Olympic Plumbing response 10-20-09.pdfOLYMPIC HILLS Plumbing, Inc.
14221 Lake Rd. • Lynnwood, WA 98037
(425)775-5829 • Fax (425)742-4970
6IL b 2ao610 ).2z
`to 5 �a—ACV- j ve �
54- Jj i 'F 1.(i4s 4,+
aS G.- 4 k,,,
sl�v�, Zvi vule>�e�
`f2 -
RESUB
n 7f09
i��.�1_.CSIPd� 1:3F'P%'gt=;"1'l1lf�t�f"i
Vra4
Ar
tinu 1n g Drainage
Waste anct Vent
fol e f f i c i e �
pressers control
By
Dr. Mk -had
MSc ry ciii PY L.Eny I'VICIBSE tvfl i*
Lecturer in Architectural Engineering and tenured Research
Fellow of the Drainage Research Group, The School of the
Built Environment, Heriot-Watt University
Scotland.
Email: m.gormleyCm1sbe hw.ac.uk
Prof. John A. Swaffield
BSc MPhil PhD FRSE CEng FOBSE MRAeS
William Watson Chair in Building Engineering, Fellow of the
Ro},al Society of Edinburgh, Vice President of the Chartered
Institution. of Building Semices Engineers (CISBE) and Head of
the School of the Built Environment, Heriot-Watt University,
Scotland.
Email: J.A.Swaffield@.hw.ac uk
rwiNO R.-TIIOENT
, ',G' i"ly'ili�.�'"J,
January 2007 J Heriot-Watt Unhvrsity
'8-4, rfdrrt . DYrE,i Inge 011asfe and V-elk., syxti rnjsz
li3fea=i�,e°°. c� 11cienrPr-x sare t "rra€rr 1
Contents
Contents i
Stunmary ii
1. liatL on 1
1.1 An historical perspective 1
1.2 Water in building drains 4
1.3 Airflow in buildin; drains 4
1.4 The requirements of a well designed system 6
2. Pressure transients in plumbing systems
2.1 What are pressure transients 7
2.2 What do these pressure transients do in a
building drainage system? 8
2.3 How to overcome pressure transients a
3. Designing for best results
3.1 Alleviating negative pressure transients. 9
3,2 Alleviating positive pressure transients 10
4. Building case studies
4.1 Modeling flows in drainage netwoCks 11
4.2 Two story building 12
4.3 10 story buildings 13
ivtfl elusions 15
6. References 17
fl
YD?a;`id,?' a=_f'{ L'Ii;ntji3gds.`-Var€'CisilErol
Summary
There are few real mysteries remaining about the mechanisms at play in building
drainage and vent systems. This has been: well understood tom the beginning of
modern sanitary engineering at the end of the 19"' Century. The description of
Building drainage and vent system operation is best understood in the context of
engineering science in general and fluid mechanics in particular.
Early researchers in the field were well aware of this and many examples of the
application of sound. fluid mechanics are available as evidence. Much research has
been ear; ■ed out since the end of the ` llorld War 11, where, particularly in Eampe,
extensive reconstruction work prompted the quest for more efficient approaches to
drainage and vent system design.
At the center of the system's integrity is the water trap seal, which stops foul air from
entering a habitable space from the sewer. The water trap seal is ;:sup-Ily IV;: ate' 2
inches in depth depending on the fixture it is protecting.
It comes as a surprise to many that the flow of air- is as important, if not -more
important, than the flow of water, to the safe operation of the drainage system. This
air flaw is `induced' or 'entrained' by the flaw of water. The unsteady nature of the
water flows causes pressure fluctuations (known as pressure transients) which can
compromise water trap seals and provide a path for sewer gases into the habitable
space.
Transients can be dealt with by a combination of careful design and the introduction
of pressure relief devices as close to the area of concern as possible. Long vent pipes
can be an inefficient way of providing relief due to friction in the pipe_ Distributing
air supply inlets using AAVs around a building provides an efficient mearns of venting
and it reduces the risk of positive transient generation. AAVs do not cause positive
Pressure transients, they merely respond to their by closing, and hen;,e reflect a
reduced amplitude wave.
11
Optivini Ai?efi'`.Yclent�"+if=.^usarf'' Cia`lti',ol
In tall buildings parallel vent pipes can only provide a small relief path for a positive
pressure transient (approx 1/3 if the vent pipe is tie same diameter as the main
vertical stack) thus a wave will still propagate throughout the rest of the system that
could compromise Nvater trap seals, The introduction of a positive air pressure
transient allevintinn device rrOlJides- a means to `blow oftft' pressure surges as Close to
their sotirce, thereby protecting water traps. Attenuation of up to 90% of the incident
wave car, be achieved, thus protecting the entire systemThere is little that can be
done for a system experiencing a total blockage, generating excessive static positive
pressures in the drainage system. In such circumstances the lowest water trap seal will
`blow' providing relief for the whole system. This wild occur regardless of the method
of venting employed.
