<< BACK

Issues with Dehumidification

Dehumidification, while often effective in preventing corrosion, has numerous cost and performance problems.  To illustrate the issues, the US Navy recently conducted a test on dehumidification because it wanted better and more cost effective protection than provided by corrosion preventative compounds (CPCs).  This test is illustrated at the following website http://www.lakehurst.navy.mil/p2/servlet/DocServlet?wDID=242. 

Please note that the facility used as storage was a shelter that they had to line with plastic to provide a vapor barrier.  They chose a PVC liner, which, incidentally, offgases corrosive chlorine gas.

“3.1 Shelter

The shelter at NAS JRB Fort Worth consists of an existing concrete bunker, 120 ft. long x 40 ft. wide x 20 ft. high, that was originally designed to store cruise missiles. Access to the building is provided by four 14 ft. x 18 ft. sliding blast doors. The inner walls and ceiling of the building are lined with a translucent, PVC-impregnated, polyester weave fabric to minimize vapor transmission. The fabric is self-extinguishing and complies with NFPA Standard 701. The maximum vapor transmission rate is 0.08 g/100 in.² in 24 hours.”

Some of the study’s comments include:

• Compared with the previous Level I preservation efforts employed, the dehumidified storage system has a 10-year return on investment of $96,552.20. The break-even point is 4.8 years. Refer to the Cost Analysis for complete data.  <That comes to a dehumidification cost of $46,000.>
• The DSS experienced several component failures during the evaluation period. The contractor addressed problems promptly when they occurred.
• The process air relay on the dehumidifier repeatedly tripped itself and had to be manually reset. Since this happened before the telephone line was installed, no one was notified of the failure. It is suspected that power outages caused by thunderstorms may have contributed to this problem.
• The dehumidifier was replaced with the spare unit to continue system operation. The malfunctioning dehumidifier remained onsite, and the manufacturer supplied a relay for site personnel to install.
• During a contractor site visit, the ambient temperature sensor was found to be inoperative. The sensor was removed and replaced with the spare.
• It was suspected that the CPU of the remote computer in the personnel office was infected with a virus, and the entire operating system was destroyed  

The cost of the Intercept Shrinkfilm solution would have been about $5000, which would include materials, labor, and travel.  That would include a layer of plastic draped over the equipment to more quickly help “soak up” the corrosive gases trapped inside the large volume.

Additional dehumidification equipment problems associated with expense, oversight, and potential for failure can be demonstrated in the following study conducted by the National Renewable Energy Lab report at http://www.nrel.gov/docs/fy01osti/26131.pdf.

Dehumidifiers usually use desiccant wheels, which can have some of the following issues:

• Some wheels use proportionately more or less regeneration air than others. The mass-flow ratio typically ranges between 0.25 and 1.0.  Fan power is typically a one consumer of primary energy (<20%) in a system. The other is that typically only about 20% of the system fan-power requirement can be attributed to pressure drops across the desiccant wheel and heat exchanger combined. The bulk of the pressure drop comes from turbulent airflow through the cabinet and ducting. These figures assume a 1.0 mass flow ratio exists at some point in the system. This is the case for designs that employ a heat exchanger (i.e. most cooling applications where supply temperature must be minimized), so even for systems that exhaust a portion of the regeneration-side airflow prior to the regeneration heater; the wheels consume a fraction of the total fan power. Therefore, wheel pressure drop is more important a factor to seal performance than overall energy consumption.
• Rotary heat/mass transfer devices produce very spatially non-uniform air temperature distributions.
• Improper ducting can present a very non-uniform air distribution that will degrade performance. Introducing inlet air too close to the wheel or at an odd angle through too small a duct can cause blow through. It starves some portions of the wheel, and raises flute velocities in others for a net negative effect on performance.
• Co-sorption is the potential for desiccants to adsorb other chemicals with the water vapor. If the desiccant were able to pick up considerable amounts of undesirable chemicals from an exhaust flow and dump them back into the supply air, this would create a much more powerful carryover effect than wheel rotation could produce, and essentially concentrate the pollutants in the storage container.
• As the matrix rotates out of the regeneration airflow, it carries with it both regeneration air trapped in the flutes and heat, contained in the air and in the matrix itself. This amounts to a small, constant .rotation leak or carryover from RI to PO. Purging purposely misaligns one of the seals on the RI/PO face of the wheel to eliminate this leak by forcing a purge leak from PI (process inlet) to RI (regeneration inlet).
• Wheel matrices are generally not perfectly uniform, in either open area or desiccant loading, and excess desiccant or compressed flutes will tend to restrict the air passages. This means airflow resistance varies with circumferential location. If the wheel has sufficient authority in the airflow circuit, its rotation will cycle the flow rates in synch with its frequency. It also means that performance can vary the same way.
• Leakage across face seals is a common condition that prevents moisture mass balance. The seals on commercial units typically will allow balance when face differentials are kept below 2. w.c. A balance of less than 1.0 usually indicates leakage from RI to PO, and degradation in MRC. The bone-dry PO air is very susceptible to small leaks of wet regeneration air. If the test system does not employ four fans, it may be necessary to induce a pressure drop on the PO ductwork to stop the leak.
• Leaks from inlets to outlets affect actual face velocities and contaminate outlet flows. In the field, fans are often arranged in blow/draw configuration to preserve grain depression in the supply air. Supply air is blown through the wheel, and regeneration air is drawn through. This prevents any regeneration air from forcing its way into the process side of the cassette, which can seriously degrade performance.
• Standard baffles, screens, or mixing vanes accelerate thermal mixing, as shown in Figure 4, and can help shield sensors or sampling trees from radiative heat exchange with the rotor. A drawback of these devices is that the pressure drop they add is not adjustable, and may adversely affect face pressure differentials at times.
• Due to the unique nature of a desiccant dehumidification rotor, the moist air properties of the air streams leaving one of these devices will be quite different from other HVAC equipment. The process outlet air will typically be single-digit relative humidity while the regeneration outlet will be hotter and more humid than naturally occurring, terrestrial environments. Because of this, careful selection of appropriate humidity sensors is required.
• The air leaving an actively regenerated desiccant rotor is very non-uniform in humidity. It is postulated by Reynolds Analogy that once the air is thermally uniform, moisture uniformity is also achieved.

