in/sec to mm/sec conversion

Good morning, all.

This may be a dumb question, but I'm still working on my first cup of java here. I have a US customer with engineers residing in Europe who are used to discussing vibration in mm/sec. My questions are:

1. Is there a way to toggle these units in RBMware?

2. Failing that, can I just take the overall vibration reading in inches/second and simply convert the inches portion to mm and be done with it?

Thanks in advance,

Original Post
Providing your IPS portion isn't IPS^2 or IPS E2 and wanting to convert to soley mm from one of those EU's, else John's straight forward answer is basically correct.

I would further qualify the OA baseband as RBMware isn't quite according to Hoyle and maybe no one else's computation either. If they are used to mm over the specific given baseband frequency, then they will probably be un-nerved with CSI's level.
So it is sounding like if my reading is 1.0 IPS-PK, I can convert it to 25.4 mm/sec (PK), though I should add in qualifiers mentioning that the original overall came from CSI's RBMware to cover my six on Sam's post.

Here's a follow-up: Would it make a difference in the conversion if I was gathering analog (unfiltered) overalls or is that just one more thing I should stipulate?

From the main menu, select database set-up, then Database Global Information.
There is a dialog box, one of the options is "System for Data Units".
You can choose English or Metric.

See the attached document.

The selection of Pk-Pk or RMS is found under Diagnostic Plotting. from the Tools menu, select Globals, then the Engineering Units tab. Here you can change the units to your preference


Thanks to everybody for the help and the education in unit conversion, not as simple as I thought.

Ian, your instructions were exactly what I was hoping for, worked like a charm. I had stared at that screen before and totally missed that option. You win the coveted yellow-sticky of appreciation! Big Grin
Sorry to throw in a spanner, but I see a common misconception appearing here. The relationship between rms and peak being associated with root 2 is purely and uniquely true in the case of a single sine wave. To generally state that you can convert from overall rms to peak using this conversion will almost certainly be incorrect.

best regards,
Tom - I think you are considering "true peak" in which case your comment is valid.

The subject of the thread was converting overall vibration and said nothing about true peak. If we are simply talking about converting an overall value between rms overall and pk/0 overall, then sqrt(2) is the appropriate conversion IMO.
If you are working in the US and reporting overall velocity values or spectral peaks , the "standard" (but not universal) units are inches per second pk/0 (for example as identified in NEMA MG-1) . It is a way to express the overall value, not a way to express the TWF true peak.

Take a look at attached two screens from Entek E-monitor.
Screen 1 in ips rms. The 1x vibration is 0.70 and the overall is 0.983845.
Screen 2 is in ips pk/0. The 1x vibration is 0.99. The overall is 1.39137.
The ratio between the spectral components of screen 2 and screen 1 is sqrt(2) .
The ratio between the overall values of of screen 2 and screen 1 is sqrt(2).
0.139137 / 0.0983845= 1.41421667=sqrt(2)

The ratio between ips pk/0 and ips rms (for spectral components and for overall) is sqrt(2).

You are free to choose whatever units you like. Personally, I agree that rms would be a more natural way to express an overall value than pk/0. But I didn't invent the terminology or the standard. It's a little like the debate between metric and imperial/US units. Plenty of people use the imperial/US units even thought they can make a problem more complicated. The important thing to understand the system of units you're using. If you're expressing an overall as a pk/0, then you have to recognize that it is that an overall expressed in pk/0 is not intended to reflect the true peak or an "approximate true peak" as you have said. It is intended to represent the overall.


The fact that Entek Emonitor does it the wrong way is not a justification.
A spectrum is effectively an rms calculation, so a spectrum in units of peak technically has no meaning anyway.
Since the time signal is the only pure, and hopefully unadulterated signal, it would make sense perhaps to store the occasional time signal so that an overall peak velocity could be calculated in the future if needed.

Best Regards,
Tom Murphy
NEMG MG1-2003
7.8.1 General
The following limits of vibration are for machines running at no load, uncoupled, and resiliently mounted
according to paragraph 7.6.1. For machines tested with rigid mounting, these values shall be reduced by
multiplying them by 0.8.
Vibration levels shown in the following paragraphs represent internally excited vibration only. Machines as
installed (in situ) may exhibit higher levels. This is generally caused by misalignment or the influence of the driven or driving equipment, including coupling, or a mechanical resonance of the mass of the machine with the resilience of the machine or base on which it is mounted. Figure 7-6 establishes the limits for bearing housing vibration levels of machines resiliently mounted for both unfiltered and filtered measurements. For unfiltered vibration the measured velocity level shall not exceed the limit for the appropriate curve on Figure 7-6 corresponding to the rotational frequency. For filtered vibration the velocity level at each component frequency of the spectrum analysis shall not exceed the value for the appropriate curve in Figure 7-6 at that frequency. Unfiltered measurements of velocity, displacement, and acceleration may be used in place of a spectrum analysis to determine that the filtered vibration levels over the frequency range do not exceed the limits of the appropriate curve in Figure 7-6. For example, for the top curve in Figure 7-6 the unfiltered velocity should not exceed 0.15 in/s peak (3.8 mm/s), the displacement should not exceed 0.0025 inch (p-p) (63.5 microns), and the acceleration should not exceed 1g (peak).
NOTE"”International Standards specify vibration velocity as rms in mm/s. To obtain an approximate metric rms
equivalent, multiply the peak vibration in in/s by 18.

Note that they use the unfiltered vibration in lieu of filtered vibration as long as the unfiltered vibration meets the limit. The logical basis for this is that the unfiltered vibration refers to the overall and the filtered vibration refers to individual spectral peaks. Both are given here in units of "peak". There is no limit imposed on "true peak" by these specifications. (whether or not there should be is a different question).

Also note they suggest the factor of 18 (=25.4/sqrt2) for converting between peak ips and rms mm/sec.

The assumptions inherent in this terminology:
"peak overall" is NOT the same as "true peak". (twf peak)
peak overall is defined as sqrt(2) times rms overall.

If you intend to compare your vibration to the NEMA MG-1 specifications, the easiest way is to express your overalls in "peak". If you use Entek equipment, your meter and software will facilitate this comparison by displaying the spectrum and the overall in units of peak / 0. I am very sure there are many other standards and instruments that use the same conventions. I am also pretty sure that at least 75% of people in the US state their housing velocity overalls in units of ips pk/0 (as I have defined, not as true peak).

The fact that Entek Emonitor does it the wrong way is not a justification.

I have already agreed with you that rms overall is a more logical way to express an overall. BUT what is the most logical to us is not the ONLY way to do it. The fact that you happen to prefer a different terminology does not make this terminology wrong.

Add Reply