Lettinga Associates Foundation

for environmental protection

and resource conservation

Manual for the design

of UASB reactors

Date: June 16, 2007

Lettinga Associates Foundation Project number: 07 - 359

P.O. Box 500 Author Lucas Seghezzo
NL-6700 AM Wageningen Drawings
The Netherlands Revised by
Phone: +31 317 482023
Fax: +31 317 482108
http://www.leaf-water.org
Manual for the design of UASB reactors

Manual for the design of UASB reactors

Table of contents

1. MANUAL FOR THE DESIGN OF UASB REACTORS...............................................................8

2. INTRODUCTION............................................................................................................................9

2.1. SOME HISTORY......................................................................................................................................9

2.2. ROLE OF THE UASB IN A TREATMENT PLANT.......................................................................................10
2.3. ADVANTAGES AND DISADVANTAGES ......................................................................................................11
2.4. TECHNOLOGY SELECTION.....................................................................................................................12
2.5. DISCLAIMER.......................................................................................................................................13

3. FIRST STEPS.................................................................................................................................14

3.1. CLEARLY IDENTIFY THE PROBLEMS.......................................................................................................14

3.2. FORMULATE OBJECTIVES AND ASSUMPTIONS ..........................................................................................14
3.3. SET BOUNDARIES ................................................................................................................................14
3.4. DETECT ALL STAKEHOLDERS AND LOCAL ACTORS ..................................................................................15
3.5. COLLECT INFORMATION.......................................................................................................................15
3.6. IDENTIFY POSSIBLE TECHNICAL ALTERNATIVES.......................................................................................15

4. STARTING POINTS.....................................................................................................................17

4.1. WASTEWATER CHARACTERISTICS..........................................................................................................17

4.1.1. ORGANIC MATTER..............................................................................................................................18
4.1.2. TEMPERATURE..................................................................................................................................18
4.1.3. PH..................................................................................................................................................19
4.1.4. FLOW RATE......................................................................................................................................19
4.2. CONSIDERATIONS AND ASSUMPTIONS......................................................................................................19

5. THE CONCEPTUAL DESIGN ....................................................................................................21

5.1. RELEVANT WASTEWATER CHARACTERISTICS...........................................................................................21

5.1.1. PRIMARY DESIGN PARAMETERS.............................................................................................................22
5.1.2. ENABLING ENVIRONMENTAL PARAMETERS..............................................................................................22
5.1.3. PARAMETERS THAT AFFECT DESIGN AND OPERATION................................................................................23
5.2. BASIC ASSUMPTIONS............................................................................................................................24
5.2.1. HRT AND OLR...............................................................................................................................24
5.2.2. REACTOR HEIGHT..............................................................................................................................25
5.2.3. NUMBER OF UASB UNITS.................................................................................................................25
5.3. PRELIMINARY CALCULATIONS...............................................................................................................26
5.3.1. CALCULATE AVERAGE AND PEAK FLOW RATES........................................................................................26
5.3.2. CALCULATE THE VOLUME...................................................................................................................27

2
Lettinga Associates Foundation

5.4. DESIGN PROCESS.................................................................................................................................28

5.4.1. THE REACTOR...................................................................................................................................29
5.4.1.1. Area of the reactor...................................................................................................................30
5.4.1.2. Upflow velocity.......................................................................................................................30
5.4.1.3. Reactor shape...........................................................................................................................30
5.4.1.4. Reactor sides (cubic reactors)..................................................................................................30
5.4.1.5. Reactor diameter (cylindrical reactors)....................................................................................31
5.4.1.6. Check HRT..............................................................................................................................31
5.4.1.7. Calculate OLR.........................................................................................................................31
5.4.1.8. Round up the dimensions.........................................................................................................31
5.4.1.9. Recalculate parameters............................................................................................................32
5.4.2. THE GLS SEPARATOR........................................................................................................................33
5.4.2.1. Deflector..................................................................................................................................34
5.4.2.2. Biogas......................................................................................................................................34
5.4.2.3. Total area of the GLS separator...............................................................................................34
5.4.2.4. Number of GLS separators......................................................................................................35
5.4.2.5. Height of the GLS separator....................................................................................................36
5.4.2.6. Angle of the walls....................................................................................................................37
5.4.2.7. Gas liberation area...................................................................................................................39
5.4.3. THE INLET DISTRIBUTION SYSTEM.........................................................................................................42
5.4.4. EFFLUENT........................................................................................................................................43
5.4.5. EXCESS SLUDGE................................................................................................................................43
5.4.5.1. Sludge production....................................................................................................................43
5.4.5.2. Sludge discharge pipes.............................................................................................................43
5.4.5.3. Sludge drying beds..................................................................................................................44
5.4.6. SUMMARY DESIGN PARAMETERS...........................................................................................................45
5.5. START UP OF THE REACTOR..................................................................................................................46

PRELIMINARY TREATMENT......................................................................................................47

6. POST-TREATMENT.....................................................................................................................48

6.1. USE OF THE TREATED EFFLUENT...........................................................................................................48

6.1.1. RECHARGE OF WATER BODIES..............................................................................................................48
6.1.2. AGRICULTURAL REUSE........................................................................................................................48
6.2. EXAMPLE: POLISHING PONDS...............................................................................................................49

7. COSTS.............................................................................................................................................50

7.1. INVESTMENT COSTS..............................................................................................................................50

7.2. OPERATION AND MAINTENANCE COSTS...................................................................................................50
7.3. OTHER COSTS.....................................................................................................................................51
7.4. SUMMARY OF COSTS............................................................................................................................51

8. REFERENCES...............................................................................................................................52

3
Manual for the design of UASB reactors

List of Tables

TABLE 1. COMMONLY CITED ADVANTAGES AND DISADVANTAGES OF ANAEROBIC

WASTEWATER TREATMENT. SEE COMMENTS IN THE TEXT..........................................12

TABLE 2. BASIC WASTEWATER INFORMATION NEEDED TO DESIGN THE UASB

REACTOR. ........................................................................................................................................18

TABLE 3. LIST OF MOST IMPORTANT DESIGN PARAMETERS FOR THE UASB

REACTOR. Q = FLOW RATE; C = CONCENTRATION; COD = CHEMICAL OXYGEN
DEMAND; OLR = ORGANIC LOADING RATE; HRT = HYDRAULIC RETENTION TIME;
VUP = UPFLOW VELOCITY; GLS = GAS-LIQUID-SOLID SEPARATOR.............................45

TABLE 4. SUMMARY OF ESTIMATED COSTS..........................................................................51

List of Figures

FIGURE 1. NATURAL SEQUENCE OF DEGRADATION OF ORGANIC COMPOUNDS IN

POLLUTED ENVIRONMENTS........................................................................................................9

FIGURE 2. SCHEMATIC DIAGRAM OF A UASB REACTOR FOR SEWAGE

TREATMENT. FROM WWW.UASB.ORG ...................................................................................10

FIGURE 3. EXAMPLE OF WASTEWATER TREATMENT SYSTEM INCLUDING A UASB

REACTOR AND AN AEROBIC POST-TREATMENT STEP, FROM THE MANY
ALTERNATIVES POSSIBLE (NOT TO SCALE)..........................................................................11

FIGURE 4. COLOR AND FONT CODES USED IN THE XL FILE............................................21

FIGURE 5. PRIMARY DESIGN PARAMETERS. COD = CHEMICAL OXYGEN DEMAND;

Q = FLOW RATE..............................................................................................................................22

FIGURE 6. ENABLING ENVIRONMENTAL PARAMETERS. BOD = BIOCHEMICAL

OXYGEN DEMAND; TSS = TOTAL SUSPENDED SOLIDS; COD/BOD GIVES AN IDEA OF
THE BIODEGRADABILITY...........................................................................................................23

FIGURE 7. PARAMETERS THAT MIGHT AFFECT DESIGN AND/OR OPERATION.........23

FIGURE 8. BASIC ASSUMPTIONS MADE AT THE BEGINNING OF THE DESIGN

PROCESS...........................................................................................................................................26

4
Lettinga Associates Foundation

FIGURE 9. CALCULATION OF AVERAGE AND PEAK FLOW RATE...................................26

FIGURE 10. ESTIMATION OF THE SEWAGE FLOW RATE BASED ON DEMOGRAPHIC

AND OTHER DATA..........................................................................................................................27

FIGURE 11. PRELIMINARY CALCULATIONS TO DECIDE WHETHER THE REACTOR

IS ORGANICALLY OR HYDRAULICALLY LIMITED. THE VALUE IN THE CIRCLE WAS
SELECTED IN THIS CASE.............................................................................................................28

FIGURE 12. MOST IMPORTANT REACTOR DIMENSIONS NEEDED DURING THE

DESIGN PROCESS...........................................................................................................................29

FIGURE 13. FIRST CALCULATION OF THE REACTOR DIMENSIONS...............................29

FIGURE 14. ROUND UP DIMENSIONS TO REALISTIC, EASY TO BUILD, VALUES, AND

COPY THESE VALUES IN THE COLUMN “FINAL”.................................................................32

FIGURE 15. RECALCULATION OF REACTOR DIMENSIONS BASED ON USER-

SELECTED SIDES (OR DIAMETER IN CYLINDRICAL REACTORS). START
RECALCULATING THE AREA WITH THE NEW DESIGN DIMENSIONS (IN THE XL
FILE, SEE LONG ARROWS)..........................................................................................................32

