I would suggest you check your wiring.  There is some similarly in the unfiltered time domain signals but they are identified as being on opposite ends of the machine.

Ow, now I see.. made a mistake in the text on that screenshot. The real flat 1x filtered is drive side, the other screenshot are both from non drive side. Note that i don't have a keyphasor, and the 1x filtered is taken via setting the speed manualy in the datacollector.

Thanks in advance!

See graphic for what I was referring to.  Because of the similarity of waveform shapes I wonder if you have a probe from drive end and a probe for non-drive end forming an orthogonal pair.  It should be easy to check but use some care to avoid a machine trip.

You mean possible physical wiring issue at the Bently rack connection, probe locaties? I have recorder these vibrations via portable datacollecor csi2140, paired them via software. Or might there be a problem with the bufferen outputs at the rack?

Really strenge that when i was measuring the vibrations were stable at 35-40 microns.. but few min later when I was outside this sudden trip happened.

Could ask to measure the motor seperatly or do some additionel testing.

You need to do what is called a loop test to verify that the probe at the machine is what you think it is where you connect the CSI 2140.  There are several ways of doing this, one being disconnecting the probe at the machine and observing the gap voltage at the rack.  CAUTION:  If trips are enabled, it is highly likely a trip will occur.  Another method is to trace as best you can the probe and extension cable to its proximitor and then measure the gap voltage at the proximitor.  Then measure the gap voltage at the rack outputs and see if it matches.  Usually the various probes have voltages that differ enough that this method will work.

Thanks, will check this asap next week.

All the vibration probes are proper connected at the correct location to the Bently rack.

- checked via disconnecting the probes and monitoring the rack display.

-> My current idea:

1) orbit shapes make us conclude of high preload pressence, reason not clear to identiy sadly..

2) aprox. 10sec before tripping, while measuring the casing vibrations at the compressor we notice this high 0.43x vibration [2.2mm/s!]. So I can conclude that the cause/problem is most likely to be found at the compressor.

** Can oil instability / whirl cause Preload at the motor and make it trip?
** Weird to see Oil whirl in combination with the high preload orbit at the motor, usualy oil whirl originates at too low loads?
(no orbit data from compressor however)

** During the summer It usualy tripped during extreme high ambient temperature / wheather (+32 degreesC outside)
- hence first thought was too high load on the motor, which was pulling 123A [compared to only 58A now].

** Sad to not see clear signs at motor which can or could cause this high preload and here after trip, not sure if the compressor vibrations would be strong enough to cause a trip on the bigger motor..

Thanks for your time and help, let us hope the problem is solved when exchanging the compressor.

Cheers

@Registered Member posted:

All the vibration probes are proper connected at the correct location to the Bently rack.

- checked via disconnecting the probes and monitoring the rack display.

Good that you have checked, just one less thing to consider and easy to verify.

-> My current idea:

1) orbit shapes make us conclude of high preload pressence, reason not clear to identiy sadly..

2) aprox. 10sec before tripping, while measuring the casing vibrations at the compressor we notice this high 0.43x vibration [2.2mm/s!]. So I can conclude that the cause/problem is most likely to be found at the compressor.

I would agree, you would not likely see fluid induced instability at the same time as evidence of a heavy preload.  You mention a trip; does the prox probe amplitude stay up high on the coast down indicating that perhaps some compensated data should be reviewed?

** Can oil instability / whirl cause Preload at the motor and make it trip?

Fluid instability typically is clearance limited meaning that the rotor whirls within the clearance.  As such it typically presents an almost circular orbit.

** Weird to see Oil whirl in combination with the high preload orbit at the motor, usualy oil whirl originates at too low loads?
(no orbit data from compressor however)

** During the summer It usualy tripped during extreme high ambient temperature / wheather (+32 degreesC outside)
- hence first thought was too high load on the motor, which was pulling 123A [compared to only 58A now].

** Sad to not see clear signs at motor which can or could cause this high preload and here after trip, not sure if the compressor vibrations would be strong enough to cause a trip on the bigger motor..

Thanks for your time and help, let us hope the problem is solved when exchanging the compressor.

Cheers

See my comments above.  Another recommendation, especially if the connection is made with a gear type coupling,  would be to check the condition of the coupling lubrication.  I have seen instances where a locked up coupling causes issues and only on one side of the coupling. 

Have you checked the runout on coastdown?

Not all rotor instabilites fill most of the bearing clearance.

Ye; only full clearance if it turns into a whip i guess.

I have not whitnessed the virbations itself at rhe rack, was outside at that moment measuring the casing vibrations.

Have not checked the runout. But now that I think about it.. we have axially moved the shaft 2mm, this for magnetic center correction. (But also at the previous axial location we had this trips).

Can you measure runout with a csi2140 easily, sadly we can not slow turn the motor..

I really hope it is the compressor; removing the freon alone will cost us 10k euro's..

Preload causes:

1),motor (electrical/mechanical)

Mechanically checked the bearings etcetera.

Electric checks only magnetic center correction, any other easy tests or likely causes to check? (Can monitor the motor current and winding temperatures)

2) compressor (mechanica, proces)

Mechanical, like to high bearing clearance.. causing oil whirl and hereafter tripping the motor. Or process issues..

# might also be that the motor would be the problem, which during the event causes the compressor drive shaft to the center in the bearing and causing this oil whirl.

“Ye; only full clearance if it turns into a whip i guess.”

Generally, not the case at all.  Both can and often will come close in amplitude to the bearing clearance and thus present a circular orbit.  One key distinguishing feature is that whirl will be a constant percentage of running speed, often 43% to 47% whereas whip will latch onto a constant frequency.  

See https://www.turbomachinerymag....d-oil-whip-and-a-rub for a good understanding of the two phenomenon.

Whirl and whip are both at natural frequencies but unstable. This is how one calculates them.

In some Bently courses, we would demonstrate this using an oil whirl rig. (I am not sure I did this for whip; this can be rough even on a Bently rotor kit.). One can do this if you have a similar kit. Often, one can see the natural frequency befor it goes unstable. (It tracks running speed--more or less--because the bearing properties change with speed.)

This is what whirl >> whip looks like in a cascade plot.

Yes, but we are not operating above critical speed. Let alone 2x critical freq, so no whip possible.

@Registered Member posted:

Yes, but we are not operating above critical speed. Let alone 2x critical freq, so no whip possible.

Just trying to be informative.

As a point of interest, you mention a 0.43X followed by a trip a few seconds later.  Are you referring to 0.43X of the gearbox pinion?  On my iPad, the frequency scale is difficult to read.

0.43x input shaft, 3000rpm.

@Registered Member posted:

0.43x input shaft, 3000rpm.

Just an FYI, generally gearboxes under load, due to the tangential and separating forces, are quite stable and don't exhibit fluid induced instability.  The gearbox connected rotor as well, when using fluid film bearings, will be operating at an eccentricity ratio of approximately 0.8.  But when a machine is tripped, those forces can be suddenly removed and the gear shaft may suddenly change its position in the bearing.

On the motor at 0.43X, has the alignment to the gearbox been checked?  I also asked earlier what type of coupling and suggested (if a gear type) the possibility of coupling lock up.

.43x sounds a lot like oil whirl, how was the alignment corrected Laser? Thermal growth? The waveform on drive end looks like a "w m" pattern, I don't know if it's possible to capture the centerline plot with 2140 without the higher end license and transient

If you have oil whirl on a motor, it should not be difficult to correct. The bearings may not be to design, worn/damaged, or improper specification (third party?).