In validated test simulations air admittance valves (AAVs) have been shown to
provide as least as good protection for water trap seals as a fully vented system, and in
tall buildings in some circumstances, even better. The fully engineered designed
active control system utilizing AAVs for negative pressure relief and Positive Air
Pressure Transient Attenu.ators (PAPAS) for positive transient relief is shown to be an
effective method for balancing the need for safety and. efficiency while maintaining
functionality inv±sible to the user.
iii
r3"f°3d"din Y3) 11ina w J-6a n- d-,ad PVIIf' 45,1-stefn.,q
L P xr,';f814 's f' i' ILt'7 ?Pi', ti' air' Contra!
1. Introduction
1.1 A historical perspective.
To most people the building drainage system lurking beneath their pristine ceramic
and stainless steel appliances presents a mystery beond their usual `nee to know"
How their sunk full of soapy water gets from their newly refurbished kitchen island to
the municipal treatment plant is of little or no iizterest, and likewise, few people
ponder the similar journey fiom the PVC, bath or bidet in the bathroom; until that is,
they are suddenly faced with a foul smell from 'somewhere down there' or are met by
it filling WC bowl which keeps on fitting and pours onto the new floor vovering. The
mystery surrounding the drainage system suddenly deepens on the presentation of an
unfeasibly costly repair bill.
In truth there are few mysteries about the operation of a building drainage system.
The Underlying principlev bo:rernu^g the flows of all fluids (water and air) have been
well describedand indeed applied. to the building drainage system for both design
`inamag the system "work and forensic analysis (finding out why it didn't wUik` for
t � Y } '� Y� � g y _. �
many years. It is worth remembering that while humans have many cultural taboos
surrounding the bathrooms, which have contributed to the myths surrounding the
drainage system, there is a strong scientific basis for the movement of :paste by means
of water which has a long tradition, going back thousands of years. However our
conceal is with models systerag and thereiure developmeists over the last 120 - 150
years are relevant.
The age in which tl.e innovation of safe and practical building drainage and plumbing
were at the cutting edge of technology was in the late 19`h Century. Many of the
important actors, of maintaining the system's tiltegray buy preventing sewer gases
from entering living spaces, the water trap seal and systems venting, had already been
introduced and much work on improving the system's response to the inevitable
pressure fluctuations encountered in a fluid transport system were well under way.
This work was initially carried out by Scientists and notable Engineers of the time, In
the UNK. the wafer trap seal was hive ated by Cumri-zings as early as 17751(11
Cummings was an Engineer and a watchmaker and resurrected the idea of a flushing
Rtfild i Dr a,ifu�),e 4`4 le con d )'`d'f% $rsieT@:m
zi)iit: i i,v F=P` t'.`Af"1 D'r6i'x'.e:ld; r3 C'o dtP.?'c
rWC originally invented by Harrington in the 17th Century. While much of the parts of
tite system had been aroundfor some titnc it wasn't until the iuid iit�� Century that
any impetus existed to sort out the poor sanitary conditions in large towns and. Cities.
in 1542 Edwin Chadwick, an English civil servant, published his `Report into the
Sanitary CorditiOns n f the Labouring Pqpdaiion q f Great Brltaill'. This report
initiated a process of reform which prompted investment in sanitation as a public
health priori-ty in the slum conditions created vy the rapid expansion of British cities
as a result of the industrial Revolution. Such was the importance of sanitation at the
time that even the eminent Scientist/Engineer, Osborne Reynolds, whose work on
turbulent flow was seminal and still considered central to any discussion of fluid
dynamics today, was moved to write a paper on `Sewer Gas and How to Keep it Out
of the House' tz�, which dealt with sanitation in the slums vfl`VliulcLLVSter, England In
the late 19th Century
While t11is cork —. s rnntinlling in Eiirone-, in the United Statc$, Architects., Scientists
and Engineers were facing their own growth problems as immigration from Europe
and. rapid ccnornic expansion prvv ucd the
e disvcr for a building bows. vvYvrk
(reported by) a notable Engineer, George Waring in his book `How to drain a hoarse,
practical information .for householders',ts) highlights the depth of knowledge
available at the tulle.
While some of :'daring's approaches are outdated, his writings edd show that lie had a
firth grasp of the link between what was going on in the drain and its relation tofluid
mechanics. The following extract illustrates this well;
"Efficiency [of the vent system] is due entirely to the admission of air fast
enough to supply t«4 cleivand fair air told thL 'vaciiLliii eallsed by water uv`riing
through some portion of the pipe beyond the trap, it is not only a question of
having an opening large enough to admit air, but of having an adequate current
led freely to the opening ......... A one inch pipe, for example may admit air fast
enough, while a longer pipe of same diameter, or a smaller pipe of the same
length would not do so"
Waring, 1895 pp 101-102
2
'i:t'?! Ti'i t°� €Xiftsf i' f`{j.:ti'ti3 and1''4"t!1 c_t't".tt.ti
What Waring is suggesting here is the importance of pipe friction and the necessity to
analyze the problem in a time — dependent and dynamic way. This is a crucial point
and one which has driven much of the computer based systems modeling carried out
in the past 30 years, Building drains carry unsteady flows which mean that they are
rapidly changing andcannot be analyzed ,using simple calcuintions based on steady,
unchanging flows, which are often used for the slower moving public sewer networks.