FIGURE 16. EXAMPLE OF A GLS SEPARATOR SHOWING THE INFLUENT

DISTRIBUTION BOX ON TOP, THE DEFLECTOR, AND THE EFFLUENT GUTTERS ON
BOTH SIDES (BUCARAMANGA, COLOMBIA)..........................................................................33

FIGURE 17. SELECT A DESIGN VUP AT THE APERTURES AND FIRST CALCULATION
OF THE GLS DIMENSIONS............................................................................................................35

FIGURE 18. CALCULATION OF THE SIDES OF THE GLS SEPARATORS..........................36

FIGURE 19. FINAL VALUES FOR THE GLS SEPARATORS...................................................36

FIGURE 20. CALCULATION OF THE HEIGHT OF THE GLS SEPARATOR BASED ON

THE “HEIGHT RELATIONSHIP”.................................................................................................37

FIGURE 21. BASIC DIMENSIONS NEEDED TO CALCULATE THE ANGLE OF THE GLS
SEPARATOR.....................................................................................................................................38

FIGURE 22. VALUES CALCULATED FOR THE CASE STUDY...............................................38

FIGURE 23. SOMETIMES IT IS BETTER TO FIX THE ANGLE AND CALCULATE THE
GLS HEIGHT.....................................................................................................................................39

FIGURE 24. SOMETIMES IT IS BETTER TO FIX THE ANGLE AND CALCULATE THE
GLS HEIGHT.....................................................................................................................................39
5
Manual for the design of UASB reactors

FIGURE 25. CALCULATION GAS LIBERATION AREA (A)....................................................40

FIGURE 26. CALCULATION GAS LIBERATION AREA (B)....................................................40

FIGURE 27. CALCULATION GAS LIBERATION AREA (C)....................................................41

FIGURE 28. ESTIMATION OF THE POTENTIAL BIOGAS PRODUCTION UNDER FIELD

CONDITIONS....................................................................................................................................41

FIGURE 29. CALCULATION OF THE NUMBER OF INFLUENT DISTRIBUTION SYSTEM.

..............................................................................................................................................................42

FIGURE 30. CIRCULAR BOX FOR HOMOGENEOUS INFLUENT DISTRIBUTION WITH

16 INLET POINTS. NOT TO SCALE. ADAPTED FROM VAN HAANDEL AND LETTINGA
(1984)...................................................................................................................................................42

FIGURE 31. CALCULATION OF THE EXCESS SLUDGE PRODUCTION.............................43

FIGURE 32. NUMBER AND POSITION OF THE SLUDGE DISCHARGE AND SAMPLING
PIPES. CALCULATION DETAILS CAN BE CHECKED IN THE XL FILE.............................44

FIGURE 33. ESTIMATED DIMENSIONS OF THE SLUDGE DRYING BEDS.........................44

FIGURE 34. BASIC SCHEME OF THE UASB REACTOR (NOT TO SCALE) SHOWING
THE MAIN DIMENSIONS (IF NOT INDICATED OTHERWISE, DIMENSIONS IN
METERS). SEE ALSO ATTACHED SPREADSHEET. NOTE: ONLY ONE GLS MODULE IS
DEPICTED.........................................................................................................................................46

FIGURE 35. SCREENS (LEFT) AND SAND TRAP (RIGHT) REQUIRED FOR
PRELIMINARY TREATMENT. NOTICE: FOR THE REACTOR OF OUR CASE STUDY,
THE SIZE NEEDED IS MUCH SMALLER (BUCARAMANGA, COLOMBIA)........................47

Glossary

LeAF = Lettinga Associates Foundation

UASB = Upflow Anaerobic Sludge Blanket
BOD = Biochemical Oxygen Demand
COD = Chemical Oxygen Demand
TSS = Total Suspended Solids
HRT = Hydraulic Retention Time
SRT = Sludge Retention Time
Vup = upflow velocity

6
Lettinga Associates Foundation

GLS = Gas-Liquid-Solid

7
Manual for the design of UASB reactors

1. MANUAL FOR THE DESIGN OF UASB

REACTORS

The Lettinga Associates Foundation (LeAF) provides in this document a manual for
the conceptual design of Upflow Anaerobic Sludge Blanket (UASB) reactors.

A conceptual design is not a detailed engineering design but it provides all the basic
dimensions of a UASB reactor. These dimensions have then to be transferred to a
construction blueprint, following the guidelines and regulations of the country or
region where the reactor will be built.

To facilitate the calculations, and to make it more practical, this manual will be based
on a specific case study: the design of a UASB reactor to treat domestic wastewater
in a Mediterranean country. However, the design principles can, in principle, be
applied to other situations, as long as some basic conditions are met.

Only a brief introduction about the UASB concept will be given. Those interested to
know more about this reactor and its history, and about anaerobic digestion and
anaerobic wastewater treatment processes, are referred to the documents provided
in Annex I, and to the extensive literature list cited therein.

In order to take full advantage of this manual, some previous knowledge on chemical
and environmental engineering is required. However, this manual can also be used
by other professionals, policy makers, or people interested in more sustainable
wastewater treatment technologies.

8
Lettinga Associates Foundation

2. INTRODUCTION
A more sustainable treatment of organic wastewater should rely on the use of natural
degradation processes, as it usually proceeds under natural conditions (Figure 1).
Anaerobic processes like anaerobic digestion and sulfate reduction can be the main
biological steps in such a more sustainable wastewater treatment system because
they remove and mineralize organic pollutants with formation of very useful by-
products like methane gas and soil conditioners.

Other methods like high-rate micro-aerobic processes or more conventional aerobic

treatment systems (like waste stabilization ponds or trickling filters) can be used as
post-treatment steps to reach specific effluent quality, according to the intended final
use of the wastewater.

When environmental conditions are suitable for anaerobic treatment, the application
of conventional aerobic treatment system makes little economic, environmental, and
social sense although it is general practice in “modern” centralized sanitation
systems.

CH4 CO2 SH2 CO2

CO2

Anaerobic
Solution Aerobic Solution with
Organic digestion, Micro-aerobic
with processes, mineralized
wastes sulfate reduction, conversions
nutrients nitrification compounds
denitrification

Soil conditioner S

Figure 1. Natural sequence of degradation of organic compounds in polluted environments.

2.1. Some history

Anaerobic processes have been used for the treatment of concentrated domestic and
industrial wastewater for well over a century. The simplest, oldest, and most widely
used process is the famous “septic tank” (McCarty, 1981).

However, in spite of their early introduction, the interest on anaerobic systems as the
main biological step (also called “secondary treatment”) in wastewater treatment
plants was scarce until the development of the UASB reactor in the seventies by
Gatze Lettinga and his co-workers at Wageningen University, The Netherlands
(Figure 2).

9
Manual for the design of UASB reactors

biogas

weir effluent
3 phase settler gas
separator cap
baffles

gas bubble
sludge granule

sludge bed
influent

Figure 2. Schematic diagram of a UASB reactor for sewage treatment. From www.uasb.org

2.2. Role of the UASB in a treatment plant

Several configurations can be imagined for a wastewater treatment plant including a
UASB reactor.

In any case, there must be a sand trap, screens for coarse material, and drying beds
for the sludge.

Although the UASB reactor is only a pre-treatment step, it can partially or completely
replace many of the process units required in conventional aerobic systems
(activated sludge, trickling filters), like primary settlers, sludge digesters, the
secondary aerobic step itself, and secondary settlers.

As mentioned before, the effluent from UASB reactors usually needs further
treatment that can be accomplished in aerobic systems like polishing ponds or in
micro-aerobic processes which present an enormous potential for the removal of
remaining Chemical/Biochemical Oxygen Demand (COD/BOD), colloidal matter,
pathogens, ammonia-nitrate conversion, and smell nuisance prevention (Figure 3).

The final result of including an anaerobic step is an overall treatment system that (a)
has a smaller footprint, (b) is easier to operate, (c) requires fewer inputs, and (d)
produces energy and by-products that can be reused.

10
Lettinga Associates Foundation

Biogas

Screens Grit chamber Flow meter Post treatment

Influent Effluent
UASB
reactors - Polishing ponds
- Trickling filters
Coarse Grit, sand - Others
materials

water
sludge
biogas Stabilized
sludge
Sludge drying bed

Figure 3. Example of wastewater treatment system including a UASB reactor and an aerobic post-
treatment step, from the many alternatives possible (not to scale).

2.3. Advantages and disadvantages

Commonly cited advantages and disadvantages of anaerobic sewage treatment, with
special emphasis on UASB reactors, are summarized in Table 1. However, the
following comments can be made on the alleged disadvantages:

 Substantial improvements have been made in tackling most of the alleged

disadvantages of anaerobic treatment, with the result that only a few of the
previously presumed drawbacks have remained, while all its principle benefits
over conventional aerobic methods still apply.
 The fact that anaerobic systems remove only biodegradable organic matter and
not the nutrients (ammonia salts, phosphates, and sulfides) is an intrinsic
characteristic of the process but, rather than a drawback, it is more and more
being considered a benefit. Appropriate, simple, cheap, and very efficient post-
treatment methods are available to remove and/or to recover these nutrients.
Moreover, if coupled to agricultural reuse, nutrients can be beneficially used for
ferti-irrigation.
 Current sanitation concepts in the public sector overlook the need to preserve
nutrients for further reuse and promote the dilution of highly concentrated human
wastes like excreta and urine as the only alternative for wastewater treatment.
Anaerobic systems can cope well with concentrated waste streams like those
present in decentralized sanitations schemes.
 The potential odor nuisance commonly ascribed to anaerobic systems can also
be completely prevented using simple methods like biological filters, contrary to
what we see in conventional systems.

