I started a job using CSI RBMWare today. I was wanting to set up each point so it collected:
a. Velocity overall
b. Velocity spectrum to 15 times RPM
C. Acceleration overall
d. Acceleration Time wave to 6 rev
e. Acceleration Spectrum to 150 times RPM
f. Demodulation spectrum
g. HFD overall

I could get one setup to give me Velocity overall and a spectrum and it looked like I could get Acceleration overall, spectrum, time wave and demod in another setup. I'm not sure if it gives me a HFD overall or not. It seems that the timewave gets the same averages as the spectra but I'm not sure. I also couldn't see how I could get all those readings without storing each point twice.

I can set up those directly in Entek and get them in one store.
DLI will only allow one overall and it can't be HFD and can get the rest in one store
SKF I have to set up multiple points but can tie them together with a @ in the discription allowing one store

I would like any help anyone has on set up for CSI RBMWARE.

That is the "beauty" of CSI. Each measurement with a different analysis parameter set will require separate measurement point definition and individual store action. Units of measurement could be switched to any type when viewing the data after raw data has been collected.

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

Mike,
Let's address b, e, and f first. You will have to take a separate reading to get each of these spectrums. You can change units in the spectrum by pressing the 1,2,3 or 4 key and it will toggle between acceleration, velocity, displacement, and stored units. In other words, you can look at a STORED spectrum differently by simply hitting the 1 or 2 key etc., but the difference in fmax that you specify in b and e will require a separate reading. As far as "f demod" goes, my advice is to skip that and go to peakvue. Peakvue will be a whole separate measurement point to itself, but well worth it.

About a, c, and d. With CSI, if you want a velocity overall, you HAVE to take an analog (velocity) waveform. If you want overalls in acceleration and velocity, you will have to take them separately. Also about d...The number of revs in the waveform is a function of fmax divided by lines of resolution. You can adjust your fmax and lines to get 6 revs and take whatever resolution that is, or you can pick your resolution and take whatever revs that is. There is a "workaround" of sorts. You can take a "special time waveform." Go into "database setup" and go to your analysis parameter set. Look on the third tab "waveform parameters." Check the "obtain special time waveform box." Look at the boxes below and set the maximum frequency and number of points to get 6 revs. What data collector are you using? The software will let you set up to 32,768 points, but a 2120 will only take 4096 and leave the rest. 4096/2.56=1600 lines. When you take a special time waveform, the data collector will take the reading as any other data collector, store the spectrum, dump the waveform, and then take yet another reading. It will then store this second waveform according to the specs you entered on the special time waveform page.

Now, did I make all of that as clear as mud?

David Eason

David_G - your idea of beauty is different than mine

David E - Let me see if I understand some of this.

I set the Velocity readings to NOT take a time wave which means I wouldn't get a overall so I need to go back and select take a time wave to get the overall....

As far a PeakVue goes - I'm not a CSI person so I didn't know what it was or how to set it up - I do know what demodulation is so I was using it.

I saw the "special timewave" option but by then I was too confused. I will work on resetting that part up.

Thanks - the mud is a little clearer

Mike,
For the velocity reading....go ahead and let it take the waveform, the data collector is going to "do it's thing" anyway. For a normal reading...suppose you want 3 averages. The data collector takes the first sample, constructs a spectrum, then dumps the wavform. It then takes the 2nd sample, constructs a spectrum which it averages with the first spectrum, then dumps the second waveform. Now it takes the 3rd sample, constructs a 3rd spectrum which is averaged with the other 2 spectrums and stored along with the 3rd waveform. The spectrum is an average of all 3 samples, but the waveform is only the last one sampled. This last waveform may or may not show the 6 revs you want; enter the special time waveform. If you set it up for a special time waveform, it will then dump that last waveform and take another reading with the parameters for the special time waveform. Just bear in mind that with a special time waveform, the spectrum that you are looking at wasn't constructed from the data you see in the waveform.
I want to strongly urge you to swap the demod for the peakvue. I know that you are in unfamiliar territory, but I would be glad to help you through any specifics you need with it here or by e-mail.

You said,"....but by then I was too confused." I know what you mean. I have been using CSI going on 8 years and I am still learning new stuff.

David Eason

This will probably muddy the waters a bit more, but....

When you say you want velocity overall, do you want the overall from the 15x RPM spectrum, or the velocity overall for the frequency range of the datacollector (about 40kHz for a 2120)? Similar question applies regarding the acceleration overall.

You can probably get a, b, and g from one reading, c, d and e on another, and you will need yet another for the demod/peakvue.
Thus, you will need 3 seperate readings.

Regarding the HFD, you can actually specify what frequency range you want to look at (by nominating Hz vHFD in the trend parameter setup) and then putting in the frequency range you want. Or you can just nominate HFD and it will calculate everything above 5kHz.

Ian

Ian
I selected Analog for overall "assuming" that would give me the sum of the instrument's range.

I'm not sure how I would get HFD overall with the Velocity readings like you mentioned. I will look over the setup again.

Mike,
When you select database setup; Analysis Parameter set information; then select the AP set you are using for your velocity reading and click ok; at the bottom of the screen under the spectrum parameters tab the last selection is Number of analysis Parameters. This is the number of parameters you can choose. If you click ok, it shows what each parameter is. I use the last one and name it what you want under description, under unit types you can select accel and type of parameter is where Ian is saying Hz vHFD, then the last to last two boxes you set the lower and upper frequency range you are interested. When you are looking at you trend plot, on the right of your screen is a tab: new parameter. This will let you trend your HFD in acceleration even if your overall is in velocity.

Ronnie

Mike,
Another way to look at you HFD is with your cursor set inside you spectrum, to the right of your screen select analyze data and then stored parameters and you will see the overall for each paramater you've set.

Hope this helps,
Ronnie

Mike,
You "assumed" correctly re analogue overall.

I have attached a brief procedure regarding finding the HFD stuff; hope it makes some sort of sense.

Ian

The_HFD_value_can_be_obtained_as_follows.doc (255 Kb, 41 downloads)

Thanks Ronnie & Ian
I will be at the CSI location again tomorrow and hope to get some of this squared away. It makes sens now but after tomorrow I will probably confused all over again...........

Mike,

Are you making progress here?
Anything I can do to help?

Danny - Thanks for the concern. I have most of the problems figured out. One of the things I'm trying to understand now is the viewing of timewaves. There are two sets of red lines on the positive and negative sides. These lines are scaled so I can't really see the timewave - so I'm trying to find out what to change in the setup.......

Mike,

Those lines (horizontal?) are probably alarm lines and can be turned off with the buttom called globals I think. I will get some time to take a look at that today. There is also a button to expand the waveform so that you can see it better and manipulate the cursors better.

Good Luck,

Mike,

While viewing a waveform, hit the escape key--that will take you to the options menu. Uncheck 'Display Alarm Levels.' Or if you want different values, type them in. Hit 'OK' or the enter key to take you back to the plot.

Or the other option, while viewing the waveform is to go to the set-up button on the bottom right hand corner of the screen (just below the green tick). Go to the Display Options tab, and un-tick the Display Alarm Levels box.

Good luck!

Ian