AERATION: Principles and Practice ( VOLUME 11 ) - Chapter 7 pot

In order for in-process guarantees to be successful, therefore, it is important thatthe following elements are accurately and clearly fulfilled:• the engineer’s specifications relative t

Testing and Measurement

7.1 INTRODUCTION

Historically, many methods have been used to test and specify aeration equipment.Over time varied methodologies have led to confusion and misrepresentation ofequipment performance Furthermore, equipment suppliers, consultants, and usersoften employ differing nomenclature when they report equipment capabilities.Performance guarantees for oxygen transfer devices have long been the topic oflively discussion by engineers all over the world It is important that the engineer/ownerhave some guarantee from the manufacturer ensuring efficient and effective perfor-mance of the proposed aeration equipment

In the design of an aeration system, the engineer/owner must first select a process

or processes that will meet discharge permit requirements There is substantial latitude

in process selection, but the choice is often made on the basis of engineer/ownerexperience, process and operational reliability, and capital and operating costs Often,several alternatives may be initially selected, and evaluations are made to objectivelyselect the best system It is likely that the oxygen transfer system will play animportant role in this selection process since it usually represents a significant portion

of the total process power cost From that point of view, it would be highly desirablefor the engineer/owner to obtain guarantees on aeration performance under actualprocess conditions

Typically, once a process is selected, the engineer may estimate actual oxygenrequirements (AOR), which depends on wastewater characteristics, mean cell resi-dence time (MCRT) or F/M, and requirements for nitrogen transformations amongother process variables (see design example in Chapter 3) The AOR is subsequentlyused to estimate the field oxygen transfer rate (OTRf) If an in-process oxygentransfer efficiency guarantee is available (usually expressed as mass/time power orpercent efficiency), the engineer can estimate power requirements for each competi-tive system Once the oxygen transfer system is selected, it is necessary to verifythe guarantee by means of compliance testing

For this scenario, the engineer must provide all process information that mayimpact aeration performance in order for the manufacturer to provide an in-processguarantee The manufacturer can then apply their equipment to the prescribed pro-cess using their most favorable equipment, layout patterns, gas flow rates, and otherphysical considerations and based upon experience with their equipment, estimatealpha and beta for the prescribed wastewater and operating conditions The manu-facturer then may estimate a guaranteed oxygen transfer under process conditions

7

In order for in-process guarantees to be successful, therefore, it is important thatthe following elements are accurately and clearly fulfilled:

• the engineer’s specifications relative to the AOR, process, physical layout,operational parameters, and wastewater characteristics

• the manufacturer’s knowledge of the factors that affect their aerationsystem performance including equipment, operation, and wastewater char-acteristics

• the verification method for the in-process guarantee, or compliance ification, which must include the test method to be used, the test protocol,and procedures and test methods for test evaluation

spec-Typically, the first two elements are technically feasible although often understood, but the third, field verification, is still in its infancy and creates the singlebiggest impasse to the successful application of in-process guarantees for oxygentransfer devices As a result, most compliance specifications are written for cleanwater performance Thus, the engineer/owner must make the decisions on aerationsystem performance under process conditions and estimate clean water performancerequirements that will meet the required field conditions

mis-At present, there are standard methods in the U.S., Europe, and other countriesthat have been written for both clean water and in-process performance testing ofaeration equipment These methods are discussed below Other testing methods arealso required for aeration equipment In recent years, there have been reportedinstances where installed fine pore diffuser systems did not meet specified require-ments when tested in full scale Since performance tests were conducted near theend of the construction period, failure to meet performance requirements resulted

in delay of start-up Recent work has produced guidelines for quality assurance offine-pore diffusers at the construction site To better understand and evaluate diffusedair devices, methodologies have also been developed to characterize diffuser ele-ments in new and used condition

7.2 AERATION TANK MASS BALANCE

In deriving the equations for the analysis of the data collected from aeration systems,

a mass balance of oxygen around a completely mixed aeration tank, Figure 7.1 isconstructed

This is more general than Equation 2.26 since it is not limited to a clean waterbatch system with the subscript “f” relating to field conditions It includes the oxygentransport rate as well as the oxygen transfer rate and oxygen uptake rate (OUR), R.

