Problem - Electrical Fluting VFD driven motor/pump application with a known history of electrical fluting.  Any information I have read talks about wash boarding on the outer or inner race, but I haven't found any information about rolling element patterns in vibration data.  Has anyone run into a situation where the data clearly indicates rolling element damage and not the outer race damage?

Bearing 6313

Shaft speed = 1032 rpm 

BPFO = 3170 cpm

BPFI = 5087 cpm

BSF = 4202 cpm

Accel FFTzoom FFTEnvelope

Dave Leach

Attachments

Photos (3)
Original Post

Your spectrum seems to go to 24000 cpm which may not be high enough to see the fault frequencies of a bearing exhibiting fluting.  The effect of the "washboard" pattern on the bearing surfaces is often seen at relatively high-frequencies, typically between 100000 CPM and 180000 CPM.  Let's just look at some numbers, say you have a inner diameter of the outer race that is 5 inches and the spacing of the irregularities caused by fluting is every 1/8 inch.  That means you have about 125 washboard bands per revolution.  The frequency would be 125 * 1032 (your speed) = 129000 cpm (2150 Hz).  This frequency may be present in itself but it is also believed that it excites a resonance in the bearing.  The spacing of the peaks may be separated by the bearing fault frequencies but are often not seen at the lower frequencies commonly observed.

The spacing for the peaks in this acceleration spectrum are at the ball spin frequency not the BPFO as would be expected.  My question still stands - has anyone ever seen the dominate frequency for shaft currents at the BSF and not the BPFO.  Accel 200k cpm FFT

Attachments

Photos (1)
tagline57 posted:

Problem - Electrical Fluting VFD driven motor/pump application with a known history of electrical fluting. Are we assuming this is fluting based on the history of this motor?  Any information I have read talks about wash boarding on the outer or inner race, but I haven't found any information about rolling element patterns in vibration data.  Has anyone run into a situation where the data clearly indicates rolling element damage and not the outer race damage?

The spacing for the peaks in this acceleration spectrum are at the ball spin frequency not the BPFO as would be expected The peaks in your, seemingly, enveloped data shows as 2xBSF, are the peaks in the acceleration spectrum below showing as 1xBSF or the same 2xBSF?.  My question still stands - has anyone ever seen the dominate frequency for shaft currents at the BSF and not the BPFO.  I do not recall every seeing rollers or balls with fluting without the presence of outer race and/or inner race fluting also.

What is the order number and cpm frequency of the circled peak in this spectrum?
Thanks and Have a Great Day,
Ralph
Fluting Question

Attachments

Photos (1)
Ralph Stewart posted:
Ralph Stewart posted:

 

What is the order number and cpm frequency of the circled peak in this spectrum?
Thanks and Have a Great Day,
Ralph
Fluting Question

Crop and stretch in word indicates something near 121700 cpm, about 118X based the originally stated machine speed of 1032 RPM.

I have asked the customer to change the bearings and send them to me so I can analyze them, however they have told me that they have changed the  bearings in this motor several times and even had a bearing manufacturer analyze them and the report they received.

9-22-2017 2-44-54 PMstated electrical fluting as the root cause.

Attachments

Photos (1)
John from PA posted:
Ralph Stewart posted:
Ralph Stewart posted:

 

What is the order number and cpm frequency of the circled peak in this spectrum?
Thanks and Have a Great Day,
Ralph
Fluting Question

Crop and stretch in word indicates something near 121700 cpm, about 118X based the originally stated machine speed of 1032 RPM.

Thanks John.

Thanks Tagline57.

Tagline,

How close is John's 121700 CPM caculation of the peak's frequency to the meter's calculated frequency? Or did I miss seeing it in your last reply?

Has the bearing displayed in your posted spectrums been changed?

Thanks,

Ralph

 

The bearings haven't been changed yet...this customer actually has three of these units and all three have a history of electrical fluting on the motor bearings.  I have recommended the installation of a product like Agius be installed, however they don't want to spend the money.  So we are forced to monitor and trend the problem, but it's their money !!  

tagline57 posted:

The bearings haven't been changed yet...this customer actually has three of these units and all three have a history of electrical fluting on the motor bearings.  I have recommended the installation of a product like Agius be installed, however they don't want to spend the money.  So we are forced to monitor and trend the problem, but it's their money !!  

Thanks tagline57.

Can you confirm the actual order number and cpm position on the circled peak I posted earlier?

I am sort of like you, never seen 2x BSF only in a possible fluted bearing. I am just wondering if the circled peak (or one of the peaks in the cluste around 120,000 cpmr) might be approximately the 40x harmonic of BPFO at 122880cpm (which is very close to 29x 4202 cpm) , modulated by 2xBSF.

You may have already checked and confirm there is not a ~40x BPFO harmonic in this area.

With the BSF showing a 2x instead of a 1x, this case might be a just a damaged ball. IMO.

Thanks and Have a Great Day,

Ralph

tagline57 posted:

I have recommended the installation of a product like Agius be installed, however they don't want to spend the money.  So we are forced to monitor and trend the problem, but it's their money !!  

The Aegis system (see http://www.est-aegis.com/index.php) isn't necessarily expensive and often can be added to the motor in the field, no removal or realignment necessary.  I have seen them added in about an hour, admittedly to a fairly common motor.  Tag out time probably took as much time as the installation.  There is a split system for motors up to about 500 HP.