11
Manual for the design of UASB reactors

Table 1. Commonly cited advantages and disadvantages of anaerobic wastewater treatment. See
comments in the text.
Advantages
High efficiency. Good removal efficiency can be achieved in the system, even at high loading
rates and low temperatures.
Simplicity. The construction and operation of these reactors is relatively simple.
Flexibility. Anaerobic treatment can easily be applied on either a very large or a very small
scale.
Low space requirements. As high loading rates can be treated, the reactor footprint is small.
Low energy consumption. As far as all plant operations can be done by gravity, energy
consumption is almost negligible. Moreover, energy is produced during the process in the form
of methane.
Low sludge production. The sludge production is low, when compared to aerobic methods, due
to the slow growth rates of anaerobic bacteria. The sludge is well stabilized for final disposal
and has good dewatering characteristics.
Low nutrients and chemicals requirement. Especially in the case of sewage, an adequate and
stable pH can be maintained without the addition of chemicals. Macronutrients (nitrogen and
phosphorus) and micronutrients are also available in sewage, while toxic compounds are
absent.
Preservation of nutrients. Nutrients are not removed from the wastewater and can be recovered
or reused in activities like agriculture or aquaculture.
Suitable for campaign industries. Adapted anaerobic sludge can be preserved without feeding
for a long time and can be used as inoculum for the start-up of new reactors.

Disadvantages
Low pathogen and nutrient removal. As any biological secondary treatment system pathogens are
only partially removed, i.e. about 1 log unit, except helminth eggs, which in many cases are
effectively captured in the sludge bed. Nutrients are not removed and N concentrations may
slightly increase during treatment owing to the mineralization of organic matter.
Longer start-up. Due to the low growth rate of methanogenic organisms, the start-up takes longer
than in aerobic processes, when no good inoculum is available. For sewage treatment in tropical
and subtropical climates, however, this time is about 3-6 months.
Possible bad odors and emissions. Hydrogen sulfide is produced during the anaerobic process,
especially when there are high concentrations of sulfate in the influent (which is not the case in
domestic sewage). A proper handling of the biogas is required to avoid bad smell. Methane
dissolved in the effluent needs to be recovered as well to minimize methane emissions to the
atmosphere.
Necessity of post-treatment. Post-treatment of the anaerobic effluent is generally required to reach
the discharge standards for organic matter, nutrients and pathogens, depending on the final use
that will be given to the treated effluent (irrigation, infiltration, discharge on superficial waters, etc.).

2.4. Technology selection

The selection of the “right” technological option for a specific purpose, in a given
location, at a certain time, should always be assessed on the basis of its contribution
to local and global, short-term and long-term sustainability, and not only on the basis
of narrow technical considerations.

Criteria like use of energy and land, emission of harmful gases, potential for reuse of
by-products, robustness in operation and maintenance, acceptability by end users,

12
Lettinga Associates Foundation

adaptability to local culture, self-sufficiency in terms of know-how, “know-why”, and

inputs, among others, have to be included in the evaluation.

Different technologies may be used, provided that the basic assumptions behind
these techniques, and their advantages and disadvantages are clear for all involved.

Caution in the selection of the methods, representative participation of all

stakeholders, transparency in the process of assessment, and responsibility and
accountability in the final decision-making are a guarantee of success.

2.5. Disclaimer
The information contained in this report is the proprietary information of Lettinga
Associates Foundation. Lettinga Associates Foundation accepts no liability for
damages that may arise from the use, illegal or otherwise, of the information in this
report.

This Manual has been developed for internal training purposes and it is intended
solely as a guideline. It is not meant to replace the advice of an expert on the subject.
Extrapolations to different types of wastewater and/or environmental conditions
should be done with care, and under the supervision of a specialist.

13
Manual for the design of UASB reactors

3. FIRST STEPS
Some issues must be sorted out before the design process actually starts. We have
to bear in mind that wastewater treatment can never be an objective in itself, but a
tool to solve a given problem, or better, to transform a problem into an opportunity.

It goes without saying that reducing or even eliminating the production of wastes and
wastewater should be the first option to be considered in a sustainable water and
wastewater management system.

In any case, before designing a treatment plant, assuming that it is the only option
left, it is important to go through a preliminary process.

3.1. Clearly identify the problems

It might seem redundant, but this step is generally overlooked. The impact of
untreated wastewater discharges depends heavily on the local setting. Heavily
populated areas or pristine environments will be strongly affected by such discharges
and, probably, the intensity of the treatment will have to be higher in those areas.

Local laws and regulations have to be complied with, and they provide us with an
indication of the minimum level of treatment that is needed in the region.

3.2. Formulate objectives and assumptions

Once the “problem” is identified, the objectives of our intervention have to be
formulated. The need or the desire to (re)use the treated effluent in agriculture,
aquaculture, groundwater recharge, landscape irrigation, etc., has to be included as
one of the objectives of the system, because it will certainly affect its design. All the
assumptions made have to be made explicit at this early stage.

3.3. Set boundaries

It is impossible to solve all the problems at the same time. Therefore, the boundaries
of our intervention must be limited and realistic. A step by step approach is usually
the best one in environmental issues, and it is wiser and easier to later duplicate a
small successful intervention than to regret a big failure.

14
Lettinga Associates Foundation

The desired degree of centralization has to be discussed here. The need to collect
sewage from an entire city and convey it to a single point to be treated has to be
challenged on economic, technical, and environmental grounds.

3.4. Detect all stakeholders and local actors

All issues have to be openly discussed by all interested actors, so a careful
identification of all (powerful and otherwise) stakeholders is very important to avoid
problems once the decision is made.

As a quick guide, we can say that stakeholders have to be selected in terms of the
following three key attributes or criteria:

(a) Power (people that have to be included) (politicians, companies, donors),

(b) Legitimacy (people that can not be left out) (consumers, end users, local NGO’s,
association of engineers), and
(c) Urgency (people that should be taken into account) (farmers who depend on
wastewater for irrigation, people leaving downstream, neighbors of the future
treatment plant).

If the right people are adequately and openly involved in the decision making,
something that can look like a bit cumbersome at the beginning, the overall process
will be faster and smoother in the end.

3.5. Collect information

It is advisable to gather all possible information about the city or town for which the
plant will be built. The information needed includes demographic, climatic,
geographic, and topographic data, water consumption, sewage production, age and
types of pipes installed, habits, cultural aspects, solid waste disposal system,
regulations and local environmental standards, even some history, previous attempts
to solve the problem, and so on and so forth. The more information we can collect,
the fewer mistakes we are likely to make in the design.

It is also important to have a local partner who can easily interact with all relevant
institutions on location.

The collection of information has to be complemented with at least one site visit.

3.6. Identify possible technical alternatives

15
Manual for the design of UASB reactors

There are probably two or more potential solutions for the problems detected. It is
advisable to take all of them into consideration, weighing carefully their advantages
and disadvantages in the local context. Expert advice will help make a sound
decision, but the final decision must always be taken by local stakeholders, as long
as there are many aspects to consider (social, cultural, institutional, legal,
environmental, economic), and not only technical ones.

16
Lettinga Associates Foundation

4. STARTING POINTS
For the purposes of this manual, it will be assumed that a decision has been taken to
build a sewage treatment plant in a given (small) city, and that the UASB technology
has been selected as the best option under local circumstances.

Therefore, from now on, a very practical, technical approach will be followed.

The design of the UASB reactor is based on two things:

a) Wastewater characteristics, generally provided by the client, and

b) Considerations and assumptions, made by the designer.

4.1. Wastewater characteristics

Sewage temperature, composition, and flow rate pattern are crucial for an accurate
design. Therefore, all necessary efforts have to be taken to make sure that these
values are correct.

Average values are not always enough for an accurate and safe design. The daily,
weekly, monthly, seasonal, and even yearly variations are sometimes decisive for the
selection of the right design parameters. A sound sampling method should be applied
to collect accurate information on each one of these parameters.

Population growth, changes in consumption patterns, availability of drinking water,

and many other factors are also important for design purposes. For domestic
wastewater, these data are usually difficult to gather and one has to rely on
extrapolations from similar cities, or assumptions based on expert opinions.

For industrial wastewater, more reliable information is generally collected and it is

easier to adjust the design to the specific conditions of the industry.

When no data whatsoever are available, the designer has to resort to general
information like drinking water consumption, population in the area, satellite images,
etc.

The basic wastewater characteristics needed to design the reactor for this case study
are presented in Table 2.

In the coming paragraphs, some hints are provided as to how these specific values
must be interpreted for design purposes.

17
Manual for the design of UASB reactors

Table 2. Basic wastewater information needed to design the UASB reactor.

