My 2130 is out for calibration, so I pulled my 2120 out of mothballs in the meantime.

I thought it seemed like a lot of my machines exhibited lower WF energy with the 2130 when I started using it early last year, and now I see why. The AP on this motor has an fmax of 50.5 orders in RBMware; the 2130 spectra go out to ~ 51 orders, but the 2120 goes beyond 60 orders, resulting in higher WF levels. No changes have been made between surveys.

Also, I notice I don't have to be nearly as careful with my cable to avoid the ski slope issue seen with the 2130.

2120_-_2130.doc (103 Kb, 55 downloads) data

The peak in the top spectrum is probably a rotor bar pass frequency. Most of the motors I see have in excess of 50 bars/slots. All of my waveforms are defined as "special" which you have to use to get acceleration waveforms using analog integration, so I haven't noticed this problem. Where is your Fmax defined?

quote:

Originally posted by rustythevibeguy:
The peak in the top spectrum is probably a rotor bar pass frequency. Most of the motors I see have in excess of 50 bars/slots. All of my waveforms are defined as "special" which you have to use to get acceleration waveforms using analog integration, so I haven't noticed this problem. Where is your Fmax defined?

The fmax of the data above is defined in Database Setup > Analysis Parameter Set Information at 50.5 orders. I do not use the special WF function, I get acceleration WFs by using digital integration. I am just noticing the 2 analyzers do not agree in terms of the fmax, with the 2130 closely obeying the AP set directive, and the 2120 not so much. I thought it may have been due to the machine above being a variable speed machine, but as I collect more data it doesn't seem to matter. I trend Max Pk WF, so this issue is screwing my trends

Further, on some machines the analyzers agree on fmax and on others they don't, which really doesn't make sense. The only common thing I see is that it appears that on machines that include PeakVue points the fmax issue is greater than on machines that do not have PeakVue points. I have had to turn the dual channel mode off in my 2120 due to the difference of the channel grouping between the analyzers, I don't want to change all those and then have to change them back in a week.

BTW, I had forgotten how slow the 2120 is compared to the 2130. It's taking forever to run a route, especially without dual channel collection

There are discrete fmax values for the 2120. When you use orders to set your fmax it will default to the next highest discrete fmax. There are separate sets of fmax's for normal, PeakVue and TSA collection.

I don't know how this applies to the 2130 but during the boot up, your 2120 will have an indication of hv1, hv2 or hv3 filters. This is a reference to the discrete fmax values. I am attaching a file with them. There are also rules as to what data will be collected simultaneaously and what will be sequential. They can be found on CSI's website but like you said, it would take you longer to correct them than it will to get your 2130 back.

fmax.xls (20 Kb, 12 downloads)

Thanks, Danny. After my last post, I bit the bullet and contacted CSI support, which gave me the info in your Excel file as well as info on the 2130, which I have attached below.
In short, the 2130 has a lot more fmax's available than my HV1 2120, so the 2120 defaults to a higher fmax more often. In my case above, because 50.5x the run speed was between 1 kHz and 1.5 kHz, the 2120 defaulted to 1.5 kHz, or 90 kCPM, or 60+ orders. The 2130 has a discreet fmax at 1.25 kHz, so it stops there, much closer to 50.5 orders.

I knew there was a reason for it, just thought there might be a setting to get around it. Not this time, apparently.

2130_Fmax_values.doc (52 Kb, 11 downloads) 2130 fmax info

Michael, using digital integration, are you not having a problem with low-end ski slope with the 2130? If not, are you doing anything "special" or different than when you used the 2120 to collect data?

quote:

Originally posted by rustythevibeguy:
Michael, using digital integration, are you not having a problem with low-end ski slope with the 2130? If not, are you doing anything "special" or different than when you used the 2120 to collect data?

Yes, I have more ski slope w/ the 2130 than with the 2120. The 2130 seems much more sensitive to cable movement.

As long as the vibration level is above, say, ~.02 IPS overall, I don't see it much if I hold the cable coming from the accel perfectly still. I try to hold the cable against something solid, like the machine frame, until all cable movement ceases; between my hold point and the 2130, cable movement (within reason) doesn't cause slope. However,if the signal is < ~.02 IPS, often nothing I do will prevent excessive slope. The WF will act noisy, like a bad cable, and it is not unusual for the spectrum to scale to .3 IPS but show an overall level <.02 IPS. I have always looked at acceleration WFs, so analog has never interested me. And on the occasions where I have temporarily switched to analog to avoid slope, my experience has always been that it mattered little. I'm not sure I have tried it w/ the 2130, tho.