Hi all,
it is my first post here, thank you in advance for all the answers.
I have been monitoring 45 kW pump, ~1480 RPM and i have seen strange harmonics approx. at ~6,31 RPM, strongest on DE pump and motor side. I have checked bearings in manufacturer catalog and it says that these should be :
Motor : 6313 DE/NDE
Pump: QJ 313 DE / 2213 NDE

None of the bearings fault frequencies match with 6,31 harmonics. The closest one are 2x BPFO motor bearing at 6,142, and pump BSF at 6,661.
Pump vane pass is in my opinion 7x RPM.

Another curious thing is that motor DE has stronger vibrations vertical, while pump DE has stronger vibrations horizontal. I have attached spectrum in PDF.

These harmonics of 6,31 RPM are clearly visible in all directions DE and NDE in velocity spectrum in pump and motor, but have much lower amplitudes.

We have two other pumps of this model, but they are in other place and they are working fine, with almost no vibration and much less noise.

What can be cause of these harmonics? Are these amplitudes dangerous?

Regards, Kamil U

Attachments

Files (1)
Original Post

Hello. I am sorry to say that but your spectrums are a mess and from that point you are talking all wrong. Why do you use Orders for X axis, after that Hz, and again Orders...? You are saying a couple of times you have harmonics of 6,31 RPM. This is not true. You have harmonics of 6,31 x Orders. It is close to bearing frequencies you mention but could also be close to vane pass frequency. You should check the exact number of vanes, verify your rotation frequency. For a motor bearing check I suggest you take acceleration, envelope and acc. time-waveform of 1s in horizontal direction. And acc. time waveform 1 s for pump also. You have to unify your units!

Becar posted:

Hello. I am sorry to say that but your spectrums are a mess and from that point you are talking all wrong. Why do you use Orders for X axis, after that Hz, and again Orders...? You are saying a couple of times you have harmonics of 6,31 RPM. This is not true. You have harmonics of 6,31 x Orders. It is close to bearing frequencies you mention but could also be close to vane pass frequency. You should check the exact number of vanes, verify your rotation frequency. For a motor bearing check I suggest you take acceleration, envelope and acc. time-waveform of 1s in horizontal direction. And acc. time waveform 1 s for pump also. You have to unify your units!

Thanks for your reply and sorry for my mistakes. I'm an intern and i try to learn vibration analysis on my own using only Internet and books.
I've made new measurments with higher number of lines but lower frequency for velocity spectrum and figured out that shaft speed is 1486 RPM what makes these harmonics 6,28 X Orders.

Now the closest bearing frequency is pump DE BPFI - 6,547 X Orders.
But i have to tell you that i'm not sure if bearings installed are the same that in manufacturer catalog. Sadly it is very difficult to obtain such informations in my company. Do you think these spectra can indicate bearing fault?

I have basic version on PRUFTECHNIK VIBSCANNER with magnetic mount accelerator, that can measure only average vibration, spectrum and envelope spectrum... So i can't measure time waveform nor phase. But i attached acc. spectrum and envelope of motor horizontal direction. Envelope readings are low amplitude in motor so i think there is problem with pump.

I can see clearly 7 X Orders peak in all pump spectra so i think that pump has 7 vanes. But again i cannot be 100% sure.

Are such amplitudes as 2g peak-peak envelope acc. are dangerous to such pump bearings??

Thanks

Attachments

It is a pitty you don't have the ability of time waveform. From the data available, I would say the worst thing is the envelope spectrum on the pump (the last one). The look is similar to bearing fault frequencies which can't be confirmed from your info. The question is, why those frequencies are more obvious on the motor velocity signal than on the pump. It could be because of some construction instabilities or because of a bad coupling alignment. However envelope spectrum on the motor looks OK. If it is a very important pump, I would change the pump bearings, make alignment. Otherwise I would repeat a measurement after some months.

BECAR,
thank You for your advise. Recently, we've changed pump DE bearings, and in the PDF file there are photos of them. Certainly they were the reason of high vibration. And what is more, these bearings are different type than those in catalog (3310 and 2310)...
I've told people responsible for this pump to pay more attention to greasing.

Attachments

KAMILU,

You did well with assistance from BECAR. That is a benefit of the forum. That is one lesson you learned well; getting feedback by inspecting the bad bearing. This should gain respect for you from the maintenance staff. Now you can move on by improving your measurement technique and diagnostic abilities.

Walt

Add Reply

Post

HACKATHON in association with IMC-2019

×
×
×
×