A contemporary of Waring, the Boston Architect J. Pickering Putnam went fiirther in
his 1911 book `Plumbing and household sanitation' ta> in which he doubts the
necessity for any venting on properly designed systems with anti -siphon traps — he
even suggests the use of mechanical air vents in close proximity to water traps in
order to overcome siphonage problemsEa,�,i691 Putnam's conclusions fouowed years of
experimentation on water trap seals and venting arrangements based on sound fluid
mechanics principles. The point raised by Waring above was further promoted by
Putna_r+? foLowing a series. of f— periments on pipe friction catTied out by the
Massachusetts Institute c?f Zechnology (NIIT)�arzsa7 Putnwu's 718 page book
concludes with a paper delivered to the 44' annual convention of the American
Institute of Architects in San Francisco, Jan 18, 1911, entitled. `Better Plumbing at
half the Cost' in which he suggests a single pipe system for multi-storey buildings
based on an economic, argument and the years of experimentation and experience of
the author.
This work on the single pipe system was further investigated in the U.K by the
Building research Station in the ZO years or so following World War H. Again, the
driver was a rapid expansion in building projects as the war torn country was rebuilt.
Work published by Wise in 1957tsi concluded that the single pipe system (known as
the Single stack system in the U.K.) was a robust, safe and economical option and
that, if properly designed, building drainage systems do not require every trap to be
vented..
Against this historical background this report will explain some of die long
establisihed principles of the operation of building drainage waste and vent systems,
and will illustrate options for effective venting using the modern method of computer
9
_i`^slzt?;ll l f dd:fkiLty{< `%`at5.s' idr$x`: b` F'' f s7:' d':§;
€�pi 'rPs -2p e ai.'ti'P .ig tvJis"t? f`"€tr""f#
�.,............................. .. ..
based simulation to represent and predict the rapidly varying flows found in building
drains_
1.2 Water in building drains
When a WC is flushed or a bath or lavatory is emptied., the water flows in the
horizontal pare of the drainage system and carries with it solids from the WC or,
perhaps solids which had deposited in the pipe from a previous flush- When this water
reaches a vertical stack pipe, it pours in, in a curved fashion until it strikes the back
wail of the verticr I pipe.(`) The at -
then swirls around the inner surface
and fails down the pipe, under
gravity, clinging to the pipe wall,
this is called annular water flow (see
figure 1). The water film on the
inner surface of the pipe is
surprisingly thin, even at high flow
rates producing little more than '1¢
inch film thickness_ The solids
Figure I Water discharging from a branch
Hcrir, —It im vrivcrSdfp
fall, under gravity, in the care of this pipe.
1.3 Air in building !rains.
While most people are aware of the presence of water in a building drain, because this
is what the user is trying to get out of their house or office, few are aware of the
important role played by air in the system. Of these two important fluids (air and
water) it is the regulation and control of the air flow which poses the greatest
challenge for designers, installers and code authorities alike. The whole process isn't
helped by the general lack ofund.erstanding surrounding the subject. So, how does air
come to play a role at all in the building drain.
When water starts to flaw in a pipe, as described above, air is entrained along with it.
This phenomenon is more marked when water falls down the veit3cal drainage pike,
where air is drawn down from the upper termination.(') This is due to the shear
between the water and the air which acts to produce ail airflow. The air pressure,
4
Budding 1)ra ridge iWG.stvfIn f`i'x'ig,s}`.4}fems,
which is assumed to be Tap of stack
atrnospheric at the upper
pressure
termination (where the air drop at
comes from) is subject to water iiLet
`losses' on the way down
the pipe. These losses can be
due to separation (,at, the pressure
regain
termination itself), friction Tract
Traction'
(in the dry part of the pipe)
or simple pressure drop
across a branch to stack
junction when water is
pouring in,
possible
positive
pressure
at base of
stack
Negative pressure I Positive pressure
Atmospheric �
Pressure
Figure 2 Pressure Profile in the Stack
Stack
Height
These losses reduce in the pipe to sub - atmospheric and therefore place a suction
force on a portion of the system,
The pressure in the pipe below the discharging branch follows a different pattern.
Since the water iaduees an air flow the dominant force on the air is traction rather
than frictioii(s). This ha a tendency to inake the air pressure move in a positive
direction (or a reduction in suction pressure) this moves the pressure back towards
atmospheric at the base of the stack. This pressure at the base of the staCk- Can go
above atmospheric pressure in certain circumstances, this is known as back pressure.
The pressure profile usually associated with this process is shown in Figure 2. It must
be remembered that this is only a representation of the pressure `signature' associated
with a speciac event at a sing1c point to time, it is in effect a `temporal snapshot' of
the pressure distribution in the vertical stack, and is probably best applied to taller
buildings. In reality this profile will change rapidly with time sending pressure
transients up and down the stack communicating these changes as described. below.
