Data-Based Management of Compressed Air Systems

by: R. Scot Foss, Plant Air Technology

Solving problems requires more than just adding another compressor; decisions should be based on good information and a knowledge of how the system operates.

A majority of plant engineers and maintenance managers have little idea what is going on in their plant compressed air system.  At best, the system may be meeting an undefined level of acceptability that says, “If production isn’t complaining, it must be OK”.

What an odd, but rather normal standard of measure this is to use for compressed air.  This unique approach towards air, which is not true of any other utility, is confused by the consistent acceptance of several facts.

·        There is substantial waste in the plant air system.

·        No one really knows how much air is needed or used (regardless of the amount of compressed air equipment that is running).

·        Operating cost and consumer accountability are nonexistent.

·        Corrective actions are taken without adequate problem definition.  At best, sophisticated guesswork is used.

·        Production has the authority to demand more compressed air with no responsibility for how it is used.

·        Facilities are held responsible for how air-powered production equipment works with no authority to do anything about it.

·        “On” compressors is an indication of cost, but not necessarily requirement.

Knowing these circumstances and not being able to do anything about them frustrates facilities professionals.  Obviously information is missing in the process.  Often problems are approached at the wrong level, because compressed air systems are not dealt with as a system.

Organizational Problems:

Inappropriate assignment of responsibility is usually where problems begin.  Verifying that equipment meets the manufacturer’s claims before it is shipped is of limited value.  How compressors work is a function of how the system is configured and operated. Factory testing with no systems standards or information is of questionable value.  Fudge factors, throttling capabilities, and lack of systems data easily mask a weak design.

Too often, production is not held responsible for what is bought and installed in a system or how it is used.  If a light bulb is flickering, the first thought is not to call the local power company to change the way it is delivering electricity.  Yet, when delivered air fluctuates, the facilities department is held responsible.

This leaves the application of energy and resulting increase of operating expense as the available solution to all problems.  Other than process applications, there are seldom any measurements of system performance in a plant air system.  If there were, plant personnel might have a better idea of what reasonable benchmarks and standards should be applied to optimize compressors, motors, and production results at the lowest cost.

Compressed air seems to go unattended as we move into the age of information.  Capital and operating expense are thrown at difficult-to-understand problems when production is not satisfied with air system results.  When the outcome of a plant air improvement project is simply minimum acceptable results, it is difficult for engineering managers to enthusiastically approach management with proposals.  The result is that plant air problems are moved to the back of the line until they are intolerable.

Most electric utilities serving industry are asked by regulators to assist clients in aggressive energy conservation.  Electric utilities know that compressed air ranks as high as chillers and lighting as major energy-saving opportunity.  They know plant air systems operate so poorly that even if air ranks number three in total energy consumed, its conservation potential is the highest.  The problem is substantiating saving to rationalize the investment in time and/or money.  Practically no one has an installed database, so unexplained requirements get attention.  Fixing the condition, then having the same situation reoccur, is often the undesirable result.

Until now, there has been a curious approach to measuring how well plant air systems work.  A quality of air compressors are bought and installed.  All of them are turned on except one, which is intended to be a standby.  One day, production calls to say the air supply is insufficient.  This usually means that some one machine in production, which uses air, is not performing up to expectations.  Could it be a plugged filter element, leak in a hose, or increased production cycles with no change in the installation?  These and other problems cause pressure to drop and effect production.  Since people at the supply end of the system cannot tell how the system is working, they assume responsibility for the workability of the equipment.  Hey have only two choices available; raise pressure or turn on another compressor.

What was the real problem?  Did the systems pressure drop or did the point-of-use pressure drop?  Did the person who called know anything about compressed air?  One of the two choices, or both, should quiet the irritated caller.  The standby compressor is turned on.  There were two compressors, and now there is no standby.  Another compressor has to be purchased.  The problem is solved until the same situation happens again.  What did this decision cost?

Where Money and Air Goes:

There are two attention-getting facts about compressed air.  One is cost.  At $0.06/kw hr for electricity, compressed air costs between $1.50 and $2.50/100 scfm at 100 psi/hr of operation including maintenance, water depreciation, and labor.  These numbers assume the system is operating efficiently.

Most plant air systems operate around the clock, or 8,760 hr/yr.  That means 100 scfm costs between $13,140 and $21,900/yr to operate.  The cost to turn on a compressor to solve a poorly defined problem could be $100,000.  If a company operates at a typical 5% profit, every $1 wasted in the process of making compressed air for production requires $20 of production-generated revenue.  Every $100,000 of plant air waste requires $2,000,000 of production to offset it.

If there are two compressors on line, which were modulating, and another one was loaded into the system, it would elevate the operating pressure of all three.  The same flow is divided across three units instead of two.  This causes the pressure to rise and amps to fall on each unit.  Flow increases as a function of the increased demand volume of all unregulated points of use, including leaks.  A database would show supply energy and pressure increase in conjunction with demand flow and pressure indicating this was not the right decision.  Without it, all that is seen is a slight increase in demand pressure.

The attention-getting fact about compressed air is a breakdown of the typical constituents of demand in most plant air systems:

·         Well-applied production use of compressed air: 20% to 50%.

·         Poor applications for plant air (open air blowing, vacuum venturi’s, etc.), which could be done better with electricity: 10% to 25%.

