Chiller Thriller

With the last moments of warm weather dwindling and Halloween fast approaching, let’s take a look at the performance of our new Lyndhurst chiller plant over the course of an extremely hot summer. For the background info on the project, check out the Chiller Thriller Part 1 blog post. Part 2 will focus on the savings achieved and how we measured them; or, in fancier terms, our monitoring and verification (M&V) process.

I’ll start off with the AWESOME RESULTS before putting you to sleep with the nitty gritty details.


Check out those savings numbers – peak demand reduction of almost 50%, and energy consumption reduction of over 80%!! Annual cost savings of over $60,000!! Toronto Hydro has been a valuable partner for many of our electricity saving projects and this one is no exception – based on the demand reduction of 139.8kW peak, UHN has been approved for an incentive of $111,840. The rest of this blog will show how we got to these numbers.

Chillers, pump motors, and fan motors are all electric devices, so the M&V plan is entirely focused on electricity savings. The performance of the original plant was measured using a portable energy meter. The chiller performance was measured and logged over a period of one month to evaluate the range of operating conditions. The old pumps and fans were spot measured because they were running at constant speed and, therefore, their electricity consumption was fairly continuous.

The new plant has variable frequency drives (VFDs) on the chiller, pumps, and fans. The VFDs have built-in capability to report power consumption and we have these outputs connected and tracked on our building automation system. I collected all of this data from the original plant and the new plant, along with the outside air conditions, and developed a performance profile for each system according to the outside air conditions.


Interesting to note that the old plant used to run in the 10-15 Celsius range, but we have found the new plant doesn’t need to run in this range. This may be due to the fact that the new magnetic bearing chiller has no friction in the compressor and therefore no heat generation requiring cooling. The old chiller may have been running at these temperatures just to cool itself when it didn’t actually need to be running with low cooling demand in the building.

Based on historical weather data, we know how many hours occur in each of these bands on average, so we can calculate the total energy used over the course of the year. This gives us an apples to apples comparison that takes into account the effects of changing weather year over year. The savings have been so great that they clearly show up on our utility bills despite this summer being way hotter than last year. Check out the chart below!


The peak demand reduction calculation was much simpler. Using the same data, I targeted the three peak (i.e. the hottest and most humid) hours occurring during regular building occupancy for both the pre- and post-retrofit cases. The measured demand of the major plant equipment during the three peak hours was averaged to give an accurate picture of the average peak demand before and after the project.

Here’s our nice new plant as seen on our building automation system:bas-1

Next year, we should see even more savings because we are modifying the air handling unit cooling coil valves so that they operate more efficiently. The current operation uses a valve that bypasses water back to the chiller when demand is lower, but we are planning to get rid of the bypass lines so that the total flow is reduced when demand drops. This change will enable the chilled water pump to modulate its speed lower and further reduce electricity consumption. The chiller should also run slightly more efficiently because the water coming back from the AHUs will be warmer; the chiller operates more effectively at higher temperature differential.

A similarly exciting project is coming up this winter at Bickle and I look forward to reporting on that one as well. Thanks to the following for all their work on the Lyndhurst project: Angelo Suntres from UHN, Richard Rollox and staff at LC, Mark Baniuk from HH Angus, Andrew Peeters from the contractor Geo A Kelson, and Toronto Hydro for their support of the project.