It is very useful to ineaswie pressure in drainage systems in term of `head' - Where
pressure is referred to as uL equivalent water depth, for example `coluinu inches of
s!y
t.. uES::{.in, :i':`(q`:t S7YW�-Se a ?+Fa`ut t: cJy,ti sfi°:3,..3.
water', or simply inches of water. The advantage of using depth of water as a
reference for air pressure is that a su+iion pressure of 2 inches of water will remove a
tromp 2 inches deep and. is therefore a useful equivalence.
1.4 The requirements of a well designed system
Put simply, the main requirement of a well designed system is that it should operate
without the user beinb aware of its existence. However, this is a tall order and there is
therefore a need to more fia]]y specify some regtiireirsentc which can lend to the
`invisible system'. The following requirements are essential in achieving a safe,
usable and relnable Chmiiiiige syst,,=;
• The system shouldremove all waste as quickly -as possible
• Long horizontal pipe runs must be self cleansing
+ There must be minimal loss of water trap seal to z-nsurc there is a. bander for
the ingress of sewer gases
Other requirements which are less critical «re
a Minimal noise from the system
Minimal Odor from the appliance side (WC design)
• Ease of maintenance
Code resul xtluns were esse,.—, deslgned in order to ensure that installations meet
these requirements, and to protect iillnabitants against any possible health risks from
contact with contaminated fecal material, in developed industrialized countfics the
majority of installations meet these standards and the health risks &ci m drainage
systems are still very low. As with most fields of engineering, sanitary equipment and
techniques have benefited f-oin scientific and engineering re,e rch which lnus
improved understanding of system operation and helped develop new innovate and
cost -etTecfive ways of achieving tine foal of safe, reliable drainage systems with no
increase in health rick,
i
L
'.'as'Ffi �i `; � C: i bh';1, : a ��'' qq. '$ � tr,' ro„
,.,5 A.���. ���'tl Vt'f E.t��;/ P >:"y' $. t�•rii7 i� :'1'i:: ,}%Yf._��..
2, Pressure transients in plumbing mtents
2.1 What are pressure transjents?
Any discussion on the challenge of draining a building would be incomplete without
reference to Etlr plressure transients, but what are they? Pressufc trur113Gn1t5 axe very
simply the physical communication of a condition at one point in a system to another
point. This means that if there is an event at point A then this information is
coirJ unieatedto. point H soiLc distance amo.—, by means of a pi'ess�ure :wave, The wave
moves much faster than the air in which it travels and can move in any direction, not
necessarily in the flow direction. In a pipe the speed at which an air pressure transient
travels is the acoustic velocity, approx 1050 ft/sec. A negative transient communicates
a need for more air and. represents a suction force while a positive transient
the need to reduce the «i: flowing and represents a pushing, force. A
negative transient can be caused by air leaving the system (hence the need for more
air' and a positive transient can beCaU.Sed by iit iE11- rear i[ing a CloSe(l end Wtc7p the
air there's no escape route)
An analogy :t?a y Delp to vise ahze Eznw Chic u nrLrc in i rtiCe. hmaaill driving along a
Highway at rush hour when cars are traveling at a modest 40 MPH nose to tail. The
rC/aU is iVrlg aCEU winding Wltl a slight lnClllle, it is dark ;so JEi CQ.Ell of C[I.IliFghl$ Can
easily be seen for several miles ahead. At some point in the journey, a car, ztow out of
sight, is forced to stop. The driver is forced to apply the brakes. At this time you. are
still try--ng at 40 11APH. Up ahead. in the distance you car. see t1 brukke !fight s
illuminating as drivers respond to the event out of sight. The `wave' of brake lights
-works its `way bncit trough the raf h; until you are forced. to apply your `ui'akcs arld
stop. The illuminating lights are analogist to a pressure transient communicating to
you that there has been an event up ahead (which you can't see) and that you must
stop. This "gositiv-" t;Y- preaaurc wave trwvcla much i'aater than the 40N2H that you
were traveling at before braking (although in this case the speedof the wave is
deterr—rainedd buy the response of dElverS to ScCing brake lights tip aheau), Y'nen the road
is cleared up ahead the reverse happens as brake lights go out and drivers find
themselves with a space to drive into as the car in front moves away, Again the
ML-rmation to move !� comint:r�icated by the "rlepati 6" jVpe pressure wave.
7
iWh6ul. n,r ,!D aina e `kt.we &--tad Pt Vi is as.3.4l!'ngs-,
w p io- a.l f-' f, ti'
it IS lilicrc5ting t7 CC+il5idcl i1ic` i oTiScquciLCeS 1f t, is Car Spccii. i� inCrcziScd. if tii2 car,
were traveling at 70 WR and the first car stopped abruptly then there is a good
chance of a pile up, the driving equivalent of a Jowkowsky type pressure surge.
�Jowkow ky deternm ned that the magnitildc of a-es&ure surge is dependent on the
product of the velocity of the fluid., its density and its wave speed]
2.2 What do these pressure transients do in a building € r ainaa-Fe
system?
