For those who have experienced problems in performing the analytical potion of phase analysis, a simple document that has been popular over the years (many copies of this have been distributed over three continents).
Archive for April, 2007
Phase Analysis: A Simple and Methodical Approach
Sunday, April 15th, 2007Un Acercamiento Metódico Hacia la Fase
Friday, April 13th, 2007
Phase Analysis Proliferation…
Thursday, April 12th, 2007…and lack thereof…
When doing open sessions for various others, I have at times sought to establish statistics on the percentage of people who performed phase analysis when they encountered a dominant 1X RPM or even a 2X RPM or 3X RPM predominance within developing “problems”.
I’ll admit that my initial attempts at arriving at an assessment seemed to return too low a number, so I eventually moved the “instructor asking questions” to a latter moment in a course, to benefit from the now warm feeling that we might share (a little humor and a ready openness tend to facilitate communications). However, to my great dismay, the previous statistics held and returned an equal answer of 15-17%…
Why is this worrisome? Most instructors spend time demonstrating not only the usefulness but also the absolute necessity of phase analysis. Yet, only 1 in 6 analyst / technicians / engineers seek to use this method to establish or confirm their findings?
Many participants repeatedly point to the difficulties in completing a phase analysis as a cause for their failure to execute the procedure. Progressive protective covering of any potential reference point and the quasi inevitable loss of any glued reflective tape certainly do form hurdles to measuring phase. However, by now, many vendors offer relative phase measurements between two channels off analyzers and data collectors; the excuses slowly evaporate.
One repetitive problem can disappear when performing phase confirmation: unbalance WILL generate 1X RPM vibration, but 1X vibration does not necessarily come from unbalance.
Other problems might yield that same 1X RPM vibration… That fact should always stay close at hand to pop up whenever 1X RPM amplitudes perk up.
The morrow should allow us to include a short but useful (or at least well-liked) methodical approach to phase as well as a basic interpretation of results.
© 2007 by François Gagnon
Zeroing In on a Reference: Complex Spectrum Derived ODS
Wednesday, April 11th, 2007Operating Deflection Shape can be completed using a phase collection method such as a laser or photocell reference, but is usually best or easiest when the method relies on complex spectrum collection.
The underlying reason strikes us as remarkably simple: a system collecting amplitude and relative phase data for orders of running speed (typically 10)usually provides little data until download, whereas the complex spectra can be observed and readily reviewed for any anomaly potentially affecting the data and should any reactions be present relative to a frequency other than synchronous (perhaps excited from a nearby source), the relative responses can also be plotted out and/or animated.
Complex spectra force us to choose a reference location where one accelerometer will remain while the other(s) roam. Other(s): single when dual-channeled, but plural when using more than a two-channel instrument.
Where is the best spot? For the reference location to be useful throughout collection, it must obviously provide a usable reference for EACH frequency of interest. The best procedure therefore calls for an initial data collection (most likely only partial in scope) to determine where all frequencies are present with significant amplitudes and stable, repeatable relative phase. From such a reference, creating the animation will impose fewer hurdles in terms of accuracy.
An ODS remains naught but an animated representation of a phase analysis. In that sense, it is eye-candy… Admitedly useful eye-candy, but still… The animation does not contain any information above and beyond what can be extracted from the comparative study of the various signals / measurements (when carefully / adequately) collected on the structure or machine. It does provide a quick view of aberrant relative behavior, if the animation is observed at the frequency or frequencies for which abnormal responses are present. Essentially, an ODS renders visible and comprehensible phenomena which might otherwise mystify a client (in the large sense of the word: the plant manager might be maintenance’s client for purposes of comprehending structural work).
© 2007 by François Gagnon
Vibration Scorecard DRAFT
Tuesday, April 10th, 2007
John S. Mitchell’s Vibration Scorecard initial document is now available for those who might wish to comment and participate in furthering the cause (so to speak).
Send an email with name and affiliation to fgagnon@vibra-k.com to receive the draft copy in pdf form. Contributors will be credited (unless they wish otherwise).
Please keep in mind that the document in its present form is not meant for redistribution.
New Articles Likely Difficult this Week
Sunday, April 1st, 2007Management apologizes to fans everywhere: the week should prove difficult for posting, but several articles (should they warrant such a name) in progress should be finished (and posted) by next week-end.
Unthinkable, Amazing, Mind-Boggling…
Sunday, April 1st, 2007A brand new conundrum baffles us as Dr. P.Hish, professor emeritus at the dept of inventions of the engineering faculty of the University of Mensa-Cola, succeeds where no other man has ever had any success: his tribble-driven sensor delivers an instantaneous and 100% accurate self-contained diagnosis of the machine every time something is wrong with the monitored asset.
Within the scope of this effort, it is believed many consultants will be put out of work as the “tribblesensor” delivers its extraordinary assessments instantaneously, quite exceeding any need for a similar judgment from external contractors.
Only a few primitive analysts dispute the claims of accuracy by saying that a diagnosis of “Something is wrong with the machine” does not suffice in this day and age…
If you have any doubts as to the accuracy of the information, please note the date of posting!
© 2007 by François Gagnon