All,

I collected data on a diesel driven fire pump today. The timewaveform plot is from the pump inboard bearing (vertical direction) and shows impacting spaced at 0.5X. The spectrum plot (pump inboard vertical) shows running speed harmonics and fractional running speed harmonics(0.5X, 1.5X, etc.). Any thoughts about the source? We suspect wear/excess clearances with the engine journal bearings. Could a damaged coupling show these characteristics?



Thanks

Fire_Pump_222.doc (48 Kb, 37 downloads)

What does your fire pump look like? I have attached typical vibes from the pump vertical on our fire pump and a graphic of our typical points. Your data looks like it could also be a gear problem. Do you take any data on the engine?

This message has been edited. Last edited by: Ed Hudson,

Doc2.doc (134 Kb, 34 downloads) fire pump info

Steve,

Arne Lindohlm sends the following information :

"...this is fully normal with
4-stroke engines. [regarding 1/2 harmonics]

Most diesels show x3 fairly high. One must also be
careful not to accept resonance. Levels below 4.5 mm/s rms
is what we accept over here. But driven machine must
of course be lower.

Many emergency diesels shake so much that they do not survive
even testing periods. The nuke station pumps here must
be operated 3 days pumping full once a year and show
no high vibes at all. That should be a natural thing for
any serious fire emergency pump installation."

He his having difficulty in joining this board, but hopes to work it out soon and wanted to give you this information.

Good Luck,

I agree it is common to see 0.5x on diesel engines - I have seen it on machines at our plant although I'm not responsible to analyse them.

I vaguely understand it has something to do with the fact that the cyclinders fire differently every other revolution, giving a fundamental frequency of 0.5x. Can someone could explain it to me in more detail?

Arne - hope to see you here soon!

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

All - Thanks for the replies.

The pump is a horizontal mounted direct-drive (no gearbox).

With a four-stroke engine, the power stroke of each cylinder occurs every two revolutions which equates to 0.5X. That would explain the presence of the fractional harmonics.

I'm concerned about the timewaveform characteristics and the overall amplitude of approximately 0.6 ips.

Thanks again.