As a minimum have the site personnel carefully go through http://static.schneider-electr...rives/8800DB1501.pdf with careful attention paid to proper grounding.

 

Big J posted:

Anyone have any experience with these to solve fluting issues?

http://www.coolblue-mhw.com

I've seen them but never directly involved in their use.  The technology has been around for a long time, often the word "reactor" is used instead of choke.  They supposedly help mitigate the problem but don;t necessarily eliminate it.  So if you go from a bearing life of 1 year to 3 years, is that good enough?  See http://www.emersonindustrial.c...ues/PDF/Reactors.pdf for the Emerson take on the devices.

This is the followup from a post I did a few months ago. I may have just solved a VFD/fluting issue for a customer by uncoiling the VFD cable and shortening it. The first motor change out lowered vibration for 3 months, but it came back. We will see in 2 or 3 months if this solves it permanently.

We have over 2500 wireless sensors in 3 different facilities now. Scheduled to add 13 more facilities next year. 30,000 plus wireless sensors total. They have many more VFD/Fluting issues to solve.

5220

 

Attachments

Photos (5)

First time on this blog and I see these postings are a couple years old, but here goes it anyway:

I reviewed the pictures.  The bearing is a bit blurry making fluting detection more difficult.  If I had to guess, there is likely more than just fluting occurring.  The fluting could be a side effect of a mechanical/lube issue at play.  That said, incidence rates of bearing failure on VFD motors is at least 60% by 2 years and only continues to climb from there.  It is a well-accepted and pervasive problem that particularly plagues operations that run 24/7.  The following information is based on long-term field and lab studies over 30 years.

1) Capacitive discharge through the bearing is a local event caused by the VFD gated mechanism switching on and off.   VFD brands, altering the carrier frequency, installing induction absorbers on the wiring, shunts, conductive grease, etc. are not a fix for this phenomenon.  A shaft grounding device is the only known way to economically eliminate capacitive discharge through the bearing.

2)  Circulating currents can be present on larger framed motors and are inherent to the motor.  This is different than capacitive coupling as described above under #1.  Our field research has shown that motors 100-200hp and with rpm <=1200 or motors >=200hp regardless of rpm are susceptible to circulating currents.  Incident rates of circulating currents being present is 25-35% for the motor parameters described above.  Insulated bearings will not take care of circulating currents as the current will find a parallel pathway and take out bearings in connected equipment.  A typical set-up to deal with circulating currents is to either provide shaft grounding on both ends of the motor or use an insulated bearing on the non-drive end and a drive end shaft grounding system to break the current loop.

3)  Not all shaft grounding devices are equal.  It is certainly confusing with the different available products on the market.   The following are key principals that cannot be ignored while researching shaft grounding products.  Always have these concepts and questions at the forefront of your mind when researching shaft grounding options:

The grounding system MUST be a better competitor to the bearing for grounding the voltage/current away from the bearing.  While most shaft grounding systems lower shaft voltages right out of the box, not all shaft grounding systems lower voltages below the threshold to prevent current flow through the dielectric (bearing grease) then onto the bearing.  Moreover, most shaft grounding systems cannot maintain for the long-term voltages below the known threshold.  One major manufacturer of shaft grounding rings (SGRs) was asked the question during a webinar "how do you know the grounding system is working?" and their response was if the shaft voltage readings are below 10v, then the grounding system is working.  Problem with that response is shaft voltages needs to be much lower than 10v to prevent breakover through the bearing grease.  While 10v will certainly slow down electrically induced bearing damage, it will not eliminate the problem.  We've extensively lab tested the above referenced shaft grounding products and found that out of the box, the shaft grounding ring drops shaft voltages from ~30v peak-to-peak (no shaft grounding) down to 5-7v peak-to-peak.  Even the 5-7v peak-to-peak isn't low enough to prevent dielectric breakover.  As our test motors ran 24/7, the 5-7v creeped up to ~15-17v after just over a year of continuous operation.  These results are better than no shaft grounding, however, the test motors' bearings fluted after 3 years of operation; far less than the motors service life.  

In short, don't be afraid to asked the following questions.  They will cut to the core of how a shaft grounding system performs: 1) what shaft voltages does the shaft grounding system achieve?  2) how does the system maintain the shaft voltages below that threshold known to cause breakover through the bearing grease? 3)  If the environment is dirty, then how does the shaft grounding system ensure a path of least resistance for current to flow away from the bearing?  It doesn't take a lot of contamination to negatively impact shaft grounding performance.  

I hope this information is help.  Feel free to let me know if you have any follow-up questions. 

Sam P.

www.shaftgroundingsystems.com

 

Attachments

Cable length VFD to motor may be important. I had a case with 6 OH slab grinders where 3 were near VFD cabinet installation and 3 had twice and more longer cables. Those with long cables had severe fluting problems, those with shorter had no problems (as long as I saw them).

Oli,

Sometimes with longer runs there are junction boxes or other discontinuities in the ground that can cause the problem.  If both power cable setups and grounding systems were identical, there is no logical answer for this one.  EDM is ALL predicated on insufficient grounding.

No this was in steelworks hot slab grinding, brand new, no connections, no boxes. Maybe cheap cables.... Yes there are logical reasons, this is HF, radio freq. it is no longer regular AC. You can have resonances, standing waves and infinite Murphy ways to mess up enhance and amplify the noise and hence improve the charge voltage at the shafts.... So I find it strictly logical.

Add Reply

Post

HACKATHON in association with IMC-2019

×
×
×
×