  I have worked on some VFD using gearboxs that while in a limited speed, there was some gearbox clatter and it was reflected back at the motor.  There were multiple units, maybe about six; for most we decided just to avoid the speed range but one had some form of a controller issue.
electricpete

Sr. Member
Registered:
Posts: 634
Reply with quote  #6 
In theory the force driving the non-load dependent 2*LF (which is most common) varies with the square of the no-load flux level which in turn varies in proportion to voltage and inversely proportional to frequency.

Another way to say it is that the force varies in the same direction as the volts-per-hertz ratio.  

The hertz part of that varies almost proportionally to the speed (neglecting slip).

As speed (and frequency) decrease below rated speed (and rated frequency), most vfd controls would ramp down the voltage to keep the volts/hz constant (to avoid overexcitation)

As speed (and frequency) increases above rated speed (and rated frequency), most vfd controls would keep the voltage constant (to avoid breaking down the insulation), so the volts/hz decreases and vibration decreases. 

Here is a voltage vs frequency curve which combines these two regions. 

Based on the previous three bullets, I personally would expect 2*LF to be roughly constant up to rated speed, and then decrease with speed above rated speed as a result of the v/hz decrease.  This applies to the no-load-dependent variety of 2*LF and just applies to the exciting force....doesn't consider any possible resonances. 

If this is a 6-pole motor rated for 50hz, then this speed variation is above rated speed and the behavior I described explains what you're seeing.  If this is a 6-pole motor rated for 60hz, then the speed variation is below rated speed and this doesnt explain it.  Also if this is a 4-pole motor operating at really low speed this wouldnt' explain it. 

There can also be load related 2*LF vibration, but that is not as common.



Noknroll

Avatar / Picture

Sr. Member / Supporter
Registered:
Posts: 837
Reply with quote  #7 
Thanks all for your interest
Big Al
thanks for saving me some typing to explain to seandd
seandd
what Al said
John
No gearbox, direct coupled to pump.
Alex
I think you might be saying what Pete said in more detail?
Pete
Thanks for your as usual detailed input, I pulled a few paragraphs from your post that pretty much said what I thought might be happening. So I gather that my observation in the OP is probably normal?
electricpete

Sr. Member
Registered:
Posts: 634
Reply with quote  #8 
> So I gather that my observation in the OP is probably normal?

I think it's probably normal when speed is changing above the rated speed.

If speed were changing below rated speed then 2*LF vib should be relatively constant. 

That's what I'd think based on theory, I don't have much experience monitoring vfd motors (our motors are mostly fixed speed).  
OLi

Sr. Member
Registered:
Posts: 1,874
Reply with quote  #9 
OEM's say that going over nominal speed reduce efficiency dramatically and if you need same or higher torque you need to increase magnet-field and things happen and in combination with if you have the magic cos fi compensation turned on it may be like this?  
__________________
Good Vibrations since early 1950's, first patented vibrometer 1956 in the US.
http://www.vtab.se
ssimmon1

Member
Registered:
Posts: 46
Reply with quote  #10 

In practice, I've seen widely varying amplitudes of "2*LF" (just over # of motor poles: ~2.03x, ~4.06x) in PeakVue at a wide range of speeds (almost all less than rated). Never made any calls based on it nor experienced any failures associated with it.

Quote:
Originally Posted by Big Al
Nok,


 The same peaks are often seen as side bands around RBPF in the velocity data.


I've seen it show as sidebands around RBPF....but I've also seen it as sidebands around a much higher (200-260kCPM) frequency, which I assume to be the VFD carrier frequency.

 

 

Previous Topic | Next Topic
Print
Reply

Quick Navigation:

Easily create a Forum Website with Website Toolbox.