Compressed Air System

How to conduct compressed air audits on a budget

Low-cost, low-tech ideas to improve compressed air system operation and efficiency.

By Ron Marshall, Compressed Air Challenge

Is your budget so tight you can’t afford to bring in a compressed air auditor? There are a number of low-tech and low-cost ways to assess your system to get a good idea of your current operating costs, system efficiency, and level of system waste. Once you know your numbers you can even predict possible savings for some common compressed air efficiency measures. Why wait when you can do it now and start the process of harvesting some low hanging fruit?

A definite first step in understanding your system is to understand the supply side and how the compressed air is produced. Effective assessment methods should look at both the supply and the demand parts of the system. A very good way to learn about compressed air systems is to take part in one of Compressed Air Challenge’s Fundamentals of Compressed Air Systems seminars somewhere near you. Some of the methods discussed in this article are also presented in this seminar.

Drawing a block diagram

The first step in assessing your system is to draw out a block diagram of your supply side and some elements of the demand side so you can start to understand how things are connected and how the compressed air flows through to the end uses. On your block diagram you will need to collect and record relevant information about your compressors, air dryers, filters, and storage receivers (Figure 1). This will give you a resource from which you can do calculations or ask questions of your compressor supplier or service provider. One important piece of information is compressor nameplate data. Through this information you can learn the type of air compressors you have, their pressure ratings, their rated power consumption, and how much air they can produce.

cac1
Figure 1. The first step in assessing your system is to draw out a block diagram of your supply side and some elements of the demand side so you can start to understand how things are connected and how the compressed air flows through to the end uses. (Source: Compressed Air Challenge)

Control mode

One of the things students of the CAC’s Fundamentals seminar learn is the differences in the various compressor control modes. These modes of operation and the way each compressor is set up to operate within a system are often the most important elements in producing compressed air efficiently. In fact, one of the key ways to optimize your compressed air system is to produce the compressed air in the most efficient manner possible.

For the purposes of this article we will deal with lubricated rotary screw compressors, the most common type of compressor in the industrial market. This leaves out centrifugal compressors and multi-stage reciprocating units. Lubricated rotary screw compressors can operate in one of five capacity control modes:

  • start/stop
  • inlet modulation
  • load/unload
  • variable displacement
  • variable speed.

To complicate matters some compressors can operate in a number of these modes at the same time. Let's assume for our purposes that any compressors used in examples will be in one mode only.

An air compressor has to be controlled because, if you think about it, it is very rare that a fully loaded compressor will exactly match the plant load in a facility. If the compressor were left uncontrolled it obviously would not start when there was low pressure; if it were manually started and left to run at full load it would push the pressure up to extreme values until something blew up. Of course this is not what we want, so the compressor manufacturers have figured out various ways to automatically control the compressors by limiting the output in some way to match the compressed air load. Determining which control mode you're using is important because each has different characteristics. If you don’t know, you should ask your compressor supplier.

Calculating baseline energy and cost

Making a block diagram tells you what you have; the next items you need to calculate are how much energy the system is consuming and how much it's costing. The second step in the road to improvement involves creating a baseline, determining your energy consumption. This step involves taking basic electrical measurements or estimating the electrical consumption. In addition, annual operating hours need to be determined. For more accurate cost estimate results, you’ll also need a copy of your most recent electrical bill.

For most operating modes, it's fairly easy to get a rough idea of each compressor’s energy consumption. The tricky part is estimating how much compressed air flow each compressor is producing so you can estimate the supply system efficiency, expressed as specific power (kW per 100 cfm produced). The method of measurement depends on how accurate you want to be, with the highest accuracy costing the most money.

Measuring the compressor power consumption is best done using a three-phase kW meter. If measuring an air compressor for baselining without a kW meter there is a standard formula to use to estimate the power consumption from measured Amps and Volts:

fig1

 

 

Where:
A = average Amps of all three phases
V = average line-to-line voltage
PF = measured or estimated power factor (Power factor at full load can often be taken from the main compressor nameplate. If not known, use 0.85 at full load and 0.6 at in the unload position.)

Measuring electrical parameters shall always be done by qualified personnel using the appropriate personal protective equipment and approved safety procedures.

To estimate the power consumption of compressors in start/stop, modulation, capacity control, and VSD modes, a number of measurements need to be taken at various times to estimate the power for the full operating profile. For these control modes the power factor used in the formula remains fairly constant and would be near nameplate values. For compressors in load/unload mode if the Amps falls below about 70% of full load, the power factor used in the formula should be reduced to about 0.60. Measurements can be done manually, but for best accuracy data loggers should be used and average Amps and voltage determined from the data output based on many measurements over a long period of time.

To calculate annual kWh consumed by the compressor you need to determine how long in a year the compressor is running in the average conditions. This can be estimated using the compressor hour meters, if these have been recorded for maintenance purposes, or simply by observing the compressor operating hours and sitting down with a calendar and counting the days of operation per year. From this, the annual hours and energy costs can be calculated:

Annual cost = average kW x annual hours x blended power rate

The blended rate can be estimated by looking at your monthly power bill and doing some basic calculations. Taking the total billed amount and dividing by the number of kWh used is a good estimate to use for initial cost analysis. Of course power companies complicate matters by charging for time of use, charging different rates for different blocks of power, and applying demand charges, but for these rough blended cost calculations we will ignore this.

