1、3535789 0023838 b7T f DEPARTNENT OF THE ARMY ETL 1110-2-239 Office of the Chief of Engineers DAEN -CWE - HD Washington, DC 20314 Engineer Technical Letter No. 1110-2-239 - 15 September 1978 Engineering and Design NI TROGEN SUPERSATURATION 1. Pur ose. The purpose of this letter is to provide guidance
2、 hydraulic structures having the potential to produce nitrogen supersaturation. for + t e evaluation and identification of those projects with 2. Applicability. This letter applies to all field operating agencies having responsibilities for the design of Civil Works projects . 3. References. a. ER 1
3、130-2-334 b. ER 15-2-11 4. Bibliography. a. ER 1110-2-1402 b. EM 1110-2-1602 C. EM 1110-2-1603 5. Discussion. a. Nitrogen supersaturation and associated fish mortality due to gas bubble disease has occurred at Corps of Engineers projects on the Columbia River in the North Pacific Division (NPD) and
4、more recently at the Harry S. Truman project in the Missouri River Division. Nitrogen supersaturation can result at any hydraulic structure from entrained air introduced by the spillway-stilling basin action. As the flow is subjected to hydrostatic pressure in the stilling basin, a portion of the en
5、trained air is driven into solution before it has the opportunity to rise to the surface and escape into the atmo- sphere. characteristics of the flow within or downstream of the A potential problem situation will exist if the :i 1310 -7 Provided by IHSNot for ResaleNo reproduction or networking per
6、mitted without license from IHS-,-,-3515789 0021839 506 ETL 1110-2-239 15 Sep 78 stilling basin are such that the flow does not have the necessary turbulence to degas or purge itself of the excess dissolved nitrogen. ducive to rapid equilibration with the atmosphere are shallow, turbulent streams. T
7、he reaeration and gas transfer characteristics of deep, slow moving rivers or downstream reservoirs are relatively small. Generally, fish will not suffer from gas bubble disease so loig as they swim in depths below 15 feet. At those depths the external and inter- nal gas pressures on fish are approx
8、imately equal. If the fish swim to the surface, however, the internal gas pressure exceeds the external gas pressure on the fish resulting in gas embolism or gas bubble disease. The tolerance of fish to levels of nitrogen supersaturation depends upon the time of exposure and the age and species of t
9、he fish; however, dissolved nitrogen levels referenced to surface pressure above 110 percent are generally considered to be harmful. (Figure i.) Flow conditions below projects con- b. The phenomenon of nitrogen supersaturation below hydraulic structures is complex and depends upon a number of factor
10、s. Normally the problem of nitrogen supersaturation has been associated with aerated flows plunging into deep stilling basins with slow moving downstream flow conditions. If the hydraulic jump in the stilling basin is a free jump, suffi- cient turbulence should be present to degas the flow so that d
11、issolved nitrogen levels referenced to surface pressure will not exceed 110 percent. If the hydraulic jump is submerged, the flow may plunge to the bottom of the basin. With submerged hydraulic jump flow conditions, the change in momentum of spillway or outlet works releases due to a typical 50 foot
12、 radius toe curve subjects thc flow to a pressure about 1.16 times the hydrostatic pressure on the apron due to the downstream tailwater. The jump will become fully submerged when the tailwater depth is greater than approximately 125 percent of the theoretical dp value. It should be noted that rolle
13、r bucket stilling basins are designed for tail- waters greater than 125 percent of d2. In general, if for a given discharge the tailwater exceeds a depth of 25 feet and if the tailwater depth is greater than 110 percent of theoretical d2 (p artially submerged jump) and if flow conditions downstream
14、of the project are not conducive for degassing the flow, the potential for nitrogen supersaturation exists and should be investigated. c. Nitrogen levels can be determined by measuring total gas content with a gas saturometer and subtracting dissolved 2 Provided by IHSNot for ResaleNo reproduction o
15、r networking permitted without license from IHS-,-,-m 3535789 0023840 228 m ETL 1110-2-239 Sep 78 L 02 ir) O ir) c ld30 W Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-3515789 002l1841 Lb4 ETL 1110-2-239 15 Sep 78 oxygen content measured or by usin
16、g a calibrated gas chromato- graph. Techniques to estimate the percentage of nitrogen supersaturation below a hydraulic structure have been developed by NPD and by the U.S. Bureau of Reclamation (USBR). Inclosure 1 gives a summary of the development and evaluation procedure for the NPD method. Inclo
