BSF and FTF in PeakVue on several new Solids-Handling Pumps

Some time ago I was called in to perform acceptance testing on 7 new pumps in a wastewater application.  These are vertical centrifugal solids-handling pumps with a jack shaft going upstairs one elevation to the drive motor.  All of them operate at or near 900 RPM, with 5 of them on VFDs (the 2 largest on soft starts).  There were lots of problems with air locking and venting of the lines to get the pumps to start pumping.  Once we did individual runs of the pumps, I found that pumps 1/2 and 3/4, which are different sizes, were exhibiting some impacting on the thrust bearing, which is a Timken tapered roller on the top side of the bearing housing.  The impacting was most evident in PeakVue (was also visible as 2xBSF in normal high frequency data) and was highest on pumps 3/4 (16-25 g-s TWF).  All frequencies matched ball spin of the associated bearing with sidebands of cage and harmonics of cage.  I called this out and recommended checking lube and preload on the bearings as this was a common pattern to all 4 pumps.

The OEM became involved and got a second opinion, which indicated no faults.  Not long after the #4 pump had an over-temperature issue with the pump upper/thrust bearing reaching 400F.  The bearings were pulled and inspected and found to visually look fine.  The bearing manufacturer identified excessive preload, or lack of endplay, as the cause.  I did not see the actual report.  The preload was supposedly fixed and the bearings were replaced.  Not long after, the same thing happened to the #3 pump with similar results.  Nothing has happened with pumps 1/2, which are smaller but have exhibited similar BSF/FTF patterns albeit at lower amplitudes (8 g-s).

I have recently gone back in to the facility and collected data again and found the same pattern, but with #4 exhibiting levels of impacting in PV at around 42 g-s.  I also could hear cavitation in the pumps and recorded data on the pump impeller casing and found significant broadband high frequency energy.  This was most obvious on pumps 3/4, the biggest pumps (15 MGD each).  IR thermal profiles on the pump bearings were very normal.

I have discussed with the customer my concerns and also my hesitation to recommend pulling these bearings given the condition the last set appeared to be in when they were pulled.  I find it unusual that the common fault frequency is BSF and cage.  Normally PeakVue has been a great backup to normal data collection, especially once you look at slower speed shafts or have to sift through lots of flow turbulence for fault frequencies.  Could this be a case where the endplay is still improperly set and the end result are these frequencies?  Note that the pumps in question have about 10 run hours on them to date.  They have not been greased since the OEM replaced the bearings.

I appreciate any thoughts.

Original Post

The bearing you talk about reaching 400F and you found nothing wrong with bearings? Can you elaborate on what you did see?

Assume 3 has same issue with wrong preload, did the bearing also get hot and look fine when inspected?

What do you know about the pump operation, good area of pump curve, number of impeller vanes, meeting design spec? If max speed is 900, assume min speed is 700?

Yes setting bearing wrong can cause ball spin and cage rattle, neither are a good thing to have....

What about false brinnelling from other pumps operating? Have you taken data on a pump that is not running while other pumps are running to see if there is a lot of vibration on the machine that is off?

I did not inspect the bearing; I was out of the loop at this point and was getting feedback 3rd hand from sales.  The rotating assembly was pulled and sent back to the pump manufacturer for teardown and inspection.  Their visual inspection, as well as that of the independent witness from the builder, indicated the bearing appeared pristine.  Timken examined said bearing and indicated the cause of the overheat was excessive preload.  I did not see this report.  In what images I did see, the grease on the bearing looked crystallized, almost like metal shavings.  The report from the OEM showed the bearing installed and were not suitable for me to determine any metal damage.

Pump 3 "failed" on overtemp while being monitored in situ by the OEM.  Bearing temperature was observed to climb past 200F, at which point the run was terminated.  I understand the results were similar.