There are also other items in your compressor room that consume power, the most significant of which are the air dryers. If the air dryers are the refrigerant type, the power consumption of these should be measured and added to the baseline power and energy calculations, too.

Baseline pressure profile

The most important issue in a compressed air system is providing the end user with adequate pressure to do the intended job. There will be one or two end users that seem to need higher pressure than all the rest, these can cause the compressor discharge pressures to rise to higher levels. The higher the pressure, the more it costs to produce the air because for systems operating near 100 psi, for every 2 psi in higher pressure, the air compressors consume about 1% more power.

Pressures can be measured at the various points as indicated in Figure 2. These measurements should be taken at the same time and frequency as the power readings; therefore, the best accuracy is gained using data loggers, but careful manual readings can suffice. Accurate calibrated pressure gauges should be used; digital units are fine, but often quick variations in pressure cannot be detected without using standard mechanical gauges.

cac2
Figure 2. Pressures can be measured at the various points and should be taken at the same time and frequency as the power readings; therefore, the best accuracy is gained using data loggers, but careful manual readings can suffice. (Source: Compressed Air Challenge)

The more readings that are taken the more accurate the profile will be. Since the worst case pressure profile is what you are looking for, because this is what sets the required compressor discharge pressure, the pressure readings should be taken during the highest system flows. These readings are useful in determining what savings might be gained by optimizing your system through reducing pressure differentials and compressor discharge pressure.

Be aware that there are very many end users connected to the system. In order to truly optimize the system, the most pressure critical end users must be found out of hundreds, perhaps thousands, of compressed air consuming equipment. To find the needle-in-a-haystack involves asking a lot of questions and thoroughly going through the system. The values in Figure 3 show a typical worst case pressure profile in an industrial plant where the compressors are producing pressures of more than 100 psi, but the final end user is only getting a pressure of 70 psi.

cac3
Figure 3. This system pressure profile shows significant pressure differentials are affecting end use pressure. (Source: Compressed Air Challenge)

Estimating or measuring flow

Estimating average flow is often a difficult process, especially for compressors operating in modulation or variable displacement modes. If you have these types of compressors, the best bet is to purchase and install a low-cost thermal mass flow meter. If this is not possible a rough estimate of compressor output flows can be done using average Amps/kW readings or by observing the average output pressure at the discharge of the compressor.

Compressors with modulation control operate within a fixed pressure band, typically 10 psi wide. At the low end of the pressure band a compressor with a functional and properly maintained control system will be outputting full load. At the high end of the pressure band the output of the compressor will be at its fully modulated flow. This relationship can be seen in the chart in Figure 4. Note that fully modulated flows vary with compressor type; consultation with the manufacturer may be required in creating an accurate relationship for your particular compressor.

cac4
Figure 4. Compressors with modulation control operate within a fixed pressure band, typically 10 psi wide. At the low end of the pressure band a compressor with a functional and properly maintained control system will be outputting full load. At the high end of the pressure band the output of the compressor will be at its fully modulated flow. (Source: Compressed Air Challenge)

The calculation of the estimated compressor flow is fairly straightforward using simple ratios. For example, if the compressor is halfway into its modulation band (5 psi in the case of a 10 psi range), it will be producing the rated output at the halfway point of its curve. If the type of compressor you have modulates between 0% and 100%, then the pressure/flow relationship would be extended to the full compressor range, not between 40% and 100%, as shown on the chart.

Either pressure or Amps/power can be used to roughly estimate the output flow of a modulating compressor using this typical curve:

fig33

 

 

Where:
Average Amps = recorded average Amps during the period, not including off time
Full Modulation Amps = Amps when the inlet modulation valve is fully closed
FL Amps = Amps when inlet valve is fully open
Range cfm = flow range in the modulation band = full load cfm – full modulation cfm
Modulation cfm = flow at full modulation (varies with compressor modulation setting)

Or if average pressure is used:

fig44

 

 

Where:
Average psi = recorded average psi during the period, not including off time
Full modulation psi = psi when the inlet modulation valve is fully closed
FL psi = psi when inlet valve is fully open
Range cfm = flow range in the modulation band = full load cfm – full mod cfm
Modulation cfm = flow at full modulation (varies with compressor setting)

These formulas are only valid for operation within the range of the compressor modulation control. For example, if the compressor unloads or turns off during the period measured the result of the calculation is incorrect. Also, if the compressor is in draw down, that is, if it is at full output but cannot keep the pressure within the modulation range, then the formula is inaccurate and will produce a flow higher than the rated capacity of the compressor. As a result of these limitations the interpretation of the results is tricky; this is the area where good compressed air auditors earn their keep.

