Seasonal shoulder months require periodic chiller use. Manage startup carefully to save money.

Last Thursday was a beautiful winter day in the Bay Area, with temperatures on the last day of January reaching 68 degrees — just about the point at which most buildings have to start their chillers. A warm day in January is not surprising. What is surprising is that 15 minutes of chiller use can drive 20% of your energy bill.

Let’s take a closer look at how two Bay Area buildings performed last week. The buildings are similar in size, construction date, use, and cooling system design. First, the demand curve from Building A:

Daily electric energy data demand curve for a chiller in San Francisco for a single day

Building A is typical. The chiller starts at 1:30pm in response to rising temperatures. Water in the system is warm, and the powerful chiller quickly ramps up to satisfy the cooling demand. The result is a 75kW spike at 1:45, after which demand falls back to a steady state level. With demand costs at $12/kW, that 15-minute spike costs the building an extra $900!

Now let’s contrast that with chiller startup in Building B:

Chart of a chiller electric energy data demand curve for start-up sequence in San Francisco

This building uses demand-limiting features on the chiller itself to control spikes at startup. Notice the smooth ramp to a fixed demand level for the entire time of operation. Chillers are actually more efficient at partial loads, so in addition to reducing demand charges, demand-limiting a chiller improves overall performance, especially with condenser water reset on mild days.

Demand charges are a grenade going off in your utility bill. Starting your chiller is pulling the pin.

Note also that Building B starts its chiller much earlier than Building A. Starting early helps reduce overall load from equipment not affected by chiller demand-limiting, such as pumps and towers. While the demand savings are nice, how do they stack up against the excess use costs incurred by running the chiller for three extra hours? The short answer is: very favorably. Building A suffered a $900 demand spike. The three hours of extra operation cost Building B only $37 in use charges. In fact, as long as you avoid more than a whopping 72 hours of excess chiller run time per month (over 3 hours per working day!) building B is ahead of the game by starting the chiller early. Especially during shoulder seasons, when chiller use is sporadic, the early startup strategy pays back handsomely.

These examples are timely, but the basic operational insights apply throughout the year:

  • Chillers set demand peaks. One of our customers describes a demand charge as a grenade going off in your utility bill. If that’s the case, starting your chiller is pulling the pin.
  • Demand is a year-round issue. Chiller startup is relevant even in the winter. Chillers have massive power requirements, and are typically sized for the hottest days of the year.
  • Chillers have demand-limiting capabilities. Most chillers have demand-limiting functionality that can help. Consult your manuals and talk to your service rep. Just be cautious not to override the minimum levels.
  • Look for offsetting loads. The chiller is only part of the load of the chilled water plant. Some building operators report success in further reducing demand at startup by slowing fans and tower operation while the chiller starts.
  • Know your rates. The examples above are for PG&E territory. Check your rates and help your team understand the tradeoff between demand and use.
About Tom Arnold

Tom Arnold is co-founder and CEO of Gridium. Prior to Gridium, Tom Arnold was the Vice President of Energy Efficiency at EnerNOC, and cofounder at TerraPass. Tom has an MBA from the Wharton School of Business at the University of Pennsylvania and a BA in Economics from Dartmouth College. When he isn't thinking about the future of buildings, he enjoys riding his bike and chasing after his two daughters.

0 replies on “Managing chiller startup”

You may also be interested in...