Analyse vibratoire des machines sur un banc d'essai - cas d'un disque tournant

ANNEXES

PROBLEM TYPICAL . SOURCE SPECTRUM REMARKS

HYDRAULIC AND AERODYNAMIC FORCES A. BLADE PASS & VANE PAS S

1 X v(

BPF BPF . 2BPF

IX 0 BLADES X RPM (IPM0,^41^111011111116idiffusers sIpEBM1bs \ _ ` i i

i

~+ Blade Pass Frequency (BPF) . No. of Blades (or Vanes) X RPM. This w frequency is inherent m pumps, fans and compressors, and normally does not present a problem. However, large amplitude 9PF (and harmonies) can be generated In pump i1 gap between rotating vanes and stationary is not equal all the way around. Arse, BPF (b harmonie) sometimes can coincide Min a system natural frequency causing high vibration. High BPF can be generated it impeller wear nog seizes on shaft, or i1 welds fastening drtluser vanes lad. Also. high BPF Can Cie Caused by abrupt bends in pipe (or duel), obstrvclrons which disturb flow, damper settings or if pump Or tan rotor is positioned ac0entrically within housing

RANDOM Flow Turbulence oten occurs blowers due to variations in pressure VIBRATION ar velocity of the air passing IMu Mea Or COnneeted ductwork flow

B. FLOW TURBULENCE

tan This BPF = BLADE OR disruption causes turbulence which wiu generate random, ^lowhequency 1 X BPF VANE PASS vibration, typically in the range O1 50 to 2000 CPM. If surging occurs FREQUENCY within a compressor, random broaduand high frequency vibration Can occur. Excessive turbulence can also ascite broadband high frequency

C. CAVITATIONGaation Cavitation normally generates random, higher frequency broadband RANDOM HIGH energy which is sometimes superimposed with blade pass frequency FREQ, VIBRATION harmonies. Normally indicates insuffipem suction pressure (starvation) vft can be quite Left e destructive ro pump internals uncorrected. Il

can particularly erode impeller varies.n of When present. ten sounds as it 1X B J. II 120KM 'gravel' is passing thru pump, Cavitation is usually caused by insufflaient Ai CA inlet flow. Car occur durrlg one survey, and be absent the next survey ( d

changes ^insuction valve settings are made),

GEARS A. NORMAL SPECTRUM

Q GMF. aT xq NT X Normal Spectrum shows Gear & Pinion Speeds. along with Gear Mesh = a tea' r RPUe Frequency (GMF) and very small OUF harmonics. OUF harmonics its common) will have running speedsidebands around them. Al-K yrig peaks are tier of law amplitude, and no natural frequencies of gears are excited FA,,,r 7GkiF sour ofat 3.25X OUF (minimum) when · teeth are known If loath

3.25X GMF count a not known. set F, at 200X RPM on ascii shah

B. TOOTH WEAR 51 T/26T Y \~,. n^''-

EAR ,J r ~ ^0 , E

NATUgAL FREOUENCv ur sour Tour ++ l~

Key indicator al Tooth Wear is excitation o/ Gear Natural Frequency ^(1,,),.S, along with sidebands wound it spaced at the running speed of trie bad gear. Gear Mash Frequency (GMF) may or may not change in amplitude, atmough high amplitude sidebands and number of sidebands surrounding GMF usually occur when wear is noticeable. Sitleband5 may be better 1 wear indicator than GMF frequencies themselves Also. high amplitudes ! commonly Occur at either 2XGMF or at 3XGMF (esp. 3XGMF). even when

i \

3,25X OUF GMF amplitude is acceptable.

C. TOOTH LOAD

. g

GMF- GMF Gear Mesn Frequencies are often very sensitive to load, High OUF amplitudes do nor necessarily indicate a problem, paritularty if sideband frequencies remain foie level, and no gear natural hequenCies tour 3sare excited. Each Analysis should be perlormed with system at Maximum operating load for meaningful spectral comparisons

D. GEAR ECCENTRICITY AND BACKLASH · ï

g i~ rs x f y 0 x p

76MF

sour 3.25X GMF Fairly high amplitude sidebands around OUF harmonies often suggest gear eccentricity, backlash, or nonparallel shafts which allow the rotation of one gear to 'modulate either the GMF amplitude or the running speed of the other gear. The gear with the problem is indicated by the spacing al the sideband frequenciea,Alsso, 1X RPM level of eccentric gear wet normally be hign if eccentricity is the dominant problem Improper backlash normally excites OUF harmonics and Gear Naturel Frequency. both of which will be srdebanded at 1 X RPM. OUF amplitudes will often

3.25X OUF decrease with increasing bad it backlash is the problem.

