Turn-Down Ration for Centrifugal Compressor? 4

[Sabergg1981]

Hi All.

I am looking for a suitable turn down ratio for centrifugal compressors? Does it depend on compression ratio?

For a Centrifugal compressor with ratio of about 4 and a capacity of 350000 kg/hr ,what turn down ratio do you suggest?

 

[zdas04]

If you're doing 4 compression ratios on a centrifugal compressor then it is either multi-stage or you spend a lot of time on the wrong side of the surge line. It is an amazing coincidence that they work at all, and they work very well on their system curve. Off the curve they don't do much. If you have a very constant suction pressure, a very constant discharge pressure requirement, a constant suction temperature, very stable fluid properties then you can build a centrifugal compressor that will be an amazing workhorse for a very long period of time. Change suction pressure by more than 5%, change suction temperature (in absolute units) by more than 2%, change discharge pressure more than 15%, change the required mass flow rate more than 5%, or change molecular weight of the gas by more than .05 SG units and the world starts getting really ugly really fast.

"Turndown" has a lot of definitions, but if we use maximum mass flow rate at maximum compression ratios divided by minimum mass flow rate at minimum compression ratios then the answer for centrifugal compressors is so near to 1.0 as not to matter.

David Simpson, PE
MuleShoe Engineering

"Belief" is the acceptance of an hypotheses in the absence of data.
"Prejudice" is having an opinion not supported by the preponderance of the data.
"Knowledge" is only found through the accumulation and analysis of data.

 

[pmover]

your answer can be obtained by inspecting the compressor performance curves furnished by the manufacturer - preferably the proven/actual performance curves or the predicted will suffice in the absence of actual curves. if the compressor is an api 617 compressor, check the data sheets or use this definition: "1.5.51 turndown: The percentage of change in capacity
(referred to rated capacity) between the rated capacity and the surge point capacity at the rated head when the unit is operating at rated suction temperature and gas composition."

good luck!
-pmover

 

[zdas04]

What the heck does "1.5.51 turndown" mean?

David Simpson, PE
MuleShoe Engineering

"Belief" is the acceptance of an hypotheses in the absence of data.
"Prejudice" is having an opinion not supported by the preponderance of the data.
"Knowledge" is only found through the accumulation and analysis of data.

 

[Sabergg1981]

Our operation condition varies with years. Suction pressure changes from 70 bar to 25 bar after 5 years and then remain constant for 10 years. discharge pressure shall be constant 91 barg. Molecular weight remains approximately constant.but our rate is too large about 60 MMSCD.

 

[pmover]

zdas04,

relax & thanks; the clarification for "1.5.51 turndown" is the definitions paragraph in API-617, 7th edition.

Sabergg1981,
with increasing compression ratio (Rc), the flow reduces in the centrifugal unit. due to decreasing inlet pressures, there comes a point at which the centrifugal unit can no longer increase the Rc due to speed limitations (variable unit) &/or the compressor is operating at or left of the surge control line. this then becomes an inefficient process/operation and it is time to re-wheel the centrifugal unit or add another compressor (+HP) designed for the lower inlet pressures. this is expensive.
if flow rate is relatively constant over the years, then seriously think about the recip compressor. have the ariel or DR folks send you some preliminary data (sizing) on recip compressors based on your predicted conditions. they may be able to configure a compressor unit with cylinders for single and 2-stage operation depending upon inlet conditions. you will need to be certain of sufficient HP (driver) for all conditions.

good luck!
-pmover

 

[Sabergg1981]

I'd forgotten to mention our capacity falls down from 60 MMCSD to 4 MMCSD. I think you may be right pmover. it seems we must change our mind and think to positive displacement compressors. why do you think screw compressors are not suitable for our project? and will turn down ratio is employed for rotary compressors? Is there any references about how to arrange posotive displacement compressors for gas busting station? Tnx

 

[zdas04]

Maximum suction pressure on a flooded screw varies by manufacturer but it is rarely higher than 4 bar. That should pretty much kick screws out of the mix.

In the early days (71 bar to 91 bar) you are doing about 1.3 compression ratios and hp requirements are slightly less than 30 hp/MMSCF or 1700 hp. Late days you're doing about 3.5 compression ratios and your hp requirements go to 85 hp/MMSCF or 340 hp for the design MMSCFd. I'd do this with a single stage recip with something like 1400 hp (assuming you can tolerate 72-74 bar in the earliest days) and variable speed control (either a speed changer on an engine or a VFD). Late days you may need to put a backpressure regulator on the discharge to waste some hp to keep the driver on its performance curve

David Simpson, PE
MuleShoe Engineering

"Belief" is the acceptance of an hypotheses in the absence of data.
"Prejudice" is having an opinion not supported by the preponderance of the data.
"Knowledge" is only found through the accumulation and analysis of data.

 

[Sabergg1981]

Is There any gas compressor station employed reciprocating compressor?

