Sign up Calendar Latest Topics Donate
 
 
 


Reply
  Author   Comment  
Noknroll

Avatar / Picture

Sr. Member / Supporter
Registered:
Posts: 843
Reply with quote  #1 
  • Fan is a center hung axial flow. Variable pitch blades. 998rpm
  • attached readings are from fan out board/Non drive end Horizontal (on triax block mounted at top of rolling element bearing housing)
  • Fan came back from an outage at higher overall and 1 X mm/sec than historic levels at the outboard bearing. Hub, blades and actuator rebuild
  • balance was checked soon after return to service at the inboard/drive end bearing and found to be ok at 1.2mm/sec. (Access restrictions to outboard bearing)
  • route readings for NDE @ 1 X are 4 mm/sec +/- 
Peak and phase were measured at various %  blade opening, all the time both peak and phase were unsteady in Red text

The figures in red are approximate max & mins, on occasion, they did read even higher and lower.

Mean figures are just for graph purposes

Whats the first thing that comes to mind when you see these figures.
Due to the phase swing I think I've ruled out imbalance?

 
Attached Files
xlsx peak & phase.xlsx (13.22 KB, 75 views)

Dave Reynolds

Member
Registered:
Posts: 53
Reply with quote  #2 
First thing that comes to mind, where is your vibration data?

Do you notice a beat frequency?

Is the phase going around in a circle, swinging 90 degrees only?

What changes the pitch of the fan blades, is there a long bar with an actuator that is connected to front side of fan blade? controllable-pitch_axial-flow.jpg 
Have seen air flow/surging affect the 1x on this style of fan, changing blade pitch will affect static discharge pressure. If you building/system is not asking for more air, when you change the pitch of the blade, the air flow is going to go up which could affect airflow downstream of the fan.

Is the fan spring mounted? Was any machine work done? Are blade tips rubbing on the inside of fan housing?
Dave

Noknroll

Avatar / Picture

Sr. Member / Supporter
Registered:
Posts: 843
Reply with quote  #3 
Thanks Dave,
response to your q's in no particular order
+1 correction, I said center hung but its actually overhung, sorry.

No beat frequency. Fan is a Howden made Variax type
I only saw the phase swing 90 degrees, no circles
pitch is actuated hydraulically directly through piping to centre of hub on swivel connection, no long bar.
Fan is load balanced with identical fan on the boiler. While we ran the B fan through the vane % opening range the A fan was in auto and compensating. We were at approx 400MW load at the time out of a full capacity of 660MW
No blade tips rubbing and no spring mounting. 1B FD post.JPG  1B FD trend.JPG  FD 1B pre.JPG  FD Fan.PNG 


RustyCas

Avatar / Picture

Admin
Registered:
Posts: 1,810
Reply with quote  #4 
Was the phase test ever done before? If not, how would you know this behavior is abnormal? If the 1X component is very small, the phase may not "lock onto" 1X and your phase can drift around. If the phase is steady, but changing, as your speed varies, then that is probably due to the system "stiffness" changing with blade angle. For a given imbalance condition, if the phase is the same every time you put the fan at a particular blade angle, then it's likely normal. In other words, is the phase repeatable at a given blade angle. As the stiffness changes, the lag angle of the imbalance response changes. Do you know what the critical speed of the fan is? During your outage, was anything done to the ducting? Strain on the housing can change the natural frequency of the fan, even though they are not directly connected (physically). The change in vibration before/after the expansion joint failure points to this.
__________________
"The trend is your friend"
Walt Strong

Sr. Member
Registered:
Posts: 889
Reply with quote  #5 
"Whats the first thing that comes to mind when you see these figures."

Something loose on/in fan hub. The details are for your detective work.

I had a project to resolve high vibrations on a centrifugal PA fan that had unsteady vibration levels and phase angle. They were blaming the foundation and not the fan. The short story is that I found enough sand (sandblast) inside the fan hub to more than fill a 5 gallon bucket! The moving mass inside the hub was causing both amplitude and phase changes. After the sand was removed, the technician was able to balance the fan.

Rotating moving mass or looseness – same thing!

Walt
GaryVibe

Member
Registered:
Posts: 16
Reply with quote  #6 
This looks like a Howden Variax? The hydraulics on these are intricate.
Was the service work done by Howden?
Is the photo attached of the actual fan?
Most of the Variax fans are sleeve bearing, not rolling element, but it is possible. First thing that comes to my mind is looseness as well. Was the Hub removed from the housing for service? Who did the rebuild? 
I used to work for Howden and these are very intricate fans and somewhat of a maintenance "wh**e". The coupling and drive shaft can also be a contributor, they need to checked to make sure of proper setup, which could contribute to the "Looseness" symptons as well. 

Lots to think about....The Howden Field Service Reps are a pretty good group, maybe a call to them would help...

No I do not work there anymore......

