User manual | Leak detection systems in Complex DX equipment (VRV/VRF systems)

Leak detection systems in Complex DX equipment
(VRV/VRF systems)
Graham Wright MInstR, Daikin Airconditioning UK Ltd
Abstract
Complex DX systems have been available within the UK for nearly 30 years and their popularity has grown
due to the system flexibility, ease of installation, efficiency, and reliability. During this period however, we
have seen a significant change in the use of refrigerants, moving from R22 to R407C then to R410A. This
has been driven firstly by removing the Ozone Depleting R22 from use through to regular F-Gas inspection
for the current class of refrigerants whose Global Warming Potential is the current focus of attention.
Manufactures have updated their systems to reflect not only significant EU legislative changes but also the
requirements of local incentive programs and regulation. This has led to important modifications to the way
systems operate and the safety / prognostic features embedded within them. This paper looks at how
current leak detection processes have evolved within this type of system over the past few years and will
expand on some of the technology currently being used.
Introduction
Complex DX systems such as VRV/VRF systems have grown in popularity since their first introduction in
1985. The first of these systems were relatively small, being limited to one outdoor unit connected to a
maximum of 8 indoor units. This combined with relatively short interconnecting pipe runs limited their
application. However, it soon became apparent that one of the main issues that needed to be resolved with
this technology was ensuring that the installation was carried out correctly. This highlighted the fact that if a
system was installed badly one of the foremost problems would be leaking refrigerant from the system,
which would normally lead to several operational issues resulting in compressor reliability problems.
Equally, it was seen that a good installation would give 15 to 20 years of trouble free operational life.
To resolve these issues suppliers started to recommend that installation practices adopted in EN378 should
be used and this combined with the change to R410A more complex electronics and tougher regulations in
the form of F-Gas regulation created a step-change in how systems were installed and resulted in far
better operational results. One of the factors manufactures have focused upon to improve system reliability
are embedded leak detection systems. Where they used to rely upon low pressure switches as the main
means of detecting a lack of refrigerant within a system more complex techniques monitor the whole
system ensuring that should a leak occur it is reported and the system is shut down is a safe mode until the
fault has been resolved.
It should be noted that a lot of the information within the text is based upon data from Daikin
Airconditioning Ltd but it is understood that other manufactures operate similar systems to those
described with in this paper.
Why should we worry?
Containment of F-Gases is a legal requirement of the F-gas regulation and management of leaks and
refrigerant handling is an audited process. The F-gas regulation is being re-negotiated with the expectation
that phase downs and more stringent test for leaks will come into force.
However, there is evidence that an incorrectly charged HFC system can emit more CO2 due to
inefficiencies of the system throughout its operational life. Studies carried out by Daikin Japan (Daikin
Industries) suggest that significant impacts on efficiency can be expected if systems are under or over
charged. This can be seen with test carried out on a 3.0 kW hi-wall system with 6meter pipe run.
G. Wright
1
©Institute of Refrigeration Annual Conference 2013
Rating point
Figure 1 Effect of charge error on EER performance
Figure 2 Effect of charge error on COP performance
These results indicate that even a small error in charging a system or a system with a small leak would
directly affect the system’s predicted efficiency and, therefore, increase its CO2 emissions.
Life Cycle C02 emissions tons
6% Under Correct 3% Over
Charge
6.95
6.67
6.88
10 years
10.42
10.01
10.32
15years
Table 1 estimated CO2 emission changes due to charge variation
Whilst the differences may not be significant on this system it would indicate that similar results on larger
systems could have an important impact on the expected performance of the equipment.
G. Wright
2
©Institute of Refrigeration Annual Conference 2013
Complex DX system overview
During the 1980’s (in Europe) the market was limited to combinations of systems of between 5- 8 indoor
units connected to one outdoor condensing unit. Market demands and technology changes have seen this
grow to a combination of outdoor and indoor units that are far larger with the current limit of having
combinations of up to 64 indoor units connected to 3 to outdoor units (in one group) with up to 1,000
meters of pipe work. Whereas this is the extreme, it would indicate that designers do have a considerable
amount of latitude when specifying systems within larger buildings. An example of a typical system can be
seen in figure 1
Figure 3 Example of complex DX heat recovery system with rejected heat being redistributed to the
space and to a thermal water store as low temperature heat.
