Search

Achieving power redundancy - UPS configurations explained

n 1

Achieving power redundancy - UPS configurations explained

Huawei UPS5000-H Series

n 1

Huawei UPS5000-H Series (1)

Understanding the HTM: Modular UPS

When installing a tertiary power system, there are numerous considerations and a myriad of decisions to be made. These decisions are of utmost importance and could be the difference between an electrical infrastructure that can support all the functions of a healthcare facility and one that buckles in the face of a power failure. After exploring and understanding the power requirement, the next decision to be made is the UPS topology.

Traditionally tower UPS (also known as monolithic, static or single UPS) are used in either a single (N) arrangement or parallel configuration. Single arrangements provide the most cost effective arrangement while paralleling 2 tower systems in an N+1 or N+N arrangement provides greater resilience.

As part of the 2017 update of the HTM 06-01, section 11.58 includes guidance on how modular UPS can now be integrated into the design. In this blog, we’ll look at both advantages and disadvantages of installing tower and modular topologies into healthcare estates.

Modular UPS Overview

When it comes to UPS investment, modular UPS solutions offer a flexible and scalable approach with reduced operating costs and easier overall maintenance. UPS engineering works can be quickly undertaken due to the hot swappable modules within the frame, allowing them to be removed or replaced without disturbance to the load. This in turn can lead to a more reliable solution over the lifespan.

Great consideration should be given to the rationale behind choosing a modular UPS architecture over other topologies before the decision is made. For example, some key questions to consider are:

Do I require redundancy within the frame or within the infrastructure?
If in the frame, how many modules are required to cover the design load N (this could be theatre sockets, IPS loads etc)?
How many redundant modules are required?
If redundancy is required within the infrastructure (switchgear, dual feed load, etc.) then would a tower, or parallel tower solution be more suitable?

Understanding these key points will ensure the design, consultancy and product procurement process runs smoothly.

Modular Redundancy

UPS redundancy is central to healthcare installations as it provides resilience. Various methods of achieving resilience are discussed in section 11 of the HTM 06-01 and modular UPS is a topology that satisfies the needs of those specific requirements. Modular UPS provides redundancy within the UPS frame using integrated multiple power modules. Additional modules can be installed to increase capacity from the day one amount, to provide the required power rating (N) and additional redundancy (+1). The initial N can be made up of one or more modules.

For example, 100kW can be built using four 25kW modules and a fifth added to provide redundancy, hence N+1. However, as well as providing 100kW N+1, the five 25kW modules may be set up to supply 125kW total load.

In addition to the power modules being redundant, other critical parts of the UPS can also be modular. For example, the static switch, communication and driver cards, PSU cards, etc. This provides additional built-in resilience when compared with a tower system.

Modular UPS Have A Reduced MTTR (Mean Time To Repair)

The Mean Time to Repair is reduced due to the power modules and other components being modular. The fault finding process is also quicker as faults are quickly identified and isolated to a single module. Additionally, the physical time spent on sit can be reduced because in many cases the modules are hot swappable meaning the test and repair time required is reduced.

In comparison, a traditional tower UPS requires full isolation from the load by locking off in bypass to carry out an in depth and sometimes time consuming fault finding process. Once on bypass, components and boards then need testing, repaired or swapped over. Testing of the new components will still be required with the load in bypass before transferring back to UPS support.

Spare parts on site will also be easier to manage as parts will be complete modules rather than having to stock multiple components and PCBs. This provides peace of mind for the hospital with easier and quicker repair times, thus reducing downtimes of theatres.

Smaller Footprint

Modular UPS systems can be constructed within a smaller footprint than a traditional tower UPS. The power density and architecture mean the modular solution can reduce the footprint of the UPS by up to 50% with a reduced requirement for ventilation and maintenance access over an equivalent tower UPS.

Scalable

Modular UPS are available as either vertical or cross scalable configurations. However, within the HTM, UPS are specified to support the design load without the need for future expansion. This doesn’t mean it can’t be allowed for, but in most cases, the infrastructure is designed for a day one maximum potential load.

