Root Cause Analysis -Out Board Vibration

[Glanville]

I have a pump which had been experiencing increased Outboard vibration, prior to shaft failure.

What do you think could have been the cause of increasing OB vibration on the pump, could it have been due to defective bearings? The pump operates at about 70% efficiency, so I do not think that the shaft was experiencing any un due forces? I did however find out that the shaft material was changed from AISI -4140 to SAE 4340-

Maybe it could be the shaft material?

Any suggestions would be helpful.

John

 

[bklauba]

Increased vibration might be due to wear of the impeller, causing an out-of-balance condition. Hence your bearing failure. Has anyone checked the balance of the pump rotor?

Others are more qualified to address the difference in shaft material, but, it would not seem to be the cause of the problem you describe.

BK

 

[Carburize]

4340 has better hardenability and toughness than 4140 so provided it was processed correctly I doubt that the material is the cause of the problem.

 

[unclesyd]

Comeback with the heat treatment conditions of both the 4140 and 4340.
What is shaft diameters?
At what point on the shaft is it failing?

As stated by Carburize;
Heat treated correctly the 4340 should perform better.

 

[bilter]

What are you pumping? Do you have access to a virbration spectrum and if so what frequency has an icreasing amplitide in relatiopnship to the frequecy of ratation? If you can get this information you ? will be easy to answer. A high 1X running speed indicates imbalance, high 2x with harmonics indicates misalignment. If you have bearing geometry you can also ID specific bearing frequecies as the cause IE BPFI (ball pass freqency inner race) BPFO, BSF (ball spin frequency). Also vane pass of the impeller.

 

[Glanville]

Thanks Bliter. I did look at the vibration spectrum for this pump prior to failure. The coupling was 180 degrees out of phase, and we are also getting 2X running speed harmonics.

Do you think that misalignment could cause a 4340 shaft to fail, this pump operates at 70% efficiency, so the loading on the shaft would not have been unusual.

Finally a BHN of 302 corresdponds to what hardness on the Rockwell C scale?


FYI - We are pumping slurry.

Thanks,

John

 

[Carburize]

302 BHN is equivalent to approximately 32 Rockwell C

With sufficient number of cycles of load reversal all materials will eventually fail - in tougher materials the fatigue crack will grow a little further before final failure but there will not be a significant effect on total life. In fatigue cracking situations about 80% of the life is involved in crack inititiation and then 20% in crack growth and final failure.

 

[Glanville]

Is 32 an appropriate hardness for shaft material? The shaft is 4 inches in diameter, and failure occured near the impeller end of the shaft.

This shaft had been in for only 18 days...that is a short time, for a pump which operates very near BEP.

 

[Carburize]

Typical shaft hardness range would be 285 - 331 BHN which equates to strengt properties of around 110,000 psi yield strength and 135,000 psi UTS. Your shaft is right in the middle of the range.
At 4-inch diameter you don't really need 4340 to make the hardness 4140 or 4145 would have been OK but the material is probably not the cause of the failure.
Where did the fatigue failure initiate? I suspect is was from a change of section with a fairly sharp radius.

 

[Glanville]

Yes there is a sharp radius of 1/8" R at the end of the shaft just before the impeller. The diameter of the shaft at that section is 1.875".


You are correct, in that the diameter at the previous section is 3.5" Probably resulting in concentrated stress build up at that point(section change) - resulting in gradual crack propogation, then eventual failure...


but, the material may be of a lower grade , than specified, which would have affected the crack propogation characteristics.

I'll know further after the shaft returns from analysis.

 

[sreid]

Let us know what the failure analysis says. Another failure possibility is fretting corossion. This commonly happen to shafts with mounted hubs if micro movement is possible. The fretting corrosion causes stress risers followed by complete shaft failure and it can happen almost unbelievely fast.

 

[unclesyd]

A 1/8" radius if blended is a generous radius. A lot of time the machinist will a leave a very small step or groove at the toe of the radius legs. This can kill the best radius real quick.

What type drive does the pump have?

Have there been more than a normal number of starts and stops.

 

[plasgears]

Sounds like a pressure surge due to out of design operation. Examine pump specs, pipe sizes, material pumped, and impeller design. Excessive back pressure on an undersize pump may produce surges.

 

[MachineryWatch]

I would offer one more thing to investigate. The change in materials may have been just enough to tune a rotor resonance to one of the pump forcing frequencies (1X RPM and harmonics, Blade Pass Frequency, etc.). Exciting a resonance can greatly accelerate fatigue.

As you stated the vibration frequency and phase indicate coupling misalignment. Misalignment can put tremendous loads on a shaft and bearings no matter the "load" rating of the pump (amount of work being done). If there is a problem with improper radiusing, as others have previously mentioned, the increased load on the shaft due to misalignment would have made matters worse.

Skip Hartman

http://www.machinerywatch.com