A negative transient will attempt to suck water out of a water trap seal. The pressure
may not be sufficient to completely evacuate the water in one go, but the effect can be
cU[n111-01ve: Positive air prey -mire transients causc .air to be forced through the water
seal from the sewer side to the Habitable space inside.
2.3 How to overcome pressure transients?
The need to communicate an increase or decrease in the air flow and the finite time
that this takes is central to the requirements of providing a safely engineered drainage,
system_ The absolute key to maintaining a state of egzlilihritlm it, a clrainaga. system 1s
to provide pressure relief as close to the source of an event as possible. In the case of
our stream of traffic above, a diversion around time roau blockage as close to the
blockage itself would cause the ininimum amount of disturbance. The point raised by
Gcorge Wating in 1884 (see Introduction above), referring to the relief of suction
presstares is still true; air miist be. pr vided as fast as possible and lorig pipe runs mean
a time delay and subsequently a possible compromise of water trap seals.
8
Euihfil 0yaitF.z4" €ylkyf3' and 1•`d.'.1# S*j—ti'te°f:ix:
S. Designing for best practice
,j.i Auev>at ng negative transients
As described above, negative transierts are the system's way of conamutnicatint:, the
need for more air. This tali for air can be caused by a number of phenomenon;
i A branch pipe tilling up with water (full bore flow) cause siphonic action to
produce a Vacuu into which the ?Plater from a trap L-;cal is sucked..
• The pressure losses associated with water falling down a vertical stack will
ind'Uze negative trGinJients wbl ilih will propagate around the uystcifa at the
speed of sound. Some of these transients can be of sufficient suction pressure
to evacuate water from a trap seal (induced siphonage).
• tljly inC:reace in airflnw (fnr wh_ita..Ver rea�C,n',n) will pr+oduce negative air
pressure transients in the system as the need for more air is co.rrinunicatedto
tiiG lGillilnCllioll `1NSSGSG liiG Clli LotLFe3 11uin
• Air leaving the system will cause a negative transient (either into the sever
or from any other interface point e.g. the top of the stack)
The most efficient way of dealing with this tali for increased airflow is to simply
answer it as uickl as possible. This means providing ttie extra air as c ui�lil as
y kr p i" y
possible. In a drainage system this equate to having a termination as close to the point
of need as possible, in effect distributed venting using AA -Vs. allows this to happen in
the. most efficient way. if a trap is 30 ft away from an air inlet to the system then it
will delay the arrival of air and quite possibly compromise a water trap seal.
If this is the case then why do people not experience foul odors on a regular basis in a
fully vented system? Well, as mentioned earlier, work carried out by Wise in Post -
War Britain, proved that ifpipework was stA to the correct slope and was of sufficient
diameter to carry required loads over a specified distance, trap seals would not be
8 (�t. E F. al. . \ s a
caliLYroxiUSed �. This system LEiie single stack yr vile pipe sys ern) I -as operated very
successfully in. Europe for 50 years with little increase in risk to system integrity.
Distributed venting provides alternatives for modern building design where distances
from, appliance to the sewer may, be longer than those antiei p--_� atecl 50 year- ago,
9
g',eyFh1'ZJ'l- .D3aip}a: 7:1-e ffratsl+''.'i tad' l''$''ni S'S;iems:
$i,`i.T.'z€'-$S:ti °€,!' d:�f,tf�`i�t;�X �,�'"J�:':',9?fi €':� z*�„+('t�'�•{t�
3.12 Atlevtating gf131`Live Pressure Tl r anstrenis
if nega ive p ress'are trail s1.-M. a—, a. call for more air then positive—.1stze tral7ulents
are a call to stop sending ail'. Because pressure transient analysis follows a set of well
d.efirxed rules (reinenfibcr there are rio meal ,Mysteries) their source can De established
and are given below;
Changes in the water/air flow rate produce positive as well as negative air
presRUC transient.,
a A sudden closure at a system termination, for example a surcharge in the
sewer, resulting in a stoppage of the airflow out of die system will cause a
positive pressure wave to be produced and propagate throughout the system
• A blockage or major clog in the system
Positive pressure transients travel at the sfulhe speed as negative pressure transients,
the speed of sound, and represent a deceleration force on air and water in its path. So,
the consequences of a positive air pressure transient reaching a water trap seal would
be that air is blown throi gh the trap t=to the buildina Oat hest) or all the water in it e
trap is forcedinto the habitable space.
It is important to note here that a positive pressure wave; produced at the base of a
drainage stack, will not be alleviated by an open top on the stack. This is because the
Pressure Wave must travel the lengtlh of the stack in order to ape the building .'.t the
top. It will meet water traps on the way which, if it has sufficient pressure, will blow
t' t tL_ 1__L.'._Li_ space.
fill So relieve t11C Jysteril 111Lo tlt4 11tdUildUiL" JpYLLG.