Parameter Average value
Temperature (°C) 20.0
pH 7.0
COD (Chemical Oxygen Demand) (mg/L) 900.0
BOD5 (Biochemical Oxygen Demand) (mg/L) 600.0
TSS (Total Suspended Solids) (mg/L) 300.0
Flow rate (m3/d) 1500

4.1.1. Organic matter

The organic matter present in the wastewater shows a relatively high
biodegradability, with a BOD that amounts to almost 70% of the COD. Biological
treatment processes are, in principle, feasible under these circumstances1.

It is good to notice here that the BOD is an indication of the aerobic biodegradability
(it is a test performed with aerobic bacteria). However, it can, in most cases, and
sewage is certainly one of them, be used as an indication of the anaerobic
biodegradability as well. For other, more complex wastewaters, this assumption may
not be valid, and a specific Anaerobic Biodegradability Test must be performed
before deciding for the feasibility of anaerobic treatment.

The amount of Total Suspended Solids (TSS) will not detrimentally affect the
treatment efficiency2, but the production of excess sludge will be higher than would
be the case for a completely soluble organic wastewater.

4.1.2. Temperature
The yearly average influent temperature is above 20°C, which is, in principle,
adequate for anaerobic digestion. However, it is important to note that sewage
temperature during winter time is crucial, and the duration of the cold period has to
be taken into account for design purposes.

Based on consultations with local engineers, it can be assumed that sewage

temperature fluctuates between 15 and 19°C during the coldest period of wintertime,
which is relatively short (a few weeks). It was concluded that this period of low
temperature can be well accommodated by the sludge bed of the UASB reactor
without a major deterioration of the treatment efficiency. The poorly stabilized sludge
accumulating during this short, colder period will be immediately stabilized during
summer time resulting in a higher biogas production. There is an extensive body of
literature on the effects of temperature on the treatment of different types of
wastewaters.

1
EXPERT OPINION: It is assumed that, if biodegradability is 50% or higher, biological treatment
methods are attractive.
2
EXPERT OPINION: Although high in suspended solids, it is well known that, within certain range,
sewage can be treated anaerobically.

18
Lettinga Associates Foundation

4.1.3. pH
As it is usually the case for sewage, the pH is close to neutrality (around 7.0), and
enough bicarbonate buffer capacity will be present in the reactor. For some industrial
wastewaters, pH correction is needed.

4.1.4. Flow rate

The flow rate that will be used for design is 1500.0 m3/d, which represents, in this
particular case, the sewage generated by approximately 15000 inhabitants
(assuming a water consumption of around 100 L/inh.d). Judging by the water
consumption, it can be inferred that the city is located in a very dry area, and that the
average income is probably low, typical of some developing countries.

Water consumption is a highly variable parameter, and it depends on many factors

like climate, income, water availability, culture, etc. Standard values used in many
countries go from 150 to 250 L/inh.d.

Flow fluctuations and the extent of flow peaks have not been determined, but they
will most likely reproduce the pattern generally found for domestic sewage, with a
daily peak after the main meal and secondary peaks early in the morning and after
the working day is over and people return to their homes. A well-designed and
properly-operated UASB reactor can easily cope with such fluctuations.

For industrial wastewater, the peaks tend to be much more pronounced, especially
when production is not constant throughout the day, or throughout the week. In those
cases, the flow needs to be equalized in an equalization tank. Guidelines for the
design of equalization tanks will not be given here.

4.2. Considerations and assumptions

No matter how detailed an accurate the data available might be, the designer always
has to make a number of assumptions on which the conceptual design will be based.

If the assumptions are proven not true by subsequent measurements or new insights,
then the conceptual design has to be reviewed. The assumptions are sometimes
made explicit for legal reasons, or to minimize potential liability claims.

The assumptions made are as important as the data on wastewater characteristics,

in terms of their influence on the final design of the UASB reactor.

For this particular case study, the following assumptions need to be made for a
sound design (some of these aspects were already touched in the preceding
sections):

19
Manual for the design of UASB reactors

1. The influent can be pre-treated anaerobically without any previous conditioning

(like acidification, pH control, heating, and addition of nutrients). As in any
sewage treatment plant, only preliminary treatment steps are needed (removal
of coarse materials, grit, and sand).

2. The flow rate fluctuations don’t represent any serious limitation to the
performance of the UASB reactor and, therefore, the average flow rate will be
used as design parameter. Urban runoff is assumed to be diverted to a separate
drain. This assumption has to be confirmed on site.

3. The biodegradability of the influent total COD is 70%.

4. The amount of TSS in the influent will not hinder the operation.

5. The average working temperature will be around 20°C, with a colder period during
wintertime which will not affect significantly the efficiency of the reactor (to be
confirmed by further measurements).

6. There are no disturbing amounts of fat in the influent.

7. The influent contains no toxic or inhibitory compounds.

8. Removal efficiencies for COD and BOD will be in the order of 65 to 85%.

9. There is no need to add macro and micro-nutrients because sewage is known to

contain them in sufficient quantities.

10. The reactor will be constructed above-ground or partially interred, but no pumping
or external energy supply will be needed for the treatment process.

11. The reactor will be designed as hydraulically limited and the conceptual design is
based on an average Hydraulic Retention Time (HRT) of about 10 h. According
to previous experiences under similar climatic conditions, we consider that this
HRT is long enough to guarantee a safe sludge age, or Sludge Retention Time
(SRT). The demonstration plant will provide valuable information to confirm this
assumption, one of the most critical in the design of UASB reactors.

20
Lettinga Associates Foundation

5. THE CONCEPTUAL DESIGN

The UASB reactor conceptual design will be explained following the data of the case
study selected. This Manual has to be read in parallel with the XL file provided,
where all calculations can be checked by the user.

Before starting, it is useful to adopt a clear set of color and font codes to be used in
the XL file in order to facilitate the design process and prevent from inadvertently
deleting formulas. These codes are shown in Figure 4.

Cells in black font are the only ones where you can manually enter data. Cells in red
and/or blue fonts are formulas that change automatically once a value is entered.
Don’t write in them or you will loose the formula contained therein. However, once
you are confident enough with this file, you can start changing these cells in order to
“tailor-made” the file according to your specific needs. Always keep the original
version safe for consultation.

If you detect an error, or make an improvement to the layout or contents of the file,
we would be happy to hear from you.

Figure 4. Color and font codes used in the XL file.

5.1. Relevant wastewater characteristics

Sewage temperature, composition, and anaerobic biodegradability are variable from
place to place. From a theoretical point of view, many wastewater characteristics are
important for the design of anaerobic treatment processes. Not only organic matter
concentration (COD, BOD), but also the type of organic matter (sugars, proteins,
lipids, volatile fatty acids, etc.), suspended solids, temperature, pH, alkalinity,
presence of toxic compounds, flow rate variations, inhibitors of anaerobic digestion,
macro and micro nutrients, etc.

However, from a very practical point of view, aiming to make a conceptual design of
the reactor, we can divide the design parameters in

(a) Primary design parameters,

21
Manual for the design of UASB reactors

(b) Enabling environmental parameters, and

(c) Parameters that might affect design and/or operation.

This classification is made only for teaching purposes and should not be taken too
literally.

5.1.1. Primary design parameters

In Figure 5, the very basic design parameters used to determine the dimensions of
the reactor are shown. As you can see, once that the feasibility of applying anaerobic
treatment to the specific wastewater under investigation is assumed to be possible,
the number of basic parameters needed to actually design the reactor is surprisingly
low.

Note that, before starting the design process, units have to be converted to kg, m 3,
and d (or sometimes h), which are customary used in wastewater treatment.

Figure 5. Primary design parameters. COD = Chemical Oxygen Demand; Q = flow rate.

5.1.2. Enabling environmental parameters

Figure 6 shows a list of parameters which, strictly speaking, are not design
parameters per se. However, if these parameters are not within a given acceptable
range, the entire process may fail.

Sewage temperature, for instance, can certainly affect the size of the reactor as long
as it will influence the rate of biological conversions and, consequently, the residence
time required in the reactor.

When no anaerobic biodegradability data are available for the wastewater under
study, the ratio BOD/COD can be used as an indicator, assuming that compounds
degradable under aerobic conditions will also be degraded under anaerobic
conditions, which is usually the case for sewage. The anaerobic biodegradability is
defined as the percentage of the COD present in an organic sample that may be
transformed into methane under anaerobic conditions. It is also known as the
percentage of methanogenesis, and can be seen as the anaerobic analogous of the
BOD, which represents the aerobic biodegradability of a sample. Knowledge on the
anaerobic biodegradability of a particular sewage can be a first indication of the
potential applicability of anaerobic treatment. In the case of sewage, it is safe to
assume that about 70% of the COD is anaerobically biodegradable.

22
Lettinga Associates Foundation

The performance of an anaerobic reactor under field conditions depends not only on
biological processes, but also on the physical removal of suspended particles.
Therefore, the initial amount, and then the removal of suspended solids becomes
very important. This removal occurs by physical processes such as settling,
adsorption, and entrapment. Subsequent degradation of the removed particulate
fraction depend mainly on temperature.

With the exception of some extreme cases (e.g. very dry countries where water
consumption is extremely low), it can be assumed that the amount of suspended
solids in sewage is acceptable for anaerobic treatment.

Figure 6. Enabling environmental parameters. BOD = Biochemical Oxygen Demand; TSS = Total
Suspended Solids; COD/BOD gives an idea of the biodegradability.

