Experimental and Numerical Study for a Horizontal Wastewater Heat
Recovery System
DRAGOȘ PURGHEL, CĂTĂLIN TEODOSIU
Faculty of Building Services and Equipment, CAMBI Research Center,
Technical University of Civil Engineering,
122 – 124 Lacul Tei Bvd, Bucharest 020396,
ROMANIA
Abstract: - This study aims to numerically and experimentally investigate the performance of a drain water heat
recovery device. Consequently, a heat exchanger prototype was developed to integrate with classic rectangular
shower trays of small dimensions. The proposed drain water heat recovery (DWHR) system was tested using a
specially designed experimental stand. In addition, the experimental data were compared to numerical results
based on Computational Fluid Dynamics (CFD) simulations. The values of effectiveness and Number of
Transfer Units (NTU) achieved for the proposed DWHR system, based both on experimental and numerical
data, are similar to those from the literature for analogous configurations of horizontal DWHR systems.
Consequently, the heat exchanger prototype analyzed could represent a pertinent solution for heat recovery
from drain shower water with minimum investment costs.
Key-Words: - Heat recovery, drain water, energy economy, experimental set-up, CFD modeling, heat
exchanger effectiveness, heat exchanger Number of Transfer Units (NTU).
Received: May 19, 2024. Revised: September 5, 2024. Accepted: October 9, 2024. Published: November 26, 2024.
1 Introduction
Energy saving is a fundamental concept of
contemporary society, and it has been researched in
an exhaustive manner in the last two decades. As a
measure to reduce harmful gas emissions and
create a suitable habitat, the researchers' attention
has been directed towards identifying all
possibilities for sustainable use of resources. In this
context, it is worthwhile to mention that heating
and domestic hot water (DHW) for buildings in the
European Union (EU) represents approximately
75% of their total energy consumption, [1].
Consequently, there is a major interest in building
energy efficiency. Concerning the DWH, it is
known that the sewage system represents an
important loss of energy, [2]. As a result, the
research focussed on solutions for heat recovery
from wastewater has developed more and more in
the last 40 years, and drain water heat recovery
(DWHR) systems are constantly improving, trying
to reach their thermodynamic limits, [2]. DWHR
units are normally based on heat exchanger
devices, co-current flow, or counter flow [3],
transferring heat between hot wastewater and cold
fresh water. The two principal types of DWHR
systems are vertical and horizontal systems, [2].
Vertical DWHR units can lead to adequate
performance, with heat recovery efficiency of 20-
75% depending on the heat exchanger
configuration and working conditions (e.g.
temperatures, water flow rates), [4], [5]. However,
vertical DWHR units have limited applications as
they demand more space, [6]. Consequently,
horizontal DWHR systems have also been
developed for configurations with reduced space.
For instance, a horizontal DWHR system where the
cold-water pipe is immersed in the larger
wastewater pipe (Figure 1) has been analyzed,
reaching an efficiency of over 50%, [7].
Furthermore, a financial analysis dealing with two
solutions of horizontal DWHR systems has been
carried out, [8]. The results showed that the
configuration which leads to intensified turbulent
flow for the wastewater allows major financial
savings based on recovered energy. On the other
hand, it has been shown, based on a short review of
studies dealing with different types and
applications of wastewater heat recovery systems,
that about 50% of thermal energy could be
recovered, [2].
In this context, a prototype for a horizontal
crossflow heat exchanger to be integrated under
shower trays was designed within CAMBI
Research Center - Technical University of Civil
Engineering of Bucharest. This heat exchanger
prototype was integrated and tested using the
WSEAS TRANSACTIONS on HEAT and MASS TRANSFER
DOI: 10.37394/232012.2024.19.8
Dragoș Purghel, Cătălin Teodosiu
E-ISSN: 2224-3461
73
Volume 19, 2024
experimental setup of the CAMBI Research Center,
which is related to studies dealing with energy
recovery applications from wastewater. It is
worthwhile to mention that the experimental
studies were performed both for unbalanced and
balanced configurations (depending on the cold
water flow rate in comparison with the wastewater
flow rate), [9]. Moreover, Computational Fluid
Dynamics (CFD) numerical models were
developed for the proposed prototype of the
DWHR system, and simulations were carried out
for several configurations. Consequently, the heat
transfer effectiveness and the number of transfer
units (NTU) for the heat exchanger prototype were
experimentally and numerically assessed.
