Variable Speed Compressor Concerns 2

[wk390107]

Hello All,

I am currently working on a proposal to replace the plant compressors. The plant currently has one 100hp load-unload compressor and one 60hp inlet modulation compressor. The idea was to replace both of the compressors with a single large VS compressor. However during a recent air survey I have noticed an extreme flow fluctuation in the air usage. The flow ranges from 50-150 cfm for ~10 seconds and then increasing to ~400-550cfm for ~20 seconds. This is a constant cycling, due to the plant having such intermittent demand. I ensured the flow were measured after the dryers so the low flows are not dryer purge.

This large range of flow concerns me because from research I learned that variable speed compressors do not operate efficently below ~20% and in order to size a compressor large enough to conver the 500CFM flows the compressor would have to operate below 20% for a significant amount of time. I am aware of another option to size the VS compressor small enough to cover the low flows efficently and use it in conjuction with a direct drive machine to cover the high flows, but I am concerned that I would end up cycling the direct drive too much.

Thanks in advance for any guidance.

 

[TD2K]

What size compressor are you looking at installing? If your peak air flow is 550 cfm I'd think you'd be looking at about a 700 cfm compressor for a standalone unit. 20% turndown on that is 140cfm which isn't too far above the lower amounts of air. You could also go smaller if you are going to keep your existing compressors available as backup if they are needed.

What size air receiver do you have? The air reciever is designed to help smooth out swings between production and consumption by storing or releasing air as the system pressure rises or falls. How successful is this with your existing compressors under the same demand requirements?

 

[zdas04]

Also, what technology are you looking at? I would be fine with a flooded screw in this application, not sure I'd be as easy with a recip or a dry screw.

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.

 

[BigInch]

Size the compressor best operating point for the time-weighted average of all normal flowrates experienced in your process during a typical day, ignoring transient flowrates. Specify the flow range required for your compressor as the range of min to max normal flowrates, perhaps any flowrate lasting over 1 hour, or some other time period, 15 minutes, etc. You should not design for transient flows of a few seconds, only provide the receiver tank, as TD2K suggested above, to smooth out those transient flowrates so that the flowrate finally delivered to the process is still within the min to max normal range.

"People will work for you with blood and sweat and tears if they work for what they believe in......" - Simon Sinek

 

[ccfowler]

First, I presume that the interest in a variable speed compressor is the result of the current fad to install variable speed drives on just about everything possibly including pencil sharpeners. This subject had been beaten to death in this forum, so I won't bother with a tirade. Air compressors are almost universally poor candidates for variable speed drives.

Second, I would want to pay serious attention to the reliability requirements of the loads. Are any critical to safety or avoiding harm to equipment, production, emissions, ...? Unless a complete loss of compressed air can be tolerated for whatever period may be required for compressor maintenance or repairs, multiple compressors should be installed. I would want to install at least two compressors sized large enough for each to comfortably serve the critical loads with a suitable margin for reasonably anticipated future critical load increases.

Third, the receiver should be sized to take care of all of the relatively high frequency demand fluctuations such as the 10 to 20 second duration flows that you mentioned.

You know your space constraints, so there may be some controlling limitations on receiver sizing, compressor dimensions, and weight. These may play important roles in determining practical options.

If frequent changes in capacity are unavoidable, load-unload operation of lubricated recip(s) would be my first choice for handling these variations. The real energy usage penalty will almost certainly compare favorably with the parasitic losses of variable speed drives. Assuming that your compressed air system operates somewhere in the vicinity of a nominal 100 psig, I would never use a recip of less than two stages. For recips, the larger and slower they are, the more efficient and trouble-free they can be. Constant-speed, variable capacity flooded screw compressors may be a good choice, too.

Unless there is some compelling reason, I would avoid non-lubricated compressors whether recip or screw.

Valuable advice from a professor many years ago: First, design for graceful failure. Everything we build will eventually fail, so we must strive to avoid injuries or secondary damage when that failure occurs. Only then can practicality and economics be properly considered.

 

[TenPenny]

I don't want to sound like a broken record, but it bears repeating - you need to look at your receiver. If these short-duration demands are causing frequent cycling, I would venture that your receiver is not sized properly. Last week in a purge, I ditched all of my notes from the IR desk engineer course, so I don't have any reference material, but it's easy to find the right tools on line. Work out the criticality of your demands as noted above, and also the volume time span of those peak demands, and you'll be able to size the receiver based on that. I agree that one single compressor will be inherently less reliable than two units, and also that VFDs are generally not very useful on compressors.