Again the best way to provide relief against positive air pressure transients is to locate
a pressure rclibf dc._ ,., .;uch as the PAPA a5 clove to tic source as possiblc< So in the
case of a transient produced at the base of a stack, relief is needed at the bottom, not at
the top, 1'<u41lcl vent pips vuly divcr`t a portion of the wave and will provide best
relief if the diameter of the vent pipe is equivalent to the diameter of the stack. But
this will only reduce the magnitude of the pressure by 1/1 In laboratory tests PAPAs
have been shown to reduce the m abli:.:de ^.f.a o5iti;iY. air re ssi rc tr icnt i try
p p-w.,,.�.., ans_ by 1p
10
}',
f;er'yi>t. t ,Ff., t.trk-eliE Y1ric,sx1€€e � rsghro
3.
.;tn_mm.,W_•-a-ismssa��_��.��.zmw..�.. `_�_-r-'s�.a._�— .,y+c..�,�=�-W . ......... � .. ....--Ya.r..n
90% 1€1)'t�t1. Effectively the device allows the diversion of the airflow and its gradual
�ccciciaLTV another eXutu 2 vl L f 4 c C2iri v7i Luc Lu�uway anaLv�Fy.
Do AAVs produce positive air pressure transients? Quite simply No. AAVs respond
to positive air pressure waves by closing and si,nply reflect a % of the incident wave.
AAVs will also produce a small negative transient as the inflow is closest off'.
The magnitude and ferocity of positive air pressure transients can be limited by
distributing the air venting around the building_ Since the magnitude of a positive air
press -,we waive ig a f jnctiorl of tl�e veltcity of th= nirfloW stopped, and hence airflow
rate itself, it is better to reduce the risk of stopping a large flow by installing a number
of air inlets with small airflows around the building, Mel-e y IMIRing the ixLsgnitildB of
any potential air pressure transient produced. This is best done by installing AAVs
around the building.
4. Building Case Studies
4.1 Modeling flows in drainage networks
Research and analysis of real building drainage systems is complicated by the
difLicuiiy In obtaining data froth `J'ivc' buildings. Most areas of engineering employ
soine forin of modeling teclu-lique in research and development in their `look and see'
approach to development. In DWI research there are few models capable of dealing
with the complex time. depenclent transient flows, The computer model A!R_MT is
such a model acid as far as the authors are aware, the only validated model(s)'(12),(13
1
Capable of such a complex lash, nt tLie iieiui of the AIRNET model is the
mathematical technique known as the method of characteristics. The technique allows
the propagation of waves to be predicted along the length of a pipe at different time
Steps. This is a very powerful and tinigite way to `loop and see' what iS ?Ctllally a ,
on inside a building drainage system, the simulations in this section were carried out
using AMNTI T.
.;�an���r�n3�;°�' .�.��'�t��'l:as�%"• f.+'��5;`s:i:> �'Z��i� � 4F�T-`� <��3:�`i.�d'`€�.'k.�':
'KM#.'HM.WN.kYW4W.'AS'WRARYaHfiAe.4FHCAFHAkPa:HVH,SXVIK/HA'R.WRRM1'CH0.H?.Y.4HIF.UCY4EH.R1�'b'.twA'A'.RAXWAI4NGY'K':CNdCI%WeN�SINY�.l'b.1tlY Yn'lYi(.VdS�:lY�S:JFxxY.Y-S'YAY15%YY.kiY.vvY.vwvWs'YaY.'.n]tlnbM'.9YSfi1W.WNtlMY: )HAA'M3kMY.W.4':<YUA4:tYXN
4.2 Two story building
As stated above, a two story building drainage system can operate sufficiently well
With minimal additional ventilation as longasas it is designed and installed properly.
This is borne out UV re ;renc,e to tile. installation shown iri Figures 3 and 4 Ur'ilUW. T life
building represents a fairly common bolas~ N-ith a number of bathrooms and a grorip
branch in a Kitchen 1 Wandry area_ The simulation Was run in two different scenarios.
System with Open top
2. System with an A AV at the top of the stack
t1 discharge flow, late was itiLlr ulCLLGU from, ffie Lop floor consisting of a rcornUlilG f UVd
fram a WC and a bath. This dischmArge, :ua�' Simulated from the upper floor aPA the
effect on the -water trap indicated by shading N.,=as recorded from the output data. It can
be seen froin the bar graph that little water has been last as a result of the operation of
system devices in either scenario.
'1nination
,nt pipe
WC
Tr1b seal
Figure 3 Fully vented systent with open tole and
parsillel vent pipe
2
AAV
Figure 4 Two story house with AAN's on
branches and an AAV termination at the top of
the stack
Mg.
! Etna,
e
evl
4.2 Two story building
As stated above, a two story building drairhage system can operate sufficiently well
=rith niinimal additional .rMtilati— as long as it is &sig:3ed and installed p:Upert,r
This is borne out by reference to the installation shown in Figures 3 and 4 below. The
building represents a fairly cotr morn lhoust; sviui a number of batrroi3ms and a group
branch in a kitchen l laundry area. The simulation was run in two different scenarios.