Fig. 1: Cross section of the proposed DWHR unit,
[2]
2 Methodology
2.1 Drain Water Heat Recovery Systems
The operation of a wastewater heat recovery
system is based on the use of hot water (37÷41°C)
as a primary agent to heat cold water (the
secondary agent) with a temperature of 10-15°C
within a heat exchanger. Depending on the
configuration, there are two main types of heat
exchangers: parallel or countercurrent. It is
considered that for a crossflow heat exchanger,
theoretically, an efficiency of 100% can be
achieved, while for parallel equipment the
efficiency is limited to a maximum of 50%, [9].
Fig. 2: Crossflow heat exchanger
Fig. 3: Parallel heat exchanger
Figure 2 shows a crossflow heat exchanger
while a parallel heat exchanger configuration is
represented in Figure 3.
2.2 DWHR System Proposed
As the DWHR system was designed to be
integrated under shower trays with reduced
dimensions, the geometry of the equipment consists
of a parallelepiped box measuring 610 x 460 x 78
mm (W x W x H) to which two tubes have been
added, with a diameter of 40 mm and 32 mm
respectively for the inlet/exit of grey water (Figure
4).
Fig. 4: Proposed DWHR system
Taking into account the intake and exhaust
locations of the connections, the prototype can be
considered a crossflow heat exchanger.
Perpendicular to the length of the box, nine copper
pipes were inserted (diameter of 15 mm and length
of 475 mm), which were joined at the ends with
two bigger pipes – distributor/collector (diameter of
35 mm and length of 710 mm). Finally, it should be
also noted that analyses dealing with the
optimization of lifecycle cost [10] were considered
in designing the geometry and choosing the
materials of the proposed DWHR system.
WSEAS TRANSACTIONS on HEAT and MASS TRANSFER
DOI: 10.37394/232012.2024.19.8
Dragoș Purghel, Cătălin Teodosiu
E-ISSN: 2224-3461
74
Volume 19, 2024
2.3 Experimental Setup
The integration of the heat exchanger prototype in
the experimental set-up is presented in Figure 5
(the heat exchanger is placed under the shower
tray).
Fig. 5:
Experimental setup for the proposed DWHR
system
A digital clamp-on ultrasonic flow meter
(Siemens Sitrans FUP1010) with 1011 Universal
sensors was used to measure the water flow rate
through the two sides of the heat exchanger. In
addition, 18 K-type thermocouples were used to
determine the temperatures at different points of
the DWHR system (Figure 6).
Fig. 6: Position of the thermocouples
Finally, the Almemo 5690-2 data acquisition
system was used to record the values of
temperatures.
2.4 Numerical Model
The CFD model was developed using the general-
purpose, finite-volume, Navier-Stokes solver Ansys
Fluent (version 15.0.0). The numerical model
summing up is presented in Table 1.
It is worthwhile to mention that exhaustively
grid-independent solutions analyses were carried
out, based on several meshes. An overview of the
final computational domain discretization taken
into account (2,763,722 tetrahedral cells; 6,140,803
faces; and 785,796 nodes) is shown in Figure 7.
Table 1. Main features of the CFD model
Characteristic
Description
Fluid
Water
Flow
Three-dimensional, steady state,
non-isothermal, turbulent
Computational
domain
discretization
Finite volumes, unstructured mesh
(tetrahedral elements), 2,763,722
cells
Turbulence
model
Shear Stress Transport (SST)
turbulent kinetic energy-specific
turbulent dissipation rate (k-ω),
with low-Reynolds corrections
Boundary
conditions
Inlets: water flow rate, temperature,
turbulence parameters (turbulent
intensity and hydraulic diameter)
Outlets: outflow (continuative
boundary, zero normal derivatives
at the boundary for all quantities)
Numerical
solution
Second-order upwind scheme;
Velocity-pressure coupling:
SIMPLE algorithm; Convergence
acceleration: algebraic multigrid
Fig. 7: Computational domain discretization
The justification for using the turbulence model
k-ω SST - Shear Stress Transport [11] is based on
the fact that this turbulence closure modeling was
successfully validated against different industrial
turbulent flow configurations, [12], [13].