5.1.3. Parameters that affect design and operation

Parameters like flow rate fluctuations, pH, and the presence or absence of
macronutrients (N, P) and micronutrients (Fe, Cd, Zn, Co, among others) may have
some influence on the design and/or the operation of the anaerobic reactor (Figure
7).

Figure 7. Parameters that might affect design and/or operation.

Flow rate fluctuations, if too large, may force the designer to include an equalization
tank or to increase the volume of the reactor. Normal variations observed for sewage
are normally small enough, in magnitude and duration, and we can neglect them. On
the other hand, the UASB reactor has shown a high degree of resilience to cope with
flow rate variations. However, this may not be the case for many industrial
applications.

23
Manual for the design of UASB reactors

pH control or the addition of nutrients (in cases where pure organic substrates have
to be treated), affect the operation of the reactor but will not fundamentally change
the physical design of the UASB reactor.

5.2. Basic assumptions

The most delicate point in the entire design process is the selection of the right
Hydraulic Retention Time (HRT) and/or the appropriate Organic Loading Rate (OLR)
on which the design will be based. Nothing can replace here the advice of an expert
on the subject with experience on the anaerobic treatment of a similar type of
wastewater. As this advice is not readily available most of the times, these two basic
values are generally obtained from a careful literature review. Needless to say that
access to up to date, comprehensive databases or bibliographies on anaerobic
treatment is required for this task.

Failure to select the right values for these parameters will compromise the reliability
of the entire design process.

5.2.1. HRT and OLR

The HRT is the time the wastewater will stay in the reactor, and should be enough to
ensure a proper treatment.

V
1) HRT =
Q

Where

HRT = hydraulic retention time (h)

V = volume (m3)
Q = flow rate (m3/h)

On the other hand, the OLR is the amount of organic matter that is fed to the reactor
per unit of reactor volume:

Q⋅C C
2) BV = =
V HRT

Where

BV = organic loading rate (kgCOD/m3.d)

C = concentration (kgCOD/m3)

24
Lettinga Associates Foundation

The OLR can also be expressed in terms of amount of organic matter fed to the
reactor per unit mass of volatile suspended solids (VSS), assuming that the VSS are
an indication of the amount of viable bacteria available to treat the wastewater. This
is called the specific OLR, or the sludge load:

Q⋅ C
3) BX =
V⋅X

Where

BX = sludge load (kgCOD/kgVSS.d)

X = active sludge concentration in the reactor (kgVSS/m3)

The values for HRT and applicable OLR can vary dramatically depending on the type
of wastewater, climatic conditions, flow rate, and other factors. It is beyond the scope
of this Manual to elaborate on this issue or to list the vast body of information
available on this topic.

For the scope of this Manual, the values shown in Figure 8 will be used as starting
point in the calculation process.

Values from 6 to 12 h are commonly used for sewage treatment, depending on its
temperature (the higher the temperature, the lower the required HRT). Maximum
applicable OLR for sewage is less known, but it was assumed that this value would
be around 3 kgCOD/m3.d.

5.2.2. Reactor height

It was also assumed that the UASB reactor will have an internal liquid height or 4 m,
a very standard value for full-scale reactors treating sewage (Figure 8). For industrial
wastewaters, this height can go as high as 10 m, or even more in some of the new
types of UASB reactors.

For a more detailed discussion on this aspect, refer to Annex I, where a

comprehensive review on sewage treatment was included.

5.2.3. Number of UASB units

It was also assumed that only one UASB reactor will be built (Figure 8). The need to
build more units has to be discussed on a site specific basis. The number of reactors
depends greatly on the flow of wastewater to be handled. For small flow rates
(around 500 m3/d) only one unit is enough. However, if the flow rate is higher,
probably two units are more convenient for a number of reasons: (a) start up might
be easier to accomplish in a small unit that can later provide seed sludge for the

25
Manual for the design of UASB reactors

second unit; (b) maintenance is facilitated as one reactor can be taken out of service
while the other is working; and (c) construction can be done in phases to account for
growing population or new connections to the sewer system. Costs may be higher if
the plant is over dimensioned but, as investment money for big sanitation projects is
generally available only once, it is better to spend it in a long-term sewage treatment
plant than risk money shortages in a later phase.

Figure 8. Basic assumptions made at the beginning of the design process.

5.3. Preliminary calculations

5.3.1. Calculate average and peak flow rates

By using the data provided by the client, calculate the expected average and peak
flow rates. It is useful to have these values in two types of units to make later
calculations easier and more understandable.

Figure 9. Calculation of average and peak flow rate.

This step, as simple as it might seem, can be very tricky if no good data are
available. In that case, the expected flow rate has to be estimated based on data like

1. Current population,
2. Life time of the treatment plant,
3. Expected growth rate during this period,
4. Percentage of the population which is actually sewered,
5. Drinking water supplied to the network,
6. Amount of water that reaches the sewers after use, and
7. Losses of both drinking water and sewage in the respective networks.

En example of these calculations is given in Figure 10 (see XL file for the formulas).
Input values are fictitious, and were only chosen to fit the flow rate provided by the
client. Outcomes can vary greatly depending on the input values used. Population

26
Lettinga Associates Foundation

might be underestimated due to outdated information. If the area under consideration

belongs to a central district of the city, probably the sewerage coverage and/or the
water consumption are higher than the city/region/nation average, or may be the
network losses are smaller because the pipes are relatively new and well maintained.
In our specific case study, the flow rate provided by the client will be used. The area
selected belongs indeed to a central city district and no significant population growth
is expected in any of the variables in the foreseeable future.

Figure 10. Estimation of the sewage flow rate based on demographic and other data.

Even when apparently reliable measurements are at our disposal, it is wise to double
check the flow rate value provided with demographic information to avoid mistakes
that will look very, very stupid once the plant has been built (there are pathetic
examples that don’t need to be mentioned here).

The same goes for virtually all other field data (like sewage concentration, for
instance). If our estimations are way different than the data provided by the client, it
is advisable to confront the client and suggest her/him to verify the reliability of the
information.

It is important to emphasize here that lack of information or the use of inaccurate

information can be disastrous for the design of the treatment plant.

5.3.2. Calculate the volume

Decide whether the reactor is limited by the hydraulic load or the organic load by
comparing the reactor volume calculated in the following two ways:

4) V = Q × HRT ,

using the formula of HRT, and

27
Manual for the design of UASB reactors

Q× C
5) V = ,
BV

using the formula of OLR.

Where

V = volume (m3)
Q = flow rate (m3/h)
HRT = design hydraulic retention time (h) (minimum)
C = concentration (kgCOD/m3)
BV = design organic loading rate (kgCOD/m3.d) (maximum)

Compare the volumes obtained in the calculations. The higher volume will indicate
the type of limitation. In our case study, the limitation is hydraulic. Besides, as the
flow rate peak is not too large, the design of the reactor could be based on the
average flow rate (selected volume indicated with an arrow in Figure 11).

Once a decision on the reactor volume is taken, the detailed design process can
begin. In the XL file, this process is described in the second page, under the title
“Design process”.

Selected value

Figure 11. Preliminary calculations to decide whether the reactor is organically or hydraulically
limited. The value in the circle was selected in this case.

5.4. Design process

In this section of the Manual, the detailed design process needed will be explained
step by step, following the example in the XL file provided.

Figure 12 shows schematically the most important dimensions that will be calculated
during the design process.

28
Lettinga Associates Foundation

LGLS

n
a N
hd

Wd hGLS
2p
p α

q
hr

o p
WGLS

Lr V

Wr

Figure 12. Most important reactor dimensions needed during the design process.

5.4.1. The reactor

In Figure 13, the first steps to calculate the actual dimensions of the reactor are
presented. The calculation process starts in the green-shaded cell under the column
“Calculated”, where the volume selected (see Figure 11) must be copied. The gray
arrow pointing downwards shows the pathway of the subsequent calculation process.

Start in this column

Figure 13. First calculation of the reactor dimensions.

29
Manual for the design of UASB reactors

5.4.1.1. Area of the reactor

The area of the reactor is calculated as follows:

V
6) A =
h

Where

A = Area of the reactor (m2)

V = Volume (m3)
h = reactor height (m)

5.4.1.2. Upflow velocity

Once the area is calculated, the liquid upflow velocity inside the reactor can be
calculated in the following way:

Q h
7) Vup = =
A HRT

Where

Vup = upflow velocity (m/h)

Q = flow rate (m3/h)
A = area of the reactor (m2)
h = reactor height (m)
HRT = hydraulic retention time (h)

5.4.1.3. Reactor shape

There are two options: cubic (rectangular) or cylindrical. The choice of the shape
depends on factors like cost, size, availability of local providers and materials,
previous experiences, etc. Small reactors up to about 300 m3 can be cylindrical. For
volumes exceeding this value, cylindrical reactors are less economically attractive.
Bigger reactors are generally rectangular for construction reasons. The gas collectors
are also simpler and easier to standardize for rectangular reactors.

