Eng-Tips is the largest engineering community on the Internet
Intelligent Work Forums for Engineering Professionals
-
Congratulations waross on being selected by the Tek-Tips community for having the most helpful posts in the forums last week. Way to Go!
4-Run Method of Balancing 1
- Thread starter mtigos
- Start date
GregLocock
Automotive
Is this for school?
Cheers
Greg Locock
SIG
lease see FAQ731-376 for tips on how to make the best use of Eng-Tips.
Cheers
Greg Locock
SIG
lease see FAQ731-376 for tips on how to make the best use of Eng-Tips.- Thread starter
- #3
No, this is not for School..I want to balance a Flue-Gas Fan that keep coming out of Balance. I just want to know whether finding the correction mass has the same relation as when you use Mils for the vibration parameter.
Thanks.
I am a Graduate Engineer (Mechanical)
Thanks.
I am a Graduate Engineer (Mechanical)
GregLocock
Automotive
Save yourself hours of agony and get a proper analysis system that uses phase as well as magnitude.
The units you use don't really matter.
Cheers
Greg Locock
SIGlease see FAQ731-376 for tips on how to make the best use of Eng-Tips.
The units you use don't really matter.
Cheers
Greg Locock
SIG
lease see FAQ731-376 for tips on how to make the best use of Eng-Tips.-
1
- #5
Is this a continuation of the posting that begin on Jan 26th?
At that time you had 'experts coming in to do the balancing' and now you are doing it yourself?
Anyway....to answer your questions
a)mils/sec is not a sensible unit for vibration measurements
mils are 0.001 inches displacement
mm/sec is a recognised unit in terms of vibration velocity
Do you understand the difference? If you can 4 run balance with mils you should be able to balance with other units.
b) Unbalance is unbalance no matter what units you measure vibration in - simply put, the objective of balancing is to get the centre of mass to coincide (or get close to) with the centre of rotation - you do this by adding mass (ounces or grams, depending on your background) to the rotor assembly
Greg is right - get a setup that uses phase angle and you can balance much more quickly. if your vibration is not due to solely to unbalance (thermal effects/cracks/misalignment)and you go for a 4 run method, you are going to waste a lot of time and effort.
At that time you had 'experts coming in to do the balancing' and now you are doing it yourself?
Anyway....to answer your questions
a)mils/sec is not a sensible unit for vibration measurements
mils are 0.001 inches displacement
mm/sec is a recognised unit in terms of vibration velocity
Do you understand the difference? If you can 4 run balance with mils you should be able to balance with other units.
b) Unbalance is unbalance no matter what units you measure vibration in - simply put, the objective of balancing is to get the centre of mass to coincide (or get close to) with the centre of rotation - you do this by adding mass (ounces or grams, depending on your background) to the rotor assembly
Greg is right - get a setup that uses phase angle and you can balance much more quickly. if your vibration is not due to solely to unbalance (thermal effects/cracks/misalignment)and you go for a 4 run method, you are going to waste a lot of time and effort.
- Thread starter
- #6
TPL
Thanks mate, and yes I am still on that January balancing issue. I have decided to do it myself and learn on my own through help from guys like you. Thank you so much, I am now confident to go to the Section Engineer and let him know that I will attend to the problem. As for other issues like misalignment, cracks and heat/thermal, I have checked and made sure, but though I have suggested the support structure to be stiffened, I am going to go for the balancing and hopefully get it right. The support structure is only going to be re-done in the next shut-down, 20th April.
Thank you all for your valuable input. I owe you all.
Thanks mate, and yes I am still on that January balancing issue. I have decided to do it myself and learn on my own through help from guys like you. Thank you so much, I am now confident to go to the Section Engineer and let him know that I will attend to the problem. As for other issues like misalignment, cracks and heat/thermal, I have checked and made sure, but though I have suggested the support structure to be stiffened, I am going to go for the balancing and hopefully get it right. The support structure is only going to be re-done in the next shut-down, 20th April.
Thank you all for your valuable input. I owe you all.
I am now confident to go to the Section Engineer and let him know that I will attend to the problem.
I do hope that you Section Engineer recognises that your confidence far outweighs your ability and decides to call in someone who really does know what they are doing.