Gary K. 
Dave Reynolds

Member
Registered:
Posts: 53
Reply with quote  #7 

How about the process data? Air flow/pressure/amps.

When you state "flexible joint blowout", does this mean the discharge pressure became so great is blew the joint? Or was it mainly old age?

I notice a fairly large decrease in vibration just by fixing the joint?

If the image you provided is the actual fan, is there a large bearing cartridge? What type of fan blade? 

Need more info on the fan design, how the fan operates within the system, vibration at each bearing location/axis and any phase data

It is surprising to see the vibration levels were quite low and now they have increased dramatically. This is not a good situation...have seen locked couplings, bent shaft, cracked shaft, broken fan blades, damaged/broken inlet/discharge guide vanes, missing air turning vanes, loose bearing fit, loose bearing support all create high 1x rpm....

Dave

Noknroll

Avatar / Picture

Sr. Member / Supporter
Registered:
Posts: 843
Reply with quote  #8 
Thanks all for your replies
Once again, answers to your questions in no particular order

The fan pictured is the actual fan
Rusty ive been contracting at this station for 5 of the last 7 years and to my knowledge these phase checks have not been done before, so quite possibly normal. I'm thinking i should request the same test on the sister fan to compare
This is a fixed speed and I dont know the critical speed

Walt, something loose on the fan has been considered and I've suggested stopping the fan for a physical inspection. The ID fans here get "dust in hub" on a regular basis and need to be cleaned out to bring the fan back in balance, but the subject FD fans dont have the dust in hub problem

Gary, GE did the rebuild and worked on a lot of other equipment during the outage and did terrible work and arent being invited to tender for the next outage. So QA could be an issue here. And yes these variax fans are quite intricate

Dave, the expansion joint blew out from old age.
To describe the fan blade hub arrangement would take too long to type here but if you like you can view this link, you know what they say about a picture and a thousand words
https://www.howden.com/products-and-services/fans/variax-axial-flow-fans
All the faults you listed are things that can be added to the list of things to check if the fan is allowed to be stopped for an internal inspection
trapper

Sr. Member
Registered:
Posts: 77
Reply with quote  #9 

As to what Dave & Rusty mentioned above, the phase data shift could be coming from the change in the angle of incidence of the air flow as you changed blade angle. This also could be influenced to some extent by any surge and/or stall effects on the flow through the fan.

What caught my eye were the waveforms. The 2016 waveform, to me, looked more like something structural with looseness contributing due to its more sawtooth shape. The positive peak down to the negative peak takes around 2/3 of a rotation. It appears whatever work was done in the reconditioning of the fan fixed many of these influences.

On the 2017 waveform, you have a flattened top on the positive peak. The 2x and 3x peaks don't look to have large enough amplitude to have that much influence on the waveform shape. It almost looks like the end of the range of travel of a stuctural movement. Similarly, on the negative portion there is an abrupt change in slope from the point where you have the cursor to the negative peak -- almost like you have a discontinuity (break or rub) at that point in the rotation. Do you see something similar in the other radial direction or on the other bearing?

I would inspect the structure all around the fan support and the ductwork in addition to what was mentioned above.

Noknroll

Avatar / Picture

Sr. Member / Supporter
Registered:
Posts: 843
Reply with quote  #10 
Hello members,
a bit of follow up for anyone interested.
Physical inspection of the 30 blades revealed .
blades 1-15 showed slight radial looseness of 0.1 - 0.2 mm's
while the majority of blades 16-30 showed 0.0 mm's of radial movement.
effectively under 990 rpm of centrifugal force and a diameter of approx 5 metres creating an imbalance. that how it appears to me anyway.

RustyCas

Avatar / Picture

Admin
Registered:
Posts: 1,810
Reply with quote  #11 
Believe it or not, you could have a resonance in the pedestal that is influenced by the strain on the housing from the expansion joint.  At the power station in my former life, we had 7500 hp vane-axial fans (Novenco) that ran near a resonance.  They first "detuned" the resonance by hooking come-alongs to the outlet diffuser and jacking it around.  It was only connected to the fan housing by the expansion joints.  They didn't know they were detuning a resonance, but when I started balancing them, I figured it out.  These fans ran very close to resonance and called for huge amounts of weight.  Sometimes simply adding a balance weight would move the resonance from above to below running speed (or vice versa... 25 years ago).

Vibration would start drifting up over a few months, and we'd go out and simply put pressure on the diffuser using the jacking bolts on the rollers (the diffuser was rolled away from the fan during major maintenance).  Vibration would drop back to normal.  These were boiler ID fans, so yes, dust accumulated.  I could do this about 3 times in the 18 months between shutdowns so we didn't need to do mid-cycle balancing.  I think they eventually had someone come in and do testing and determine that the pedestal/foundation was indeed resonant.  Not sure what they ever did about it.

__________________
"The trend is your friend"
Previous Topic | Next Topic
Print
Reply

Quick Navigation:

Easily create a Forum Website with Website Toolbox.