In Equation 7.2, t 0 is the detention time in the aeration tank based on the total influentflow, Qi, to the aeration tank, including the primary flow, Q P, and the return activatedsludge flow, Q R

7.3 CLEAN WATER PERFORMANCE TESTING

Consensus procedures for the evaluation of aeration equipment in clean water are now

in place in the U.S and Europe and have been adopted by a large number of engineeringfirms and manufacturers worldwide The ASCE Standard-Measurement of Oxygen Transfer in Clean Water (ASCE, 1991) was first published in 1985 and was reeditedand adopted in principle in Europe as a European Standard in 2000 (CEN/TC, 2000).The method covers the measurement of the oxygen transfer rate (OTR) as a mass

of oxygen per unit time dissolved in a volume of water by an oxygen transfersystem operating under given gas and power conditions The method is applicable

to laboratory-scale oxygenation devices with small volumes of water as well as thefull-scale system with water volumes found in activated sludge treatment processes.The process is valid for a variety of mixing conditions and process configurations.The ASCE method also includes measurement of gas rates and power

A schematic of the clean water testing technique is given in Figure 7.2 The test

is conducted using clean (tap) water under batch (nonflowing) conditions The steady-state method is based on dissolved oxygen (DO) removal from the test watervolume by the addition of sodium sulfite in the presence of cobalt catalyst Thesesteps are followed by transfer measurements of reoxygenation to near saturationconcentrations Test water volume DO inventory is monitored during the reoxygen-ation period by measuring DO concentrations at several points selected to best

non-FIGURE 7.1 Mass balance on a completely mixed aeration tank.

represent the tank contents These DO concentrations are measured in situ or onsamples pumped from the tank The method specifies minimum sample number,distribution, and range of DO measurements at each sample point.

Equation 2.26 describes these conditions Letting D = – C L and dD = – dC L

provides the following

(7.3)

Analysis of data using the above equation is referred to as the “log deficit”technique and is one of the oldest methods used in the field Due to difficulties ininterpreting results from the above approach when exact values of oxygen saturation

FIGURE 7.2 Clean water test schematic.

L

K at L

0 0

are not known, the ASCE Committee on oxygen transfer has recommended usingEquation 7.3 in terms of concentration.

(7.4)Data obtained at each sample point are then analyzed using a nonlinear regressionanalysis of Equation 7.4 to estimate three parameters including the apparent volu-metric mass-transfer coefficient (K L a), the equilibrium spatial average DO saturationconcentration ( ), and the initial DO concentration (C 0) The nonlinear regression,NLR, computer program developed by the ASCE committee to fit the DO - timeprofile measured at each sampling point during reoxygenation also provides statistics

on the best-fit parameters and the residuals to the model equation For a viable test,

no trend in residuals should occur Typically, the coefficient of variation on K L a will

be < 5 percent and the standard deviation on < 0.1 mg/L

Figure 7.3 shows the use of both “log deficit” and NLR techniques on a typicalset of clean water field data The NLR fit is excellent with very low residuals Notethat if any lingering effects of sulfide addition exist in the system, a lag in the expo-nential increase will occur giving an initial “S” shape to the curve This initial datamust be truncated during data analysis since only the exponential portion of the curve

is analyzed by Equation 7.4 The log deficit results depend on the choice of thesaturation value When the value is too high, the semi-log plot tails upwards as thedeficit approaches zero The reverse is true when is too low Errors in K L a, between

13 and 23 percent, occurred for this data set for the <1 percent change in saturationvalue However, when the log deficit is performed on the measured DO data using onlyvalues up to 80 percent of saturation, as recommended by Boyle et al (1974), then anerror of only 2 to 4 percent in K L a occurs This result is shown in Figure 7.4.From the above results, it is recommended that the NLR technique always beused in final data analysis For rapid on-site estimates, the log deficit techniqueshould provide K L a values within 5 percent of the NLR value when data up to ~ 80percent of saturation is analyzed