I do hope that you Section Engineer recognises that your confidence far outweighs your ability and decides to call in someone who really does know what they are doing.
- Thread starter
- #8
I understand TPL. My confidence is not coming out of reading the help I have got from you all only, though it was very insightful. I have a Balancing Jig at my disposal and throughout this week I have been working on it and learning balancing issues (mostly single plane). I have learnt a lot. I believe in challenge and strive to better myself always. I saw you mentioned "mils/sec is not a sensible unit for vibration measurements", it was a typo on my part, meant mm/s. I may sound like the wrong kind of guy for the job to you, but I assure you, I will not try/do anything that I cant do.
The 4-run method seems easier and faster than what one guy was doing on the FANS initially, i.e; putting weights randomly and hoping that by some trial-and-error stroke of light he gets the balance right.
There was not even a log where he kept recordings of what mass he placed where and what at what angles. I was not involved at all, just observed. But I took it upon myself to try and do a better job. We do not have a set-up that gives amplitude as well as phase, only amplitude.
Thanks.
The 4-run method seems easier and faster than what one guy was doing on the FANS initially, i.e; putting weights randomly and hoping that by some trial-and-error stroke of light he gets the balance right.
There was not even a log where he kept recordings of what mass he placed where and what at what angles. I was not involved at all, just observed. But I took it upon myself to try and do a better job. We do not have a set-up that gives amplitude as well as phase, only amplitude.
Thanks.
GregLocock
Automotive
"putting weights randomly and hoping that by some trial-and-error stroke of light he gets the balance right."
Yes, I've done a fair bit of that myself, particularly on 2 plane balancing.
OK, you need to identify 4 unknowns, phase and mag of the original out of balance (P0 and M0), and the phase and mag of the response of the system to a known trial mass in a known position (Pt and Mt). You might be lucky and be able to assume that Pt is 'obvious' but that is not true in general.
I suggest you make up an example where you know what each of those four unknowns are, and plot them on the phase plane. Then see if you can back-derive that plot from the 5 amplitudes - I am told there is a geometrical construction to do it but did it by observation and trial and error myself.
Cheers
Greg Locock
SIGlease see FAQ731-376 for tips on how to make the best use of Eng-Tips.
Yes, I've done a fair bit of that myself, particularly on 2 plane balancing.
OK, you need to identify 4 unknowns, phase and mag of the original out of balance (P0 and M0), and the phase and mag of the response of the system to a known trial mass in a known position (Pt and Mt). You might be lucky and be able to assume that Pt is 'obvious' but that is not true in general.
I suggest you make up an example where you know what each of those four unknowns are, and plot them on the phase plane. Then see if you can back-derive that plot from the 5 amplitudes - I am told there is a geometrical construction to do it but did it by observation and trial and error myself.
Cheers
Greg Locock
SIG
lease see FAQ731-376 for tips on how to make the best use of Eng-Tips.If you have your heart set on the 4-run method, then here is an online calculator for it:
In no way do I recommend that Web site nor do I suggest you use the 4-run method for a large fan. You would be better served by getting a vibration analyzer capable of measuring speed, filtered vibration and phase angle and then either use a single plane or 2-plane solution. I use a CSI 2120 analyzer, but there are plenty of others on the market. I always try to find the heavy spot, so that my trial weight will typically reduce vibration level and may be left in place and then a final trim weight is added. The 4-run method is OK i you are balancing a ceiling fan in your house, but it is not good for large fans.
Walt
In no way do I recommend that Web site nor do I suggest you use the 4-run method for a large fan. You would be better served by getting a vibration analyzer capable of measuring speed, filtered vibration and phase angle and then either use a single plane or 2-plane solution. I use a CSI 2120 analyzer, but there are plenty of others on the market. I always try to find the heavy spot, so that my trial weight will typically reduce vibration level and may be left in place and then a final trim weight is added. The 4-run method is OK i you are balancing a ceiling fan in your house, but it is not good for large fans.
Walt
GregLocock
Automotive
Incidentally even if you only have a two channel scope you can figure out the phase, there is really no excuse for not measuring it.