5.4.1.4. Reactor sides (cubic reactors)

The sides of a cubic reactor are calculated as:

30
Lettinga Associates Foundation

8) L = A

Where

L = reactor sides, both length and width (m)

A = area of the reactor (m2)

5.4.1.5. Reactor diameter (cylindrical reactors)

The diameter of a cylindrical reactor is calculated as:

4⋅ A
9) d =
π

Where

d = reactor diameter (m)

A = area of the reactor (m2)
π = number pi

5.4.1.6. Check HRT

To check for errors, recalculate the HRT with the dimensions obtained so far. If the
reactor is hydraulically limited, it has to coincide with the value assumed to calculate
the volume, as long as it was the starting point of the calculations (this is the case in
our example). However, if the reactor is organically limited, this calculated HRT will
be higher than the minimum applicable HRT from the basic assumptions.

5.4.1.7. Calculate OLR

Calculate the OLR to be applied to the reactor. If the reactor is hydraulically limited,
which is the case in our example, this value should be lower than the maximum
applicable OLR of the assumptions (check). However, if the reactor is organically
limited, this value should coincide with the OLR of the basic assumptions.

5.4.1.8. Round up the dimensions

The calculated dimensions are probably not suitable for construction and they must
be rounded up to more realistic numbers (ask your engineer!). In the example, the
sides were rounded up to the values indicated by the gray arrows, and the diameter

31
Manual for the design of UASB reactors

was rounded up following the light blue arrow (Figure 14). In this example, it was
decided that the shape of the reactor will be rectangular. However, for comparison
purposes, the dimensions for cylindrical reactors will also be calculated throughout
the training. The values entered for length and width in the column “Design” will be
the final values, and they can be copied in the same row under the column “Final”.
Only copy in this column the values calculated for the rectangular reactor.

1) Round up 2) Copy here

Figure 14. Round up dimensions to realistic, easy to build, values, and copy these values in the
column “Final”.

5.4.1.9. Recalculate parameters

As the reactor sides (and diameter for cylindrical reactors) have been rounded up,
the area, the volume, the HRT, and the OLR will change and they have to be
recalculated starting by the new area, as indicated in Figure 15. If no further changes
are made, these values are the final design values.

1) Recalculate (rectangular shape)

2) Copy
Figure 15. Recalculation of reactor dimensions based on user-selected sides (or diameter in
cylindrical reactors). Start recalculating the area with the new design dimensions (in the XL file, see
long arrows).

32
Lettinga Associates Foundation

5.4.2. The GLS separator

The GLS separators are, together with the influent distribution system, the most
characteristic and most important device in the UASB reactor.

They serve many functions:

(1) they collect the biogas escaping from the liquid phase;
(2) they allow settling of suspended solids in the upper part of the reactor;
(3) they help to keep the effluent solids concentration low;
(4) they create a space above the separator for the sludge bed to expand
temporarily due to high hydraulic loads; and
(5) they allow some sludge thickening of the sludge inside the digester part of the
reactor.

The GLS separators are composed of a set of gas collector elements at the top of the
reactor and a layer of gas deflectors beneath the apertures between the gas
collection elements. An example is shown in Figure 16. Details of the design and
construction have to be decided by local engineers.

Distribution box

Effluent gutters

Gas collector

Deflector

Figure 16. Example of a GLS separator showing the influent distribution box on top, the deflector, and
the effluent gutters on both sides (Bucaramanga, Colombia).

33
Manual for the design of UASB reactors

5.4.2.1. Deflector

A number of deflectors need to be constructed to direct all gas towards the GLS
separator. The distance between the deflector and the bottom of the gas collector (q
in Figure 12) should be at least equal to the distance between the reactor wall and
the bottom of the gas collector (p in Figure 12).

There must be an overlap between the gas collectors and the deflectors of about
0.20 m all around the reactor to ensure that the biogas is effectively derived to the
GLS separator and collected (o in Figure 12).

5.4.2.2. Biogas

On top of the gas collectors the biogas leaves the reactor via the biogas outlet pipes.
A 3-way valve is installed here to allow venting of the gas hood during filling and
emptying of the reactor. Failure to open these valves can result in damage to the
GLS separators. The gas pipe passes a water lock to allow pressure build-up inside
the gas collector, preventing water to enter the biogas pipe. An extra outlet could be
installed on top of the gas collectors for emergency cases (e.g. clogging).

The basic parameters for the design of the GLS separator are

a) the maximum allowable liquid upflow velocity at the lowest point of the GLS
(Vup GLS),
b) the angle of the walls (α),
c) the proportion of the reactor height that is occupied by the GLS separator,
known as the “height relationship” (H).

5.4.2.3. Total area of the GLS separator

The area available for passage of liquid between GLS units is referred to as the area
of the apertures (Aaperture). This area is located at the lowest point of the GLS and it is
the smallest area available for liquid passage inside the reactor. The Aaperture is
calculated assuming a maximum allowable liquid upflow velocity (V up GLS) at this
point (Figure 17):

Q
10) Aaperture =
Vup GLS

Where

Aaperture = area of the apertures, needed to comply with the design Vup GLS (m2)
Q = flow rate (m3/h)
Vup GLS = maximum allowable upflow velocity at the aperture (m/h)

34
Lettinga Associates Foundation

It is generally assumed that too high a value for the Vup GLS would prevent
suspended solids that might eventually reach this point to go down into the reactor.
However, it has to be said as well that solids present at this high point in the reactor
are probably not worth retaining and that they would better be washed out. In any
case, it is accepted that the Vup GLS should not be higher than 3 to 4 m/h and that,
whenever possible, it should be around 2 m/h, which was the value selected for our
design3.

Once Aaperture is known, the total area needed for the GLS can be calculated as:

11) AGLS = Areactor − Aaperture

Where

AGLS = area of the GLS separator (m2)

Areactor = total area of the reactor GLS (m2)
Aaperture = area of the apertures (m2)

Design Vup at Q
A aperture =
this point VupGLS

A GLS = Areactor − A aperture

Figure 17. Select a design Vup at the apertures and first calculation of the GLS dimensions.

5.4.2.4. Number of GLS separators

The number of GLS separators depends pretty directly on the reactor size. As we will
see below, a compromise must be struck between the angle of the walls and the
height of the GLS separator, and both parameters have to be within certain range.
When the reactor area is beyond a certain value, it is not possible to comply with
both restrictions by using only one GLS unit (see value out of range in Figure 22 for a
cylindrical reactor with just one GLS unit). There is no fixed rule to decide for the best
number of GLS units, but easiness of construction is a very important aspect to take
into account. In our example, we decided for 4 units. The dimensions of the GLS
3
Some people with field experience claim that this parameter is not relevant, and that it is sometimes
more important to restrict the passage of water at this point as a way of minimizing the wash out of
suspended solids.

35
Manual for the design of UASB reactors

separator are calculated assuming that one of its sides (the length) will be the same
as one of the UASB reactor sides. As we already calculated the total area needed for
the GLS separators, knowing one side we can calculate the other side (Figure 18).

LGLS fixed (equal to Lr)

Don’t round up when A GLS

WGLS =
more than 1 GLS LGLS

Figure 18. Calculation of the sides of the GLS separators.

This value should not be rounded up at this stage if we decided to build more than
one GLS units. It has to be noted that this value is the theoretical width of a single
GLS separator.

As we have more than one, this value has to be divided by the number of GLS
separators in order to obtain the width of each one of the units. As shown in Figure
19, this width has to be rounded up and the final dimensions of the GLS separators
are finally calculated.

5) Final values
1) Width of each 4) Recalculate
GLS separator

2) Round up
3) Area of each
GLS separator
Figure 19. Final values for the GLS separators.

5.4.2.5. Height of the GLS separator

36
Lettinga Associates Foundation

The height of the GLS separator can be calculated in two ways. The first way is by
assuming a certain height relationship (H). The value for H suggested in literature
goes from 0.15 to 0.30. That means that the GLS separator should occupy between
15 and 30% of the total reactor height. Assuming H = 0.30, and a reactor height of 4
m, the height of the GLS separator would then be 1.20 m (Figure 20). The height
calculated for the GLS separator can be rounded up and the final H is recalculated.
In our example, it was considered that there was no need to round up this value.

4) Recalculate
1) Design value

3) Round up
2) Calculate
hGLS = h × H

Figure 20. Calculation of the height of the GLS separator based on the “height relationship”.

The GLS device position might be slightly moved upwards or downwards if practical
considerations so dictate. However, it is important to bear in mind that the total height
of the GLS separator below the water level (hGLS in Figure 12) must be about 30% of
the total reactor height. If the GLS separator must be moved upwards, an additional
piece can be added at the base of this device to compensate for this distance. In this
case, the maximum allowable height of the additional piece is the distance between
the top of the GLS and the water level (h d in Figure 12). However, there is no such a
limitation if the GLS needs to be moved downwards because the height of the dome
above water (a in Figure 12) can be adjusted at will. This dome of the GLS separator
can stick out of the water level as needed for practical reasons like access, easiness
for pipes connection, and so on. To easily strip the gas from the liquid, particularly
when there is some scum layer formation, it is essential to maintain a sufficiently
large liquid/gas interface inside the gas collector. The liberation of gas is guaranteed
even if there is no internal overpressure in the GLS separator because the area of
the dome calculated to be able to release all the produced biogas remains the same
above the point calculated as hGLS - hd (see in Figure 12). However, the gas pressure
of the biogas in the gas dome could be up to 40-60 cm above the external air
pressure.

