Hello guys

i have experiencing a quizzical phenomenal on an engine+gearbox+pump train.

This is a crude pump with Caterpillar engine,Lufkin Gear and a pump (ONE to be updated).

This machine behave normally in terms of VIBRATION during running 800-1200rpm but it goes up to 10mm/s rms on most of the horizontal axis vib when running at 1300rpm and mostly having DOMINANT 3x peak including engine,gearbox and pump side. MAchine will behave ~ normally if running above the 1300 rpm.
Engine rated speed up to 1800 rpm.

I cant thinking much on how does the 3x vibration really high. Other info is both coupling is a flexible coupling, which having shimpack to cater for slight misalignment.

Could anyone had facing same problem? Pls share and appreciate other inputs & explanation from the experienced guy out there.

THANKS

Axial float on a Thomas shim pack coupling is a normal phenomenon that has to be dealt with.

Pump? Sleeve Bearings? Gearbox? Set all floats in their correct running position and set coupling air gap. Also sounds as though your pump is hunting. Any vib problems with the gearbox?

Vibuser,

I am working on a CAT engine now with 3.5xSS vibration. What is the CAT model, and what type of coupling between engine and gearbox?

Walt

dear sam,

firstly thanks for your comments.

anyway the gearbox is double helical gear type and having anti-friction bearing on slow speed and sleeve on high speed shaft. pump having both sleeve berings.

Yes, mostly yhe gearbox will produced abnormal noise if engine run abt 1300rpm. but the impacting is not repetitive.

will post some vib spectrum for viewing

thanks

The first thing that comes to mind of course is resonance since
1 - you mentioned vib directional/horizontal (by the way, how much higher was H than V)
2 - you mentioned the vib is much higher at one speed.


Another thing that comes to mind is whether the common base is resonant since it is seen on all 3 machines.

Perhaps check for vib on the base. Try some bump tests or temporary bracing.

It would also be interesting to distiguish between two patterns:
A - pattern of harmonics of 1x peaking at 3x? TWF shows fundamental frequency of 1x
B - pattern with 3x / 6x/ 9x higher than other running speed harmonics (perhaps on log scale). TWF shows fundamental frequ3ency of 3x.

dear walt,

i am sorry i couldnt saw your post initially. Btw it is Caterpillar model G3306 6 cylinders
175HP. My other problem is that i dont have any firm vibration limit for this. I take 7mm/s and 11 mm/s to be as guidelines of alarm and danger.

Attach is the vibration spectrums for the unit.You may find the 3x mostly on all measrement including pump side!

Pete,
you may view the vibration tabulation, ~ 2:1 ratio between H & V of the engine

Thanks

P804_March_08.doc (120 Kb, 29 downloads)

Hello Vibuser,

Is this the same Engine where you were enquiring about vibration dampener issue that you thought was the problem, in another thread?
Looking at the spectral profiles you provided, it looks like you have mechanical looseness/rub and hence, resonance related excitation at 1300 rpm.
I'd be checking for anchor bolt looseness and weld cracks (particularly at the Gearbox location) on the skid, followed by an alignment check; ensure bolts are torqued to spec.
Is this unit installed at off the shore of Miri??

Good Luck....Rajan Muthukrishnan

You mentioned that there are shimpack couplings to accomodate slight misalignment.

What is considered slight?

Can you inspect the coupling with a tunable strobe? Look at the shim packs and check for distortion.

I would suspect something with a natural frequency at 3900 cpm which is excited by the 3 x running speed from the coupling due to misalignment between the engine and gearbox.

Edited to add:
I posted before I saw the data you attached. I think I would add looseness also although I don't have much experience with engines. Is this possibly a 3 point mount?

Vibuser,

The 3xSS is probably dominated by firing frequency that is equal to half the number of cylinders for a 4-cycle 6-cylinder engine. The amplitude varies with load. You may also be seeing the effect of a resonance (either lateral or torsional) that causes the big change in amplitiude for only a 100-rpm increase in speed. Here is what I try to do on an engine installation:

1) Variable speed test measuring speed, vibrations, and torsional vibrations (or dynamic torque)
2) Impulse-response test to measure natural frequencies in H-V-A directions on engine and driven components

Walt

RajSha,
Yes, of of them. But the Eng NDE Horiz 1x peak decrease a bit this month. But the other unit giving this problem, and i found most of the crrude pump have same behaviour.

I agreed with Dan that the system could have 3900cpm natural freq. But uncouple engine test show H,V,A of 6;1;1. I guess that the engine is the root cause, but which part?

I agree also to conduct structure-dynamics studies on the unit, but its may takes time for hiring contractor. (procurement process/budget)

My temporary solution would be changing the engine harmonic dampener / engine vib damper. Would it reduce the current vibration problem guys?

Vibuser,

I can think of only a few reasons to change the damper at the front of the engine:
1) The damper is known to be defective
2) The wrong damper is installed, based on checking part number and other info.
3) There is a torsional natural frequency, and a damper size change could possibly change the frequency -- this assumes that the torsional frequency is dominated by engine compnents and not the connected components (gearbox and pump)

How can you ask if changing the damper would solve the problem, when you don't know what the problem is? I have tried and tested several dampers on a V-8 2-cycle 1000-hp diesel engine.