Nice find Walt, but it is still clear as mud to me.
Cheers
Greg Locock
SIGlease see FAQ731-376 for tips on how to make the best use of Eng-Tips.
Nice find Walt, but it is still clear as mud to me.
Cheers
Greg Locock
SIG
lease see FAQ731-376 for tips on how to make the best use of Eng-Tips.
electricpete
Electrical
It seems to be a popular method mentioned recently on maintenanceforums.com Victor Wowk is a big proponent. Why some people preferit, I'm not sure.
The method is spelled out several places in cookbook fashion. The reason why it works is not obvious. With some help from WFoiles, I put together a vector analysis which shows why the method works (attached).
=====================================
Eng-tips forums: The best place on the web for engineering discussions.
The method is spelled out several places in cookbook fashion. The reason why it works is not obvious. With some help from WFoiles, I put together a vector analysis which shows why the method works (attached).
=====================================
Eng-tips forums: The best place on the web for engineering discussions.
I always thought that the 4 run method was a solution along the lines of a Venn diagram setup.
The 4 run method will only work if most of the vibration is due to unbalance AND the unbalance remains constant throughout the 4 runs - if there is any change from run to run it won't work and then you are left with lots of head scratching.
I want to balance a Flue-Gas Fan that keep coming out of Balance
If you have a phase reference (tacho pulse or keyphasor or whatever you wish to call it) you can use previous balancing efforts to go for a one shot balance each time the vibration goes high - that is the way to impress people, otherwise you just have to go for a 4 run balance each time (although with a bit of experience, you might be able to modify it down to a 3 run exercise).
If this machine is continaully going out of balance, you don't really want to be welding weights on - I suggest the first step would be to drill a number of holes (say 12, to include the 4 needed for the 4 run balance and the others to get as close to the final solution as possible) around the circumference of the fan (or one on each blade) and use nuts bolts and washers as the weights
The 4 run method will only work if most of the vibration is due to unbalance AND the unbalance remains constant throughout the 4 runs - if there is any change from run to run it won't work and then you are left with lots of head scratching.
I want to balance a Flue-Gas Fan that keep coming out of Balance
If you have a phase reference (tacho pulse or keyphasor or whatever you wish to call it) you can use previous balancing efforts to go for a one shot balance each time the vibration goes high - that is the way to impress people, otherwise you just have to go for a 4 run balance each time (although with a bit of experience, you might be able to modify it down to a 3 run exercise).
If this machine is continaully going out of balance, you don't really want to be welding weights on - I suggest the first step would be to drill a number of holes (say 12, to include the 4 needed for the 4 run balance and the others to get as close to the final solution as possible) around the circumference of the fan (or one on each blade) and use nuts bolts and washers as the weights
- Thread starter
- #14
Guys I am thrilled to report that I have done the 4-Run method on a big Fan (Flue Gas), and it worked!!
The Fan initially tripped and my Boss called me and asked if I was confident I would do it before knock-off time, I said YES. It took me about 2hrs and the Fan vibration levels dropped to 2mm/s (NDE) and 3.8mm/s (DE). The trip occured at 44.11mm/s (NDE) and 48mm/s (DE). So you can see that was indeed a dramatic improvement, i.e; comparing it with 16mm/s (NDE) and 12mm/s (DE) that the Guy from South Africa always managed. I do agree that we need to get a set-up that measures both maginitude and phase to make it easier but my company is a bit skeptical about spending cash since the economic slump we all going thru.
But thanks guys for the constructive critism (TPL) and help (TPL,GergLocock and Strong).
This is indeed an informative house.
The Fan initially tripped and my Boss called me and asked if I was confident I would do it before knock-off time, I said YES. It took me about 2hrs and the Fan vibration levels dropped to 2mm/s (NDE) and 3.8mm/s (DE). The trip occured at 44.11mm/s (NDE) and 48mm/s (DE). So you can see that was indeed a dramatic improvement, i.e; comparing it with 16mm/s (NDE) and 12mm/s (DE) that the Guy from South Africa always managed. I do agree that we need to get a set-up that measures both maginitude and phase to make it easier but my company is a bit skeptical about spending cash since the economic slump we all going thru.