·         Leaks: 10% to 30%.

·         Uncontrolled or artificial demand above production requirements created by operating at elevated operating pressures: 10% to 20%.

·         Energy required to overcome resistance to flow: 5% to 15%.

·         Inefficient operation of the compressors individually and systematically: 10% to 30%.

·         Air lost in drainage applications:  3% to 8%.

·         Purge air required by desiccant dryers:  3% to 14%.

Consider the first item as the only one essential to production.  Should another compressor be turned on, or should demand be controlled?  Diagnosing comparative data at various conditions reveals what is online in production and what is a source of waste.  Without such information there is nowhere to start.

Essential Data:

To begin an analysis, start by drawing a flow diagram of the system.  It should be simple to visualize if everyone is expected to easily work with it (Fig 1).  Then track the most important information: energy efficiency, systems workability, production results, and maintenance.

There are several minimum data requirements for managing a compressed air system.

Kilowatts – Express electrical information in kilowatts (kw).  If power factor information is required, use kw transducers.  The fact that the inlet on a compressor is wide open is not a measure of its load in terms of potential mass.  The compressor can appear to be fully loaded and be producing only 70% of the pounds of gas capacity.  The idea is to minimize pounds of gas at the lowest kw level and to continuously express total kw.  Depending on the size of the system and its components, include dryer and tower equipment in total kw or load in accessory data based on the equipment’s status (on/off).

In addition to expressing electrical cost, kw is also used to express unit status.  If kw is known when a unit is unloaded, partly loaded, of fully loaded, show that information (Fig 2).  Monitor the condition of the compressor based on trended kw.  Ring or cylinder wear, valve lift, or displacement efficiency problems are detected long before they show up as a unit failure.  Compare benchmarks when the units are in top maintenance condition against actual condition.

Supply Pressure – This is the pressure resulting from compressor output versus the demand volume production calls for.  This can also be signal pressure to the compressors.  This information, in conjunction with other information form the system and kw, shows what is occurring in the system.  The system is constantly in a state of dynamic change, shifting from demand exceeding supply -- to supply exceeding demand.  It is easy to confuse the effect of adding a compressor and reducing demand.

Compressor Status – Is the compressor motor ready to start, running, or off?  Is the compressor unloaded, partly loaded, fully loaded, or in a fault condition?  If there is automation, is the unit op0erating on local, manual, or automated controls?

Demand Volume Expressed as Mass – This is measured in scfm at pressure and temperature of pounds per minute.  Volume without regard for actual conditions is misleading.  The amount of energy used should be compared to work energy of compressed air measured on the downstream side of the supply equipment.  Demand volume should be measured downstream.  It is easier to measure volume when the density is controlled at variable mass.  Density must be measured at a lower density than the lowest compression density to produce consistent results in production.  By establishing benchmarks for various production conditions, change is detected and corrections made before it either creates a problem for production or loads the nest available compressor.

Interpolate compressor performance against supply electricity by comparing supply and demand pressure against demand flow.  In some instances flow meters on the supply side of the system check compressor flow.  What compressors are displacing is a mute issue in com0parison to what demand is using.  In a supply managed rather demand-managed system, the volume of unregulated users increases as total demand requirements fall.  This relation gives a rather distorted picture in a “needs” assessment.

Demand Pressure – This should be measured at the downstream side of demand flow.  This pressure is critical for production and diagnostic evaluation.

Sector Pressure – Transducers should be located downstream in critical use sectors of the system.  They should measure in hundredths of a psig.  By comparing them to each other and against demand pressure and volume, where flow is being used and where the real problems are in a system are determined.

Demand Pressure Dew Point – Measure this downstream of demand flow measurement.  Pressure dew point is only relevant to actual demand pressure.  Many dryers are applied at dew points, which are much too low, because their performance is measured on the supply side of the system.

Relative Humidity and Ambient Temperature – All compressed air systems operate as a function of ambient conditions.  The results in production and amount of power needed are expressions of change in conditions.  Differentiate between condition changes and maintenance problems.

By data manipulation, useful information is developed that is far more important to systems management than individual pieces of information.  Several items are important.

Demand Events or Rate of Change – These are expressions of demand change in production. They should reflect only the actual changes in demand in scfm or lb/min as compared to a previous condition.  This is the best status check for what the system is doing and how well it is supporting production.  This information is obtained by comparing the pressure change in the system to the time register in the database and the constant storage capacity of the system expressed in scf per psid.  One of the most important uses for this data is measuring usage of production equipment.

Total Kilowatts – Simply add the individual kw from each compressor.  This information is used to determine hourly cost.  Combined with peak demand and time, this could be inputted to a spreadsheet and used to generate a monthly billing service.

Volume per Kilowatt – This number should be expressed to two or three decimal points.  It is a constant tracking of how well supply and demand are managed.  If compressor mix or efficiency changes, it is expressed against previous recorded demand flow at a lower efficiency level.  For the busy plant engineer or maintenance superintendent, monitoring of this one number could easily be an alert to difficulties in the system far in advance of a crisis.
 

Reprinted with permission from R. Scot Foss, president of Plant Air Technology, Charlotte, N.C., a company specializing in system auditing and design.  This article is based on his book, "Compressed Air System Solution."  A portion of the proceeds from sales of the book is donated to children’s charities.  To order a copy of the book, please contact Southern Corporation.