Sag Mill Syncronis Motor high axial vibration non drive end bearing 2

[Robbie59]

On site we have a Sag Mill Driven by a 5500KW Motor. We have a vibration issue right across the Mill. The highest vibration is 30mm/sec RMS axial on the motor non drive end bearing.(White metal Bearing)
The gear mesh frequency is 4200 CPM. The dominant frequency is 2400 CPM. We can not relate this to anything on the Mill or drive train but with the high energy vibration the number 1 suspect is the 5500 Kw motor but need something to confirm this
The vibration has high energy vibrating platforms etc. Where ever I take velocity measurements the dominant frequency is 2400 CPM . Measurements on the pinion bearings motor bearing , trunion bearings, the concrete foundations etc.
As mentioned the highest measurement is the motor non drive end axially 30mm/sec . This is a synchronis motor 200RPM . 200RPM 30 pole motor 50Htz

We have inspected pinion bearings ,the air clutch, covers removed from motor for a visual check. Nothing found.The pinion and girth gear yes are worn but thermo graphic and visual visaul indicates this is not the issue.

I believe the vibration is coming from the motor possibly at a coil pass frequency which I have read has high energy vibration.
I am in a very isolated ara and hard to get access to specialists.
I am trying to get the name plate information and hoping someone has photographed it in the past
Have a look at the attachment with spectrum and an overall model of the Mill Vibration
Has anyone had a similar experience. Is it possible to have a coil pass pass frequency of 200 x 12 to = 2400CPM .
Very hard to get information about this motor. It is a GE Canada Electric motor . I have attached a data sheet.
I would appreciate any feed back or indications where to look.
Find the 12 x 200 RPM shaft speed we find the issue
All help would be highly appreciated

 

[Skogsgurra]

I wish that you stopped clinging to those 2400 CPM.

Remedy the faults you have found. And, please, get used to the idea that there are other mechanisms than pure gear ratio phenomena. One is the interplay between exciter frequency, number of diodes/thyristors and their possible malfunctioning plus rotational speed of the motor.

As I said before: Facts first - explanations later.

And don't let the fact that you cannot understand (or explain something to yourself) stop the rectification of obvious defects. Like an air gap that varies between 1.7 and 4.7 mm when the nominal gap shall be 3.2 mm

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[Robbie59]

Skogsgurra (Electrical) 14 Jan 15 07:01

[highlight #EDD400]I wish that you stopped clinging to those 2400 CPM[/highlight]. I do not appreciate these sort of comments. I am asking for advice not comments like this
I am not sure what industry you work in but not mining. Your suggestion just fix what you know. In the case of the exciter air gap it has been there from day 1 as i have found out. Ok we do it your way fix the exciter straight way $500,000 US lost production. It is not the problem,, say to the boss sorry for that
We find something not right again and we fix it with longer down time this time maybe $1M US ,, again sorry boss I thought we fix what we find.
$1.5M US and have not found the answer but lets fix what we know.

The 2400 CPM is that dominant it blocks out all other frequencies. Not important irrelevant !!!!!.

I am looking for proof of the vibration issue to go to the root cause.
All advice appreciated but as i said have a low tolerance for comments not relevant
I appreciate everyone help with this issue and wish all the best in the future
I am out of this forum
Have a nice day

 

[Skogsgurra]

Dear Robbie,

I am sorry that you are sorry.

I have been working with drives from my first employment as a design engineer for drives at ASEA, now ABB, at Västerås headquarters in 1963 and then with Siemens. I have been working with steel works, paper mills, power plants, ships and, yes, mines. LKAB Kirunavaara and other places. I dare say that I have enough experience and that I have solved enough drive problems all over the world to have a reason to say what I say.

As long as you haven't corrected obvious defects, you cannot hope to solve a problem.

I have been present at so many "Let's talk the problem away" sessions that I know the symptoms. They are almost always very similar to what you describe. There is a phenomenon that no-one can explain because it doesn't correspond to the existing knowledge. I also dare say that I have left those sites with the problem solved.

