Seismic vs Proximity Probe 1

[norzul]

Hi,

We just procured gas turbine. Wonder whether is there any particular standard on using seismic or proximity probe for the alarm/tripping and protection of the machine.

Someone told me that using proximity probe is not that common for tripping the GT? Is it true

Thanks

norzul

 

[rob768]

I assume you mean accelerometer? It gives an earlier warning (increase in vibration level) than a proximity probe. If you measure an incresae in deflection, it is usually late or too late.

 

[norzul]

Yes, seismic probe is based on acceleration (even though the unit is normally converted to velocity i.e. mm/s) whereas proximity probe is based on displacement (micrometer) i.e. eddy current

In our case, the vendor had proposed that the seismic probe will be located on the compressor side casing of the GTG. This is for tripping (safeguarding) purpose. And the proximity probe (X-Y) at each radial bearing for radial vibration monitoring and thrust bearing to monitor the axial displacement.

Is this a common setup for the GTG?

Our company technical stds require that shutdown protection shall also be provided as well for the bearing (radial & thrust). So, what the supplier intend to provide is a clear deviation to our company std...

Just checking whether the deviation is genuinely technical or due to some commercial reasons...

 

[byrdj]

I'm not aware of current CT recomendations. For ST, the prox probes are becoming very popular (and used for protection).

You could review the make and model of the prox probe equipment and see if relay contacts are provided that can be used.

I would think the desired display from prox probes would be in V since the desired D for tripping would be a function of speed.

Another item to know would be if the prox probes are absolute or relative. since relative measures gap between journal and bearing, it is possible for some bearing designs to vibrate in phase with journal and relative reading be very low for a severe condition. (a function of bearing support stiffness) Absolute usally have an acceleramoter included.

I would also expect the thrust prox probes are for actual position (not axial vibration) monitoring to see a thrust bearing failure (wear) has occurred.

 

[JJPellin]

We don't currently have any gas turbines in our plant. But I believe the protection requirements would be the same as a large steam turbine. The machine uses journal bearings which will tend to dampen out much of the shaft vibration. Case probes (seismic) could miss a potentially disastrous failure until extensive secondary damage had occurred. Typically, two radial probes on each bearing should be set up to trip the machine in a dual voting (2 out of 2) configuration. If both probes agree that the shaft displacement is above the danger level the machine should trip. The monitor will need to be set up so a failed probe or proximeter will not be counted as a vote to trip. Dual thrust probes should be set up the same way. The case probes are nice to have since the prox probes really are a relative measure (shaft versus housing). With a case probe, it is possible to subtract out the case movement and get absolute shaft movement. Also, prox probes have a high frequency limit and are not very good at detecting extremely high frequency events that might occur at blade pass or gear mesh frequencies. We would not normally set up the case probes to trip since they could be bumped in the field. On some machines, we also set the prox probes up to trip on gap in case the shaft centerline drops without substantial increase in overall radial vibration. We set up alarms for acceptance region that will alert us if the vibration levels drop or if the phase angle changes greatly.

 

[turboco1]

Displacement probes for shaft movement in the vicinity of the probe. They cannot measure shaft bending away from the probe. Used to indicate problems of unbalance, oil whirl, misalignment.

Velocity pickup detectors have flat response of amplitude as a function of frequency. Alarm setting is unchanged regardless of speed. Diagnostics is limited due to being very directional. If you change its direction of pickup the value of the same force will change.

Acceleromters mounted on the casing pick up the spectrum of problems transmitted from the shaft to the casing and offer velocity regardless of frequency. Used to id hi-freq problems ie blade flutter, surge, etc..

 

[rob768]

As far as I know there is a difference between a seismic probe and an accelerometer, allthough both use the same principle of measmurement. Seismic probes are used in the low frequency range, and as such function as last alarm when an installation is really breaking down. Accelerometers can be used for trending and giving early warming of wear, increasing unbalance and such. Proximity probes can measure shaft displacements that may accompany such phemonema.
I'd personally go for vibration monitoring through accellerometers

 

[TPL]

The fundamental choice of transducer depends on the type of bearing fitted to the GT - for industrial gas turbines fitted with journal/sleeve/tilting pad bearings (where the shaft is supported by an oil film and is able to move relative to the bearing surfaces) the preferred choice would be proximity probes, but for aeroderivative GTs, generally fitted with rolling element bearings (no shaft movement relative to bearing and nearly all forces transmitted to casing) a seismic type transucer would be the first choice - for a variety of technical reasons, moving coil velocity transducers have generally lost ground to solid state acclerometers.

If a gearbox is present and fitted with journal type bearings, then both accelerometers (to measure high freuqneices associated with gear mesh frequencies) and proximity probes might be fitted

 

[norzul]

TQ guys for the explaination...apreaciate very much. From all the explanations and my own research, below is my current understanding on the issue of "Seismic vs Proximity probes"

There are 2 primary types of vibration probes or transducers:-
1) Contact or Seismic Type - normally attach on the casing or bearing housing i.e. external
a) Velocity Probes
b) Acceleration Probes (Accelerometer)
2) Non-Contact or Proximity Type (eddy current) - located inside the bearing housing i.e. internal

Yes I do agree with TPL, the choice of probes largely dependent on the type of bearing (i.e hydrodynamic, roller, ball) fitted on the machine. However, it is also depending on the supplier. In our case, GE/NP who is the supplier for our GT, strongly recommend to use the seismic - accelerometer probe on the compressor side casing for protection/tripping. It is their standard practice for Frame 5 GTG (industrial). In addition, for monitoring purpose only they also have 2 proximity probes (X-Y) on each radial & thrust bearing (Bently Nevada 3500).

BTW, Bently Nevada is part of GE now.

API 616 (for gas turbine) specifies proximity probes for radial & axial shaft vibration monitoring. It also specifies casing vibration (seismic) for monitoring. The protection or tripping based on proximity probes is stated as optional not mandatory...

API 612 (for steam turbine) specifies proximity probes for radial & axial shaft vibration monitoring and protection.

Thanks

norzul

 

[abeltio]

norzul,
what you mention is and has been the GE standard for decades.

seismic probes are used for monitoring and protection (trip function)

proximity probes are used for monitoring and diagnostic (Bently Nevada ADRE)

The main reason GE uses the seismic probes for protection is that it allows for redundancy:
e.g. brg#1 has two probes that individually detect vibrations in the vertical plane

while the proximity probes require the composition of the X and Y probe and redundancy at the position is not possible.

lately the use of newer versions of the control systems allowed for the use of the proximity probes for protection (trip function), but reserved for the high end of the product line, i.e. F and H class units... unless so specified by the owner.

HTH

saludos.
a.