 

[pmover]

yes, there are hundreds of them in service.

with the reduction in capacity, are there economics to justify such a project?

good luck.
-pmover

 

[zdas04]

No more than a few thousand stations in the Rocky Mountains of the U.S. More in the Gulf Coast and mid-America. Upstream guys tend to stay away from centrifugals because centrifugal compressors don't play well with the blood, guts, and feathers that we get from reservoirs. It is after plant processing where people feel confident enough in pressures and compositions to consider a centrifugal compressor. That confidence sometimes results in installing centrifugals sometimes it still ends up using recips. A big reason is that recips have so much better turndown ratio and anyone who thinks they can predict required flows and pressures over time is lying to themselves, their bosses, and/or their clients. Every time you think you have it nailed, the gas-control guys or the marketing guys say "I know you designed this for 60 MMSCFd, but I found this other gas source and committed to moving 30 MMSCF/d of their gas at 20 bar suction, you can do that right?". With recips, you drop in another million dollar machine and go for it. With centrifugals, you start a 5 year Engineering and manufacture process that costs $25 million. That may be overstating a bit, but not all that much.

David Simpson, PE
MuleShoe Engineering

"Belief" is the acceptance of an hypotheses in the absence of data.
"Prejudice" is having an opinion not supported by the preponderance of the data.
"Knowledge" is only found through the accumulation and analysis of data.

 

[Sabergg1981]

I'm a little confused. Do reciprocating compressors exist with such a large amount of capacity?
I think fewer capacity of recip compressors than centrifugals lead in more recip compressors in number.
Can you give me a gas busting compressor with recips details? name, location,....?
i am wondering can i find a reference about how to build a gas compressor station with reciprocating compressors? do they arrange in parallel? what about turn down ratio? and other design problem...

 

[zdas04]

It is not a very big application. It is consistent with an Ariel JGJ/6 with six 4-5/8 HJ cylinders for whole range of over the life of the field. Ariel Performance says that early life it will do the job with 854 hp running at 1800 rpm and late life it will use 333 hp at 900 rpm.

I don't know what in the hell you are talking about when you say "gas busting compressor" and the term sounds kind of derogatory.

David Simpson, PE
MuleShoe Engineering

"Belief" is the acceptance of an hypotheses in the absence of data.
"Prejudice" is having an opinion not supported by the preponderance of the data.
"Knowledge" is only found through the accumulation and analysis of data.

 

[pmover]

Sabergg1981,

your reference is PHMSA 192.

based on your questions, it is apparent you do not have the expertise nor knowledge to build a gas compressor station yourself, much less determine HP size and type. I recommend you seek the expertise from reputable engineering firms.

good luck.
-pmover

 

[Sabergg1981]

Thanks Zdas04 from your patience and about information you gave me.

 

[Sabergg1981]

A will be thankful if you name me some upstream( field) gas compressor station as an example.

 

[Sabergg1981]

@ Zdas04

Can you reference me about criteria you mentioned above: "Change suction pressure by more than 5%, change suction temperature (in absolute units) by more than 2%, change discharge pressure more than 15%, change the required mass flow rate more than 5%, or change molecular weight of the gas by more than .05 SG units and the world starts getting really ugly really fast."

I appreciate your concern

 

[zdas04]

Not really. All of those numbers came from private conversations with operators, field Engineers, Tech Reps, design Engineers, and Manufacturers over my career. If a general statement of the performance envelope of either recips or centrifugal compressors is in a book anywhere I haven't found the book.

David Simpson, PE
MuleShoe Engineering

"Belief" is the acceptance of an hypotheses in the absence of data.
"Prejudice" is having an opinion not supported by the preponderance of the data.
"Knowledge" is only found through the accumulation and analysis of data.

 

[rotw]

To the OP: by turndown do you refer to API 617 definition or what exactly ?
If you want to have a feeling on how large are operating curves of centrifugal compressor, I would suggest to pay much attention to the MW, knowing that Inlet Pressure and Temperature would also impact.
Hydrogen service have often wide and flat curves.
Wet gas, coker gas or other heavy MW compressor have very narrow curve range with steep curves.
Typically you want a good turndown to avoid recycle, if your turndown case is specified for which you want no recycle, situation might end up with the rated case being selected to the right of the curve with penalty on efficiency and even bigger driver sizing.
For typical natural gas 17 ~20 MW application, I would say if you set the design point on the rated case you may have around 30-35% turndown but really from the top of my head!

 

[Sabergg1981]

@ Dear Zdas

Do you remember our discussion under this topic? we are starting basic engineering and i want to challenge our guys about change our minds to reciprocating compressors. Also some reservoir parameters change.

Here, we want to make up pressure drop of a gas reservoir by building a gas compressor station. capacity of this reservoir is 68 MSMPD. after 4 years this capacity falls down and in last years( 20 years) it reaches to 4 MSMPD. Suction pressure of this station is 74 barg at start and after 4 years drop to 21 barg and then remains constant. also molecular weight varies about 10% ( i dont know exactly, but our process guys calculated so).

you had told me about Ariel JGJ/6 with six 4-5/8 HJ cylinders for whole range of over the life of the field. but they only give about 1860 hp. from power approch for compressing of 68 MSMPD we need about 250 MW power for this station. considering a 2 MW ariel compressor we need a large amount of compressor. how it could be? or some concepts are mistaken here by me?