Sign up Calendar Latest Topics Donate
 
 
 


Reply
  Author   Comment   Page 2 of 2      Prev   1   2
electricpete

Sr. Member
Registered:
Posts: 651
Reply with quote  #16 

Good point by Barry and Walt that motor starting considerations would often be a huge factor arguing to start with flow restricted for large squirrel cage fans.   A motor driven radial flow fan will start faster with flow restricted, resulting in less heating of the motor (especially rotor).   Repeated DOL starting in a short time is an especially cruel type of abuse for motors since the motor may not cool sufficiently between starts.  And worse yet, if you have induction disk / electromechanical relays (like our plant), the relaying doesn't even protect you from this damage… i.e. if you were inclined to start the motor over and over in rapid succession, the rotor would eventually fail (melt in the case of cast aluminum rotor) and the protection would never trip the machine until the failure occurred.  So procedural limits on repetitive starting are very important for plants that have older relays like us.  Newer digital relays have a thermal model of the motor built in which remembers the recent history of starting and protects (trips) the motor to prevent this type of damage (although you don't necessarily want to rely on that protection).   

The best case for motor starting is actually the exact opposite if we are considering AXIAL  flow pumps/fans (they start faster when flow restriction is reduced) as compared to radial flow pumps/fans (start faster when flow restriction is maximized).  That's probably not the typical fan most people would be concerned with. We only have our axial fans in ducts and they are relatively small (starting restrictions are not as severe with low inertia fans). 

One more piece of trivia that most people don't know:  if there is a specified large motor starting limit (example 1 start per hour), then you are better doing a cycle of run 55 minutes, stop 5 minutes run 55 minutes stop 5 minutes etc than run 5 minutes stop 55 minutes run 5 minutes stop 55 minutes etc if you can manage it.  Most people think this is counterintuitive because they focus on the heating which occurs while running.  But it is the heat removal while running that trumps this for repetitive starting when we are concerned with removing the heat generated by the previous start.   Typically the manufacturer repetitive starting limits something like starts per hour but they don't tell you what part of the duty cycle the motor should be run between starts (it's an odd omission imo). 
 

As far as hydraulic unbalance, I'll defer to your experience if you say it doesn't apply to typical fans.   Googling some more I see this reference which supports that:

 

 https://books.google.com/books?id=nYJKAAAAQBAJ&pg=PA603&lpg=PA603&dq=hydraulic+unbalance&source=bl&ots=3Fi3NEB3LZ&sig=h__IxhIoe1E3LhG7za67BigBSxU&hl=en&sa=X&ved=2ahUKEwjqsLe3kMXdAhUClawKHVA8BqY4ChDoATAAegQIAxAB#v=onepage&q=hydraulic%20unbalance&f=false

 

Quote:
Centrifugal Pumps by Gulich

10.7.3...
9. Geometric deviations between the individual impeller channels create a nonuniform pressure distribution at the impeller outlet, which rotates with the rotor speed and causes synchronous vibrations. The resulting radial hydraulic force has the same effect as a mechanical unbalance and is therefore called “hydraulic unbalance”. If impeller casting quality is inadequate, the hydraulic unbalance can be excessive. Typical casting errors are: variations in the outlet width between or within the channels, non-uniform blade pitches, deviations in blade angles at the outlet or inlet and deviation of the true hydraulic center of the impeller from the center of the shaft bore. The hydraulic unbalance usually exceeds the mechanical unbalance when the head per stage is larger than roughly 400 m; it becomes an important factor for the reliable operation of the pump if the head per stage is above 500 to 600 m. The synchronous hydraulic unbalance grows with increasing flow rate; it is very small near shut-off. The reason for this behavior is that the hydraulic unbalance is caused by differences in lift on the individual blades. Lift differences diminish in separated flow, where the shape of the blades has little influence, Chap. 5

That was an interesting thrumbrule they threw out "the hydraulic unbalance usually exceeds the mechanical unbalance when the head per stage is larger than roughly 400 m [570 psi]"   It seems like they are saying the hydraulic unbalance results from differing pressure around the impeller and so high dp pumps would be more susceptible to pressure-related radial forces than low dp fans.  (even though the area that the pressure acts on might be a little bigger for fans, the pressures are often orders of magnitude higher for pumps).

