Selecting a UPS With High Availability

- Jeff Stieman, spokesperson for TrippLite (www.tripplite.com), says:

Are you responsible for distributed networks, branch offices, retail store locations, IDF network closets, or a small datacenter?  If so, then you are responsible for either choosing or specifying the type of equipment needed for those locations.  Furthermore, you are responsible for service and support needed to keep equipment functioning. Part of the planning phase is to conduct a “needs assessment” to determine overall criticality – including redundant options – while considering up front and back end costs and budget constraints.  It is also important to determine how each piece of equipment will need to be supported after deployment and the costs of associated – even in the case of equipment failure.  Some support strategies include manufacturer onsite service level support or perhaps for increased deployments, signing a third party support or professional services contract.

One of the essential pieces of equipment to maintain critical equipment uptime is a UPS.  While there are many known factors that go into specifying a UPS (VA/wattage capacity, required runtime, electrical considerations, network management scheme, etc.), selecting a UPS with high availability can minimize, not eliminate, costs down the road.  What are some strategies or considerations you should think about when selecting a qualified UPS?  But don’t give your “final answer” just yet!!

A lesson in UPS Specification

A large retail chain with 2,000+ stores nationwide issued a very detailed UPS specification.  Part of the UPS  specification called for not only a 10kVA UPS, but also 1) a junction box, 2) a fused disconnect switch and 3) a service bypass panel – all of which required an electrician for installation.  In addition, the specification called for an expensive onsite service contract where a “factory trained technician” would be dispatched to fix, repair or replace failed equipment.  Not that bad of a plan, huh?  Well, keep reading.

The majority of the specification that was written was already in place before they re-issued it because that was how it was done in the past.  This strategy worked very well, but was way more costly to implement due to additional hardware, additional electrical work (time is money), and an expensive service contract. 

What was Tripp Lite able to offer?   

The junction box, fused disconnect switch and service bypass panel components were eliminated when using the Tripp Lite SU10KRT3U UPS.  The components were eliminated since the UPS already contains an isolation transformer that houses a “make before break” bypass switch – allowing power to bypass the UPS and battery while keeping the output panel hot and swapping out the failed component.

Here’s how it works

The “make-before-break” bypass switch is a detachable component that has been integrated with the UPS and is utilized during UPS replacement.   The bypass switch directs the power either through the UPS power module or around it.  In normal mode, the power comes into the transformer, is routed through the UPS power module and back out of the transformer.  In bypass mode, power is fed into the transformer input, out of the transformer output and to the connected equipment.  It is in bypass mode that you have the ability to replace the power module and battery.  Because of the make-before-break nature of the bypass switch, the power module can be replaced while the equipment being fed by the service panel remains on. 

The transformer is a fixed component.  Because of this, the only components that may need replacing are the power module and battery.  The battery will give notification that it needs replacing without having it affect the system, so emergency and immediate battery replacement is highly uncommon.  Because the power module is designed to fail if a surge or other anomaly damages it, there is potential for an immediate and emergency replacement.  The power module is attached to the transformer via a quick snap in connection.  Once the UPS is put on bypass, the faulty power module is removed and the replacement power module is connected.  This action can be completed and the UPS brought back online in less than three minutes.  Since no wires are touched, replacement does not need to be performed by an electrician or facilities/maintenance staff.

Since the UPS is component based and is “hot-swappable,” there’s no need to add on a junction box, a fused disconnect, or a separate bypass panel.  It also eliminated a costly service contract.   All other UPS’s require an electrician or at least a technician to be onsite to address any issue, since the UPS usually have to be taken off line and the electrical input and output service need to be disconnected in order to fix or replace.