Vir­tu­al Expo - ASME Turbo Expo 2020

21 - 25. Sep­tem­ber 2020 | Vir­tu­el­le Tech­ni­sche Kon­fe­renz IfTA Paper: Root Cause Ana­ly­sis of Tor­sio­nal and La­te­ral Vi­bra­ti­ons on a Com­pres­sor Rig

Auf­grund von Co­ro­na wurde die ASME Turbo Expo 2020 in Lon­don ab­ge­sagt. Nun hat der Ver­an­stal­ter eine On­li­ne Kon­fe­renz auf die Beine ge­stellt! IfTA trägt mit einem Paper zu den On­li­ne Event Tagen bei: 



In this case study an ex­ten­si­ve root cause ana­ly­sis was per­for­med on a com­pres­sor test rig fol­lo­wi­ng the ob­ser­va­ti­on of un­ex­pec­ted high am­pli­tu­de vi­bra­ti­ons du­ring the rig com­mis­sio­ning. Typi­cal­ly, the lo­ca­ti­on and the fre­quen­cy of an ob­ser­ved vi­bra­ti­on is not enough to pin­point the cause of the be­ha­vi­or in a sys­tem with mul­ti­ple com­ponents. The test rig com­pri­sing of an elec­tric motor, ge­ar­box and com­pres­sor rotor may be sub­ject to ro­tordy­na­mics, tor­sio­nal dy­na­mics, me­cha­ni­cal ei­gen­fre­quen­cies and fluid dy­na­mics in the com­pres­sor. In order to un­der­stand the sour­ce of the vi­bra­ti­ons ob­ser­ved at the ge­ar­box, la­te­ral vi­bra­ti­ons of the ca­sing and tor­sio­nal vi­bra­ti­ons of the rotor were mea­su­red si­mul­ta­neous­ly. In par­al­lel, the sys­tem was mo­de­led in MADYN 2000 in order to match the ob­ser­va­ti­ons with theo­re­ti­cal tor­sio­nal ei­gen­mo­des of the dri­ve­train. The mo­de­ling of the sys­tem re­sul­ted in a tor­sio­nal ei­gen­fre­quen­cy lying wi­thin 10% of the ob­ser­ved ge­ar­box vi­bra­ti­on, ma­king it a via­ble can­di­da­te to ex­ci­te ge­ar­box vi­bra­ti­ons through a va­ry­ing tor­que. To be able to mea­su­re this mode shape, a tem­po­ral mea­su­re­ment setup for tor­sio­nal vi­bra­ti­ons had to be in­stal­led. The mea­su­re­ments of the test cam­paign re­vea­led that a tor­sio­nal re­so­nance is in­de­ed pre­sent at the fre­quen­cy of the ob­ser­ved vi­bra­ti­ons. Ho­we­ver, a lack of co­he­rence bet­ween the tor­sio­nal and la­te­ral vi­bra­ti­ons sug­gests that the tor­sio­nal re­so­nance is not the root cause of the ge­ar­box vi­bra­ti­ons. Two se­pa­ra­te re­so­nances were ob­ser­ved as the gear mesh fre­quen­cy cros­sed the tor­sio­nal ei­gen­fre­quen­cy and the ob­ser­ved high am­pli­tu­de ge­ar­box vi­bra­ti­on. Evi­dence of a struc­tu­ral ei­gen­fre­quen­cy of the ge­ar­box was found through an im­pact test. Alt­hough the tor­sio­nal re­so­nance is ruled out as the root cause, it is pre­su­med that the two re­so­nances do ex­pe­ri­ence a po­si­ti­ve coup­ling, causing the par­ti­cu­lar­ly high ge­ar­box vi­bra­ti­ons.


Wei­ter zur Re­gis­trie­rung für die On­li­ne Kon­fe­renz