Just curious as to why (according to my observations) 2xLF almost always so cleanly is present in demodulated spectra on motors powered via VFD?

At the same time no typical RBPF with 2xLF SBs in velocity spectrum.

David

This message has been edited. Last edited by: David_G,

VFD_motors_2LF.doc (96 Kb, 76 downloads)

I have noticed this phenomenon too.

David,

Quite a coincidence that we are facing this as well. Only in an area of the plant that is just under expansion, the motors have been in operation less than a year. They are exposed to the elements on the roof of the plant, with the drives quite a ways away. I'm not sure of the distance, but it may be a factor as well as the manufacturer.

These all had shaft grounding kits installed, so we are not sure why this is so.

There are six supply fans in doghouses with the drives within ten feet of the motors and none of these show the 2xLF.

It is very common to see the fundamental frequency x2 in peakvue, spike energy, etc. name your poison, but if you have 7200 cpm, there is only one way to get that if the VFD is not on 60 Hz fundamental frequency, and that is a different ground potential at the motor than the drive.

Ron said,

"It is very common to see the fundamental frequency x2 in peakvue, spike energy, etc. name your poison,"

Ron messed up. The correct frequency is the fundamental frequency x the number of poles.
In the attached data, the motor is a 6 pole motor. At 60 Hz fundamental frequency it would turn 1200 rpm. At 37.2 Hz the 1x speed of the motor is 744 rpm. The predominant vibration in spike energy is 4485 CPM. This is the fundamental frequency x 6. Remember, this is not 6x running speed, but 6 times the field speed, since this is still an induction motor with slip. If we multiply 6 times 744 rpm, we get 4464 CPM.
Just remember that if this motor was on 60 Hz, the predominant electrical frequency from many issues would be 7200 CPM. Then, just take the ratio of the VFD fundamental and multiply it out for what you should see on the motor.
744 rpm/ 1200 rpm = 62%
62% x 7200 = 4464 CPM!

VFD_Spike_Energy.doc (80 Kb, 22 downloads)

Ralph,

Regarding HP filters (1000 HZ =60,000 cpm in this case)... Yes, what appears to be RBPF is below 60,000cpm, but apparently there is 2xRBPF which is being above Fmax used in velocity spectrum and therefore not captured. The reason 2xRPBF is in there being the carrier frequency is due to the fact that 2xLF showed up so nicely in the PeakVue.

But this is not the point I was trying to bring up in my original question. The main point I wanted to make is that for motors fed via VFD, 2xLF shows up the demodulated spectra more often and more clear then that in a motor fed across the line. IMO, it has to do likely with the shape of the magnetic field in the gap of an induction motor.

Ron,

I think in your example one has to view 4485cpm as 2xLF of the VFD (LF=37.3 HZ)

David

quote:

But this is not the point I was trying to bring up in my original question. The main point I wanted to make is that for motors fed via VFD, 2xLF shows up the demodulated spectra more often and more clear then that in a motor fed across the line. IMO, it has to do likely with the shape of the magnetic field in the gap of an induction motor.

David,

I was 110% sure you already knew what I was saying.

quote:

At the same time no typical RBPF with 2xLF SBs in velocity spectrum.

Your comparison to the velocity not having the RBPF or 2xLF SBs showing but being in the Peakvue is what triggered my reply.
Sorry to have been too in depth, but it wasn't for a well versed analyst's benefit but more for someone trying to gain a little more insight as to what is going on. After all, everyone who views this board and its postings are not always on the same page as some of the older, more experienced ones (myself included). I just wish I knew half as much as some. Looks like after 26 years I would be a little sharper.

Since my previous post seems to be have offended you, I will deleted it. Sorry for not understanding your original post, as I stated at the beginning my first reply.

I have seen this in VFD driven motors for quite some time but my question is: Is there any way to use this information to asses the health of the motor or drive? If these 2XLF peaks are not useful for assesing machine health would there be a way of keeping them out of the spectrum so they would not mask other problems?