E. GEAR MISALIGNMENT

g GMF

msharmonies Gear Misalignment almost always excites second order or mgher OUF which are sidebanded at running speed Often will show only small amplitude 1x OUF, but much higher levels at 2X Or 3X GMF. Important to set Fuky high enough to capture at least 3 GMF narmomcs, 3GUF Also. sidebands around 2XGMF will often be spaces at 2X RPM Note that sideband amplitudes Olten are riot equal on left and right side el OUF and

3.25X GMF OUF harmonies due to the tooth misalignment_ Causes uri an weer peaem.

F. CRACKED/BROKENI A Cracked or Broken Tooth will generate a high amplitude at IX RPM of this gear only n Me tine waveform plus it will excite ~I+-~,k--WÂF~M r gear

TOOTH

cG ie iii

naturale r If II IF f^requency (Id sidebanded at as running speed. His best detected in Time Waveform wr4Gh will show apronounced spike every lime the problem i ·e · ^II

y G .1-4-11-1 0 U G A r A tooth trios 10 mesh with teeth on^tthe mating gear. Time between impaCt$ K 1 or ocaa vnTR BROKiN (p) will correspond l0 1/RPM of gear with the problem. Amplitudes of ` t AL OR cwtocEo morn Impact Spikes in Time Waveform often will be lOX to 205 higher than that

^-°-i at 1X RPM in the FFT t

G. GEAR 600 pPyl ,

PHASE cue ASSEMBLY PROBLEMS W.T est. s i tsT.T' 1sT.it3 -^.11030 N^.s RPM ~ · ue 15.000

GAPF. GMF o. Gear Assembly Phase Freq. (GAPF) can result in Fractional Gear Mesh ^~ · 1 Frequencies (if NA,> 1). It literally moans (TOIN) gear teeth vela contact Z ,e (TF(N) pinion teeth and will generate I4 wear patterns, where N, in a given p pp _ K. , 7G 100th combination equates Ire product of prime factors Common to the ^x + 1 $ r GMF number of teeth on the gear and pinion (NA a Assembly Pnase Factor) GAPF + + + GAPF (Or harmonies) Can show up right from the beginning it there were 1GAPF eCr,+tP manufacturing problems. Also, its sudden appearance in a periodic survey t I spectrum can indicate damage if contaminate particles pass inrougn trie

N E 4 mesh. resuming in damage IO the teeth in mesh at trie rime of ingestion just uhPF. 6000 CPM . 020X GMT (FRACTIONAL DMF) as they enter and leave meshing or trial gears have been reoriented

H. HUNTING TOOTH M Hunting TOOth Frequency (I,.,) OCCMS when (auks are present on DOM Me PROBLEMS + + iNT = (GMF)(N) gear and pinion which might have Occurred dunng the manufacturing

DRNER DRIVEN 1000 RPM a a57 RPM a + }

-m 3 ^170EAR TeirriON) process, due to mishandling, or in the field h can carne quite high RI ys + t SteANOs r` vibration, but since h Occurs al low frequencies predominately less then ço ^ 1- 600 CPM. it is Often missed, A gear set with this tooth repeat problem / \ normally emits t 'growling sound from the drive. The maximum affect 1:."...,....^---,...,..,..... ~: occurs when the faulty pinion and gear teeth bath enter mesh et me same

e NA . 1 is the ideal assembly lime (on same drives. this may occur only 1 of every lO to 20 revolutions, 8`17(2X31 · depending an the t,,, formula). Note that T,,,,,, and T4 refer to number of 7.1X7 J NA.1 phase lapor in geai design teeth on thegear endpinion, r (6X loopj (1) 1000 respectively N, is the Assembly Phase Factor 1,,,= 143 Pulse

CPM (One Per 7 Pinion Revolteions) defined above- Wilt often modulate both OUF and Gear RPM peaks le)(7) 7

Analyse Vibratoire des Machines sur un Banc d'Essai - Cas d'un Disque Tournant -

82