The answer to the 3rd question can be wordy, but I will keep it condensed.  OEM states the pumps are operating in the preferred operating region of the curve.  I have personally heard cavitation coming from pumps 3/4 and observed broadband, high frequency energy in regular data and in PV.  Currently no explanation for this unless suction line is not properly sized or some other throttling action. Wetwell level was plenty high and no signs of aeration or vortexing. 895 RPM full load speed, no VFD, only a softstart. 3-vane impeller and met design flow conditions when baseline tested.  All pumps initially had problems with air-locking and venting.  This appears to have been addressed by performing manual venting of the discharge line versus relying on the air relief system, which was unreliable.  

There are significant problems introduced / transmitted by the discharge piping arrangement and support system.  Vane pass frequency from certain pump combinations does transmit down the common discharge header and excite other non-running pumps further down the line.  Enough so that the overall readings on the non-running pump exceed the values on the running pumps.  We are examining this as a separate phenomena as the BSF issue and overtemp has occurred on pumps during solo operation and all pumps have less than 10 hours runtime.  The pump that was significantly vibrating while off has not been physically examined but has also not shown any bearing fault frequencies in its data.

"Yes setting bearing wrong can cause ball spin and cage rattle, neither are a good thing to have...."  Can you expand on this?  


You state these are vertical solids pumps, whether the pumps have roller, taper roller, ball or thrust bearings, again over 200F with such short run time is a severe issue. With improper bearing clearance and unstable liquid flow, the pump shaft will want to move all over the place. This can cause catastrophic cage failure, cage pocket stretch, worn bearing cage. If PV shows specific frequencies, you can trust you will find the fault in the bearing. I would expect that the bearings were set too tight, so when the pump runs for X amount of time the bearing will heat up rapidly.

Keep in mind PV increases with cavitation...what about velocity readings, are they low, higher in one direction versus the other, what about axial vibration is it higher than radial?

So do the pumps run on VFD's or not, still confused on your explanation. If pump 1/2 are smaller pumps, they have different operating curves compared to 3/4. Do all the pumps discharge into a common header? Can you provide a basic sketch of your installation?

Has someone verified that the bearing temp is getting that hot with a handheld temp gun?

Sounds like to me you are in knee deep in a rats nest. Engineering sounds suspect, multiple pump operation is not a simple issue to resolve. Vane pass is typically associated with system/pump curve issues. All the pumps need to have same size impellers, operate at the same speed and a flow study of the pit should have been done?



When operated solo, each pump runs within HI specs.  When operated simultaneously, and depending on the combination/speed, overall velocity levels can exceed 0.6 IPS-RMS in the cross-discharge plane @ vane pass (confirmed resonances, loose baseplates).  Axial vibration is low.  We feel we have a pretty good understanding of what is happening in this regard, with much of the problem being augmented/transmitted by the discharge piping from pump to pump (loosely supported).  Slight speed differences are resulting in vane pass beats which influence overalls as well (captured via transient TWF).  It is complex.

As for the arrangement, there are 7 pumps that share a common discharge header.  The first 3 are smallest and on VFDs (pumps 1-3).  The next 2 are mid-size and also on VFDs  (pumps 4 -5).  The last 2 are large and on softstarts (pumps 6-7).  The header is split by a normally closed valve that separates the first 3 smallest pumps from the other 4.  In this regard, the first 3 pumps flow to the right down their own line and the other 4 flow to the left down a separate line.  Something like this.


   ^          ^        ^          ^                ^        ^       ^

   7         6       5         4                3        2       1              

Pumps 4-7 are the ones exhibiting BPF and FTF with 7 being the highest amplitude in the TWF.

Bearing temps were initially confirmed with a temperature gun and later checked with an IR camera.  Pumps 6 and 7 are the ones that exhibited high thrust bearing temps but seem to be operating fine after the bearings were replaced and endplay adjusted.

Sounds like you have different size pumps between 4/5 & 6/7? The above links talk about using different size pumps and different pump manufacturers in parallel system. 

Does the output of pumps 123 influence/combine with the output of 4567 even though they go different directions?

