PERSONAL CASE HISTORY –CRACKED FAN/BLOWER IMPELLER

-

 

PERSONAL CASE HISTORY – CRACKED FAN/BLOWER IMPELLER

In the early years as a Vibration Analyst (2000) I was working at a processing plant in Bathurst NSW. This fluid bed dryer had a fan/ blower around 130Kw, 2 pole motor 2850 RPM. This was located in a fan room with 7 other fans/ blowers.

The vibration at the plumber block bearings were starting to rise and at 10mm/sec RMS and there was a concern. So dynamic balancing was scheduled. There was only access to the fan impeller through a small side hatch. The fan could not be fully isolated from other fans so you had moist hot air coming through, which made it hard to see and work. It was impossible to inspect and assess the impeller condition.

The inlet ducting could not be removed easily as it was welded which was a poor design.

The balancing, instrument was IRD but I also made calculations with polar plots and vectors but the unbalance could not be improved. The vibration amplitude climbed to 15mm/sec RMS over the next week and another attempt was made to balance the impeller and still no success.

The vibration a few weeks later as at 20mm/ sec RMS. A balancing expert was brought in. He checked my vectors and calculations and all were ok. He also attempted to dynamic balance with no success. The balancing expert-recommended changing out the complete rotor assembly.

The next shut down was 3 weeks away. A few days later the vibration climbed suddenly to 30mm/sec RMS and the concerns were at a critical level.

Production was not keen to shut down. I taped off the entrances to the fan room with danger tape and a no entry sign. The Maintenance Manager took notice, the fan shut down and the inlet ducting removed but had to be cut by oxy acetylene.

I was not very experienced at this stage in my VA career and did not know what to expect when inspecting the impeller. I thought there may be distorted or worn blades, obstructions, etc. What was found was not expected. The hub was cracked in an oval shape with only 50 mm in that radius fully intact. The impeller was approximately 1 meter in diameter and with the RPM 2850. The potential for a catastrophic failure, injury to workers, or even worse was very real.



Conclusion

At the time there was a sudden climb in vibration to 10mm/sec RMS the inlet ducting although welded should have been removed to allow a full inspection of the impeller. The cracking in the impeller hub would have been detected at this time and replaced. 

Vibration analysis is important for many reasons but the most important is it reduces the chance of a dangerous situation in the workplace. Safety is always number one.


Robert is qualified to VA Level 111 and a Condition Monitoring Technical Specialist. Robert has held roles as Maintenance Supervisor, Plant Inspector (Oil & gas) Predictive Systems Analyst, Reliability Supervisor, and Reliability Superintendent.


Contacts

Email: robbieccs@gmail.com & robertmva@yahoo.com.au

Stay safe

Robert 

Comments

Post a Comment

Popular posts from this blog

Vibration Analysis of a Bent Shaft.

-
Image
 Vibration Analysis of a Bent Shaft. A bent shaft predominantly causes high 1X axial vibration. The dominant vibration is normally at 1X if the bend is near the centre of the shaft, however, you will see 2X vibration if the bend is closer to the coupling. The axial measurement is the key indicator. Spectrum : Look for a high 1X peak in the axial plane.  If the bend is close to the coupling then you will see a higher 2X peak.  You may also observe higher than-normal 1X and 2X peaks in the radial direction. The phase is also a good indicator to diagnose a bent shaft. The phase at 1X measured in the axial directions at opposite ends of the component will be 180° out of phase. Please note if measurements are taken at the face of the component they will be in phase. TimeWave Form : The TWF is not a good indicator, however, a sinusoidal waveform is expected in the axial plane. Bent Shaft Spectrum in the axial direction. Please note the 2Xrpm will tend to appear if the shaft bend is closer t
Read more

Vibration data for Shaft Misalignment - Angular and Offset

-
Image
  Vibration Data for Shaft  Misalignment  Offset/Parallel Misalignment 1X,2X,3X & sometimes 4X & 5X in the radial direction at Motor Running Speed, RPM Angular Misalignment: 1X & 2XRPM in the axial direction. 1X,2X & sometimes 3X in the radial direction. This can depend on the type of coupling in service. Offset/Parallel Misalignment The Centerlines are parallel but offset.  Offset Misalignment Spectrum. Be aware 3X & 4X RPM also be present. Note the 180 deg out of phase across the Coupling This is why phase measurements are needed to help confirm parallel misalignment. The 2X RPM is usually higher than 1X RPM. Readings taken radially in the horizontal and vertical planes should be similar. Angular Misalignment Spectrum :  Basically stated, angularity is the gap difference across two coupling faces-normally top to bottom or side to side. Angular misalignment in the radial direction Spectrum: Angular misalignment produces a bending moment on each shaft, and this g
Read more

Autocorrelation Instructions for AMS Machinery Manager Version 5.5 or higher. Simply explained

-
Image
Autocorrelation Instructions for AMS Machinery Manager Version 5.5 or higher. Images from Emerson for demonstration purposes In this above example, we have two Acceleration Time Wave Forms (TWF) from two different machines. The first or top TWF is from a Blower inboard bearing in the vertical plane. The second is a cooling water centrifugal pump inboard bearing in the vertical plane. Notice in both TWF’s it is hard to detect if there is any repetitive impacting To Auto correlate, select the time waveform. In the above example, we click on the first TWF blower. Once selected then right-click – (see below Autocorrelation,) After the right-click the above menu box will appear. (Select Auto correlate. See results below) In this example for the blower, the autocorrelation is near 1 so the impacts are repetitive and is most likely related to a bearing defect. The second example below is a centrifugal cooling pump. Again, once the TWF is selected we again right-click and select autocorrel
Read more