17、sure 2 gives a summary of the USBR method, The technique developed by NPD was based on projects in the Columbia River Basin. The spillways are all gate-controlled ogee crests and with the exception of The Dalles, they have similar stilling basin characteristics. The NPD method should be used to eval
18、uate the effects of structures similar to those in the Columbia River Basin. The coefficients for this technique are based on these types of structures. The technique developed by the USBR is more general than the NPD technique and utilized data from a wider variety of hydraulic structures. The USBR
19、 technique should be used to evaluate the effects of structures other than the type found in NPD. Both techniques compute downstream nitro- gen concentration values by considering such variables as upstream concentration, headwater and tailwater elevations, head loss, angle of the jet, residence tim
20、e of the bubbles, and pressure conditions in the basin. d. If measurements or estimates indicate that a potential for nitrogen supersaturation problems exists, then detailed model studies of the project may be necessary to develop alleviation measures. Assistance in the studies can be obtained from
21、the Waterways Experiment Station. Also, tech- nical assistance can be obtained from both the Federal Interagency Steering Committee on Reaeration Research and the Committee on Water Quality (reference 3b). Requests for the services of either of these committees should be coordinated through HQDA (DA
22、EN-CWE-H) WASH DC 20314. 6. Action Required. Review all reservoir projects, foowin the procedures outlined in Inclosures 1 and 2, to determine potential for nitrogen supersaturation problems under all operating conditions including interim conditions during con- str uc t ion. a. Existing Projects. R
23、eport results and proposed corrective measures in Annual Division Water Quality Reports (reference 3a). b. Projects under Planning, Design or Construction. Report recul ts sand proposed alleviation measures if required in 4 Provided by IHSNot for ResaleNo reproduction or networking permitted without
24、 license from IHS-,-,-3535789 0021842 OTO = ETL 1110-2-239 15 Sep 78 appropriate portions of Survey-Feasibility Reports, Design Memoranda, Detailed Project Reports, etc . FOR THE CHIEF OF ENGINEERS: - 2 Incl as Chief, Engineering Division Directorate of Civil Works 5 Provided by IHSNot for ResaleNo
25、reproduction or networking permitted without license from IHS-,-,-m 3515789 0021843 T37 m 1315 ETL 1110-2-239 15 Sep 78 DERIVATION OF THE SPILLWAY-STILLING BASIN MODEL* Consider the conceptual representation of the stilling basin shown below. P- r I CONCEPTUAL REPRESENTATION OF SPILLWAY -STILLING BA
26、SIN COMBINATION The water parcel indicated in cross-section by the shaded area moves through the stilling basin, decelerating and increasing in height. It extends laterally the full effective width, t3 of the stilling basin as illustrated in Figure 3 of the main report. We now make the following ass
27、umptions for the water parcel and stilling basin: 1. For that length of spillway that is in operation at a given time, the discharge is uniform along the “Final Report, Water Resources Engineers, Inc., under contract to US Army Corps of Engineers, North Pacific Division, January 1971 qaken from: “A
28、Nitrogen Gas (NIL) Model for the Sower Columbia River, I. 4 - -3 Inclosure 1 - Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-W 3515789 0021844 973 W ETL 1110-2-239 15 Sep 78 crest (this is equivalent to assuming that the properties of the water par
29、cel are constant along any line parallel to the spillway crest). 2. The value Yo is the initial depth of the spill before the jump. It is computed as: - where q = discharge per foot along the crest E = total reservoir head above the stilling basin floor. 3. The only effect of the roller which overli
30、es the main flow is to increase the static pressure within the water parcel by an amount ugi. A given mass of air M is entrained as discrete bubbles into the water parcel at the point x = O and remains uniformly distributed within the water parcel as it passes through the s ti 11 i ng basin. 5. The
31、distribution of the mass of air among the various bubble sizes remains unchanged during the water parcel S journey through the stilling basin. 4. 6. 7. The dissolved nitrogen within the water parcel is uniformly distributed. Rate of nitrogen dissolution - in the water parcel is governed by Fickian d
32、iffusiondtas: 1316 - - (D + Y,) Now let Rewriting equation A-21 with these substi tutions gives K- K- dc+c- AP*o, water at the surface of a pcol will hold SO percent less qas than water at a depth of 34 ft (10.4 in). *arometric pressure is basicallv controlled by the elevation at which the - -I 1327
33、 2 b I I ! Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-3515789 002185b b5 structure is conditions. are not large evaluation of used when ava ETL 1110-2-239 15 Sep 78 ocated, with daily fluctuations that result from atmospheric he effects caused b