But thanks guys for the constructive critism (TPL) and help (TPL,GergLocock and Strong).
This is indeed an informative house.
electricpete
Electrical
TPL - I don't understand your comment. Was this a comment about my proof? If so I would point out it is more of a proof than a cookbook recipe for the graphical solution. But if you study it you can see the recipe.TPL said:
=====================================
Eng-tips forums: The best place on the web for engineering discussions.
electricpete
Electrical
I guess one more comment is required. If you want to understand how to do it, the best way is to google and find a step by step procedure. Once you know how the step by step procedure works, if you want to know why it works my proof explains that.
Sorry I am not trying to be picky or pedantic or whatever. Just wanted to clarify what my attachment was.
Done beating my dead horse.
=====================================
Eng-tips forums: The best place on the web for engineering discussions.
Sorry I am not trying to be picky or pedantic or whatever. Just wanted to clarify what my attachment was.
Done beating my dead horse.
=====================================
Eng-tips forums: The best place on the web for engineering discussions.
I would recommend hiring a professional that allow you to follow what they do, a professional who doesn't use the 4 run method. The 4 run method is very inefficient, and if you have to trim balance you have to start over with another 4 runs.
If something changes, keeping track of phase and the vector changes helps greatly. Vector plots of the runs helps explain what has happened and can be good in a report that can be used in the future for other balances of this or similar machines; try that with the 4-run method.
Regards,
Bill
If something changes, keeping track of phase and the vector changes helps greatly. Vector plots of the runs helps explain what has happened and can be good in a report that can be used in the future for other balances of this or similar machines; try that with the 4-run method.
Regards,
Bill
MachineryWatch
Mechanical
Either what Bill says, or actually invest in one of the 2 or 3 day rotating machinery field balancing courses offered. Not only do you learn the manual vector calculation method for single and multiplane balancing, but the instructors have usually been doing this for years and have a lot "tips" to share that will help make balancing easier.
Finally, the instructor, after making sure you know how to do the manual balance calculations, will tell you that you really should be using one of the "canned" balance routines that are available with most analyzers as an option. Being able to manually balance using the vector diagram methods will serve you well as a reality check of results when you make a mis-step with one of the "canned" programs.
The four run method can work, there is no question of that. The main problem is that it is a method to use when you have no phase reference information. Having no phase data indicates that you are going to do a balance without truly knowing if the machine really has a balance problem or if it has high vibration levels due to one of the many other possible sources of high vibration amplitude at 1X RPM. If you have a soft foot, or resonance issue or an eccentric rotor or a bent shaft, the attempt at balancing will only frustrate you! Worse than that, you will suffer a loss of credibility at your facility that will be difficult to overcome.
Fully 75% of the times I have been hired to do field balancing of rotating machinery in my 30 years of experience I have found other problems to be the dominant source of the vibration. Most times the balance procedure was not even necessary.
Finally, the instructor, after making sure you know how to do the manual balance calculations, will tell you that you really should be using one of the "canned" balance routines that are available with most analyzers as an option. Being able to manually balance using the vector diagram methods will serve you well as a reality check of results when you make a mis-step with one of the "canned" programs.
The four run method can work, there is no question of that. The main problem is that it is a method to use when you have no phase reference information. Having no phase data indicates that you are going to do a balance without truly knowing if the machine really has a balance problem or if it has high vibration levels due to one of the many other possible sources of high vibration amplitude at 1X RPM. If you have a soft foot, or resonance issue or an eccentric rotor or a bent shaft, the attempt at balancing will only frustrate you! Worse than that, you will suffer a loss of credibility at your facility that will be difficult to overcome.
Fully 75% of the times I have been hired to do field balancing of rotating machinery in my 30 years of experience I have found other problems to be the dominant source of the vibration. Most times the balance procedure was not even necessary.
What is the theory behind the 4 run no phase method? It is not intuitively obvious to draw the second circle centered on the circumference of the first. I've looked and looked and can only find the how to. I've used the method successfully but didn't know the basis for the different steps.
Thanks Klyde
Thanks Klyde
- Status
Similar threads
- Locked
- Question
- Locked
- Question
- Locked
- Question