Reliability

As well as the topology providing internal redundancy, modular UPS by the nature of their construction uses mass-produced components in comparison to tower systems that are built in smaller numbers due to the array of power ranges requires. The smaller number of interchangeable modular components provides greater flexibility of power ranges and spare parts for multiple UPS systems.

Although there are numerous benefits of using modular UPS systems, there are further considerations that should be given before concluding that a modular UPS design is the most suitable architecture.

Price

Although the TCO of a modular UPS may be lower for a modular solution, the initial outlay costs of installing a modular UPS are more expensive than tower UPS. The increased number of mechanical and power components adds to the complexity of design and construction which has a knock-on effect on the overall cost. Replacement parts can also be more expensive when full modules need replacing in comparison to individual parts in a tower system.

Do Modular UPS provide true N+1 redundancy?

A modular UPS provides redundancy within the frame. However, this will usually mean there is still single points of failure. These will be present in the input supply, input fuses, cable terminal and infrastructure supporting a single modular frame.

To achieve true N+1 redundancy matching that of a tower system, multiple frames would need to be connected in parallel. Like a tower system, modular UPS can have additional dual input supplies or multiple battery strings to reduce points of failure. However, a single tower (modular or not) will not provide the level of resilience of multiple towers.

When considering N+1 configuration, thought should be given to the whereabouts of the required redundancy in the design.

N+1 could be:

within the load, for example, dual fed sockets in an operating theatre
within the switchgear or IPS (Automatic Transfer Switches, ATS)
Within the infrastructure
Bypass arrangement

UPS

Depending on these and other factors, a modular UPS may be more advantageous than a paralleled tower system or vice versa.

Infrastructure

Most decisions to install modular UPS are driven by the amount of redundancy achievable or the requirement of future expansion – which is why the maximum frame capacity is most likely larger than the initial design load (N). The ultimate question is, should the electrical infrastructure be designed and sized to the design load or the maximum UPS frame capacity?

Unlike modular UPS which are only usually available in 2 to 3 frame sizes of 100kW, 200kW and 300kW, tower UPS are available in conveniently stepped power ratings such as, 100kW, 125kW, 150kW, 200kW, 300kW and so on. When sizing a modular UPS, the actual required capacity may be different from the frame capacity, i.e., 100kW N+1 modular (five 25kW modules) could require a 200kW frame.

In most cases, the electrical infrastructure should be designed around the load requirement and distribution. Within a hospital, this could be the number of theatres or isolated power systems (IPS). Doing this ensures the UPS is then designed to support the load demand rather than the electrical infrastructure designed to cover the UPS maximum capacity.

Oversizing the infrastructure to support the maximum frame capacity is sometimes preferred. Although, in this case, consideration should be given to the fault clearance capacity of the modular UPS. The fault clearance depends on the day one UPS capacity and type of static bypass switch. For example, a centralised modular UPS uses a single static switch sized at the frame capacity. Even though this is a single point of failure it will have greater fault clearance capacity than a decentralised system where the static switches are housed within the individual power modules.

Consultation is required to ensure the UPS is specified correctly for the required clearance. However, as the modules are housed in one casing with single fuses, breakers and cable arrangement achieving true N+1 redundancy with a modular UPS may not be as simple or cost effective as using a Tower UPS.

Understanding the HTM: Modular UPS

Building Resilience into Healthcare Facilities

n 1

Building Resilience into Healthcare Facilities

Prisons & Correctional Facilities

n 1

Prisons & Correctional Facilities

Maintenance terms & conditions

n 1

Maintenance terms & conditions

Terms & Conditions

n 1

Terms & Conditions

What data centre tier ratings mean for UPS?

Established by the Uptime Institute (TUI) in the 1990s, the data centre Tier classification system is used to organise specific types of data centre infrastructure in a consistent way. Based on a set of operating criteria, the system comprises 4 progressive classifications starting at Tier 1 and ramping up to Tier 4. The Tiers typically take Roman Numeral form on the certification.
The Uptime Institute have left some ambiguity around how the Tiers are defined, however, they have made the most important metrics common knowledge. These metrics include, power availability, uptime guarantee, cooling capacity and concurrent maintainability, to name a few.

