Power – it’s a funny thing. We rely on it so much to do such fantastic things. But, it’s also a big huge headache when trying to figure out power requirements to meet the growing demands within a company.

It usually goes something like this: We have to upgrade our server. We probably should upgrade our whole setup to deal with the power requirements. Hmm… budget’s tight. We already have server racks, each with a single or multiple UPS units installed. Could we re-arrange the power distribution and save some cash? How would that work?  Please pass the aspirin…

The good news is that you might actually be able to – with the right understanding of power redundancy. Contrary to popular belief, this isn’t strictly limited to large infrastructure installments, and can also be easily incorporated into single server-rack installations. By rearranging your current equipment, you can unlock the full potential of your existing hardware, greatly increasing your up-time, and limiting the occurrence of a power failure interruption.

Why use redundancy in power distribution?

A UPS has a big job to do – ensure reliable power delivery and protect your equipment – and they do it well. Until, they don’t. UPS units, like everything else, need maintenance and with the popular (but not advised) “set ‘em and forget ‘em” mindset, they sometimes fail. So, there’s a good chance that at some point your equipment will be offline for a period of time.

With at least two power sources, a redundant system eliminates single point failures. In the event that one source should fail (or require routine maintenance), the second will ensure your equipment is still protected and supplied with reliable power.

Here’s the “how-to”:

Your Options: Basic Power distribution models

Capacity or “N” system (no redundancy)

An N system is a system comprised of a single UPS module, or a paralleled set of modules whose capacity is matched to the critical load projection. This type of system is by far the most common of the configurations in the UPS industry. The small UPS under an office desk is an N configuration, as per Figure 1.0.

Parallel “N+1” system (redundant system)

Parallel redundant configurations allow for the failure of a single UPS module without requiring that the critical load be transferred to the utility source. A parallel redundant configuration consists of paralleling multiple, same-size UPS modules onto a common output bus.

These configurations can include additional components to better improve single point failure elimination:

• Automatic transfer switches (ATS)
• Static transfer switches (STS)
• Power distribution units (PDU)
• External maintenance bypass switches
• Additional Battery packs

 

Examples:

• If your system draws 100W, and you install a single 100W UPS, this can be described as an “N” type configuration.
• If a system draws 100W, and two 100W UPS’s are installed in parallel operation, the configuration is considered an “N+1”.
• If the load is 100W, and there are three paralleled 100W UPS’s, the configuration would be “N+2” …and so on.

It isn’t uncommon to find expressions such as “2N”, “N-1”, “2(N+1)”, but they’re generally representations of larger power distribution schemes, or lesser commonly used denotations.

Redundant configurations

Single-Corded design

This design will offer protection for equipment with a single power supply or input.
Represented by arrows in Figure 2.0, the power cords from the equipment are connected to a Power Strip or PDU within the rack installation. The Power Strip is supplied directly from an ATS which is installed as an intermediary between the UPS’s and the Power Strip. In the event of scheduled maintenance or a single UPS failure, the ATS will automatically redirect 100% of the available power from the remaining UPS to the Power Strip.

Dual-Corded design

This design will offer protection for equipment with dual power supplies.
Represented by arrows in Figure 3.0, the power cords from the equipment are connected to two separate Power Strips or PDU’s. Redundancy in this system is achieved by dedicating a UPS to each Power Strip. Under normal operation, the equipment will draw 50% of its power requirements from Power Strip 1 and 50% from Power Strip 2. With the loss of UPS1, UPS2 will continue to supply Power Strip 2 and the equipment will redirect 100% of the required power draw through the remaining active power supply.

Greater redundancy

In the pursuit of eliminating single point failures, these pointers can greatly improve upon the redundancy of a power distribution system:

• Use modular UPS’s. These devices can be easily scaled-up to improve redundancy.
• Always pair a Power Module with Battery Module within a Modular UPS.
• Install additional External Battery Packs.
• Make backup Generator power accessible to your system through the implementation of a STS or ATS.

 

Feel free to reach out to us about your specific power needs. We love this stuff.

-Eddie

Additional resource:
– Comparing Availability of Various Rack Power Redundancy Configurations