34、y daily fluctuations in atmospheric pressure but they may be significant and should be considered in the C,. In this analysis measured barometric pressures were lable. If measured values were not available a standard atmos- phere was assumed and barometric pressures were computed according to elevat
35、ion. The depth of water over the flow in which gas is being dissolved is generally dependent on the depth of water in the stirling basin. the tailwater elevation will have some effect. water depth equal to two-thirds of the basin depth was used to compute satura- tion concentrations. from a spillway
36、 or outlet would penetrate to the floor of the stilling basin. The flow would then be deflected downstream and out of the basin. As the flow moved through the basin it would be diffused and its velocity reduced. This diffusion would be linear and result in a triangular pattern with the average depth
37、 through the diffusion being two-thirds of the total basin depth. Bubbles rising from the flow and incomplete flow penetration would tend to reduce this average depth, but the two-thirds depth was considered representa- tive and therefore used in the analysis, A major point of support for the two-th
38、irds depth assumption is the fact that later applications proved the assumption reasonable. If the flow being studied does not penetrate to the bottom of the pool the maximum depth of flow penetration may be used in this calculation in place of the basin depth. Thus, variations in Throughout this an
39、alysis a It was thought that initially the fairly compact jet Evaluation of C, is achieved by summing the barometric pressure and two-thirds of the basin depth (expressed in m of Hg) and dividing this total pressure by standard atmospheric pressure (760 mm of Hg) to obtain the average absolute press
40、ure on the dissolving bubbles in terms of atmospheres. absolute pressure is then multiplied by the dissolved gas saturation concentra- tion at sea level, for the desired water temperature, to obtain C,. This average The next parameter from equation 1 to be considered is the time, t. It is representa
41、tive of the length of time that the inflowing jet with entrained air is under pressure in the stilling basin and, thus, the length of time that gzs is being dissolved in tne flow. Consideration of time revealed two possible limitations that could control its value. First, it would seem that given su
42、fficient time the entrained air bubbles would rise out of the flow and end the dissolving of gas. In some cases it would seem that an evaluation of this bubble rise tine could be used to represent time. On the other hand, situa- tions miqht occur where the flow with entrained air wou:d pass through
43、the basin and be deflected to a shal!ow depth in a fairly short time. Therefore, the actual length of time required for the flow to pass through the basin could represent t. of these tinir periods might be critical in specific situations. For each fiuw ccndition and structure studied, t was evaluate
44、d for both limitations. The mi11 ler of the two computed Valilfs was considered 2pplicdblo to the particlar situation and was used in the reiiiiiinder of the analysis. During this 2:rrslysis the assmptiori was nade that either Eubble rise tine. - Evalujtion of t based on the bubble rise time, ti, wo
45、uld be, if strictly purswd, a very complex comput.3tion which wcuid probabjy produce questionable results. The vertical dimension of the jet (thickness of jet that the bubble would rise through) is never constant. The time, t, based On bubble rise time, tl, was evaluated by dividing the calculatea v
46、ertical Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-I = 3535789 0023857 523 - EL 1110-2-239 15 Sep 78 thickness of the jet at the tailwater surface by the terminal rise velocity of the bubble. By trial and error, it was determined that an assumed
47、 0.028-inch (0.7-m) diameter huhble with a theoretical terminal velocity of 0.696 ft/s (0.2 m/s) yielded the most consistent results with respect to observed prototype conditions. from two divensionless parameters, it was found that the 0.028-inch-diameter bubble yielded predicted values of K that w
48、ere consistent with the predicted values of K based on the basin retention time. Also, when an analysis was developed that predicted K (equation i) Basin retention time. - Computation of the flow retention time, t2, in the basin is accomplished by dividinq the path length of the flow by the average
49、flow velocity alonq the path. The path lenath is generally controlled by the basin shape. The path length is the distance from the point at which the jet enters the tailwater DOO to the point at which the majority of the flow is directed toward the surface and, therefore, into a lower pressure zone. If a large portion of the flow is deflected upward at a point by baffle piers, for example, this point would be