Varying vibration levels with different pump combinations. If pump 6/7 are operating at full speed with x flow and x psi, when you start 4 or 5 they need to be able to equal/overcome the psi of 6/7 to pump liquid. Then if 6/7 are both running and you try to start 4 or 5, again they will need to overcome the psi of 6/7.


From the link to pdf....."Because of the slope of the system curve, the pumps in this arrangement will each operate at a lower flow rate when operating together, than they would if they operate alone on the same system. This is particularly relevant on multi-pump arrangements, and it requires careful selection to ensure the most efficient and stable operation."


Thanks for the replies, Dave.  It is a design I have not encountered before.  The shared header and various operating speeds combined with different vane pass frequencies and loose mechanical supports are causing a lot of head scratching.  There is also audible cavitation on pumps 6 and 7, visible in the data too.  I will try to update this post once we come to resolution or next steps.  There is a lot to consider.

Walt, see the screen grabs attached.  I am trying to disposition these frequencies as recorded in regular processing and in PeakVue.  The data I am providing was all recorded on the same bearing at roughly the same time.  Note this bearing had been replaced several weeks earlier due to overheating to 400F.  The pump itself has less than 10 run-hours on it.  This bearing has a fraction of that time.

I uploaded as a Word doc and a pdf for ease of resizing etc.


I think it is coincidence that BSF is numerically very close to 3xSS that is caused by pulsation from three impeller vanes. I am not convinced about the cage frequency being present either.

The regular acceleration levels are below 4-g (0-2000 Hz BW?), while the PeakVue values are 40-g. This suggests very high frequency vibration above 2000 Hz.

The data does not indicate an actual bearing fault (cage or balls), so I would not be surprised if the bearing appeared without visual flaws during disassembly inspection.

The high frequency PeakVue data may indicate poor lubrication, overload, or looseness impacting in the bearing. I assume that Cavitation or other impacting components are not affecting the vibration data, but perhaps they are!


The cage is more subtle and appears more as a sideband around the ball spin.  The harmonic family identified by the cursors are non-synchronous (speed verified) with the second and fourth harmonics being more pronounced and match very well with the Timken defect frequencies for BSF.  Autocorrelation agrees, with majority contribution being periodic.  Do I understand correctly you believe these are vane pass harmonics in the PeakVue data?

"Do I understand correctly you believe these are vane pass harmonics in the PeakVue data?"

You have confirmed to my reasonable satisfaction that the PeakVue data is associated with bearing fault frequencies and not pulsation.

Was the grease line feeding thrust bearing purged to remove potential contamination when thrust bearing was replaced?

Is the thrust bearing overloaded based on gravity rotor weight, liquid/solids within pump rotor, pump static pressure, and dynamic pressure?

Are there any destabilizing up-thrust forces acting on thrust bearing due to pump operation?

Where did the iron come from (bearing or contamination)?


Walt Strong posted:

"Was the grease line feeding thrust bearing purged to remove potential contamination when thrust bearing was replaced? 

- I do not know the answer to this. It is a good idea all the same.

Is the thrust bearing overloaded based on gravity rotor weight, liquid/solids within pump rotor, pump static pressure, and dynamic pressure?

- I do not know the answer to this.  This would be a design issue, I imagine.  The only 3rd hand feedback I had received was that the endplay on the overheated bearings was inadequate (too little) and determined to be so after the fact.  Based on the fact that I am once again seeing the same frequencies, I wonder about the current status of the endplay.

Are there any destabilizing up-thrust forces acting on thrust bearing due to pump operation?

- There is some mild cavitation.  It is audible and shows up in the normal data and in PeakVue readings taken on the volute.  The PV impacting was observed with the pump running solo and after 30 minutes + runtime.

Where did the iron come from (bearing or contamination)?

- Wear debris analysis indicates these are rubbing wear platelets and are likely consistent with bearing component wear.  There was no sign of weld splatter or grinding swarf.  We did not get into using a SEM and doing a full-blown failure analysis.


Responses in-line above.  Thanks for the reply.

Add Reply