Walt

Hi Vibuser,

I completely agree with Walt, if you are referring to the Engine Vibration Damper; there is no reason to change it without first knowing what the problem is other then just vibration. The other issue is you are saying most of the other similar machines (COTPs) having the similar fault signs. Look at what is common to these machines, like if they are installed on the same area/platform, structural could be a potential source and you need to do a buunch of data collection/analysis.
But before you hire consultants, do some basic checks in-house, as suggested earlier and in previous threads. Such as, Check the Vibration Isolators located at the Engine foot to ensure there are no visible defects; Check for loose or missing bolts at the Engine/Gearbox/Pump; Torque all bolts around the Train to specifications, using a torque wrench, not just a regular wrench. These were suggested before...have you tried this at all? If so, what was your findings? As stated before your vibration profiles indicate mechanical looseness/rub related resonance.
Keep us posted.

Cheers...Rajan Muthukrishnan

This message has been edited. Last edited by: RajSha,

Dear Rajsha

Yes we have done alot of alignment check,uncouple,check coupling hub to hub distance,gear backlash and axial play. The alignment shims were also reduced since too many shims were installed previously. And even Newly purchase Gearbox was installed but giving the same abnormal sound and higher vibration. Mostly on engine horizontal and gearbox axial.

The quick temporary soulution is to check/change the vib dampener of one of them, since we do not know what is the exact problem. Thats why i am posting here seeking an opinion.

Thanks

Vibuser,

Your data clearly shows significant 1/2xRPM harmonic activity in all measurement points - strong evidence of rather bearing fit looseness ( are all units in the power trane share common base? ) This in turn suggests newly appeared resonances. You'll have to poke around in order to narrow it down to a specific unit.

David

Vibuser,

We understand that you'd like to resolve the vibration fault on the COTPs but, your quick temporary solution (Damper Replacement) may not be the answer to your problem and you may end up loosing credibility as a result; just as the replacement of the Gearbox did not solve the issue or re-alignment for that matter.
As David mentioned, the presence of 1/2X harmonics indicate looseness/rub condition that may be exciting the 3X resonance, which is why suggestions are made to check loose/missing anchor bolts, weld cracks. I know of cases where skid jack bolts were used to adjust for alignment of piping and were not removed afterwards which caused mechanical seal, bearing failures and site personnel were hiring Contractors to repeatedly re-align and re-balance the machine; without success.
Either way you are the one who needs to decide on the course of action based on the opinions given here and what the vibration data indicates.
All the best.

Cheers...Rajan Muthukrishnan

0.5x and it's harmonics is perfectly normal and expected on some reciprocating machines, depending on how many pistons and strokes. We have it on quite a few of ours and we ignore it. I don't remember at the moment exactly how to figure out which machines it is expected on (does anyone else?)

ePete,

I would tend to agree with you if the amplitudes of the 1/2x and it's harmonics are rather low and inactive. The attached profiles at 1200 rpm when compared to 1300 rpm indicates active 1/2x harmonics coupled with raised floor noise (broadband) and relatively high amplitudes; along the machine train not just at the Engine locality.
And so I'd be a little more concerned, as to what's causing such energy to excite the 3x. I've seen similar phenomena on different engine makes...Cat, Cummins, Wakesha, etc. driving Recips/Pumps/Compressors, caused by improper/lack of bolt torquing, missing bolts, etc; especially on the Offshore Platforms.
My 2 bits worth to share past experiences.

Cheers...Rajan Muthukrishnan

Those are good points.

I went to wikipedia to refresh my memory on what I was trying to say before. It is probably basic stuff for you guys.

It applies to a four-stroke engine: See the animation here:

http://en.wikipedia.org/wiki/Four_stroke_engine

Looking at a single cylinder, it fires once every other rotation which is a natural way to generate half orders and half order harmonics. OK, it's true there will be some other cyclinder firing in the alternate every-other revolution (resulting in three cylinders firing each revolution in this 6 cylinder engine) but it will not be applying force at the same location as the first and may not have the exact same combustion parameters as the first. So every other revolution we have a different cylinder firing at the same time within the revolution creating a slightly different pattern and creating half orders.

And a 6 cylinder four-stroke engine has some cylinder fire 3 times per revolution... the 3x is probably simple cylinder firing frequency as Walt said, especially since 6x and 9x are also slightly prominent above their neighboring harmonics.

The fact that it occurs on all points... doesn't seem to change the conclusion. I would not think all machine points would have similar high 3x unless they were all being driven by the same excitation and the excitation seems to be the engine imo.

From my perspective, I still don't see the half orders and associated harmonics as a problem.

This message has been edited. Last edited by: electricpete,

The origin of the firing frequency (3xSS in this case) is actually a torsional vibration that couples with the engine case and gearbox to create lateral vibrations. I just measured a diesel where the firing frequency (6xSS) excited a torsional resonance. It also created lateral vibrations in all directions; some higher than others, since it depends on structural response. I find that it is now essential to measure engine torsional vibrations, if you want to understand lateral vibrations. The 4-stroke engine produces multiples of 0.5xSS. THe 1/2 harmonics are capable of exciting a torsional natural frequency. For the 12-cylinder engine I measured yesterday, there were three torsional critical speeds: lowest speed excited by 6xSS, next speed excited by 4.5xSS and the highest speed was excited by 3.5xSS. Working on an engine withou torsional data is like working with a blindfold and one hand tied behind your back!!

Walt