Continuous Emission Monitoring. James A. JahnkeЧитать онлайн книгу.
– freeze‐protection line) can be used to transport the sample to the analyzer.
Typical temperatures for extractive‐system heated sample lines are higher than 180 °C. Typical sample line materials are high‐density polyethylene (HDPE – maximum temperature 110 °C), polytetrafluoroethylene (PTFE – maximum temperature 260 °C), perfluoroalkoxy Teflon (PFA – maximum temperature 260 °C), and stainless steel (ss – either uncoated or treated coated with glass (SilcoNert®)). Heated umbilicals can be operated either at a constant power density or can be self‐limiting (self‐regulating). In self‐limiting lines, a specified minimum operating temperature will be maintained if the ambient temperature falls (e.g. sub‐zero conditions in winter) and will not exceed a specified maximum temperature if the ambient temperature rises. A cross section of a self‐regulated line for a dilution extractive system, with the heating element in the center of the bundle for freeze‐protection, is shown in Figure 3‐9.
This umbilical consists of the following components:
One 3/8″ O.D. PFA Teflon tube heated sample line
One 3/8″ O.D. PFA Teflon tube heated dilution air line
Two 1/4″ O.D. PFA Teflon tube (unheated calibration gas and purge air and spare)
Two 1/4″ O.D. PFA Teflon tube (unheated spares)
Two 3/8″ O.D. PFA Teflon tube (unheated spares)
Electrical lines in the umbilical include the following:
Sample line heater (120 self‐regulating heating cable)
Sample line Type K thermocouple
The low‐temperature self‐regulating heater in the line operates to maintain a temperature of approximately 50 °F, to guard against water condensation in the flue gas sample if the ambient temperature decreases below the dew point of the diluted sample gas (<−40 °C). The internal heated portion of the bundle is enclosed by a thermal barrier of an aluminum heat transfer tape. The outside of the bundle is enclosed by a flexible, black polyvinyl chloride insulation jacket. In this case, the calibration gas lines are located in the outer, unheated jacket of the bundle. Another option is to provide a separate, unheated umbilical to house the zero air and calibration gas lines.
Although much information has been cataloged on the relative chemical resistance of sample line materials to corrosive gases (McNulty et al. 1974; Podlenski 1984), 316 stainless steel or PFA Teflon are commonly used. Problems can occur with Teflon if the line temperature is too high since its softening temperature is near 250 °C. Efforts to increase sample line temperature to avoid condensation or adsorption sometimes melt the line and the umbilical cable. Furthermore, some gases (such as CO) can permeate through polymeric sample lines (Dunder and Stone 1995), especially at higher temperatures.
One solution to minimizing adsorption of volatile organic compounds has been the development of glass‐coated flexible stainless‐steel tubing (Gerhab and Schuyler 1996). This combination of fused silica and stainless steel can be heated to desorb gases from the tubing surface, if necessary. The company SilcoTek, (formerly, Restek) located in Bellefonte, PA, is the principal supplier of coating services for tubing, parts, fittings, chemical reactors, and process industry components. Proprietary materials such as SilcoNert®, Dursan®, and others are deposited by chemical vapor deposition to a depth of typically 2 μm (SilcoTek 2017).
Figure 3‐9 Umbilical line cross section for a dilution extractive system.
Chemical attack and wall adsorption effects can be a problem when sampling reactive gases. For example, Marshak (2015, 2019) and Hoard et al. (2014) have examined the problem of NH3 retention in detail, using an FTIR spectrometer. At a flow rate of 3 l/min, they found that the time it takes for an NH3 calibration gas value to stabilize at the analyzer can be significant, particularly when using untreated stainless‐steel sample lines. They also found that NH3 adsorption of the sample system walls was dependent upon line temperature, flow rates, and the material. However, line length and diameter were not found to have a statistically significant effect. It was also found that PFA and SilcoNert® stainless steel lines work well for NH3 with a flow rate of > 5 l/min, where 10–15 l/min may be preferable. Also, humidification of the sample and higher line temperatures reduced adsorption. SilcoNert® stainless steel was found to be the best for sample probes, and filters composed of borosilicate glass were found to be better.
Umbilicals are usually custom‐made and can contribute significantly to the expense of an extractive system. Umbilicals are usually priced per running foot and a total fixed price is generally not quoted by CEM system vendors unless the exact length is provided in a CEM system bid specification. Heated umbilicals are generally less than 250 ft in length, although longer lines can be installed if sufficient attention is paid to their construction. Freeze‐protect lines for dilution extractive mercury monitoring systems have worked well even at a length of 1000 ft; however, costs can be significant if, and when, they need to be replaced.
The time that it will take for gas to flow through a sample line will depend on the sample tube diameter, its length, and the flow rate. This lag can be roughly calculated from the expression (McNulty et al. 1974)
(3‐1)
where
t = the lag time (in minutes)
V = sample line volume (in liters)
Qsl = volumetric flow rate of gas through the line (liters per minute)
This expression is oversimplified since it assumes that there are no frictional wall effects restricting the flow. It shows, however, that if the sample line length is increased, the lag time will increase as the sample line volume increases. Typically, for a flow rate of 1 standard liter per minute (l/min), the lag time for a 100‐ft section of a 0.25‐in. internal diameter tube at a pressure drop of 152 mm Hg is only 30 s.
Hot spots should be avoided when installing a heated umbilical. Hot spots can be caused by the loss of temperature control through communication or temperature sensor failures or damages due to flexing problems. Hot spots can also occur if the bundle is clamped too tightly to a support, if the tubing makes contact with itself, or if multiple heated bundles are tied together, or if account is not taken for the temperature gradient between the inside and outside of a CEM shelter, in which case multiple heated zones may be necessary. The line should have a continuous downward slope from the probe to the conditioning system (5° is recommended) in case of condensation within the line, avoiding sags where condensate might collect, yet allowing for sufficient freedom for strain relief to allow for thermal expansion and contraction, especially when starting up or shutting down a unit. In the case of short circuits, damages may be minimized by incorporating ground fault circuit interrupter (GFCI) circuit breakers, tripping at 20 MA, into the system (Robinson 2010).
Condensation by water, acid, and other gases in cooler “pockets” can increase system corrosion or line plugging if fine particles are not adequately filtered at the probe. Sample lines can also become contaminated with reacted materials, which are difficult to remove. Plugs of particulate matter are difficult to remove in the heated sample line, and if a heater wire breaks or burns out, it is often difficult to find the break. On occasion, entire umbilicals will need to be replaced when these problems cannot be resolved (Bembe 2019).
Proper design and installation of the umbilical is important for its operation. This is not the area where costs can