A few years ago, when energy costs were a lot lower, our supermarket client asked if it should purchase and install a desiccant system for its upcoming new supermarket. The new store was about 120,000 sq ft, with large outside air requirements due to the store's kitchen exhaust hoods, which are used in making prepared foods. The store is located in Rochester, NY, on Lake Ontario. The higher humidity near the lake was one reason to consider using desiccants. In addition, the cost of natural gas is lower than that of electric energy, for the same usage.

As with many engineers, I did not find a lot of data on the actual annual savings of energy dollars offered by desiccant units. The sales representative of the desiccant unit was helpful on its operation, size, and price, but when questioned about return on investment (ROI), the answers were given in terms of a simple savings percent at a specific design condition of the units. Consulting the factory yielded the same reply. Just as other readers quoted in past issues of ES have reported, I found that there is little available or known about annual dollar savings via installed desiccant systems.

Thus, my ROI study had to begin with a clean piece of paper. The study began with the reasoning that we needed to know the outdoor absolute humidity, in grains/lb of dry air, for each hour, or day. Then I would develop a day-by-day calculation of the savings when the desiccant was running, and also when it should not run, based on economic reasons. Then, the study would provide data on the following:

  • Savings of electrical energy;
  • Savings in electrical demand; and
  • Additional cost of natural gas.

Reducing the electrical demand requires changing the schedule of the compressor sizing. This often may lead to additional significant savings, perhaps even equal to the savings from decreased electric energy consumption.

Desiccant systems can reduce energy expenses only when the outdoor absolute humidity is high, and when the use of natural gas can reduce the grains of water/lb of dry air without using the hvac system - or in this case, without using the electric hvac or electric refrigeration compressors. The monthly ROI must be calculated based on the outside absolute humidity levels. Then the electrical energy savings can be calculated using humidity-type bins.

Table 1. Bins by average grains of moisture, with the number of occurring days.

The Desiccant Study

At the time I conducted this study, ASHRAE did not offer annual bins of humidity; the raw national weather data was used. Then, it was necessary to use that raw data and then make "bins" of outdoor humidity levels. The following (Table 1) represents the manually made bins of the Rochester area outside conditions, arranged by humidity level and including the average temperatures of each day.

Table 2. Dollars of hvac annual energy saved.

The Hvac Study

On the hvac side, the exhaust systems for the in-store kitchens is over 15,000 cfm, which has a significant outside-air latent load of about 40 tons. In this study, we looked at placing the desiccant unit on the roof to transfer the latent load of the outside air from the rooftop compressors, by first sending the outside air through the desiccant unit.

Since we now know the number of days, and the absolute humidity of the outside air in each of these "humidity" bins, we can calculate the monthly electrical savings, and the new gas usage for each of these "absolute moisture" bins. Again, the calculations measure reduction of hvac compressor electrical usage, reduction of electrical demand, and increase of natural gas usage at the desiccant unit.

These amounts, in each humidity bin, will be multiplied by the number of total days of operation to develop the dollar savings total. For space purposes, the following are summaries of the results found for each bin.

Bin A. The latent load saved was 632,400 Btuh. The hvac sensible load saved by not bringing air down to 62°F (conditions for dehumidifying) was 129,600 Btuh. The hvac load in discharge temperature of the desiccant unit was -32,400 Btuh. Thus, the total hvac load saved was 729,600 Btuh.

Converted to electrical terms, this load savings translates to 97.8 kW. The additional fan load required reduces the savings to 88.08 kW. The total natural gas used by the desiccant unit, according to specs, equals 560 MBtuh, calculating a store rh of 42%. The total hrs of operation are 99 (9 days, 11 hrs/day).

Bin B. The latent load saved was 612,000 Btuh. The hvac sensible load saved by not bringing air down to 62° (conditions for dehumidifying) was 210,600 Btuh. The hvac load in discharge temperature of the desiccant unit was 48,600 Btuh. Thus, the total hvac load saved was 871,200 Btuh.

Converted to electrical terms, this load savings translates to 116.7 kW. The additional fan load required reduces the savings to 107.8 kW. The total natural gas used by the desiccant unit, according to specs, equals 552 MBtuh, calculating a store rh of 38%. The hrs of operation are 209 (19 days, 11 hrs/day).

Bin C. The latent load saved was 571,400 Btuh. The hvac sensible load saved by not bringing air down to 62° (conditions for dehumidifying) was zero. The hvac load in discharge temperature of the desiccant unit was -97,200 Btuh. Thus, the total hvac load saved was 474,000 Btuh.