No offense meant. But I think that reality needs to be considered.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[electricpete]

I scanned back over the thread to see if anything jumps out. A few miscellaneous comments/question.
It seems odd that torsional vibration would show highest in the axial direction.
What are the clues that are pointing towards torsional resonance?

The interaction with the clutch disengaged, between the poles is minimal even though the rotor is fully excited

Can you explain what you mean by this. Which poles? Where is the clutch ? Can you give brief summary of the train if you haven’t done already

The strongest support for the motor to still be the possible source is the sinusoidal shape of the waveform (virtually no harmonics) . If gear or mechanically related the vibration data would normally be impactive.

If the impact excites a resonance, the resonant response will be sinusoidal.

I have attached an article I found on the net and it gives a new look into the possible cause.

I don’t think you attached anything.


=====================================
(2B)+(2B)' ?

 

[Skogsgurra]

Not impactive usually means that there are no higher "clanky" sounds like if hit with a steel hammer. The absence of higher harmonics (there is around 3% 2nd and 1% 3rd) points to a soft (non-impactive) excitation of the vibration. So, I still maintain that a thorough check of the working of the excitation shall be carried out. Including low-ohms check of every connection in the exciter and rotor. Plus, of course, fix that air gap issue.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[electricpete]

The strongest support for the motor to still be the possible source is the sinusoidal shape of the waveform (virtually no harmonics) . If gear or mechanically related the vibration data would normally be impactive.

If the impact excites a resonance, the resonant response will be sinusoidal.

Not impactive usually means that there are no higher "clanky" sounds like if hit with a steel hammer. The absence of higher harmonics (there is around 3% 2nd and 1% 3rd) points to a soft (non-impactive) excitation of the vibration

My point was/is exactly the opposite and I stand by it. The exciting force is impactive with higher harmonics. My statement did not refer to the exciting force but to the resonant response to that force. The resonance amplifies a single frequency within the harmonic series (which may or may not be the fundamental). Assuming the amplification of the resonant frequency is much higher than the nonresonant frequencies, then the response is dominated by that single resonant frequency (i.e. the response is roughly sinusoidal).

This is certainly not the case 100% of the time depending on many factors including how high is the resonant amplification compared to the non-resonant. But the op was ruling out mechanical/impactive cause on the basis of the sinusoidal response and I was/am disagreeing with his logic.


=====================================
(2B)+(2B)' ?

 

[electricpete]

I was/am disagreeing with his logic.

Sorry, that came out wrong. I don't necessary disagree with op (since he has many facts I don't and has not fully described his thought process). I am just pointing out a potential weakness in the thought process for him to consider.

=====================================
(2B)+(2B)' ?

 

[Skogsgurra]

Agree. If the impact is repetitive, of the right frequency (2400CPM/n) and not very energetic. There are usually sound effects associated with mechanical impacts. I think it would be difficult not to hear such sounds. Especially if they cause 10 mm/s RMS in such a large machine.

I am almost volunteering to go and find out what the problem really is. But given the boss' attitude, I feel less inclined to do that.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[Skogsgurra]

Sorry, it was 30 mm/s RMS, actually. That's a lot and I don't think that the production loss will be a mere 500,000 USD - or 1000,000 if they let this machine vibrate for months and weeks until it breaks down.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[Robbie59]

To Skogsgurra I apolagise . Very frustrated as the issue is one it hard to locate
But Skogagurra is correct and correct what find. Yes the exciter air gap should of been corrected on the day. We should also check magnetic centre. We have an air clutch locking device and should make sure that is correctly set
I mention down time. A gear specialist suggested we change the pinion and grind the girth gear teeth. They admit the gears are ok but they believe it is part of the issue. One day doen is about $1M Us ,,
Agin when i thought about it and what Skogsgurra said correct every thing and worry about the frequncy later.
Will keep the group informed waht we correct and the results as it can help all

 

[electricpete]

Can you take a look at my post 14 Jan 15 13:58 and respond to those questions. Thanks.

