Our Team in Action: Defying Expectations, Driven by Data
Before working with Carbon Lighthouse, Tesla’s on-site team had brought in utility and engineering consulting firms to reduce energy use, while working regularly with equipment manufacturers and service providers like Trane for ongoing maintenance support.
Given these relationships, the Tesla team expected us to glean few additional savings, but we love a challenge that puts our experience, infrastructure, and processes to the test.
Our first step with any client engagement is to spend time on-site, learning about the facility’s unique operations from the people who know their building best.
The knowledge and observations of our facilities team partners are always a critical complement to the data we gather from building management systems (BMS).
But our engineers have consistently found that even the most sophisticated BMS will miss important measurements that reveal hidden savings opportunities.
In order to capture this hidden data, our engineers deploy hundreds of new sensors to track thermodynamic, electrical, physical and behavioral measurements for up to six weeks.
Evaluating the Chiller System
Working closely with Trane, the original manufacturer of the chillers, our engineers used data including kW/ton curves measured in the eld to troubleshoot the equipment issues — and the data after calibration showed significantly different performance curves than design or factory testing indicated.
A data mismatch between the BMS and the chillers threatened to jeopardize $30,000 per year in energy savings.
» During implementation, our engineers detected a different amperage displayed in the BMS versus the amperage displayed internally in the chiller’s current transducer (CT)
» As we dug into the issue, we found the errors were caused by a BMS CT being placed on the wrong side of the VFD as well as improper calibration
By looking at the plant from a holistic point of view and evaluating the operations of the cooling towers and chillers separately, Carbon Lighthouse delivered significant additional energy savings.
» To resolve these discrepancies, our engineers collected a set of clean and reliable data for an additional six weeks during implementation
» Our new baseline data enabled our team to take action and resolve issues with high confidence
» Working closely with Trane, the original manufacturer of the chillers, our engineers used data including kW/ton curves measured in the eld to troubleshoot the equipment issues. The data after calibration showed significantly different performance curves than design or factory testing indicated
» The Carbon Lighthouse Unified Engineering System (CLUES) gave our engineers deep insights into the complexities of the Tesla building’s energy use
» In addition to cutting overall energy use, we used our proprietary CLUES data and analysis to inform careful PID loop tuning and optimize interactions between the two chillers
» Using additional data not available to the BMS, our team optimized the pumping energy use against chiller use and cooling-tower fan use
» The data encouraged us to accept penalties in cooling tower energy in exchange for efficiencies in the chiller, balancing both against pumping energy
By looking at the plant holistically, in addition to carefully considering the operations of the cooling towers and chillers separately, Carbon Lighthouse delivered significant additional energy savings. Our team achieved this by optimizing the pumping energy use against chiller use and cooling tower fan use, using additional data not available to the BMS to accept penalties in cooling tower energy in exchange for efficiencies in the chiller, and both balanced against pumping energy.
Driven to Earn Compelling Results
Our standard practice is to use what we learn from each new implementation to continuously improve our platform. Energy models developed during our Tesla engagement are now integral to our platform, ready to benefit our future clients.
Prior to our Tesla engagement, our engineers already had over 400 buildings under their tool belts, and our energy modeling was well-tuned to actual operations.
Programs like Tesla’s, however, present exciting new challenges to our engineers and prompt us to create even more customized infrastructure in order to look at the data in the most detailed way possible.
Even with Tesla’s building occupancy in flux — 300 new people moved in during implementation — our in-house impact was so minimal that Tesla’s facility team actually reached out to ask how we were doing it.
Today, our ongoing services model continues to deliver optimizations at no cost to Tesla. For example, we discovered that converting 100,000+ sq ft of factory space to offices would guarantee savings that outperformed our original predictions.