Converted to electrical terms, this load savings translates to 63.5 kW. The additional fan load required reduces the savings to 53.8 kW. The total natural gas used by the desiccant unit, according to specs, equals 504 MBtuh, calculating a store rh of 40%. The total hrs of operation are 176 (16 days, 11 hrs/day).

Bin D. The latent load saved was 459,000 Btuh. The hvac sensible load saved by not bringing air down to 62° (conditions for dehumidifying) was zero. The hvac load in discharge temperature of the desiccant unit was -81,000 Btuh. Thus, the total hvac load saved was 378,000 Btuh.

Converted to electrical terms, this load savings translates to 50.7 kW. The additional fan load required reduces the savings to 40.9 kW. The total natural gas used by the desiccant unit, according to specs, equals 448 MBtuh, calculating a store rh of 35%. The total hrs of operation are 330 (30 days, 11 hrs/day).

Bin E. The latent load saved was 295,800 Btuh. The hvac sensible load saved by not bringing air down to 62° (conditions for dehumidifying) was zero. The hvac load in discharge temperature of the desiccant unit was -210,600 Btuh. Thus, the total hvac load saved was 85,200 Btuh.

Converted to electrical terms, this load savings translates to 11.46 kW. The additional fan load required reduces the savings to 1.7 kW. The total natural gas used by the desiccant unit, according to specs, equals 448 MBtuh, calculating a store rh of 35%. The total hrs of operation are 242 (22 days, 11 hrs/day). For this bin, I made a note to discard Bin E, since calculations showed that the natural gas will cost more than the electrical savings.

The summary of the bin results and their composite savings is shown in Table 2.

Table 3. The depression of rh (by hrs) as a result of the desiccant system.

Hvac Demand Savings

The desiccant system will handle 729,000 Btuh of latent load, or about 60 tons latent load in the 15,000 cfm of outside air. The hvac units can have the same reduction of tonnage. When considering three 20-ton compressors, this total savings translates to $4,096. About forty percent ($1,520) of this amount comes from the reduction in off-month demand costs.

Table 4. The total changes in energy costs ($) as a result of using a desiccant system in this case.

Refrigeration Analysis

On the refrigeration side, the lower in-store humidity delivered by the desiccant unit lowered the latent load on the refrigeration compressor racks. The door defrost heaters are "double" users of energy, using electric heat and the refrigeration compressor to re-cool the cases when the door heaters are on. However, this calculation would require some laboratory testing, and it was likely less than 5% of the overall savings.

The Coefficient of Performance (C.O.P.) of the refrigeration case compressors are relatively low, compared with the hvac compressors. This is due to much lower suction temperatures.

The supermarket had five refrigeration sections in the store, which saved their compressor electric power when the wetbulb (wb) temperature was depressed in the store. This information was obtained from the case manufacturers. Therefore, the use of desiccant depression of the humidity during hours of operation is a savings in refrigeration compressors. In this application, Table 3 reflects the depression of rh by hours.

The latent loads of the various refrigeration cases were determined by previous studies, and by testing near the cases. The total case latent load added up to 81,240 Btuh. The case manufacturers were contacted to determine the expected reduction in compressor operating time, with the given depressed wb temperatures. The reported reduction of operating time for the refrigeration compressors varied from 11% to 25%.

Now knowing the refrigeration loads, and the lower time of compressor rack operation when the wb was depressed by the desiccant unit, the calculations were done. Similar to the hvac calculations, it was determined that based on the hrs of the desiccant unit operation, the yearly savings were 9,972,197 Btu/yr. These specific calculations have been omitted due to their length.

In transferring that figure to electric power saved, each of the five cases of electrical savings were calculated using the refrigeration C.O.P.s described above. This amounted to 9,972 kW/hr/yr. And using the same cost of electricity at that time ($.086/kW/hr), the electrical savings was $857.

Refrigeration demand was not considered, since the compressor racks in the refrigeration system should not be changed, assuming a possible downtime of the desiccant system.

Conclusions

The total changes in energy cost as a result of using desiccants in this instance are summarized in Table 4, with total annual energy savings totaling $7,340. The study did not have an accurate cost of installing the desiccant system. Based on the rough estimate to install the system, as given by the factory representative, the installed cost was reported to be approximately $135,000 (20%). Therefore, the desiccant unit had a simple payback of 18 yrs. However, as energy cost has risen in recent years, and may well continue to rise in the future, it is reasonable to consider another look at a desiccant unit's payback given today's utility markets. Of course, southern areas of the country have a significantly higher number of hrs of high outdoor absolute humidity, which of course would reduce the average desiccant system's payback years even further. ES