This is a flooded screw air compressor: 4 male and 6 female lobes. No timing gear. The spectrum always shows harmonics of 6x of female screw RPM. IMO this vibration is due to pressure pulsations, so harmonics are expected.

What is notable here is the fact that in certain direction 12x is always much higher then the 6x. Is it a symptom of malfunction or possible resonance?

Thanks.

David

Air_compr_vib.doc (48 Kb, 114 downloads)

David,

I'm not sure of the significance of the fact that 2 x gpf is higher than 1 times, but all amplitudes are very low for this kind of compressor.

Make sure that you get the readings under the same load every time. Variations in 1 x relative to 2 x can be observed if you monitor the spectrum during loading and unloading.

David,

I analyze refrigeration screw compressors on a regular basis and I see this a lot. The 2X lobe mesh frequency typically indicates over or under compression. Check your operating conditions verses design conditions. If trending is increasing and all other parameters are the same, check for wear on the thrust bearing. In some compressors, as the thrust bearing wears, the screws will back off the heads, creating the pulsation at lobe mesh or 2X lobe mesh frequency.
Looking at the spectrum you posted, I feel Danny is correct in that these amplitudes seem low for this type of machine.
Do your best to take data with the compressor at full load every time.

Mike,Danny,

That is correct, it is useful to think of this frequency as lobe mesh frequency (LMF) which is also responsible for air pressure pulsation. From this standpoint, an elevated 2 x LMF may signify improper shaft alignment possibly due to bearing wear. There is no trend up though amplitude wise, but there is a pattern and who knows how much of vibration is too much?

David

David,

I don't have my laptop set up right now, but next time I do, I'LL PosT SOME LONG TERM WATERFALLS (sorry, not screaming) of this type of compressor running with amplitudes much, much higher for years. I used gpf for gas pulse frequency.

Years ago, a guy from Ingersol Rand told me they use a shutdown limit of 1 ips on their SSR compressors. I don't remember if that was overall or maximum but either way, that is a long way away from where you are. I have seen them and Atlas Copco and other compressors live up to it.

I am in accord with tha last opinion, many times is normal this situation, see yhe evolution, tendence of change for this especific frecuencies.
Exist a catalogue of frick screw compresors, whit some articles about that, I will try to
see that and send you someting.
thanks bye.
wgarcia62@hotmail.com

Frick Ammonia compressors are not nearly as tolerant of high vibration amplitudes as air compressors

Here is a manual from Frick on their limits. GPF is always the highest peak I see, except when the thrust bearing goes out.
The alarm values in the manual for peak readings are pretty high, however, the compressors I monitor have resonance on them somewhere, sometimes the motor base, which gives a very high motor reading. A good place for digital overalls

Dave

Vibration.pdf (69 Kb, 94 downloads)

The attached spectra are taken from an Atlas Copco GA160 screw compressor. I have expressed concerns about it repeatedly since the in-house rebuild about 7 years ago (replacement of bearings) but it is hard to argue with a compressor that keeps running.

ac1.doc (78 Kb, 72 downloads)

Dave G,

Does the motor drive the female screw on this compressor? All but one screw compressor that I have seen has the male screw driven by the motor which produces a gas pulse frequency of 4X rpm (4 lobes on the male screw). If the male screw is driven, the female screw frequencies will be 2/3X and harmonics. All the compressors I monitor have a gas pulse frequency of 4X and harmonics. It is not unusual for the 2X gas pulse (8X) to be higher than 1X gas pulse.

Regards,
John J

Danny,

I see high 2x Lobe Mesh frequency (LMF) in the attachment being much higher then in my case, but I'm also trying to interpret it properly. Your's is still running. Are there any bearing problems visible in vibration data?

Although this is air pulsation frequency, it is also LMF possibly meaning shaft misalignment due to bearing wear. I have addressed this aspect in my previous post but it went unnoticed.

John,

4xMale shaft = 6xFemale shaft.

David

David,

I'm not sure what you mean by "4X male shaft = 6X females shaft." If the male shaft is driven, the gas pulse or lobe pass is 4X rpm of the motor and the female shaft is turning at 2/3 the speed of the male shaft. Think of the compressor screws as gears and the gear ratio is 4:6 (2/3).

Also, I noticed the plot shows 1987 rpm. Since there are no gears, is this compressor belt driven?

Regards,
John J

This message has been edited. Last edited by: John J,

I'd bet there is a set of gears. Something like 58/39 speed-up to the male rotor.

What is the make and model of the compressor?

This is a I-R SSR-250 flooded air compressor. There is a 67T/40T gear coupling from the motor to the male rotor having 4 lobes. There are 6 lobes on the female one.

David

That's what I expected. Shaft speeds of 1790, 3000 and 2000 and lobe mesh of about 12000.

Compared to what I normally see from the SSR compressors, this is very low amplitude and I would not be concerned at all even with the abnormal shape of the spectrum.

I have been monitoring screw compressors for 10 years now, so I have a few case histories I will share. Attached is data from 4 different compressors.

The first is a spectrum from a compressor that was totally trashed. The sleeve brgs had worn to the point that the rotors had contacted the housing and worn the ODs off both rotors. Baseline on it's replacement follows.

The 2nd shows another similarly trashed compressor that shows much more evidence of it's problem.

The 3rd shows a compressor that ran for years with high screw mesh levels, replaced after 10 years on a time basis.

The final data is from a compressor that may have damaged sleeve brgs, but has ran since 1998 looking like this.

My experience with these machines has been that sleeve brg wear doesn't consistently show up in vibration data. It may, but it may not, and when it does it is usually too late. Our compressor supplier recommends using oil dP across the compressor to monitor sleeve brg wear. I would put much more weight on harmonics of shaft speed and floor levels than multiples of mesh. JMHO.

Compressors.doc (61 Kb, 56 downloads)

Michael,

Thanks for sharing this info. My compressor in contrast is equipped with antifriction bearings thus having less chances to be wiped out.
Although it is hard to assess a sleeve bearing condition with a seismic accel, I believe there should be in other symptoms in your vibration data indicating wear because a dislodged CL should have affected air pulse and probably LMF. Did TWF indicated a problem?
Of course the final case TWF will make a vib analyst loose sleep. How it could be running this way for 10 years? Hard to imagine.

In my case there is no timing gear on the screws, therefore lobe meshing occurs in a manner similar manner to gearbox. That is why I put so much weight on LMF spectrum pattern. Of course magnitude is indicative as well.

David

David_G,

The amplitudes & TWF look very low for this machine type under a good load. Currently monitoring two I-R SSR-300 with 60,000+ hours and the trends are still good.

MTR SPEED 1789
1ST STAGE 57T/64T = 101970GMF
2ND STG 57T/74T = 101970GMF
LP1 6373
LP2 6890
MALE1 SPD = 1593
MALE2 SPD = 1378

The main issue noted here is oil seal leaking allowing oil in the drive motor.

quote:

Originally posted by David_G:
Michael,

Thanks for sharing this info. My compressor in contrast is equipped with antifriction bearings thus having less chances to be wiped out.


In my case there is no timing gear on the screws, therefore lobe meshing occurs in a manner similar manner to gearbox. David

David, sorry for the confusion. Our compressors employ two sets of AF thrust brgs stacked on each shaft, along with sleeve brgs at each end. No timing gear as well. My experience is that AF brgs have shown their defects readily, can't think of a case of unexpected / undetected AF brg failure in my time here.

Good luck!

Michael,

AF and sleeve bearings working in parallel? I can't see any benefits
in this setup ( sorry if it is just my ignorance). Unfortunately compressor manufacturers are not very eager to disclose their drawings.

David