Compressor Selection and Sizing 1

[nickpa]

I have 3 rotary screw air compressors used to meet plant design demand at about 925 scfm with a receiver tank capacity of 150 cu ft. Two of the compressors are sized at 640 scfm to act as two 50% compressors and the 3rd compressors is sized at 1500 scfm for back up when the other two are being worked on. The design and operating pressure are at 125 psig and 100 psig.

The problem is the plant does not currently operate at 925 scfm, operating capacity is normally 700 scfm. This allows one of the 50% compressors to basically operate as a 100% compressor. When the backup compressor is running it is short cycling due to it being oversized. I need to replace the backup compressor to one with a lower capacity. This is causing a lot of maintenance issues. The number one goal is to reduce the number of maintenance issues that may arise. I am looking at selecting a similar modal to the two 50% rotary screw compressors or replacing it with an air cooled centrifugal compressor with a blow off valve to avoid the cycling issue. I have no interest in a variable speed compressor due to negative experience at other sites. Let me know if you have any suggestions on the options I have selected.

 

[zdas04]

I have an idea--sell the 1500 scfm machine and get on with your life. If the current (and expected future) load can be satisfied with the 640 scfm machines then you have 100% redundancy. Not certain how a 640 scfm machine can satisfy a 700 scfm load, but if it is, it is.

I would stay away from small hp centrifugals, they have a terrible maintenance history and interstage cooling is rarely done well (this application would be a tough 2-stage or an unloaded 3 stage, either one is too close to a flow instability).

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.

 

[rakuday]

Nickpa,

For your requirement of 925cfm, you have three compressors, thats almost three times of your requirement...!!! Are there any plans of future expansion in your company that you are unaware of? These are capital equipments and no one would simply buy compressors and keep it as backups.

Centrifugal compressor for 1000cfm will not be economical. Further, you are spending money to compress air and then it’s being blown out!!

From what you have given above, it appears that proper energy & air audits have not been done. Before spending any money on buying additional equipments or scrapping, get your plant audited (air) by a professional. Am sure it will be worth every penny of it.

 

[nickpa]

The 1500 scfm compressor only operates when the first two are undergoing maintenance, therefore all 3 compressors are not used to meet plant demand. Additionally, this is for instrument and service air at a power plant so they need instrument air supplied at all times, even when under maintenance. So, there will never be a situation where all three compressors are running.

 

[TD2K]

What sort of turndown control do you have on the two smaller compressors to adjust their flow rate to demand or are you starting/stopping only on a low/high pressure band?

 

[nickpa]

They are starting and stopping only on low nd high pressure bands

 

[rconnor]

I'd echo zdas04's comment.

I'd also add that you should look into more storage. 150 ft^3 (1120 gal) is not enough for those units) which will cause you more rapid cycling. Also, look at your pressure band as it may be too tight (just be aware that increasing the discharge pressure will increase your power consumption by 1% for every 2 psi).

I don't know when the last time you looked at VSDs or what your environment is like, but from an energy cost standpoint they are great when compared to lightly loaded load/unload units. Running one of the 640 cfm units as a base unit and a VSD as a trim unit would be ideal (make sure the VSD is larger than the base unit to avoid control gap though). However, I understand that if you don't trust VSDs you don't trust VSDs...just food for thought.

 

[nickpa]

I forgot to mention that there are four receiving tanks at 150 cu ft each (Two for the service air system and two for the instrument air system). Also, since it is a power plant and they produce power they have very little concern for efficiency and power consumption.

 

[zdas04]

At the end of the day, a VFD and a receiver do the same thing--allow for varying/part load. With adequate receiver space there is no need for any variable speed. You run the compressor full on to take the tanks from some number (say 90 psig) to some other number (say 120 psig), then turn them off. It would be better if the span was farther so they would cycle less often, but you got what you got.

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.

 

[rconnor]

zdas04, from an operational point of view yes, a VSD and large amounts of storage serve the same purpose but not from an energy standpoint (unless the unit is an on/off unit with no unloaded run time and a very quick blowdown). Also, as you said, to increase the cycle time you can increase the pressure band but this comes at the cost of more power. This isn't a big deal in nickpa's case but for other industrial environments, there can be nice savings with VSDs and much smaller storage tank requirements (usually around 1/4).

 

[zdas04]

rconnor,
You really have to do the economics on an individual situation. On/Off machines with automated bypasses for startup going into large receivers at relatively high pressure (the hp load on a two stage recip increases slowly from 3.5 ratios/stage to 4.5 ratios/stage, and very quickly above that) is hard to beat. VFD with a small receiver can have better economics, but the small receiver really increases the urgency of repairs. Like everything in Engineering it is a balancing act.

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.