API rules about coolant temperature inlet for compressors 1

[diapensiya]

Hi all,

I have question about a statement in API standard 618 with respect to cooling systems in compressors.
It is stated that coolant inlet temperature shall be 5K above the inlet gas temperature.
Do you know the source of this rule ? Is this a general recommendation to increase the life of the cylinder ?
I mean what happens if it is 2 K above ? Does it mean gas is condensed ?

From API 618:
"The cylinder cooling system provided shall be adequate to prevent gas condensation. Coolant inlet temperature shall be at least 5K (10ºF) above the inlet gas temperature. Lower inlet coolant temperatures can cause condensation of gas constituents, which can be detrimental to the life of cylinder valves, piston rings and packing."

 

[zdas04]

It looks to me like a lame attempt to minimize condensation. At constant pressure, increasing temperature increases the mass of water that represents 100% relative humidity. On the other hand increasing pressure decreases the mass of water that represents 100% relative humidity faster than increasing temperature decreases it. I regularly see suction saturation levels above 3000 lbm/MMSCF and discharge levels of 200 lbm/MMSCF. 5K is not going to fix that.

This temperature management is a really big deal in oil-flooded screws. API 618 is about recips, so the ΔT restriction seems lame.

David Simpson, PE
MuleShoe Engineering

In questions of science, the authority of a thousand is not worth the humble reasoning of a single individual. Galileo Galilei, Italian Physicist

 

[georgeverghese]

This recip compressor you are working on - is it compressing a single chemical component ? If so, then max dewpoint temp at the max compressor discharge pressure is well known, and you can decide for yourself if you wish to drop the 5degC margin to 2degC.

If you are working on a multicomponent hydrocarbon gas mix, then the max dewpoint temp at max compressor discharge pressure as predicted by a process simulation program may not be accurate enough to take the risk to reduce the margin to 2degC, so it would be adviseable to use at least 5degC as recommended. In this case you would do the dewpoint temp simulation for the richest gas composition of course. As stated by David, we include the max water content also in the gas mix (if water is in the feed), so choose the gas composition carefully.

 

[diapensiya]

It is a multicomponent mix. Indeed I am planning to simulate to check when there is condensation

 

[EmmanuelTop]

I wonder how 2 degC can be considered a sufficient margin in the real world applications. Fluctuations of inlet pressure in the +/- 10% range will change water dew point for 1-2 degC, whether it is a single gas component or mixture.


Dejan IVANOVIC
Process Engineer, MSChE

 

[georgeverghese]

Pls remember that this concern is only valid for shutdown when the compressor cylinder walls are cooled down by the cooling water supply to the jackets, so the pressure to be considered is either

a) max settleout pressure for the compression unit, since all recycle valves would be open

or

b)if your compressor is depressured by auto sequence on shutdown, it would be the pressure the autosequence stops the blowdown.

Cylinder walls will be too hot to allow condensation in normal operations.

Yes, would agree we need to compare the dewpoint temp estimate produced by the simulator against the minimum cooling water temp in winter

 

[shahzadmeo]

Its not an authentic rule....we normally cool down compressor casing internals by lube oil circulation....
For Synthesis compressor on ammonia production plant
Gas Inlet=35c
Lube oil=35-40
For ammonia refrigeration compressor on ammonia production plant
Gas Inlet=12c
lube oil=35-40
Means is not exactly 5k....Best of Luck

 

[georgeverghese]

The other concern here that I had missed out is the startup with suction gas at a dewpoint temp which may be higher than cooling water supply to the cylinder jackets.
Dont see an easy way of resolving this concern which seems valid enough - would it be acceptable to include a auto shutoff valve on the water supply to the jackets that opens only after the compressor has developed a discharge temp which is say 20degC above suction? In case of failure of the SDV to open, an FSL may also need to be included on the water supply line to protect the cylinder jacket - the selection of the FSL may need to suit a cooling water stream that is fouling.