I will say this. Both Rockwell/Entek and Emerson/CSI have built "mock" trial databases for me to evaluate. I have also evaluated other systems that don't even come close to the analytical capability of Entek or CSI. Some of the "others" will perform the same analysis functions, but require more work to get to the end product (four or five clicks of the mouse as opposed to one or two). You must be careful when choosing based on verbal acknowledgment of functionality. For example, if you ask, can you identify pertinent frequencies in the spectrum with your analysis software? The answer may be yes, but identification may involve putting a mark on the monitor screen with a crayon. As mentioned, if you're looking to buy 12 units, inquire about a trial period for the equipment. The vendors that know they have a good product will work with you. After twenty three years, it should be hard for a slick sales person to slip something by you. Good luck!

Sid,

I think some of them must have investments in Crayola.

I don't know about Rockwell, but I would take a close look at what analysis parameters and alarm limits CSI used. I have found their set-ups to be lacking and always change what is assigned by RBMWizard (unless it was set up by one of their instructors or service guys).

I'm afraid that it wouldn't be too hard for a slick salesman to slip something past. In most cases, you are looking at something that is completely new to you and you know only what they tell you about it. Mr. Reliability (and everyone else) has gotten an indepth discussion from people who know the good, band and ugly of these programs inside out. I think he made his decision long ago when Ron and I came to such polar opinions. It isn't worth the hassle of learning all over. Only if you have to.

Ron,

Back when I was using Odyssey, I collected data similiar to you. I would collect all the loaded route and only look at the off-route info if I suspected something and then used notes. With the reporting capabilities set up correctly there was no need to review all the spectrums and waveforms. I could collect data for 4 hours and kick out a report in 30 minutes with the alarm report that could be generated. Now that I am using RBM and the reports are cumbersome and difficult to customize, my ratio is closer to 1:1. That is probably the area I miss the most.

Jason,

Have you tried using exception analysis/vibration reports?

With Odyssey, you apply filters like maybe .15 ips and set up the report to show you everything that exceeds that, right? It looks like you can do that with RBMWare too but it might be based on the values set in the alarm limits. I'll look further into it, because I would like to try it out on some of the run-of-the-mill equipment.

I use the RBM software as such a filter. If you setup really good you're OK. I have 4 color codes so I can immediately judge by a level's color regarding its severity. I'm using the data collector as a route PdM system and not a vibration analyzer. The OA levels with CSI seem weird as they are from like 40 kHz which are way out of line with Hoyle and may include many resonances including the transducer's resonant frequency. Why they ever went over 5 kHz I'll never know but really an OA level is the sq rt of the sum of the squares over a given baseband frequency. If data were processed closer to rules that exist, if there are any anymore, then I think analyzing CSI data would be much easier or my opinion anyway.

Duncan,
I started using a Nicolet UA500 in 1976! So, I have been looking at narrow band spectra for my entire career. I used to look at 'live' data, watch for transients, take run up and coast down data, etc. That was great when I was being called out to look at one problem child. It is impossible to take any scrutiny 'in the field' when being paid to collect a vibration route on an entire plant. For those vibeguys that work at the same plant day in and day out, vibration is now one job out of five that they perform. For us consultants, try to compete with the six other companies in your back yard that are charging $1.00 per point.
Fortunately, the hardware and software of the 21st century is up to the task of nailing a vibration route. Remember, 90% of what people are looking for is bearings getting ready to fail. Checking Joe's alignment or Bill's fan balance are easy as well.
I don't disagree with you that there is still a place for 'in depth' diagnosis and you shouldn't be using a quickly captured route to make a decision.

Ron is putting blame where it belongs - on the bearing as the part we're concerned with the most in PdM, and it is. Generally the rotor spins until bearing failure so bearing analyses is the order of the day. Some really screwed-up plants with no formal way of doing maintenance the correct way: has poor setup/alignment/pipe stress and so on - those problems take a toll on the bearings. So training becomes a part of the recommendation. But we have to get bearing analyses down pat.

Ron - my preference is/was to see as much data as I could while the data is collected so I don't miss anything that I might want to see. An in-house analyst can always go back easily but for an outside consultant, going back is often neither convenient or economical. There are also "company men" who go to remote sites and have the same considerations.

VENDOR WARNING - ;-))

As you may or may not know or may have forgotten, I'm the president of VibroTek, a manufacturer of automatic diagnostic, monitoring, etc. software and instrumentation instrumentation. We are especially known for our automatic bearing diagnostic software that includes providing an estimation of minimum safe operation life and, unlike other vendors, also diagnoses installation problems, so I assure you I'm well aware of the need for bearing diagnostics!

One of my personal specialties is to write data acquisition drivers for PC based hardware and I configure it to squeeze as much useful information as possible while minimizing the time required. This includes displaying as much as possible during the collection process. If you are not doing anything else except carrying the instrument, why not look at the data while you are collecting it. My experience is the customers usually remember the things that you miss so you are better off missing as little as possible.