Furthermore, the turbulence model k-ω SST
represents a pertinent choice within our CFD model
because a correct representation of both heat
transfer in the boundary layers and turbulent flows
in the regions far from solid boundaries is needed.
Steady-state simulations were carried out,
therefore there were no particular problems
regarding the stability of the numerical model.
Nevertheless, convergence criteria were strictly
monitored. The overall water flow balance within
the heat exchanger for all the simulations was less
than 0.0000002% during all iterations. In addition,
temperatures in different points within the
computational domain were observed during the
simulations to assess the convergence of the
computations. We present below the results for 4
points (Figure 8): α – at the middle of the heat
WSEAS TRANSACTIONS on HEAT and MASS TRANSFER
DOI: 10.37394/232012.2024.19.8
Dragoș Purghel, Cătălin Teodosiu
E-ISSN: 2224-3461
75
Volume 19, 2024
exchanger box (at half of its height); β – in the heat
exchanger box (at half of its height), near the
preheated water collector; γ – in the heat exchanger
box (at half of its height), close to the cold water
distributor; δ – at the middle of the preheated water
collector, in the center of collector pipe section.
Fig. 8: Convergence checkpoints
The evolution and stabilization of temperatures
in these points with the iterations for two
wastewater temperatures (37 and 41°C) are
presented in Figure 9 and Figure 10.
Fig. 9: Temperatures of points α,β,γ, and δ (inlet
wastewater temperature: 37°C)
Fig. 10: Temperatures of points α,β,γ, and δ (inlet
wastewater temperature: 41°C)
3 Results
We present below the results for the following
study case: wastewater flow rate 5 l/min and cold
water flow rate 4 l/min (unbalanced configuration).
These flow rate values were taken into account for
five different wastewater temperatures (37, 38, 39,
40, and 41°C) while the cold water temperature
was constantly around 15°C according to
measurements. The experimental and numerical
results for these five situations are presented in
Table 2.
Table 2. DWHR system: inlet/outlet temperatures
Inlet cold water: 15.60°C; Inlet wastewater:
37.00°C
Temperature
Numerical
Outlet preheated water
20.87
Outlet wastewater
26.23
Inlet cold water: 15.40°C; Inlet wastewater:
38.00°C
Temperature
Numerical
Outlet preheated water
21.26
Outlet wastewater
26.87
Inlet cold water: 15.40°C; Inlet wastewater:
39.00°C
Temperature
Numerical
Outlet preheated
water
21.55
Outlet wastewater
28.31
Inlet cold water: 15.40°C; Inlet wastewater:
40.00°C
Temperature
Numerical
Outlet preheated water
22.00
Outlet wastewater
28.12
Inlet cold water: 15.40°C; Inlet wastewater:
41.00°C
Temperature
Numerical
Outlet preheated water
22.27
Outlet wastewater
27.67
The data in Table 2 were employed to
determine the thermal performance of the heat
exchanger, in terms of effectiveness (ε) and
Number of Transfer Units (NTU).
3.1 Heat Exchanger Effectiveness
One of the most widely used approaches to analyze
the performance of heat exchangers is based on the
comparison between their real and ideal behavior,
known as “efficiency”, [14]. This concept allows
obtaining a clear image of a device’s performance
by assessing how close the analyzed system is to its
best performance. Moreover, the values of heat
exchanger effectiveness are extremely useful
WSEAS TRANSACTIONS on HEAT and MASS TRANSFER
DOI: 10.37394/232012.2024.19.8
Dragoș Purghel, Cătălin Teodosiu
E-ISSN: 2224-3461
76
Volume 19, 2024
concerning further improvements of the
investigated system. According to Eq. (1), the
thermal effectiveness (ε) of a heat exchanger is as
follows, [15]:
(1)
where:
- mass flow rate [kg/s];
- outlet/inlet cold water temperature [°C];
- inlet hot (drain) water temperature [°C];
- specific heat capacity of water [KJ/kg°C];
subscript c refers to the cold water side while
subscript min refers to the lesser of cold/hot side
quantity.