For results presentation, the K L a and values for each individual samplinglocation, i, are adjusted to standard conditions as indicated in Chapter 2

The tank SOTR is then calculated by using the estimates of K L a and adjusted

to standard conditions at each sample point

In the above equations, V is the total tank volume and n is the total number ofmeasurement locations SOTR represents the average mass of oxygen transferredper unit time for the total tank at zero DO concentration, water temperature of 20°C,and barometric pressure of 101.3 kPa (1.0 atm), under specified gas flow rate andpower conditions The test is conducted in clean water (alpha presumed to be 1.0)

as specified in the standard Results may also be presented as a standard oxygentransfer efficiency (SOTE), obtained by dividing SOTR by the mass flow of oxygen

in the gas stream (Equation 2.50), or as standard aeration efficiency (SAE), bydividing the SOTR by the power input (Equation 2.45) Although there is no way

to verify method accuracy, it is precise within ± 5 percent (Baillod et al., 1986).The foundation and key elements of the oxygen transfer measurement test arethe definition of terms used during aeration testing, subsequent data analysis, andfinal result reporting A consistent nomenclature has been established with morelogical and understandable terminology than the numerous and varied symbolsused historically

FIGURE 7.3 Clean water data analysis techniques.

The clean water compliance test may be performed in the full-scale system or

in the manufacturer’s shop test facility If performed at the shop test facility, it isimportant to ensure that the test results will properly simulate the field scale system.Scale-up would include geometric similarity (e.g., water depth, length to width, andwidth to depth ratios), gas flow rates per unit and volume, power input per unitvolume, density of diffuser placement, and distance between aeration units, to name

a few considerations Potential interferences resulting from wall effects and anyextraneous piping or other materials in the tank should be minimized Where nec-essary (e.g., long, narrow diffused aeration tanks), testing of tank sections may berequired where there is little circulation of water between adjacent sections Sealedpartitions are used to ensure that oxygen does not interchange between units.Although most projects require a shop or field test to verify diffuser performance,SOTR can also be measured in the laboratory to aid in characterizing diffusers bothnew and used These tests are not intended to be a substitute for shop or field-testing

or for predicting field OTR They are most often used to determine relative ences in performance between diffusers or to assess effectiveness of cleaning meth-ods A typical laboratory setup will include a small column, 61 to 91 cm (2 to 3 ft)

differ-in diameter and 2 to 3 m (7 to 10 ft) high The diffuser to be tested would be placed

in the column and a clean water OTE would be determined over a range of airflows.The clean water procedure would usually be determined by the ASCE Clean Water

FIGURE 7.4 Effect of data truncation on log deficit analysis.

Standard (1991) which is a non-steady-state method A steady-state method mayalso be used and is described in detail in the Design Manual, Fine Pore Aeration Systems (1989).

7.4 IN-PROCESS OXYGEN TRANSFER TESTING

The testing of aeration equipment under field conditions has been the subject ofconsiderable research over the last 30 years (EPA, 1983; Kayser, 1969; Mueller andBoyle, 1988) In 1996, the ASCE published the Standard Guidelines for In-Process Oxygen Transfer Testing (ASCE, 1996) and shortly thereafter the European standard(CEN/EN, 2000) was developed which drew on much of the ASCE standard guide-line The guidelines have been developed based on over 30 years of side-by-sidetesting of several methods to verify reproducibility of the methods The methodsselected have proven to be the most reliable under rigorous field conditions Thetechnology continue to be dynamic, however, and modifications and/or new proce-dures will likely occur in the future

The intent of the methods that have been developed for field conditions was toprovide useful information on field performance that can be used for future design(variability in oxygen transfer, alpha values, spatial and temporal variations inoxygen demand, etc.) It provides the owner with data that can be used for operationand maintenance of aeration equipment The procedures also offer manufacturersthe opportunity to develop and improve the performance of their equipment In someinstances, engineers may use these methods for compliance guarantees It should beemphasized, however, that performance under process conditions is affected by alarge number of process variables and wastewater characteristics that are not easilycontrolled for a given test condition Thus, compliance testing under field conditionscan be highly subjective and uncertain