And then there is the sensing equipment that regulates the dehumidifier. 

·        The primary advantage to a dew-point hygrometer is its ability to measure low relative humidity air while maintaining a high degree of accuracy.  Like the aspirated psychrometer, a chilled mirror hygrometer suffers from contamination.  The surface of the mirror must be cleaned periodically to remove contaminants. Unlike the other humidity measurement sensors, the chilled mirror hygrometer uses a control loop to maintain accurate measurements. At times the instrument will get lost and search for its equilibrium point. Depending on the nature of the event, the hygrometer may not be able to get back in control on its own and will have to be reset manually.

Actively regenerated desiccant rotors can have the following problems.  They include (but are not limited to):

Pressure/Flow

·         Maldistribution of air supplied to the rotor (blowthrough)
·         Air leaks between air measurement stations
·         Use of instrumentation outside of published range
·         Use of instrumentation out of calibration
·         Not allowing appropriate development lengths upstream or downstream of           nozzles
·         Poor nozzle construction
·         Poor pressure tap construction/location.

Temperature/Humidity

• Sampling of a non-uniform air stream
• Conduction and/or radiation affecting dry-bulb and/or wet-bulb measurements
• Use of instrumentation outside of published range
• Use of instrumentation out of calibration
• Allowing condensation to form in sampling tubes
• Insufficiently insulated ducts or sampling tubes
• Contaminated wicks for wet-bulb measurements
• Contaminated mirror for dew-point sensors
• Insufficient air flow across a sensor
• Requiring a dew point sensor or wet-bulb sensor to develop a temperature depression greater than their capability.
• Instrument readings contain both random and bias errors.

Actively regenerated desiccant dehumidification rotors can be regenerated with air at temperatures up to 400°F. Ducts and cabinets carrying this air should be well insulated to prevent contact with personnel. The temperature of the regeneration air should be continuously monitored by a high temperature limit controller separate from the computer to prevent the possibility of excessively high temperature air from entering the article and causing damage to the rotor or seals.

The first time air is passed through a rotor, desiccant dust and fumes from the manufacturing process may cause problems for sensitive personnel. An initiation period of approximately two hours is recommended, during which the outlet air is not sampled and is exhausted outdoors.

If exposed to high relative humidity air without sufficient regeneration, lithium chloride (LiCl) rotors will deliquesce, a phenomenon whereby the desiccant over-adsorbs moisture to the point where damage occurs.

 Active dehumidification has many things that can go wrong.

• Storage covers can emit corrosive gases (plasticizers)
• Reused storage covers can have embedded dirt and oils, which can damage the stored item
• Energy consumption can be high
• Energy prices are likely to increase
• Maintenance cost can be high
• Often dehumidification requires special pallets.
• Equipment can fail or require expensive repairs.
• May not remove external corrosive gases passing into storage container
• The slightest damage to the barrier layer may negate the effectiveness of corrosion protection.
• Too dry of humidity can cause cracking of elastomerics, gaskets, and electrical wiring insulation