1. Systern with open top
2. System with asp AAV at the top of the stag
A discharge flow rate was simulated from the top floor consisting of a combined flow
from a WC- and a bath, This discharge was simftlated from the upper floor and the
effect on the crater trap indicated by shad'1119 Nmas recorded from the output data. It can
be seen rrom the bar graph that little water has been lost as a result of the operation of
system devices in either scenario. AAV
Cinen tertn►gtatiotl
1
pipei e
I}
I I I
t
� I
' WC
r'r I
I I
1 I
I I
Trap seal
Figure 3 Fully vented system with open top and
parallel vent pipe
2
io
0.2
0
Figure 4 Two story house with AA -Vs on
hranc lea and an AAV termination at the top 6i
thestack
Fully VeW@ d AAV.
Figure 5 Comparison or water retained in the ground floor trap indicated
(Shaded On acharaY;c}
12
4.3 10 Story Building
The .I U story building scenario is shown in F igufe 6 below. There are basically three
.t:stallatro.; UF, being slr «fated mere, the ...Y v'.., system Figure 6(a) and a e.,e
pipe system with distributed venting and an AAV on the top of the stack, Figure 6 (b).
This system also includes a relief vent. Figure 6 (G) is the one pipe sys-tem witlt
distributed AAVs and PAPAS subjected to a positive air pressure transient simulated
to replicate the occurrence of a surcharge in the sewer. In each of the scenarios a.
representative water tra,- is sh,^,v n -Ma three floors up the bu:lair.� � F
,en term inat
Cross vent
rap seal __-V
Figure 6 (a)
F'iF-ure 6 M
Iv
of
t
Figure 6 (c)
Discussion The flo.v rate used in Uhl'.- simulation represents a maximum for the 4" vertical stack
in question (80 USgpm). This flow rate is unlikely to be observed in practice as the
simultaneous discharges required are a probabilistic impossibility (-Hunter 1949). The
flow rate is therefore indicative of a `worst case scenario' in order to push the
drainage vent system to its limits, and therefore shove comparisons betwecn the
options i-t-westigaied. The discharges making up the flm a rate are distributed my enly
along the stack to simulate a number of simultaneous discharges (approximated 16
USgpm from 5 different boors).
The bar graph shown in Figure 7 illustrates the water depth retained in the shaded
water trap in Figure b following this event. It can be seen that under these conditions
13
O`%i6o a.T for effi.1..i n:✓Y i.'a�.fvf Aol
rp:n:.xc�.'Ss-zsrn: -.... ....aa.�axzar...o- use..�w.�[.'s�^v—.sA....•v.•YSC.vs�.�.�
the system with AAVs installed (Figure 6b) has retained the most eater_ Wliv is this`'
Well, the maul reason is that the iiow irk the vertical stack kndJuces a negative pressure
transient as it calls for more air. This tegaiive transient propagates to all parts of the
system 'looking for air'. The negative transient represents a suction force which will
try to ctra 7x7 water out of tile trap seal_ If the negative transient is ton great it Will suck
water out of the trap. To stop this happening, air must be provided from somewhere
else. The methods shown In Figure 6(a) and Figure 6(b) show two different ine`uiuds.
In Figure 6(a) the air must travel from the top of the stack, approximately 100ft away
(but only after the negative transient has propagated to the top of the stack first so the
round trip is approximately 200tt), A_ltern:itivPly, -ir can be provided locally by the
provision of an AAV (Figure 6(b)). In this case the round trip to is only a matter of 10
ft. This means that the air can be provided eftiiaker u aft u e ugly vented system.
The bar graph also shows the influence of cross vent diameter on vent performance.
ThP smaller vent pipe is less effective than the larger vent ptpe. due. to Increased
friction. 'This is identical to the point made by Waring in 1895 (see Introduction
above).
2
1.8
1.s
3 1.4
L
1.2
,a 1
L 0.8
a
0.6
0.4
0.2
0
Fully Vented 2 inch cross vent Fully Vened 4 inch uuss vent AAV.
Figure" Compaison aF water retained in the Iowc—st water trap
(shaded on schematic) conditions based on negative transient
14
. >zdi+ft z<? <7- !}:�<Lee P. --TfS �F,�:�'ejeficientpressure Control
Figure 8 shows the trap retention on the same trap as the result of a positive pressure
transient in the system. The ;positive, transient xas generated by simulating
surcharge in the sewer, causing the airflow through the stack to be stopped. Again two
►nethods of deaiinI with this sUrnliAfiu; inc fully vented sysk-mn s-hown in Figure o(a)
and the `active control' option utilizing AAVs and PAPAS as shown itt Figure 6(c)_
The bar graph of trap retention clearly shows that the active control system protects
against this sort cf :vent, and that the AAV systemm with a relief vent provides be t;,r
protection than the fully vented system- The reasons for active control being better
are two- fold; firstly, the distribution of the air inlets reduces the maxifr[U n positive
pressure possible in the first place and secondly, the PAPA presents a volume which
can consume the positive pressure wave, attenuate it and destroy it, rendering it
harmless. This iu bonic out by the amount of water displaced by the positive pressure
wave.