5.4.2.6. Angle of the walls

When both the base and the height of the GLS separator are defined, the angle of
the wall can then be calculated. It is generally recommended that the angle of the
walls of the GLS separator should be between 45 and 60° for it to act as a good
settling device. The first step is to calculate the tangent of the angle as follows (see
scheme with the basic dimensions of the GSL in Figure 21):

37
Manual for the design of UASB reactors

h GSL
12) tan α =
a

The arctangent of this number will give us the angle in radians (by definition, the
arctangent is the angle whose tangent is this number). To express the arctangent in
degrees, we have to multiply the result by 180/π (Figure 22). See that the angle
calculated for a cylindrical reactor in Figure 22 is out of range because it was based
on a single GLS unit.

hGLS

α π/2

a
2a
Figure 21. Basic dimensions needed to calculate the angle of the GLS separator.

2) Calculate angle Value out of range

3) Copy here
1) Calculate tan
hGLS
tan α =
a
Figure 22. Values calculated for the case study.

In some cases, it could be useful to fix the angle of the walls instead of fixing the
height. In those cases, calculations have to be performed as shown in Figure 23.
However, it seems more practical for construction reasons to fix both the width and
the height of the GLS separator and let the angle fluctuate within the acceptable
range of 45 to 60°.

38
Lettinga Associates Foundation

H changes Out of range

Fix this value (only 1 GLS)

Calculate hGLS
Figure 23. Sometimes it is better to fix the angle and calculate the GLS height.

The distance between the wall of the reactor and the GLS separator is calculated
with the following equation:

Wr − n GLS × WGLS
13) p =
2 × nGLS

Where

p = distance between reactor and GLS separator (m)

Wr = reactor width (m)
nGLS = number of GLS separators
WGLS = width of one GLS separator (m)

The distance between two GLS separators will then be twice as large (2p). It is
important to keep in mind that these distances have been calculated based on the
maximum allowable Vup at the apertures. For this reason, we have to make sure that
the gas deflectors are located at a distance at least equal to this one (p) from the
GLS separators (this is the distance p indicated within a dashed circle in Figure 12).
Calculations as performed in the XL file are shown in Figure 24.

Wr − nGLS × WGLS
p=
2 × nGLS
Twice the distance GLS-reactor

Recommended Minimum (should not be limiting)

Figure 24. Sometimes it is better to fix the angle and calculate the GLS height.

5.4.2.7. Gas liberation area

39
Manual for the design of UASB reactors

The biogas produced inside the UASB reactor has to be released inside the GLS
separator. Because of the conic shape of this device, the area for gas release
decreases as we go upwards in the reactor. There is a minimum recommendable
area for a safe biogas release. If this area is respected, the risk of scum formation
and blockages in the biogas exit pipes are minimized. The recommended biogas
release area, or gas load, fluctuates between 1 and 5 m3 of biogas per m2 and per
hour. For reactors treating highly concentrated industrial wastewater, this parameter
can become critical because the amount of produced biogas may be extremely high.
Different types of GLS separators have been developed by private companies to deal
with this problem. In UASB reactors for sewage treatment this parameter is usually
not a problem and the recommended values can easily be attained. The calculation
process for the gas release area shown in Figure 25, Figure 26 and Figure 27 is
similar to that described for the calculation of the area, width, and height of the GLS
separator when the angle is fixed.

2) Gas production/Gas load

1) Recommended 3) Area/number of GLS units

4) Assuming LGLS = Lr
Figure 25. Calculation gas liberation area (a).

5) Height must be calculated 7) Recalculate

because angle is now fixed

6) Round up height
Figure 26. Calculation gas liberation area (b).

40
Lettinga Associates Foundation

9) Only needed if GLS is 10) Copy final values

moved upwards

8) hGLS – hd
Figure 27. Calculation gas liberation area (c).

The basic value needed for these calculations is an estimation of the future biogas
production during normal reactor operation. This estimation has to be performed on
the basis of the wastewater characteristics, and the expected reactor efficiency.
Based on a reasonable efficiency for the conversion of COD into biogas, a daily
biogas production above 300 m3 is foreseen (with methane content of around 80-
85%) (Figure 28). On-line measurement of the biogas production could be used as
an indication of the organic loading rate. Enough accumulation volume should be
available to store this volume. Details on the stoichiometry and the chemical aspects
of these calculations go beyond the scope of this version of the Manual. Short
descriptions can be found as comments in the XL file.

Q*C
OLR*Efficiency

RemCOD*CH4 content

CH4, field conditions

Biogas, field conditions

Figure 28. Estimation of the potential biogas production under field conditions.

41
Manual for the design of UASB reactors

5.4.3. The inlet distribution system

The influent must be distributed over the bottom of the UASB rector as uniformly as
possible. For this first demonstration plant, we recommend the installation of 96 inlet
pipes over the base of the reactor (around 1 inlet nozzle each 1.5 m2). Influent
injection into the reactor should be performed at about 0.10 m from the bottom of the
reactor (Figure 29). Feeding can then be accomplished by gravity. We recommend
installing 2 distribution boxes in each one of the 3 rows located in between the 4 GLS
separators. Each distribution box will contain 16 compartments leading to equal
number of influent distribution pipes. Pipes must be reachable from outside to
facilitate maintenance and for de-clogging when needed. To ensure that each inlet
point receives a similar influent flow, it is necessary to install several influent
distribution boxes in the spaces between the GLS separators. An example of a
distribution box is shown in Figure 30.

From literature or experience Copy final values

Based on practical Recalculate

considerations values

Figure 29. Calculation of the number of influent distribution system.

Figure 30. Circular box for homogeneous influent distribution with 16 inlet points. Not to scale.
Adapted from van Haandel and Lettinga (1984).

42
Lettinga Associates Foundation

The relatively high amount of inlet pipes recommended in this plant could be reduced
in other full-scale UASB reactors. However, there is controversy as to what the
optimum number of inlet points should be.

5.4.4. Effluent
Effluent collection gutters need to be mounted along the top of each GLS separator,
and on both sides. The effluent will flow into this gutter through equidistantly
positioned V-notches, and it will leave the reactor via the effluent collection pipes.
These pipes must be thick enough to collect all the effluent easily and minimize the
possibilities of clogging. In front of the gutter a baffle needs to be installed to prevent
floating materials from escaping the reactor with the effluent.

5.4.5. Excess sludge

5.4.5.1. Sludge production

Based on wastewater characteristics, and assuming reasonable values for the

biological conversion of organic matter in the reactor, the excess sludge production
will be less than 1 m3 wet sludge/day (Figure 31).

Empirical values

Figure 31. Calculation of the excess sludge production.

5.4.5.2. Sludge discharge pipes

Pipes should be installed for sludge discharges and sampling at 6 different heights.
The first one should be placed around 0.20 m above the injection system and the
rest should be evenly spread over the height of the reactor, with the last sludge
discharge pipe located around 0.05 m below the internal gas deflectors (not more)
(Figure 32). These values are only indicative, and small modifications will not affect
the operation of the reactor. Additional sampling ports are optional and might help for
monitoring purposes and to assess more closely the evolution of the sludge bed. The
discharge pipes should penetrate at least 1 m inside the reactor in order to take
sludge from the bulk of the sludge bed and avoid wall effects that may affect the
collection of representative samples. At the end of the discharge pipes inside the
reactor, a T fitting has to be installed in order to facilitate the sampling of the sludge

43
Manual for the design of UASB reactors

and to prevent clogging. The withdrawal of sludge has to be performed from a point
equidistant from influent inlet points to optimize the quality of the sample. The outlet
of the pipes must be easily accessible. The pipes have to be equipped with valves
and the sludge should be slowly discharged under gravity to prevent vortex
formation. The sludge discharge facilities should also allow sampling of the reactor
contents. A sludge discharge pipe might also be installed at the very bottom of the
reactor if a complete discharge of the reactor by gravity is required.

User defined See Excel file

Figure 32. Number and position of the sludge discharge and sampling pipes. Calculation details can
be checked in the XL file.

5.4.5.3. Sludge drying beds

Assuming that, under good climatic conditions, around 3 d are needed to dry out a
thin layer of sludge (0.20 m), it can be calculated that three sludge drying beds of
about 10 m2 would suffice (Figure 33). If the drying time is longer, the area needed
for drying beds will be accordingly larger.

Empirical value (depends on climate)

Figure 33. Estimated dimensions of the sludge drying beds.

44
Lettinga Associates Foundation

5.4.6. Summary design parameters

Table 3 shows a summary of the design parameters for the UASB reactor. Figure 34
shows these parameters in a schematic diagram (not to scale).

Table 3. List of most important design parameters for the UASB reactor. Q = flow rate; C =
concentration; COD = chemical oxygen demand; OLR = organic loading rate; HRT = hydraulic
retention time; Vup = upflow velocity; GLS = gas-liquid-solid separator.