=====================================
(2B)+(2B)' ?

 

[Strong]

Robbie,

Isn't this what you have in the vibration spectrum?
"Loose stator coils in synchronous motors will generate fairly high vibration at Coil Pass Frequency (CPF) which equals the number of stator coils x RPM (# Stator Coils = Poles x # Coils/Pole). The coil pass frequency will be surrounded by 1x RPM sidebands."
<

http://www.vibanalysis.co.uk/vibanalysis/electrical/electrical.html>

The motor NDE housing may be resonant at 2400 cpm, so an impact test could easily verify this. I doubt that there is a torsional or gear fault based on the data and information provided. Save your gear replacement budget!

Walt

 

[electricpete]

Walt/Strong - And as stated above, he must have at bare minimum of 90 stator coils in a 30-pole three phase machine.

Robbie – can you provide a brief description of the train. Now that we’ve been talking about it so long, I’d be interested to hear the gear configuration and tooth count if you’ve got them.


=====================================
(2B)+(2B)' ?

 

[Robbie59]

The drive train is described as follow
5,500 KW motor 50Htz. The motor is Sybcronis 200 RPM. The motor drives the pinion shaft directly through an Eatons Air Clutch
The Pinion drives the Mill through a large girth gear. The pinion has 21 teeth and the girth gear has 376 teeth.
On the Eatons Air Clutch there axial locking device which should be adjusted to the motor magnetic center

On the 12 January the motor white metal journal bearings were rolled out and inspected. Radial clearances were taken and all within specisfication Axial clearances taken may indicate the motor is held slightly off its magnetic centre.
Corrective action should of been taken at this time but the following recomendations were made for the next outage.
Run motor uncoupled fromn the air clutch and mark magnetic center, Make sure the axial locking device is correctlt adjusted
As mentioned before the air gap on the exciter is way out of spec and will be needed to be adjusted next outage. For example top 1.8mm and bottom 3.71mm
The gap shup be 3.2 mm with a 5% variance allow, Production had the final say and wanted to run and not fix these issues at this time. This is why i get so frustrated at times

The pinion bearings have been inspected and radial clearances are in tolerance , actually the bearings look in very good condition. The alignment between the motor and pinion shaft through the air clutch has been checked with a laser and is within the aceptable tolerance.

I have had a number of opinions recieved in relation to torshional resonance

A torsional oscillation needs two explanaitons. A) Why has it only appeared now and wasn’t there before? B) What is driving it?

One theory is the excitation is comming from the motor
Another theory is the pinion may have circular pitch errors which is the source of the excitaion

The gear speacialist who came to site beileves the vibration is due to pinion wear due to the ingress of slurry. If liner bolts leak on the Sag or other Mills the slurry can mix the the grease lubricant and cause wear. So they are saying the change in pitch line is the cause of the vibration.The gears yes there is wear but nothing that would not be expected for the 15 years this Mill has operated. If we did change the pinion and ground in the girth gear teeth we would be down for 4 days and at $1.2 M per day then it would be an expensive excercise. I mean say if we did carry out this recomendation and we still had the vibration issue, This is why I would like to rule out or rule in the motor as fixing up an issue and it is not the root cause then it is an awkward situation
So the difficult part is if they are not fixing issues as they find them like exciter gaps then it makes it hard. As Skogsgurra fix what you know is wrong.
first.
Thanks for all advice

 

[Skogsgurra]

I really feel the frustration, Robbie.

I need to ask a few questions, though. You mention torsional resonance and that is difficult to understand when you also talk about 30 mm/s RMS axial. That leads to
a) what is the vibration in other directions? Torsional resonance is more likely to produce more horizontal vibrations than axial and
b) at 30 mm/s, even your production manager should be convinced that the best thing is to shut down the machine before it kills itself. Is it REALLY 30 mm/s? Could there be a mistake in reading or conversion from imperial to metric units?

