Skip to main content

Replies sorted oldest to newest


Thx, it is in the drying section and you are right,  it is driven by Fabric.

but we have this case before (Fluting in a driven guide roll) just was beside this Roll ( 2 meters distance between them)

- i do not know if we have static electricity anywhere on the machine and i do not know how to test it.

but we have many OLD DC motors are suffering from fluting.


i learnt from you a lot...

Filter is (50-1000) Hz and the side band comes from Cage, little energy spectrum.

the energy now came to the velocity spectrum, how severe is the damage, Fluting / Spalling ?

the spectrum is for the Tending side of the roll.

Best Regards


I have seen a papermachine with electrostatic from the plastic type clothing. It came complete with lightnings and all. One guy had a strike in the helmet.
So if you measure on the outgoing gbx shaft of the driven cylinder and that is higher than the driving shaft and the outgoing bearing have fluting and nothing else have. It is pretty clear. So it happens, not certain it is the case here.


That's an interesting paper, thanks for sharing.


You're right about the lightning strikes, I've seen them too. Years ago I had a machine that would power down my CSI 2120. Luckily it never burned anything out, it just shut the power off.


These two bearing faults are just the immediate problem. You will continue to have bearing faults until you find and fix the source of the stray currents.


In my book the data goes hi quick for fluting, remain hi for quite long time and then pretty much fail, most statistics from VFD fails that is. Spalling have a more trendable history. In the end fluting may turn in to spalling for the end game....
DC induced or flash induced may give different trending.
DC induced mostly do trend up before fail in those cases I have seen.

Thx all for the input,


i measured all Motors, there is only one drive in the press section directly before the dryer section which has a strong fluting in the G.B input shaft.


the the motors we have r equipped with carbon elements.

i will continue monitoring the behavior..

How can i check the machine frame for a current discharge ? ?



I agree that your case might be a tough one to call

I have seen all types of paper machines, ranging from the machines that actually make the basic paper product from pulp to, and including those machines which convert the finished parent paper rolls into consumer products (tissue, towels, napkins, etc.) and all of them, at some time or another, have created static electricity, but I have never seen the case where, after the disassembling of the removed "failed" bearing, there were any signs of fluting. Not saying in it not possible.

In your case, I would lean toward, until proven different, that yours may be:         (1) showing, on the low end frequency range of 1x BPFO, as having a problem with the rollers ceasing to turn when leaving the load zone causing a "slide" defect at the entrance of the load zone's contact point.

And may also be showing:                                                                                 (2) Early stage of a conventional flaking fault, judging from the higher frequency peaks and the cage (FTF) frequency sidebanding the BPFO harmonics in the Enveloped data.

BTW, you said your Enveloped data filter started at: quote: "Filter is (50-1000) Hz and the side band comes from Cage, little energy spectrum." Is this correct? If it is, can you take an enveloped data set with a 1K to 3K High Pass Filter or one which will be starting above the 1K Hz Fmax in the conventional data and having a 1K Fmax? I guess the starting and stopping High Pass Filters can be regulated by the user, huh?

To me, IMO, this (enveloped data) is basically looking at most of the conventional data which you already have in the 0 to 1K Fmax. Enveloping sort of separates thing a little more clearly that does the conventional data that covers basically the same data area.

Just some thoughts. But I may be completely wrong.



Hi Ralph,

u said:

as having a problem with the rollers ceasing to turn when leaving the load zone causing a "slide" defect at the entrance of the load zone's contact point.

How do u get this analysis,   i love learning from u

-attached env 3 ( 500-10K) HZ, in the last measurement, it shows only noise ,

-SKF has standard fixed filters




Files (1)


Quote "How do u get this analysis,"

See the attached doc file.

Quote "SKF has standard fixed filters"

I think that the 500 Hz to 10,000 Hz filter is a little too broad to do much good on a bearing defect as small as yours.They (SKF) do not have one that is over 1K and under 10K? Like 2k to 5K or 1K to 5K?

The 50 Hz to 1,000 Hz filter shows a better picture, IMO, than does the 500 Hz to 10,000 Hz.

I think you do not have an electrical fluting problem but maybe more of just a small defect developing and a small "slide" defect present, based on the data I see.

Only my opinion and I could be totally wrong.




Files (1)

I took cleaning brush for a shotgun and connected a rod and a cable.
I hook up a oscilloscope to that and frame gnd and a std DVM (RMS) and compared the findings from a a gbx driving a driven roll and the AC voltage was higher towards the machine eg. output shaft compared to measured values on the input shaft towards the motor.

Be very carful when, where and how to apply the brush to the shaft, look with a strobe first so there are no keyways lock screws or other things. Make sure you select a smooth landing zone on the shaft and have no things on you that can hook on like rings or stuff and have the connections loose so nothing can latch on. I did it once a long time ago so there were no HR I could avoid to inform. If you do it be careful.

If you would do it properly, more than once, mount a carbon brush isolated from frame and pull out a cable when machine is stopped, then it would be a bit safer for more than a one time measurement.

Hi guys!

Based on measurements provided I agree with Ralph regarding the most probable it is not a fluting problem, it might be a issue on sliding effect. That problem may be caused by lack of radial load or an ovality issue on pillow block. The important thing is to monitor if such bearing frequencies increases over the time in the FFT and timewavefom.

I also have read on high 1x BPFO in velocity spectrums related to too tight bearing.

There is a way to identify is fluting is present on an application, SKF developed the inexpensive TKED1 device which counts the arcs events in a period of time, no contact needed.

I also have proven the PFM-711A  electrostatic fieldmeter (visit, it measures the electrostatic field without contact.


@Registered Member posted:


very useful info.

i red that the  TKED1 work only for motors since it detects change in magnetic field,  is this true?

the bearing in concern is very far from motors, could TKED1  help in my case ?




TKED1 is able to detect micro-arcs even if an electric motor is not involved. If you have VFD motors driven it worth to have one, I think.



Add Reply

Link copied to your clipboard.