Power wasted in Cenfrifugal Compressor / Is motor big enough? 1

[zt2]

I'm reviewing the operation of an existing compression system for increased throughput and slighly altered molecular weights.

One of my compressors has an absorbed power of 2185kW with a motor rating of only 2258kW. This compressor is compressing gas from approx 45barg to 145barg, the gas is approx mole weight of 24 and flowrate of gas is 25MMSCFD.

Compressor is a Man Turbo unit with electrical motor drive

My question is: how much power (other than that used for compression that I have given above) would everyone expect this size of a compressor to absorb in seals, bearings and mechanical linkage?

I've seen some previous pieces of work using a figure of 2% losses for mechanical linkage and 400kW for everything else but that was for a compression system with a motor rating of 4MW and two compressors on a single shaft processing 25MMScfd each.

Should I just use 2% for linkage and 200kW or does anyone have better sources for these kind of losses?

Thanks

Zak

 

[pmover]

is the motor directly connected to the compressor?

is there a set of gears between the driver and compressor?

are there magnetic bearings installed in driver/compressor?

the 2% mechanical loss is generally acceptable for non-magnetic bearing, direct-coupled driver-to-driven unit without gears.

is it possible to measure electric motor voltage, current, power factor, etc. to obtain reasonable values for power input simultaneously with measuring compressor data (i.e. P, T, Z, MW, vis, speed, etc.)? this way the losses can be somewhat substantiated.

good luck!
-pmover

 

[zt2]

The motor is directly connected

I'm not sure whether there is a gearbox but I suspect there is and will see next week when I go offshore to the unit.

I'm not sure of the bearing type.

I am going to determine what the actual losses are when I go offshore by comparing the motor power consumption and the power required for compression.

However I was hoping to get some general rules of thumb that could be applied to other situations where motor power consumption and other variables may not be known or the compressor is performing away from it's expected performance curves.

Our in house rotating machinery expert has steered me towards 2% shaft losses and 150-200kW absorbed in bearings and seals per compressor casing. These numbers stack up well with previous work we carried out where using 2% and 400kW for two compressor casings on a single shaft directly coulpled to an electric motor give use calculated amperage readings within 2% of actual observed amperage after the modification was made.

Does anyone have any good rules of thumb / experience to give on this matter?

Cheers

 

[pmover]

the short answer is yes.

with details . . .
your in-house rotating machinery expert(?) has good rules of thumb / experience to follow/abide by.

what was the original design criteria for mechanical losses?

for directly coupled (i.e. no gears), non-magnetic bearing CENTRIFUGAL installations, the 2% rule of thumb of TOTAL REQUIRED HP is acceptable. if RECIPROCATING type compressor, the rule of thumb is 5% of TOTAL REQUIRED HP.

if previous performance test data correlates well with the 2% hp losses, then use it. of course, measuring power consumed and power needed will definitely be helpful!

if MW is different than design, the available HP will be reduced if MW increases and vice versa. suggest contacting the compressor OEM and request a set of performance curves based on revised MW & other properties of gas.

good luck!
-pmover

 

[zt2]

Thanks pmover for the rules of thumb.

I don't normally have the luxury of getting to site when carrying out this type of modelling so they will come in handy in future to get a rough handle on whether we are likely to be power constrained.

I also don't know the original design criteria for mechanical losses, sorry.

Cheers