HEVAC sales statistics show that there are approximately 16,000 outdoor units sold into the UK each year
and this number is expected to follow an increasing trend. It should be noted that the indoor to outdoor
unit ratio is fairly constant at around 6:1 suggesting that the majority of systems are applied to smaller
applications or multiple systems on one site.
VRF
2010
2011
VRF outdoor
VRF indoor
OD/IN ratio
14,300 15,400
85,400 97,300
5.97
6.32
2012
15,800
101,900
6.45
Table 2 UK market data HEVAC stats 2013
With this number of systems being installed annually it is important to understand the installation process is
undertaken correctly and if there are any issues with the refrigerant containment of a system this must be
detected and remedied as soon as it’s detected. This process starts at the design / installation of a system
where the pipework runs and locations are recommended to be kept a short as possible.
G. Wright
3
©Institute of Refrigeration Annual Conference 2013
Figure 4 Standard design diagram VRV Express
Once an installation is completed the pipework is triple evacuated and pressure tested in the normal way.
The start-up process will begin with the commission engineers being left with a pipework schematic and
this is checked with the actual site installation. The actual charge for the system is then calculated. This is to
ensure that the correct amount of refrigerant is on site and to act as a double check for the auto charge
process.
Before the system can be charged it must be powered up and the internal controls within the system are
set via PCB inputs. This opens all the electronic valves and expansion device in the indoor units and enables
the system to be evacuated. The same settings are used in refrigerant recovery mode.
Figure 5 System configuration setup
G. Wright
4
©Institute of Refrigeration Annual Conference 2013
To understand the auto charging routine of this type of equipment we first of all have to recognise that the
volume of refrigerant change varies depending on the system load, mode of operation and ambient
conditions. The excess charge is stored in the outdoor unit within the refrigerant regulator, which in
effect is a sealed pressure vessel.
Refrigerant regulator
Figure 6 VRV internal schematic
This device manages the liquid gas separation and can store a large proportion of the total system
refrigerant charge. The system is also monitored by several pressure/temperature transducers which enable
the heating, cooling, and heat recovery to be closely monitored. Secondly, every indoor unit and their
refrigerant flow control devices also monitor the cooling/ heating effect that the system is producing. This
enables the entire installation to monitor its status intelligently, not only during the charging process but
also during normal operation.
The process for this type of system start-up is relatively simple in terms of field inputs. The system is
normally charged with a minimum base weight of 10 kg (this depends on the system size).The system is run
through a sequence that starts with auto charge
Figure 7 Commission process
G. Wright
5
©Institute of Refrigeration Annual Conference 2013
In auto charge the system detects the internal and external conditions along with the discharge and suction
pressures within the system. It monitors these constantly whilst the auto charge process is underway and
adds refrigerant accordingly.
When the auto charge is complete a learning process is run through for 2-3 hours followed by a full
running commission check which is carried out by software embedded within the outdoor unit that
monitors pressures, temperatures from indoor and outdoor units.
Once complete the system is able to monitor its charge automatically and basic F-Gas checking can be
undertaken using the base data created during this process.
F-gas Leak checking
Currently a service technician will have to initiate the process but future developments will allow for this
test to be carried out automatically when the system is in an off mode. In essence, once started the test
runs the system in cooling mode for 20 - 40 minutes. The results are given to the technician in the
following format and the routine can detect a loss of refrigerant larger than 500grams The results are
compared with the initial data from the self-charge process.
Figure 8 Diagram of coded output verifying refrigerant charge
Whilst this is not perfect is very unlikely that a service technician could detect a similar leak during a
normal service of a system unless of course they find it as part of a pipework inspection which should be
still carried out.
System monitoring
Monitoring of these systems has become an intrinsic part of the operational life of these systems. Even if
the Building Management options are not connected the outdoor units are always testing to ensure that the
operational refrigerant pressure and temperatures are kept within normal parameters, depending on the
mode of operation and the conditions the system is working in. Should these be exceeded the system will
report an alert. Once 5 alerts have been recorded indicating that there is an error with the charge of the
system the unit will enter a ‘Pump down mode”. This forces the refrigerant charge into the outdoor
unit’s refrigerant regulator and reports the fault to system controls as a leak detected.