As a result, based on the Eq. (1), the DWHR
system effectiveness variation on different
wastewater temperatures is shown in Figure 11.
Fig. 11: DWHR system effectiveness
It can be observed that the DWHR unit
effectiveness increases slightly with wastewater
temperature. Furthermore, the values of
effectiveness reached by the proposed prototype are
comparable to those from the literature for similar
configurations of horizontal DWHR systems, [9].
3.2 Heat Exchanger Number of Transfer
Units
Another dimensionless parameter typically used to
assess the performance of heat exchangers is the
Number of Transfer Units (NTU). This parameter
is useful to optimize the geometric and design of
heat exchangers, considering overall heat transfer
coefficients, transfer area, fluid flow rate, and heat
capacity. NTU can be determined as follows, [16]:
(2)
NTU values obtained based on both
experimental and numerical results are shown in
Figure 12.
Fig. 12: DWHR system NTU (Number of Transfer
Units)
The data from Figure 9 indicate the same trend
(experimentally and numerically): the greater the
gray water temperature, the higher values of NTU
can be obtained.
4 Conclusions
The main objective of this study was to investigate
the thermal behavior of a DWHR prototype to be
incorporated under classic shower trays with
reduced dimensions. The values achieved for
effectiveness are experimentally around 12-17%
and numerically between 25-27% while those for
NTU are 13-20% (experimentally) and 32-37%
(numerically). These results show that the
performance achieved for the proposed DWHR
system is satisfactory, considering the prerequisites
imposed for the prototype: limited space available
under the shower tray, low-cost construction, and
simple maintenance. On the other hand, these
results indicate that the CFD model should be
improved as there are discrepancies compared to
experimental data. Excluding the uncertainty of the
experimental data (which also contributes to these
experimental-numerical differences), the numerical
model must be improved regarding the prediction
of the heat transfer at the level of the heat
exchanger tubes. The validation of the numerical
model will allow us to extrapolate the simulations
for other (improved) configurations of the DWHR
unit investigated here. For instance, further focus
WSEAS TRANSACTIONS on HEAT and MASS TRANSFER
DOI: 10.37394/232012.2024.19.8
Dragoș Purghel, Cătălin Teodosiu
E-ISSN: 2224-3461
77
Volume 19, 2024
will be on solutions concerning increased heat
transfer surface without affecting the overall sizes
of the DWHR system (e.g. introducing fins on the
pipes of the heat exchanger) or on improving flow
turbulence by adding special elements within the
heat exchanger.
References:
[1] M. Schaffer, J. Widén, J. Eduardo Vera-
Valdés, A. Marszal-Pomianowska, and T.S.
Larsen, “Disaggregation of total energy use
into space heating and domestic hot water: A
city-scale suited approach”, Energy, vol. 291,
130351, 2024, doi:
10.1016/j.energy.2024.130351.
[2] Z. Wehbi, R. Taher, J. Faraj, M. Ramadan, C.
Castelain, and M. Khaled, “A short review of
recent studies on wastewater heat recovery
systems: Types and applications”, Energy
Reports, vol. 8, pp. 896–907, 2022, doi:
10.1016/j.egyr.2022.07.104.
[3] F. Ahmed, Md. Minaruzzaman Sumon, M.
Fuad, R. Gugulothu, and A.S. Mollah,
“Numerical Simulation of Heat Exchanger
for Analyzing the Performance of Parallel
and Counter Flow”, WSEAS Transactions on
Heat and Mass Transfer, vol. 16, pp.145-
152, 2021,
https://doi.org/10.37394/232012.2021.16.17.
[4] K. Ip, K. She, and K. Adeyeye, “Life-cycle
impacts of shower water waste heat recovery:
case study of an installation at a university
sport facility in the UK”, Environmental
Science and Pollution Research, vol. 25, pp.
19247–19258, 2018, doi: 10.1007/s11356-
017-0409-0.
[5] E. Ovadia and M.H.Sharqawy, “Thermal and
Economic Evaluations of a Drain Water Heat
Recovery Device under Transient
Conditions”, Clean Energy and
Sustainability, vol. 1(1), 10004, 2023, doi:
10.35534/ces.2023.10004.
