We are in the process of implementing in all our normal route data, the collection of some data in Peak-vue in several directions all along the machine train.
So far we have collected Peak-vue in analyze mode whenever we have found it may bring us additional useful information. Measurements have been sporadic and eventually not with the same comparable parameters, so that trending was not really possible.

We understand the peak to peak in peak-vue waveform may be the best alarm prediction indicator to trend for impending problems. Articles from CSI and Technical Associates of Charlotte suggest using just the band above.

Have others tried and developed A.P and A.L sets from which good and useful trending is obtained? We would really appreciate knowing their general guidelines, but knowing some of them in detail would be great.

Jorge,

If you select the 'Find' Tab above and search on "peakvue", then you will get 37 pages of discussion on the topic. That is a good starting point.

Walt

Jorge,

For the AP sets, we have found the set up as described in "Description of peakvue and illustration of its wide array of applications in fault detection and problem severity assessment", by James C Robinson and James E Berry, to work really well. We have been gathering peakvue data for 4 years or more using these set-ups, with good results.

As far as alarms go, we use a different technique.

I can email you the details, if you like.

quote:

Originally posted by IanS:
Jorge,

For the AP sets, we have found the set up as described in "Description of peakvue and illustration of its wide array of applications in fault detection and problem severity assessment", by James C Robinson and James E Berry, to work really well. We have been gathering peakvue data for 4 years or more using these set-ups, with good results.

As far as alarms go, we use a different technique.

I can email you the details, if you like.

Hello Ian,

I know the pa^mphlet you mention written between CSI and Technical Associates of Charlotte. I have it here next to me and read it again this week.
I certainly would like to know what specific analysis parameter sets and alarm limit sets you have been using with good results.
It may save us hundreds of hours experimenting different options.
email it to me please at j.echeverri@bluewin.ch
Thanks.

Jorge,

The main thing to take note of in the paper is the set-up of the AP sets. This includes (for most bearings) a 40 order, 800 line, single average spectrum. Trend paprmeter is P-P Wave. We do not collect a special time waveform. You need a different set-up for gears, or large bearings where the BPFI is much larger than 10.

I just tried to email you, but it bounced back.


Ian

quote:

Originally posted by IanS:
Jorge,

The main thing to take note of in the paper is the set-up of the AP sets. This includes (for most bearings) a 40 order, 800 line, single average spectrum. Trend paprmeter is P-P Wave. We do not collect a special time waveform. You need a different set-up for gears, or large bearings where the BPFI is much larger than 10.

I just tried to email you, but it bounced back.

Ian

Ian,
I have no idea in what I was thinking about, but I gave twice a wrong email this week. The right email address is j.echeverri@schubag.ch