22-Mec-B2 Environmental Control in Buildings · December 2018
Question 6 of 8: Two-branch duct system balanced against the fan characteristic
Nivaar worked solution (AI-drafted; not reviewed by a licensed engineer)
Notes on this paper
Paper format. National Examinations, December 2018 —
16-Mec-B2 Environmental Control in Buildings. Three hours, open book
(an environmental-control text and steam tables are expected; any non-communicating calculator is
permitted). Eight problems are printed and candidates solve five: Problem 1 carries
30 points, Problem 2 carries 10 points and Problems 3–8 carry 20 points each, so the printed
paper totals 160 points and a graded script totals 100. Psychrometric charts (SI and IP) and an
R-134a pressure–enthalpy diagram are attached to the paper. All eight problems are worked
below.
McQuiston, Parker & Spitler, Heating, Ventilating and Air Conditioning: Analysis and
Design, 6th ed. — the CLTD/SCL/CLF cooling-load method, duct design and the
degree-day/bin energy methods.
Jones, Air Conditioning Engineering, 5th ed. — plant psychrometry, percentage
saturation, apparatus dew point and coil by-pass factor.
Incropera & DeWitt, Fundamentals of Heat and Mass Transfer, 8th ed., Ch. 3 —
one-dimensional composite-wall conduction; Table A.3 for building-material conductivities.
Stoecker & Jones, Refrigeration and Air Conditioning, 2nd ed. — vapour
compression cycles, compressor displacement and volumetric efficiency.
ASHRAE Refrigerant Tables for R-134a (datum hf = sf
= 0 at −40 °F, the datum of the attached chart).
Canadian context: National Energy Code of Canada for Buildings (NECB 2020), Canada Green
Building Council (CAGBC) LEED v4 and Zero Carbon Building Standard, and Environment and Climate
Change Canada Canadian Climate Normals for degree-day data.
Question 6: Two-branch duct system balanced against the fan characteristic (20 marks)
bend 0.3; branch to main 0.2 (downstream $p_v$); branch to off-take 0.5 (off-take $p_v$); grille 0.4; expander 0.25 (maximum velocity); damper fully open 0.2
Fan characteristic
$P_t = 200-12V^2$ Pa, with $V$ in m$^3$ s$^{-1}$
Air density
$\rho = 1.2$ kg m$^{-3}$
Find. The total flow and the two outlet flows with the damper fully open, and then
the total flow and the damper loss factor required to make the two outlets deliver equally.
Figure 6.1 — The duct layout as dimensioned on the paper. The common run
A–B carries the whole flow; downstream of the tee at B the two legs are in parallel, so they
must lose identical total pressure while their flows add to the fan flow.
Approach. Work in total pressure. The system is one series run (fan exit expander
plus 15 m of 500 mm duct) feeding two parallel legs from B, so three conditions close the problem:
the two legs must suffer the same total-pressure loss, their flows must add to the fan flow, and the
fan characteristic must supply the sum of the common loss and the leg loss at that flow.
Set up velocity pressures. Every fitting loss is a factor times a velocity
pressure, $p_v = \tfrac12\rho(Q/A)^2$, evaluated at the area the data specifies. With
$\rho = 1.2$ kg/m$^3$ that is $p_v = 0.6(Q/A)^2$ Pa. The expander at the fan takes the maximum
velocity, which is at the smaller 400 mm × 300 mm fan exit; the expanders at
the outlets likewise take the 400 mm duct velocity rather than the larger grille velocity.
Account for the discharge. The fan draws from still atmosphere and both outlets
discharge to still atmosphere, so over the whole path the fan total pressure is entirely consumed:
$$P_t(Q_{\text{tot}})=\Delta p_{\text{common}}(Q_{\text{tot}})+\Delta p_{\text{leg}}(Q_{\text{leg}})$$
where each leg loss includes the velocity pressure thrown away at its grille, $p_v$ at the grille
area, in addition to the grille's own 0.4 factor. Omitting that exit term is the single most common
way to get this problem wrong.
Write the duct friction. The friction chart for galvanised sheet steel
($\varepsilon = 0.15$ mm) is reproduced analytically by the Colebrook–White relation
$$\frac{1}{\sqrt f}=-2\log_{10}\!\left(\frac{\varepsilon}{3.7D}+\frac{2.51}{Re\sqrt f}\right),
\qquad \Delta p_f = f\,\frac{L}{D}\,\frac{\rho v^2}{2}$$
which for these ducts agrees with the published chart to within its reading accuracy.
Assemble the two legs. Leg D leaves the tee as an off-take, so it takes the 0.5
factor on its own velocity pressure, then 15 m of friction, then the damper, the expander, the grille
and the exit. Leg C continues as the main, so it takes only 0.2 on the downstream velocity pressure,
but pays 45 m of friction and a bend:
$$\Delta p_D = (0.5+\zeta_{\text{damper}}+0.25)p_{v,400}+\Delta p_f(15\ \text{m})+(0.4+1.0)p_{v,\text{gr}}$$
$$\Delta p_C = (0.2+0.3+0.25)p_{v,400}+\Delta p_f(45\ \text{m})+(0.4+1.0)p_{v,\text{gr}}$$
Leg D is fitting-dominated and short; leg C is friction-dominated and three times as long.
