Output signal from Bently Nevada monitors

[tgmcg]

I plan to purchase a portable vibration analyzer for performing diagnostics on rotating machinery. Most machinery I work with is equipped with a permanently installed vibration monitoring system, such as the GE Bently Nevada 3000 & 7000 series systems. Most large turbomachines are equipped with eddy current displacement probes, which necessitates analyzing the buffered output signal from the permanently installed monitor.

In order to specify the right analyzer I need to know what the output characteristics are for the buffered output from the Bently Nevada monitors, e.g. voltage range.

Does anyone here know what the output characteristics are?

Thank you for your help.

 

[JJPellin]

We regularly collect data from our Bently-Nevada systems using our CSI portable data collection boxes and analyze it in the CSI software. On our older monitors, we simply plug into the face of the monitor BNC connectors and collect the raw signal from the proximeters which is output as a 200 mV/mil analog signal. But, in order to be sure that your data collection will be compatible with the exact monitors that you use, you need to contact Bently-Nevada and ask for the output signal characteristics. You should have a set of manuals that came with each monitor that includes this information, as well.

Johnny Pellin

 

[tgmcg]

Johnny,

Thank you for the information on the CSI system.

I've already spoken with GE who would no doubt prefer I buy an ADRE or Snapshot system.

One engineer at GE informed me that gap voltage (-10 vdc) plus peak AC 0-pk vibratory voltage are supplied to the buffered outputs. Another very experienced source tells me that only the AC component of the vibration signal is sent to the buffered outputs from the proximetor.

Some of the portable analyzers inputs have a published maximum range of +/- 5 volts. Some sources say the +/- 5 v limit applies only to the AC component of the signal.

I'd much prefer that the analyzer manufacturer do the necessary research regarding compatibility of their system with the Bently gear, but they never follow through. Most of the analyzer mfgs are not machinery people and just want to sell analyzers.

Kind regards,

Tom

 

[JJPellin]

Then, I would suggest that you take a meter and check the output signals from your monitors. Set the meter to AC volts and see what it reads. Then set it to DC volts and see what it reads. That should settle the question of whether the AC or AC+DC is output. I believe you will find that you are getting both AC and DC out. I know that the standard CSI data collectors can take this as input with no problems.

Johnny Pellin

 

[tgmcg]

I don't have direct access to the monitors. I should be able to reach a pretty solid understanding of the signal levels by alternative means...web search, forums, etc.

 

[TPL]

This is an ideal application for an oscilloscope - hook up to the monitor and switch between AC and DC coupled, to see what the buuffered output signal contains by way of AC and DC components. Unlike a meter which reads in RMS, you should be able to get a good estimate of the pk-pk amplitude to check that your datacollector is correctly set up.

I would caution against buying a Snaphot - the System 1 software is 'not very good' and the ADRE system is an inappropriate tool for this application

 

[tgmcg]

Prices for small, compact portable 2-4 channel PC-based analyzers are all over the map. From $4,500 to $20,000 for close to the same functionality.

The ADRE system is commonly used for the job I need to do, but it is too bulky for my purposes and too expensive compared to the competing products.

I'm not looking for a route-based data collector type system. I'm looking for a quality engineering tool to perform diagnostics such as vibration analysis and modal analysis associated with critical process turbomachinery.

 

[TPL]

I use IOTECH kit - started off with a Zonicbook and moved on to a 652u with ez-TOMAS software. Works really well for me - the software is intuitive but a little bit clunky. The system does everything I need - from in-situ of balancing of hi-speed compressors to trouble shooting machine problems.

Just in case you're wondering, I am an enduser in the oil/gas industry and nothing to do with IOTECH. And yes...I do have a Snapshot :-(

 

[tgmcg]

TPL,

The IOTECH looks like a great piece of kit. It certainly fits the bill in terms of functionality, but the analyzer costs $10k plus another $4k for the software. I'm an independent contractor and 14 smackers is more than I want to spend. If a cheaper system leads to a significant amount of work, and more functionality is required, then I'd consider upgrading to something more powerful.

I'm also an end-user in the oilpatch. I'm interested in the approach taken by vibDaq using off the shelf components from National Instruments together with LabView. Their system costs $4500 incl software. They're local in San Rafael and will loan me a demo. It very much depends on the functionality of their software.

Kind regards,

Tom

 

[GregLocock]

The best and cruellest test for any analyser is to record a sample set of signals that you wish to analyse, and then get the salesman to analyse them for you while you watch and take notes.

In my experience that rapidly thrifts out the wannabees, and frightens all of them.



Cheers

Greg Locock

SIGlease see FAQ731-376 for tips on how to make the best use of Eng-Tips.