For compressors running in load/unload mode, estimating the average compressed air output is much easier. Most modern controls have hour meters that measure the time the compressor has been loaded and running. Calculating the average output between two points in time is then a very simple calculation:

fig5

 

 

Where:
Loaded hours = number of hours between readings
System hours = number of hours the system was active (compressors turned on) between readings
Rated flow = cfm flow from the compressor nameplate or CAGI data sheet

These calculations are nice, but the best way to get an accurate picture of the flow output of a compressor station is to actually measure it with a flow meter. There are quite a few inexpensive insertion style thermal mass flow meters available on the market these days that are quite easy to use. It is possible to get a self-contained meter for less than $1,000 in 3-in. size and less than $2,000 for up to 8-in. pipe.

Thermal mass flow meters must be installed properly in a dry straight section of pipe or they won’t read correctly. On systems with highly varying loads, or in load/unload mode, instantaneous readings can be inaccurate due to the fast changes in flow. It is best to use long-term averages that match the duration used in measuring the compressor power consumption.

Leaks and end uses

Another basic way to reduce compressed air costs is to use less compressed air. To determine the savings that can be gained by reducing flow we must measure how much compressed air is being used on average and try to estimate what part of that flow is useful and what part is wasted or used inappropriately.

To figure this out, an important measurement of a compressed air system is the level of leaks. If you have load/unload compressors, there is a simple test that can be done using a few basic tools — a wristwatch and a calculator. During a plant shutdown, where all production machines have been shut down and the only remaining load is leaks, conduct a test of the compressor load/unload cycles. From these cycles the percentage loading of the running compressor can be calculated:

fig6

 

 

Where:
% Leaks = percentage of capacity of the running compressor (flow can be calculated if capacity is known)
T = compressor loaded time in seconds
t = compressor unloaded time in seconds

The leakage flow is then estimated by multiplying the percentage leaks by the rated compressor output capacity. Estimating leaks using compressors with modulating or variable capacity controls can be done using the procedure in Fact Sheet 7 on the Compressed Air Challenge website. If your compressor is a variable-speed-controlled unit the percentage flow will be a simple ratio of the speed during the test divided by the rated full speed.

The timer test can also be used to sectionalize different parts of the plant or estimate potential flow reductions in eliminating inappropriate end-use flows. During the testing if you can turn on the end uses for special tests during the leak testing you can estimate the resulting change in flows.

If you are interested in learning more about potentially inappropriate uses of compressed air, a list can be found in an Inappropriate Use Tip Sheet on the CAC website.

System efficiency

Once the average kW and average flow is measured the specific power of the system can be calculated. The specific power is like a “gas mileage” number for your system, where you are calculating how much power you are putting in to the system compared to what you are getting out. The most common units used for specific power is kW/100 cfm as defined by the Compressed Air and Gas Institute (CAGI). Most modern compressors have CAGI data sheets published showing their rated specific power numbers.

If the number you calculate is not very close to the rated specific power of your air compressors as taken from the CAGI data sheets, then there is cause for concern. For example, if your compressors are rated at 20 kW/100 cfm and you have determined your system is running at 40 kW/100 cfm, then this is an indication that significant improvement of up to 50% might be gained by changing the way the compressors are being controlled or by replacing the equipment altogether.

But specific power is not the only important number; the overall energy consumption in kWh and the percentage waste are also important key performance indicators that can be used to assess the efficiency of the system and to compare the operation of the system before and after changes.

Potential savings

If you have done some measurements, you may now be armed with some numbers about your system, but how much can be saved? It is now fairly simple to estimate the overall possibilities. For example, if it is discovered a poorly controlled, unmaintained system is consuming 105 kW while producing an average of 300 cfm for 8,760 hours per year, we can calculate a specific power of 35 kW per 100 cfm. This system would consume about 920,000 kWh per year costing $92,000 in electrical charges at $0.10/kWh blended rate. Here are some things that might be done and resulting savings:

  • By optimizing the compressor control by installing a VSD compressor, coordinating compressor control, and installing a cycling dryer, the system specific power might be reduced to 20 kW/100 cfm.
  • Lowering compressor discharge pressure by 10 psi by fixing things found in our pressure profile measurements might further reduce the specific power to 18.5 kW/100 cfm.
  • Eliminating inappropriate uses, wasteful system drainage, and artificial demand by 50 cfm might reduce average kW by 9.25 kW.
  • Fixing 100 cfm of leaks might further reduce average power by 18.5 kW.
  • Reducing the operating hours to 6,000 by turning the system off on weekends and holidays might save 2,500 hours of costly operation where only leaks are being supplied because there is no production activities.
Ron Marshall is a member of the Project Development Committee at the Compressed Air Challenge. Contact him at rcmarshall@hydro.mb.ca and (204) 360-3658.

The final energy consumption for this system to produce the remaining 150 cfm at 18.5 kW/100 cfm (27.75 kW) for 6,000 hours might be 166,500 kWh costing $16,650 in electricity costs for a potential savings of more than 80%. This is not an unrealistic goal. There are numerous systems in operation today with similar potential opportunities for savings. Many have already been optimized for proven savings in this range.

Assessing your compressed air system can be an affordable DIY project if careful low-cost measurements are taken. This can give you a rough idea of your potential savings. You can then justify further more detailed study that can get you some detailed information about what you can improve to get your system running optimally.