2
1.3
1.8
t.4
yN
L
1.2
�= 1
r Q.8
0.4
0,7
0
Fully vented AAV with relief vant AAW 6 PAPAS
Figure 8 Comparison of water retained in the ground floor trap indicated
{shaded an schematic) Conditions based on positive transient
5. Conclusions
This report has considered the implications for venting Hi building drainage systems.
The discussiotr has concentrated on the fiuildamental fluid, mechanics which so readily
describe the unsteady flows resulting from plumbing fixture discharges. The
description of the workings of a drainage and vent system In these terms is not nnv,
al
bbdtFi_F{i'ddffk{i(fm$t' PY':itiE<S rzEt{{� C'ii?w�ilSfE'{61`ti„
'1�,.di.3 Px5; ¢X.� i"Pt L'�g j��{.E�: ��r�`.¢��S'i 441f ��� t..•{!,{��'.=i
many early innovators were well aware of this, however, many codes and regulations
worldwide seem to avoid the engineering imperative of a description based on fluid
mechanics in favor of a prescriptive legalistic approach based on the evolution of the
industry rather than the science,
The fundamentals of system friction and pressure transient generation and
propagation are %2nual 10 understanding why venting Is required in the first place.
Possible solutions for alleviating pressure transients were discussed, including. the
well respected view that in certain circumstances systems operate perfectly well
without venting.
The advent of fast digital computers has resultedin the ability to model and simulate
unsteady air and water flows in building drainage and vent systems; providing the
capability of solving the well understood governing wave equations first described in
the 18"' Century, The computer simulation program A_IRNF'r has been under
development for over 20 years and has been validated. in malty laboratory and site
investigations. This report shows results frotri simulations of two building types; a
two storey building and a ten storey building. The output from the program confirms
the validity of distributed. venting utilizing AAVs and the effectiveness of the positive
air pressure atterrua#or (PAPA) at dealing with positive pressure transients.
It is hoped that taus paper has gone some way in de -mystifying the workings of the
building drainage and vent system `lurking' beneath the sink and floorboards, It is
also hoped that the work of those attempting to create a safe, hygienic environment
for people, fnr thefirst time, such as Waring, Putnam, Reynolds and Wise should be
remembered in a favorable light, not least because of their commitment (Waring died
as a result of investigations into a possible link between sanitation and Yycl-low fever),
but because their observations were based on the sound engineering and scientific
methods often absent fiom deliberations today.
16
�tritrditt� rraattet.;:� a•�'li.�llr t<��t€1 �isrttf,�,�vle:�ty.
6. References
Munro, B. (2000) 'Cerarrric Water Closets' Shire Publications, London,
2. Reynolds, O. (L972) `Sewer Gas and How io Keep It Out afHausW in Allen, M 'FROM
CESSPOOL TO SEWER: SAIVITARYREFORMAND THE RHETORIC OF RESISTANCE.
1848-1880' Vieforian Literature and Culture (201)21 30. 383-402 Cambridge Univervity
Press.
3. Waring, G. E. (1895) `How to drain a house, practical information for houscholder:s' D. 'Van
Nostrand corallany, New York.
4. Putnam, J, P, (191L) `.Plumbing and household .sanitation' Doubleday, Pagc & Co, Garden
City, New York.
5. Wise--, AX.B (1957) `Drainage pipework in buildings: Hydraulic design and pelybrmance'
111ASO, London
h. Swaffield J.A & Boldy A.P (1993) 1993, 'Pressure surge in pipe and duct .syylem.s', Avebury
Technical, England
7. Swaffteld J. A. and Galowin L.S., (1992) `The engineered design of building drainage
.system,y', Ashgatc Publishing LImited, England,
8. Jack L.B., (2000). Deq elopmenty iu the definition of fluid traction forces within building
drainage gent .systems', Building Services Engineering Research BSc Technology, Vol 21, No 4,
I)p266 273, 2000.
9. EN 12056:2000 `Gravity Drainage Systeoxs inside building:, Parr 2: Sanitary Pipework.
layout and calculadara', British Standards Institute, London
10, Swaftield, J.A.,Campbell, D.P. , Gormley, M. (2005) 'Pressure transient control: Fart 1 —
criteria for transient analysis and control' Building Services Engineering Research and
Technology, Volume 26, Number 2, June 2005, pp. 99- 114 (16)
I SwaffielJ, J.A.,Campbell, D.P. , Gormley, M. (2005) `Pressure transient control, Part Li —
simulation and design of a positive surge prorection device for building drainage networks'
*A
G!p eMns.fio P e Building Services Eagincering Research and Technology, Volume 26, Number 3, September
2005, pp. 195-212(18)
12. Swa.—theld J. A. and Carap'vdil D. P., 1992% "Air Pressure Transient Propagation in Building
Drainage Vent Systems, an Application of Unsteady Flow Analysis", Building and
Environment, Vol 27, ri°. 3, pp 357-365
13, Swaffield I A. and Campbell D. P,, 1992b, "Numerical modelling of air pressure transient
propagation in building drainage vent systems, including the influence of mechanical
botmdary coad-Itiom", Building and Voll 27, ii�, 4, pp 455-467
Fl.