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[Robbie59]

The axial reading today on the motor in the axial direction was 27mm/sec RMS. it varies i beleive due to load in the Mill.
This is the worry the readings are true values in MM/sec RMS. I have checked with another instrument and gear specailist agreed with our vibration readings comapred to his instrument.
It is appraoching 4 months since the increase in vibration was picked and a very long time to be running at these amplitudes
Also attached is a picture of the Sag Mill

The axial axial vibration on the motor non drive end bearing as reduced but this could be due to operation condition
Readings taken today
Motor Non Drive End bearing
Horizontal direction 2.34mm/sec RMS
vetrical direction 1.8mm/sec RMS
Axial Direction 27mmlsec RMS

Motor Drive End bearing

Horizontal direction 2.5 mm/sec RMS
vetrical direction 2.1 mm/sec RMS
Axial Direction 7 mm/sec RMS


Pinion Drive End bearing

Horizontal direction 3.3 mm/sec RMS
vetrical direction 5.1 mm/sec RMS
Axial Direction 6,4 mm/sec RMS

Pinion Non Drive End bearing
Horizontal direction 17.7 mm/sec RMS
vetrical direction 11.2 mm/sec RMS
Axial Direction 12.1 mm/sec RMS

 

[Skogsgurra]

There are two measurements that jump out:
Motor Non Drive End bearing Axial Direction 27mm/sec RMS and
Motor Drive End bearing Axial Direction 7 mm/sec RMS

If the motor were a solid body, this would be impossible. You should then have the same vibration level at both ends.

You measure very different levels on NDE and DE. That must (in my world) mean that something is flexing and that brings us back to electricpete's 4 Nov 14 10:00 post.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[Robbie59]

I will check the end plate on the motor non drive end today.
I actually got some good data on the motor on 9th january.
When the motor is uncoupled or with an air clutch disengaged it is running at very low amplitudes

Motor Non Drive End
Horizontal plane .39mm/sec RMS
Vertical Plane .46mm/sec RMS
Axial plane .73mm / sec RMS

Motor Drive End

Horizontal plane .1mm/sec
Verical plane .18 mmlsec
Axial plane .38mm/sec

Usuaally when you run a motor uncoupled you can still pick up any issues. In this case one would think the motor was ok. Could there be a big change in vibration amplitude under load

 

[electricpete]

As you would have guessed, tooth count did not reveal anything interesting to me. But I had to check for myself.

The thing about uncoupled motor runs: If motor vibrates heavily during uncoupled run, you have conclusively localized the problem. But if motor does not vibrate, it is somewhat murkier because many motor induced problems only show up under load. Unfortunately you landed in the latter situation and the uncoupled run is not conclusive.

Did you ever get a chance to count the number of stator coils knuckles? (which is equal to number of stator slots)

Interesting that you’re also getting high axial on driven equipment. Is it dominated by the same mystery frequency? How much does that frequency show in the base of the machine.

For most large motors with sleeve bearings, the motor has large endplay (in the neighborhood of 10-20mm) and the motor has no continuous thrust bearings. During uncoupled run the shaft shoulder axial face can contact the bearing axial face which limits the travel, but when connected to driven equipment the driven equipment thrust bearing controls the motor shaft position. Is this your configuration? Does this clutch limit endfloat? When you inspected the motor bearings, was it verified that the motor rotor is being maintained away from its limit of travel? (for that there any indication of axial contact between rotor and bearing where there shouldn’t have been?)


=====================================
(2B)+(2B)' ?

 

[Robbie59]

I tried to answer this post a few days ago but did not load up, Did not get a chance to check the number of stator coils etc.
The clutch has an axial locking device which will limit end float. So possibly the motor can be held off magnetic center
The 2400 CPm frequency can be measured and dominates on the foundations. We have a lubrication skid at goround level seperate to the Sag Mill foundations and you can pick up the 2400 CPM frequency there
With the motor bearings there are signs of rubbing on the non drive end bearing thrust face
Will keep you updated on what we find
Thanks for your help and advice