Part (a) — solve with the damper fully open. Setting
$\zeta_{\text{damper}} = 0.2$ and iterating on $Q_{\text{tot}}$ until
$\Delta p_D(Q_D) = \Delta p_C(Q_C)$ with $Q_D+Q_C = Q_{\text{tot}}$ and the fan supplying the total,
the balance converges on
$$\boxed{\;Q_{\text{tot}} = 1.873\ \text{m}^3\text{s}^{-1},\quad Q_D = 1.034,\quad Q_C = 0.839\ \text{m}^3\text{s}^{-1}\;}$$
at a fan total pressure of $P_t = 200-12(1.873)^2 = 157.9$ Pa. The common run absorbs 64.2 Pa and each
leg the remaining 93.7 Pa. The shorter leg takes 23% more air, as it must. Velocities are 15.6 m/s at
the fan exit, 9.5 m/s in the 500 mm main and 8.2 and 6.7 m/s in the two branches — all within
normal low-velocity practice.
Check the pressure budget term by term. Both legs must total 93.7 Pa, and they do
so by quite different routes:
Loss
Leg D ($Q = 1.034$), Pa
Leg C ($Q = 0.839$), Pa
Branch at B
20.31 (0.5 on off-take)
5.35 (0.2 on main)
Duct friction
27.40 (15 m)
55.37 (45 m)
Bend
—
8.03
Damper, fully open
8.12
—
Expander
10.15
6.69
Grille
7.92
5.22
Exit velocity pressure
19.79
13.04
Leg total
93.69
93.70
Common run (expander 36.54 + friction 27.67)
64.21 Pa at $Q_{\text{tot}} = 1.873$
Fan total pressure
157.90 Pa = 64.21 + 93.69
Part (b) — throttle the damper for equal outlets. The damper sits in leg D,
which is the leg that naturally takes more air, so closing it is the right adjustment. Impose
$Q_D = Q_C = Q_{\text{tot}}/2$. Leg C now has no adjustable element, so it alone fixes the operating
point: solving $\Delta p_C(Q_{\text{tot}}/2) = P_t(Q_{\text{tot}})-\Delta p_{\text{common}}(Q_{\text{tot}})$
gives
$$\boxed{\;Q_{\text{tot}} = 1.775\ \text{m}^3\text{s}^{-1},\quad Q_C = Q_D = 0.887\ \text{m}^3\text{s}^{-1}\;}$$
at $P_t = 200-12(1.775)^2 = 162.2$ Pa, with 57.8 Pa in the common run and 104.4 Pa available to each
leg.
Find the damper setting required. At 0.887 m$^3$/s leg D loses only 63.4 Pa with
the damper wide open, so the damper must absorb the shortfall,
$104.41-63.38 = 41.03$ Pa. Dividing by the branch velocity pressure at that flow,
$p_{v,400} = 0.6(0.887/0.12566)^2 = 29.93$ Pa,
$$\boxed{\;\zeta_{\text{damper}} = \frac{41.03}{29.93}=1.37\;}$$
The damper therefore moves from $\zeta = 0.2$ to $\zeta = 1.37$, roughly a $30\,{}^{\circ}$ blade
closure on a single-blade damper. Balancing costs 5.2% of the total flow and pushes the fan 4.3 Pa
further up its curve — both in the expected direction, which is a useful check that the solution
is not spurious.
Result
Damper fully open (a)
Outlets balanced (b)
Total flow handled by the fan
1.873 m$^3$ s$^{-1}$
1.775 m$^3$ s$^{-1}$
Outlet D (short leg, damper)
1.034 m$^3$ s$^{-1}$
0.887 m$^3$ s$^{-1}$
Outlet C (long leg, bend)
0.839 m$^3$ s$^{-1}$
0.887 m$^3$ s$^{-1}$
Fan total pressure
157.9 Pa
162.2 Pa
Common-run loss / per-leg loss
64.2 / 93.7 Pa
57.8 / 104.4 Pa
Damper loss factor required
0.2 (as given)
1.37
Split between outlets
55.2% / 44.8%
50% / 50%
Check: two readings taken from the layout drawing. First, the vertical drop of leg
D is taken as 15 m, equal to the dimensioned drop of leg C, because the drawing places both outlet
grilles at the same level and only the right-hand drop carries a dimension. If instead the D drop were
shorter, that leg would take slightly more air and the balancing damper would need to close further;
at 10 m the split becomes 1.072 / 0.826 m$^3$/s. Second, the 400 mm label applies to the main
downstream of the tee as well as to the off-take, so both parallel legs are 400 mm. The friction
chart is replaced by Colebrook–White at $\varepsilon = 0.15$ mm for galvanised steel, and
$\rho = 1.2$ kg/m$^3$ is assumed throughout.