[6] H. Nagpal, J. Spriet, M.K. Murali, and A.
McNabola, “Heat Recovery from Wastewater
- A Review of Available Resource”, Water,
vol. 13, 1274, 2021, doi:
10.3390/w13091274.
[7] A. McNabola and K. Shields, “Efficient drain
water heat recovery in horizontal domestic
shower drains”, Energy and Buildings, vol.
59, pp. 44–49, 2013, doi:
10.1016/j.enbuild.2012.12.026.
[8] K. Pochwat, S. Kordana-Obuch, M. Starzec,
and B. Piotrowska, “Financial analysis of the
use of two horizontal drain water heat
recovery units”, Energies, vol. 13, no. 15,
2020, doi: 10.3390/en13164113.
[9] P. Jadwiszczak and E. Niemierka, “Thermal
effectiveness and NTU of horizontal plate
drain water heat recovery unit - experimental
study”, International Communications in
Heat and Mass Transfer, vol. 147, 106938,
2023, doi:
10.1016/j.icheatmasstransfer.2023.106938.
[10] L.A. Khan and A. El-Ghalban, “Heat
Exchanger Exergetic Lifecycle Cost
Optimization using Evolutionary
Algorithms”, WSEAS Transactions on Heat
and Mass Transfer, vol. 3(1), pp.125-136,
2008.
[11] F.R. Menter, “Two-equation eddy-viscosity
turbulence models for engineering
applications”, AIAA Journal, vol. 32, pp.
1598-1605, 1994, doi: 10.25143/3.12149
[12] F.R. Menter, “Turbulence Modeling for
Engineering Flows”, ANSYS Inc.,Technical
Paper 25, 2011.
[13] M. Grioni, S. Elaskar, P. Bruel, and A.
Mirasso, “Numerical Simulation of Turbulent
Flows using the SST-SAS Model”, WSEAS
Transactions on Fluid Mechanics, vol. 19,
pp. 24-39, 2024, doi:
10.37394/232013.2024.19.3.
[14] A. Fakheri, “Heat Exchanger Efficiency”,
Journal of Heat Tranfer, vol. 129(9), pp.
1268-1276, 2007, doi: 10.1115/1.2739620.
[15] R. Manouchehri and M.R. Collins, “An
experimental analysis of the impact of
unequal flow on falling film drain water heat
recovery system performance”, Energy and
Buildings, vol. 165, pp. 150-159,2018,
doi.org/10.1016/j.enbuild.2018.01.018.
[16] C. Zaloum, M. Lafrance, and J. Gusdorf,
“Drain Water Heat Recovery:
Characterisation and Modelling - Final
Report”, Sustainable Buildings &
Communities, Natural Resources, Ottawa,
Canada, 2007.
WSEAS TRANSACTIONS on HEAT and MASS TRANSFER
DOI: 10.37394/232012.2024.19.8
Dragoș Purghel, Cătălin Teodosiu
E-ISSN: 2224-3461
78
Volume 19, 2024
Contribution of Individual Authors to the
Creation of a Scientific Article (Ghostwriting
Policy)
- Dragoș Purghel carried out the experimental
study and developed the numerical model,
writing - original draft.
- Cătălin Teodosiu was responsible for
conceptualization, methodology, supervision,
investigation of results, visualisation, writing –
review & editing.
Sources of Funding for Research Presented in a
Scientific Article or Scientific Article Itself
This work was partially supported by the internal
grant of Doctoral School of the Technical
University of Civil Engineering of Bucharest GID-
2024.
Conflict of Interest
The authors have no conflicts of interest to declare.
Creative Commons Attribution License 4.0
(Attribution 4.0 International, CC BY 4.0)
This article is published under the terms of the
Creative Commons Attribution License 4.0
https://creativecommons.org/licenses/by/4.0/deed.e
n_US
WSEAS TRANSACTIONS on HEAT and MASS TRANSFER
DOI: 10.37394/232012.2024.19.8
Dragoș Purghel, Cătălin Teodosiu
E-ISSN: 2224-3461
79
Volume 19, 2024