compressor size & type

[nefty]

Sorry, I shld've given more info and be more specific for my question on compressor type and size. Thanks for your help.
Gas comps; 95-96% CH4+1-2% C2H6 and rest is CO2 and N2
Suction Pressure is 10 psi and Discharge is 250 psi and the alternataive is SP:10 psi and DP:150 psi with rate is 2.1 MMcuft/day at onshore gas field.

 

[zdas04]

nefty,
You don't have to start a new thread to answer a question.

This sounds like a wellhead, wet gas application. My preference is that all low pressure wellheads see an oil-flooded screw as the first compression. Screws are very tolerant of fluctuating suction pressure (recips are not) and your roughly 11 compression ratios is a reasonable job for a screw. The 250 psig discharge points the selection toward "process derivative" machines (as opposed to "air derivative") like the MyCom, Kobelco, Howden, Ariel, or Frick machines which tend to be rated at a higher MAWP than the air-derivative (like Gardner Denver and most Sullair) machines. The choice of compressor frame is far less important than the choice of packager and your oversight of the design/build. The list of things that your specification needs to include is too long for this venue, but if you haven't specified a screw compressor or selected a packager for screws you probably need help.

The 2.1 MMSCF/day is a reasonable volume for the mid range of any of the process compressor product lines and the hp is not too high.



David Simpson, PE
MuleShoe Engineering

http://www.muleshoe-eng.com

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

"Life is nature's way of preserving meat" The Master on Dr. Who

 

[bcs5274]

with .65 sp/gr sweet gas at 4000 ft elev. it takes aprox. 650 hp with a 2.1 gear ratio on a 244mm screw to move 2.2 mmcf

 

[zdas04]

Yeah, and with a 370 mm screw and a 1:1 gear it takes (at 4.6 VI with 90% on the unloader) 458 hp at 1000 rpm input.

There are a lot of variables that can be adjusted at the design stage to optimize a machine for the problem. That's why you need a good packager or a good in-house compressor guy (both is even better)

David

 

[dcasto]

I'm an Ajax fan for high ratios, I stop screws at 5 ratios because they get so damn inefficient. An Ajax will consume 280 to 350 HP. You can control the suction pressure and load with a single Kimray suction valve and a recycle valve, and if you like add the speed controlled governor that adjusts rpm based on suction pressure. It'll reduce fuel by nearly 25% at high loads and 50% at reduced loads.

 

[zdas04]

dcasto,
You left out a word--"Optimized". The application is 11 compression ratios. If you optimize a 2 stage Ajax for exactly 10 psig suction and 250 psig discharge and 2.1 MMCF/day and the well/pipeline actually gives you those pressures then you are right about the significantly improved fuel usage (which is why we really care about effeciency). What happens if the well gives you -5 psig, 60 psig, or 800 MCF? With an Ajax it is down. With a screw you are hard on the unloader and still running.

Recips want to have a constant suction pressure within a very narrow range. The Ajax is considerably more forgiving than a high-speed machine, but it still wants a pretty narrow range. If you get too far from the design suction then you start having temperature, rod lod, and engine loading issues. Wells simply are not plants and they will give you what they will give you--blood, guts, and feathers are common. A multi-stage recip is not a good choice for blood, guts, and feathers.

I've had this argument with a client in Durango for 10 years and their data supports my contention that the first compressor a well sees should be a screw (their biggest Ajax proponent installed a dozen screws this year).

David

 

[dcasto]

With speed and suction control, the 2803 will do .49 MMSCFD from -3 psig and 250 psig discherge(I hate vacuum gathering, been there many-o-time) to .41 MMSCFD at -5 psig and 150 psig out to 3.0 MMSCFD at 16.5 psig suction and 150 psig discharge to 2.1 MMSCF at 8.7 psig to 250 psig. All this with a fixed pocket set at zero. And never exceed 338 HP the rated maximum load at 4000 feet.

 

[zdas04]

OK, do you really think that a 2-stage 2803 will go from 9 psia to 262 psia (29 ratios, probably 8 ratios in the first stage and 4 ratios in the second)? With 80F inlet and k=1.28, first stage discharge would be 391F, I think something should trip. My point is that cylinders and clearance are sized for a given set of conditions, change the conditions and the change is not evenly applied to both cylinders (like Cooper's program says it will be) and you end up with serious imbalances.

When I was trying to make a technology decision in 1997 for a group of very high volume CBM wells I visited over 200 field compressors in the San Juan Basin operated by 4 different production companies. I gathered pressures, temperatures, and flow rates across all of the skids. This study included integrals and separables. Not a single one was anywhere close to the values predicted by the vendor programs. I saw 3 stage compressors where the first stage discharge was higher than line pressure. I saw blowthrough in one stage or the other in a bunch of machines. I saw a lot of high temp kills disabled. What I didn't see was any evidence that either Engineering or high quality operations had ever paid any attention to the machines. Wellhead recip compressors require all of the attention that compressors inside the plant fence get, but no one has the staff to do it to a 1,000 unit fleet spread across a county. It just doesn't happen. Conditions change many times per day, but the compressors get re-evaluated once or twice a decade. Not a good fit. I elected to go with the largely-untried (at least in wellhead use) flooded screws. There were many times over the next couple of years that I questioned that decision as we climbed the learning curve, but in retrospect I couldn't have made a better decision (but I REALLY wish I had known then what I know now about packaging pitfalls).

David

 

[bcs5274]

amen, been there done that in mt. myself, screws are what we use and good operators are hard to find, there is a lot of difference in the operation and options of the screws too

 

[dcasto]

Not 100% methane, put nice rich gas it will. .8 gravity gas near dew point and entering into the compressor at 70F, will have a max discharge of 349 at 6.7 ratios first stage and 5.2 second stage, 13,000 lbs rod load out of 33,000. The VE sucks, but it works.

In 1998, we put in 258 screw units in Western KS. We eneded up 105% over budget and with a pissed producer because fuel costs we so high. The units were 5 psig in 35 to 105 psig out. They ranged in HP from 3406NA to 3408TA's. We fiquired in the analyzation that we would have come in at 50% over budget with a happy producer because of fuel had we gone with Ajax.

 

[zdas04]

Nobody knew much about how screws fit into upstream operations in 1998. There are folks that understand more today if you can find them. Some packagers are even starting to get it.

I've installed 25 Ajax units and was pretty much on budget every time, but the installation budget was 5 times the budget to install the same hp screw.

We've probably hijacked this thread enough.

David

 

[bcs5274]

rings won't last at 350 deg f, we try not to run any over 300 deg f so that they last more than 2 years

 

[dcasto]

I could always go to 3 stage.

Don't get me wrong either, just showing all sides. I like the screws at low ratio or huge acfm, the best unit was a cat3408 with a ngc 300D doing 16 in HG vacuum, discharge at 40 psig to a recip. 1.6 MMSCFD The gas was 1600 BTU/cf. Speed control, auto slide valve, cooler temp control to limit condensation.