Stage Parameter Design value

Design information Q (m3/d) = 1500.00
C (kgCOD/m3) = 0.90
Total OLR (kgCOD/d) = 1350.00
Average HRT (h) = 9.22
UASB reactor hr (reactor height) (m) = 4.00
V (volume) (m3) = 576.00
Lr (reactor length) (m) = 12.00
Wr (reactor width) (m) = 12.00
Vup (m/h) = 0.43
OLR (kgCOD/m3.d) = 2.34
GSL separator Number of GLS units = 4
LGLS (GLS length) (m) = 12.00
WGLS (GLS width) (m) = 2.30
hGLS (GLS height) (m) = 1.20
α (angle GLS) (°) = 46.22
p (distance reactor - GLS) (m) = 0.35
q (minimum distance GLS - deflector (m) = 0.35
o (overlap GLS-deflector) (m) = 0.20
Wd (width of the dome) (m) = 0.29
hd (dome height below water) (m) = 0.15
a (height dome above water) (m) = 0.50
Influent distribution Number inlet points = 96
Number of inlet rows = 3
Number of distribution boxes per inlet row = 2
Total number of distribution boxes = 6
Number of inlet points per box = 16
Distance base-injection (m) = 0.10
Sludge discharges Port 1 (m from bottom) = 0.30
Port 2 (m from bottom) = 0.70
Port 3 (m from bottom) = 1.10
Port 4 (m from bottom) = 1.50
Port 5 (m from bottom) = 1.90
Port 6 (m from bottom) = 2.30
Distance deflector last port (m) = 0.05
Length of pipes inside the reactor (recommended) (m) = 1.00

45
Manual for the design of UASB reactors

Figure 34. Basic scheme of the UASB reactor (not to scale) showing the main dimensions (if not
indicated otherwise, dimensions in meters). See also attached spreadsheet. Note: only one GLS
module is depicted.

5.5. Start up of the reactor

The start up period will last between 2 and 6 weeks. Specific advice is needed on this
critical period. The use of inoculum or seed sludge is not needed for sewage
treatment. The UASB reactor will start up sooner or later with the bacteria which are
naturally present in sewage.

46
Lettinga Associates Foundation

PRELIMINARY TREATMENT
Screens and sand trap are mandatory (an example is provided in Figure 35). To keep
accurate records and generate valuable information for future treatment plants, a
device to measure the flow rate is also highly recommended (a Parshall flume, for
instance).

Removal of fat and grease may be needed if the influent contains excessive
discharges of slaughterhouse wastewater and/or large amounts of domestic food left-
overs.

Sometimes measures should be taken to prevent the release of malodorous

compounds from the raw influent, for instance by covering all open channels
transporting raw sewage and treating the off gases in a compost filter.

Local engineers and consultants can assist in the selection of the right preliminary
treatment technologies, which are all well established in the market.

Figure 35. Screens (left) and sand trap (right) required for preliminary treatment. Notice: for the
reactor of our case study, the size needed is much smaller (Bucaramanga, Colombia).

47
Manual for the design of UASB reactors

6. POST-TREATMENT
UASB reactors do not eliminate completely the pollutants present in sewage.
Therefore, depending on local discharge standards, the intended use of the treated
effluent, and other local conditions, sometimes certain post-treatment steps are
needed to remove remaining COD, pathogenic microorganisms, and even nutrients
like nitrogen and phosphorous.

6.1. Use of the treated effluent

Options for post-treatment are plenty. Before selecting any of them, the use of the
treated effluent has to be agreed upon.

6.1.1. Recharge of water bodies

If the effluent will be conveyed to a water body like a river or a lake, local discharge
standards apply and the post-treatment must include a disinfection step to eliminate
pathogenic micro-organisms. Options available for this case are polishing ponds,
trickling filters with a disinfection unit, bio-rotors, etc. COD and BOD will also be
further reduced during post-treatment. Nutrients removal might be required to comply
with some standards. Similar considerations apply for the recharge of aquifers with
treated wastewater.

6.1.2. Agricultural reuse

Agricultural reuse of the treated effluent is a very interesting option in most
developing countries. The use of wastewater for irrigation has many advantages. It
reduces the demand for freshwater, provides a stable and reliable water source,
serves as a sewage treatment system, reduces freshwater pollution improving
human health downstream, contributes to food production and security, and
promotes sustainable agriculture, among other benefits. There are some risks
involved in this practice that need to be minimized through careful management.
Municipal authorities must ensure that the level of treatment provided not only fits
reuse needs but takes into account the protection of public health and environmental
quality.

There are basically two options: restricted and unrestricted irrigation. Restricted
irrigation is the irrigation of all crops except salad crops and vegetables that may be
eaten raw. For this type of irrigation, WHO (World Health Organization) recommends
that treated wastewater should contain no more than one human intestinal nematode
egg per liter. For unrestricted irrigation, i.e. including salad crops and vegetables

48
Lettinga Associates Foundation

eaten raw, WHO recommends the same helminth egg value, and additionally no
more than 1000 fecal coliform bacteria per 100 ml of treated wastewater. The level of
sewage treatment required to reach the unrestricted irrigation value is higher, and so
are the costs of the treatment plant. The cheapest option is always a series of
polishing ponds, but the land required is larger. Other possible methods are bio-
rotors, U.V. light, ozonation, and sometimes even chlorination, although this option
has some environmental disadvantages and can pose health hazards if not applied
with extreme care.

6.2. Example: Polishing ponds

A series of polishing ponds are always a good option (if enough land is available!).
Polishing ponds are not as big as a conventional system of waste stabilization ponds
and therefore the losses of useful water via evaporation are lower. Polishing ponds
require little operation and maintenance and the quality of the effluent can comply
with the standards for both unrestricted irrigation and discharge into water bodies.

For the treatment plant designed in this Manual, a system of polishing ponds in
series for post-treatment would have approximately the following dimensions (see XL
file for detailed calculations):

 Number of ponds in series = 6

 Total area needed = 1.5 ha
 Expected effluent concentration = lower than 1000 MNP/100 mL (useful for
unrestricted irrigation)
 Hydraulic residence time = 15 d

These figures are only estimative and have to be confirmed by more detailed
calculations in which kinetic constants measured in waste stabilization ponds under
local (or similar) conditions are used.

The concentration of pathogenic microorganisms in the influent also needs to be

measured in order to adjust the design to the specific characteristics of the sewage
under study.

For less stringent effluent concentrations, the area available will be smaller.

49
Manual for the design of UASB reactors

7. COSTS
Investment costs and the costs of operation and maintenance can vary greatly from
place to place. However, based on average numbers reported for a number of UASB
reactors built in other countries, it is possible to make a rough estimation of the
possible costs of the treatment plant, including investments needed for construction,
start up, operation and maintenance, and capacity building.

7.1. Investment costs

To calculate the investment costs, it is possible to use two different criteria.

a. On the one hand, it was observed in several examples in Latin America that
UASB reactors can cost around € 30 per inhabitant for medium-size treatment
plants (15,000 inhabitants or more). For our example, this number would
represent a total cost of the treatment plant of about € 450,000.

b. On the other hand, it has also been reported that UASB reactors can cost
approximately € 15,000 for each L/s of influent flow rate treated. If the design flow
rate is 1,500 m3/d (17.4 L/s), the investments costs would be about € 260,000.

The difference in these two estimations is probably due to the fact that the flow rate
used as example comes from a place where water consumption is lower than that
registered in those cities where the estimations have been made. The investment
costs will be between those two numbers, probably closer to the higher figure.

For further calculations, we will assume here that construction costs will be €
400,000.

Needless to say, these estimations have to be confirmed with local data.

7.2. Operation and maintenance costs

Operation and maintenance costs are known to fluctuate around € 0.80 per
inhabitant per year for medium-size treatment plants. Therefore, it can be estimated
that the total O&M costs will be around € 12,000 per year. This amount represents
only about € 45 per day.

50
Lettinga Associates Foundation

7.3. Other costs

The following costs are not included in the above estimations:

1) The costs required to build fences, facilities for employees and operators,
security premises, laboratories, and the purchase of laboratory equipment,
etc.
2) The costs involved in the design, construction and operation of the post-
treatment facilities (including acquiring the necessary land).
3) The costs of pumping raw sewage or treated effluent, if needed for reuse
purposes (pumping should be avoided).
4) The costs of any external advice needed for start up and operation.
5) The costs required for initial training of operators.

All in all, the total amount of additional costs can be estimated at € 400,000.

7.4. Summary of costs

A summary of the total amount of estimated costs is given in Table 4

Table 4. Summary of estimated costs.

Construction costs 400,000
Land 200,000
Design 50,000
Capacity building and supervision 100,000
Others 50,000
Total 800,000

51
Manual for the design of UASB reactors

8. REFERENCES
van Haandel, A.C. and Lettinga, G. (1994). Anaerobic sewage treatment. A practical guide for regions
with a hot climate. John Wiley & Sons Ltd., Chichester, UK.
Jewell, W.J. (1985), The development of anaerobic wastewater treatment, in Proceedings of the
Seminar/Workshop on Anaerobic Treatment of Sewage, Switzenbaum. M.S., ed., Amherst, USA,
17-54.
Lettinga, G., van Nelsen, A.F.M., Hobma, S.W., de Zeeuw, W., and Klapwijk, A. (1980). Use of the
upflow sludge blanket (USB) reactor concept for biological wastewater treatment, especially for
anaerobic treatment. Biotechnology and Bioengineering 22, 699-734.
McCarty, P.L. (1981), One hundred years of anaerobic treatment, In Hughes, D.E. et al., eds.,
Anaerobic Digestion 1981, Elsevier, Amsterdam, the Netherlands, 3.
Seghezzo, L. (2004). Anaerobic treatment of domestic wastewater in subtropical regions. Ph.D.
Thesis. Wageningen University, The Netherlands.

52