The classifications help data centres and customers alike to identify the level of redundancy and complexity that a data centre infrastructure provides and assists in ensuring that the correct size UPS is installed.

Looking further into the power aspects of the tiering system, having a redundant electrical path for power is crucial for achieving the higher Tier status. A data centre must satisfy the requirements of the previous Tier before moving upward to the next one

What is a Tier 4 (Tire IV) Data Centre?
To achieve Tier 4 certification, a data centre must meet all criteria set out in Tier 3 + Tier 2 + Tier 1 + additional criteria to provide customers with the highest levels of availability and 99.995% uptime per annum.

There must be zero single points of failure in a Tier 4 data centre with redundancies for every process, including the UPS (uninterruptible power supply). This is achieved by implementing a critical infrastructure with a minimum redundancy configuration of N+N (distributed redundant) or better still 2N+1, whereby there is two times the amount of power required, plus a backup supply.

Each UPS can carry the entire critical load and supports its own independent distribution system with no power connections between each one. Multiple power distribution units (PDUs) are used for distributing the electrical power between the critical load and independent UPS units. All downstream loads can be switched to one of the UPS units allowing the others to be taken offline for maintenance or load reconfiguration.

This is the most expensive option for customers to buy into and is also the most expensive data centre to design. For this reason, it is typically used by large, global businesses or public sector organisations. Customers requiring this level of facility will have the budget to sustain residence and will typically be hosting mission-critical servers.

What is a Tier 3 (Tier III) Data Centre?
With a minimum uptime requirement of 99.982% and no more than 1.6 hours of downtime per annum, a tier 3 data centre is the facility of choice for most small/medium sized businesses. It comprises all the credentials of a Tier 2 + Tier 1 + the requirement of having dual powered equipment and multiple uplinks. The critical infrastructure should be configured in a N+1 arrangement meaning it can undergo routine maintenance without affecting operations.

What is a Tier 2 (Tier II) Data Centre?
To be defined as Tier 2, a data centre must have redundancy on the most critical aspects of its mechanical structure; power and cooling distribution. This means that during planned maintenance outages, the level of uptime won’t be affected. To be certified to this level, a data centre must achieve 99.741% uptime per annum.

Tier 2 data centres are typically targeted to SMEs. They offer significantly higher levels of efficiency than Tier 1 without the expense of a Tier 3 site.

What is a Tier 1 (Tier I) Data Centre?
Considered the least reliable of data centre architecture, a Tier 1 data centre does not include any redundant on any part of its operations. These facilities allow the highest amounts of downtime, requiring only 99.671% of uptime per annum.

If a Tier 1 data centre experiences a power outage, it is likely that customers of the facility will incur downtime due to the lack of backup systems in place. Customers will also find themselves out of action at every maintenance interval as data centres under this certification are not required to have an additional power supply while the facility undergoes maintenance.

Choosing the Right Data Centre Tier
As the Uptime Institute are somewhat discrete about their ranking criteria, earning an official rank is difficult and there are no guarantees that investing in additional infrastructure will warrant a specific classification. The higher up the levels a data centre climbs, the more difficult it is to achieve the next level.

Having a UPS supplier that is knowledgeable in the classifications will make the process of designing or upgrading the data centre to the desired Tier rating smoother. It is also important to remember that It’s not just a case of installing the critical infrastructure, to ensure consistent reliability from the UPS, a maintenance schedule from a reputable UPS maintenance provider is essential.

Our modular UPS manufacturing partner, Huawei, show their continued commitment to the data centre industry by being apart of the TIER-Ready programme. Power Control is a long standing partner of Huawei, supplying, installing and maintaining their centralised modular UPS solutions. Available up to 21MW, their UPS systems are ideal for higher Tier data centres to ensure reliable redundancy.

Power Control has also partnered with modular UPS manufacturer Legrand, and stand alone UPS manufacturer, Borri SPA. To find out which topology is most suitable for your data centre’s needs, contact us 01246 431431 or email info@powercontrol.co.uk.