Should there be a catastrophic failure this system is backed up with a normal LP switch which will shut the
system down immediately.
Whilst this function will report a leak in the normal way other prognostic services are now becoming more
common place. These connected services allow the whole building AC system to be actively monitored.
G. Wright
6
©Institute of Refrigeration Annual Conference 2013
This significantly enhances the prognostic ability as all faults and especially changes in operating conditions
of refrigerant circuits can be seen almost immediately.
Figure 9 System anomalies detected by an enhanced system monitor such as Daikin DNSS controls
Whilst systems such as this help service technicians identify a change in refrigerant charge they do not offer
building inhabitants any form of indication of a possible danger of a leak within their room or office.
Designers seem to be well aware of the issue surrounding the safety aspects of EN378 and concentration
rates but it is not unusual for these to be forgotten or misunderstood by building operators and
maintenance engineers who work with complex DX equipment. Several manufactures now offer guidance
and systems that directly notify occupants of an affected area. These alarm systems are connected to the
main BMS and the embedded controls within these types of systems that use standard HFC refrigerants
such as R410a.
•
R410a is class A1 refrigerant with a limit of 0.44 kg/m3 (as defined in Annex E) in Occupancy Class A
locations (hotel rooms, hospitals, etc.)
•
Table C1 note “d” states: Other methods of ensuring safety in the event of a sudden release of refrigerant
are permitted. Such methods should ensure that the concentrations will not rise above the practical
limits given in the normative Annex E or to give adequate warning to occupants in the space
of such a rise so that they may avoid excess exposure time. The alternative method should
demonstrate a level of safety at least equivalent to the method described in box 1 ( Max charge = practical
limit x room volume)
•
Adequate warning in BS EN378 part 4 is classified as an alarm system. Clause 7 describes this as an
audible (at least 15dB above the ambient noise) and visual alarm located within the occupied space. For
Hotels and similar establishments the alarm system shall also warn at a supervised location as well as the
occupied space.
Figure 10 information provided by a typical industry guide
Conclusion
It has been established that the design, installation, and maintenance have a critical role in reducing the
possible leaking of HFC refrigerant from DX equipment especially complex systems such as VRV/VRF.
Recent changes to controls within these systems have allowed the use of intuitive algorithms that not only
aid the installation process but also enable the system to be monitored throughout its life cycle,
safeguarding that refrigerants are used safely but also ensuring that the tools are available to make fault
diagnostics as simple as possible.
G. Wright
7
©Institute of Refrigeration Annual Conference 2013
It should also be noted that the correct charging of a system is critical to ensure the performance in terms
of cooling and heating but also the overall efficiency during its working life.
References
HEVAC 2013 statistics
Daikin Industries data on refrigerant charging linked to performance
Dakin training courses SE5 and SE6
Daikin sales information on DNSS
List of Figures and Tables
Figure 1 Effect of charge error on EER performance..................................................................................................... 2
Figure 2 Effect of charge error on COP performance................................................................................................... 2
Figure 3 Example of complex DX heat recovery system with rejected heat being redistributed to the space
and to a thermal water store as low temperature heat. ............................................................................................... 3
Figure 4 Standard design diagram VRV Express .............................................................................................................. 1
Figure 5 System configuration setup................................................................................................................................... 4
Figure 6 VRV internal schematic ......................................................................................................................................... 5
Figure 7 Commission process.............................................................................................................................................. 5
Figure 8 Diagram of coded output verifying refrigerant charge .................................................................................. 6
Figure 9 System anomalies detected by an enhanced system monitor such as Daikin DNSS controls............. 7
Figure 10 information provided by a typical industry guide.......................................................................................... 7
Table 1 estimated CO2 emission changes due to charge variation............................................................................ 2
Table 2 UK market data HEVAC stats 2013.................................................................................................................... 3
G. Wright
8
©Institute of Refrigeration Annual Conference 2013

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project