 I haven't seen anyone else attempt to analyse hydraulic unbalance on the basis of working torque as I described above, but it seems very logical to me that the torque T associated with a fan is seen as a circumferential force on each blade.  If I have B blades at radius R and total torque T the force would be F = T / (R*B).   When the blades are symmetrical, all those forces sum to zero. But if symmetry were lost, there would be a rotating radial force.   If I were to completely remove one blade (an admittedly SEVERE hydraulic and mechanical unbalance) there would be some complicated redistribution of the torque among remaining blades but a reasonable order of magnitude estimate imo is that we would have a net radial rotating force associated with the previous (balanced-case) torque producing force on one blade F = T/(R*B).   It might be an interesting excercize to do a simplistic calculation to  compare the magnitude of the mechanical unbalance force to the magnitude of the hydraulic unbalance force in this type of scenario for a sample fan.  Well, maybe not that interesting... I think I'll pass.

fburgos

Sr. Member
Registered:
Posts: 670
Reply with quote  #17 
I remember a couple of fans that were very sensitive to diffuser/inlet alignment/eccentricity, and had 1xTS depending on difuser position vibration range from 4mm/s rms to 30mm/s rms
Walt Strong

Sr. Member
Registered:
Posts: 889
Reply with quote  #18 
"if there is a specified large motor starting limit (example 1 start per hour), then you are better doing a cycle of run 55 minutes, stop 5 minutes run 55 minutes stop 5 minutes etc than run 5 minutes stop 55 minutes run 5 minutes stop 55 minutes etc if you can manage it."

Large induced draft fan 1500-hp and larger can take longer than 5-minutes to stop rotating after shutdown! Add time for tagging and lockout, plus time for removing access panel, plus time for making weight, grinding and welding weight in place (may also have to remove weight(s), and then close access panel, remove tags and locks, and finally have operations start fan on their time table!!! Oh, and possibly clean fan rotor if it was not done at all or satisfactorily. This is the real world! If you have previous balance data and get lucky with fan behavior, then only one run is needed. I have been lucky a few times, but more often than not it can be a long day

Walt
electricpete

Sr. Member
Registered:
Posts: 651
Reply with quote  #19 
5 minutes and 55 minutes were chosen to illustrate the principle, there's nothing magic about those exact values.     I thought it was worthwhile to share a principle about motors that many folks don't understand, while I was talking about the subject of repetitive motor starting.   I introduced as a piece of trivia in attempt to convey it was a somewhat random/unrelated to the previous discussion.  It comes up at our plant in many contexts that have nothing to do with balancing.  Certainly it might be a challenge to implement this type of a cycle if multiple starts are required during a balancing evolutions.    Hence I wrote in the sentence that you quoted "if you can manage it".  


Sorry if I'm coming off as defensive.  Talking by internet is sometimes way harder than talking in person.


Walt Strong

Sr. Member
Registered:
Posts: 889
Reply with quote  #20 
Pete,

I get your point about trying to run motor long enough to cool it down between starts. My point is that a 5-minute shutdown is just not practical for in-place balancing of most fans of any reasonable size. A 5-minute (or less) balance trial run with a 55-minute shutdown would be nice, but the motor controller may have time restrictions to contend with!

Walt
RustyCas

Avatar / Picture

Admin
Registered:
Posts: 1,810
Reply with quote  #21 
I always try to balance with at last some flow, but typically it will be full flow because nothing special has to be done to achieve that. If imbalance is the only problem, it really shouldn’t matter assuming you have an inflexible rotor. But I can see here “looseness” might be affected by fan loading. For typical fans balanced in place, if you know what you’re doing, I believe you’ll find it doesn’t really matter.

The fans I balance most often are large baghouse fans at a steel mill. We’ve found that wwe need to run them with normal flow to keep the dust off them, especially after shutdowns when tons of dust gets shaken off the duct walls. There are other situations where you’ll either want flow, or as mentioned by others, sometimes not. You just have to think it through, based on the situation. But I agree with Bary that it’s easy to over-think it.

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

Quick Navigation:

Easily create a Forum Website with Website Toolbox.