The methods described in the ASCE In-Process Guidelines (ASCE, 1996) include

a non-steady-state method, off-gas technique, and the inert gas tracer method Thesemethods have been well developed and provide satisfactory precision for a widerange of aeration processes Additional provisional methods include a steady-stateprocedure and mass balance methods In general, testing methods can be categorizedaccording to whether DO is steady or nonsteady If the influent to the test basin isdiverted, these tests are referred to as batch tests and do not reflect the variability

of wastewater characteristics or the actual operating conditions that might beexpected If wastewater flow to the test basin is continuous, the test more nearlyrepresents actual operating conditions, but steady state, with respect to influentcharacter (AOR, alpha, etc.), is difficult to achieve

The basis of the steady-state and non-steady-state techniques is Equation 7.2.For the steady-state technique, , and the DO is constant in the tank, C L = C R,for a constant uptake rate, R

(7.6)

dC dt

In practice, both R and C Rvalues are measured at a number of equal volumesampling locations, i, in the aeration tank This technique requires using the averageoxygen uptake rate and DO concentration in the tank to determine the tank oxygentransfer coefficient Due to back dispersion and mixing in the tank, individual K L a f

values at each location are meaningless Representative in situ OUR values aredifficult to obtain in practice when a sample is removed from the aeration tank due

to substrate or oxygen limitation (Mueller and Stensel, 1990)

n

i n

In practice, both K L a f and C R values are again measured at a number of equal

volume sampling locations, i The average tank values are again utilized to determine

the overall tank K L a f Similar to the steady-state technique, due to back dispersion

and mixing in the tank, individual K L a f values at each location are meaningless

(7.11)

Non-steady-state methods estimate an average K L a for a test section by

measur-ing the change in DO concentration with time after a perturbation from steady-state

conditions This perturbation may be imposed on the system by changing input

aeration power (up or down) or by the addition of hydrogen peroxide or high purity

oxygen The procedure requires constant OUR, DO, flow rate, and K L a over the test

period, and it requires the accurate measurement of the test section DO and flow

rate It avoids the need to measure OUR and C *

∞.Hildreth and Mueller (1986) have shown that the above non-steady-state

approach can be used in advective-dispersive systems which are not completely

mixed The K value in Equation 7.9 is defined by The additional

term, K e, is a function of longitudinal dispersion and velocity of flow in the tank

For Ridgewood, NJ, fine pore diffusers in tanks 35.4 m (116 ft) long and 7.3 m (24

ft) wide, it varied from 0.1 to 0.3/h In long, 91.4 m (300 ft), narrow, 9.1 m (30 ft),

tanks at Whittier Narrows, CA, Mueller (1985) has shown that the batch equation

where K = K L a f could be applied near the end of the tank For accurate results, the

minimum distance, xmin, required downstream from a boundary in a section where

OUR and K L a f are constant was xmin = 2.5 U/K L a f where U is the forward velocity

Non-steady-state testing is the most suitable method available for mechanical

aeration systems However, it does not provide an estimate of the accuracy of the

results During a sabbatical leave in 1980, the senior author conceived of a technique

to get an estimate of how good the results were by conducting the tests twice Each

test was conducted at a different power level as shown in Figure 7.5 (Mueller, 1982;

Mueller et al., 1982; Mueller and Rysinger, 1981) Changing power level can be

used by itself or in conjunction with hydrogen peroxide addition to get a greater

spread in the non-steady-state curves Good results can be obtained with bothtechniques (Mueller and Boyle, 1988).

This provides two different K L a f and two different steady-state C R values withone oxygen saturation value The following equations are used with these values to

calculate the in situ OUR and saturation concentration.

during the tests, a difficult situation when K L a f values are low requiring a long time

for the tests

FIGURE 7.5 Dual non-steady-state analysis techniques, a) changing power levels, b) H2O2addition.