Latest / NOXTalks / Why Your Energy Savings Disappear in 18 Months

Transcript

Why Your Energy Savings Disappear in 18 Months · NOXTalks

• 0:07
  SPEAKER_00: Welcome to Knox Talks, the AI-generated podcast exploring the groundbreaking ways technology shapes our built environment.
• 0:14
  SPEAKER_00: Today we're diving into something that quietly drains energy from buildings everywhere.
• 0:19
  SPEAKER_00: Sequence drift.
• 0:21
  SPEAKER_00: And if you've ever wondered why the savings from an expensive retro commissioning project just fade away after a year or two, well, you're in the right place.
• 0:30
  SPEAKER_00: So here's the setup.
• 0:31
  SPEAKER_00: You bring in a great commissioning agent, they spec out high performance control sequences, Ash Ray Guideline 36, the industry's high performance reference for HVAC control sequences, and your VAV system is humming, energy bills drop, everybody's happy, fast forward, and somehow your kilowatt hours are creeping back up.
• 0:52
  SPEAKER_00: Same equipment, same occupancy, but the savings gone.
• 0:56
  SPEAKER_00: That, my friends, is sequence drift.
• 0:59
  SPEAKER_00: And honestly, sometimes the savings were never fully there in the first place.
• 1:03
  SPEAKER_00: Lawrence Berkeley National Laboratory, LBNL, has been pretty blunt about this.
• 1:09
  SPEAKER_00: Their open building control work found that best practice sequences in large commercial buildings are often not implemented correctly or not implemented at all.
• 1:19
  SPEAKER_00: And that failure alone can create 10 to 30% energy waste.
• 1:23
  SPEAKER_00: So before we even get to drift, a huge slice of expected savings just never shows up.
• 1:29
  SPEAKER_00: Then there's the drift itself.
• 1:31
  SPEAKER_00: Even when a sequence works on day one, performance erodes.
• 1:36
  SPEAKER_00: Operators respond to comfort complaints, reset schedules get flattened, points get put in manual, temporary overrides get left in place, death by a thousand cuts.
• 1:47
  SPEAKER_00: Let me walk you through how it plays out.
• 1:49
  SPEAKER_00: Tuesday afternoon, corner office is too warm.
• 1:53
  SPEAKER_00: Your operator hops on the front end, drops a set point, maybe puts a damper in manual override.
• 1:59
  SPEAKER_00: Complaint goes away.
• 2:00
  SPEAKER_00: Great, except that override never gets released.
• 2:04
  SPEAKER_00: Then there's another, and another.
• 2:06
  SPEAKER_00: Six months later, you've got dozens of points in manual that nobody remembers touching.
• 2:11
  SPEAKER_00: And to be fair, not every override is bad.
• 2:14
  SPEAKER_00: The problem is the undocumented, long-duration, unreviewed ones that quietly become permanent operating logic.
• 2:22
  SPEAKER_00: When analysts review trended BAS data, unreleased overrides show up as a recurring fault category.
• 2:30
  SPEAKER_00: Not an edge case.
• 2:31
  SPEAKER_00: Okay, what about the sequences themselves?
• 2:35
  SPEAKER_00: LBNL case studies and simulations show real HVAC savings from guideline 36, often in the 10 to 30% range, depending on baseline system type and climate.
• 2:48
  SPEAKER_00: One case study hit roughly 30% site HVAC electricity savings.
• 2:54
  SPEAKER_00: A single zone VAV simulation came in around 17.
• 2:58
  SPEAKER_00: The savings are real, but they're case-dependent, and they only show up if the sequence is implemented and verified correctly.
• 3:05
  SPEAKER_00: And that verification is where most projects fall apart.
• 3:09
  SPEAKER_00: Because traditional commissioning is a one-time event.
• 3:12
  SPEAKER_00: You get a report, put it on a shelf, and move on.
• 3:15
  SPEAKER_00: Which brings us to MBCX.
• 3:18
  SPEAKER_00: Monitoring-based commissioning?
• 3:20
  SPEAKER_00: This is the operational answer.
• 3:22
  SPEAKER_00: You instrument the building, trend the data continuously, run automated fault detection, review overrides, verify fixes, and keep doing it.
• 3:31
  SPEAKER_00: LBNL's benchmarking across 24 UC, CSU, and investor-owned utility partnership projects reported about 10% median source energy savings, with a payback around two and a half years.
• 3:44
  SPEAKER_00: MBCX is designed to preserve savings, not just chase them.
• 3:49
  SPEAKER_00: So why isn't this everywhere?
• 3:51
  SPEAKER_00: Honestly, the playbook exists.
• 3:53
  SPEAKER_00: It's just not owned operationally.
• 3:56
  SPEAKER_00: It lives in commissioning reports, dashboards, conference decks, but it's rarely in the monthly PM schedule.
• 4:03
  SPEAKER_00: Rarely a line in the CMMS workflow.
• 4:06
  SPEAKER_00: Rarely a line item in the facility's budget.
• 4:08
  SPEAKER_00: And until somebody's accountable for it, the building slowly reverts to whatever keeps complaints down fastest, not what saves energy.
• 4:16
  SPEAKER_00: That's the gap.
• 4:18
  SPEAKER_00: So here's what I'd push you to do: make this a cadence, not a project.
• 4:22
  SPEAKER_00: Once a month, review overrides.
• 4:25
  SPEAKER_00: Pull the list, justify each one, release the ones you don't need.
• 4:29
  SPEAKER_00: Once a quarter, verify your reset sequences.
• 4:33
  SPEAKER_00: Twice a year, test economizer logic in your trim and respond loops.
• 4:38
  SPEAKER_00: And after every control's programming change, every sequence edit, controller replacement, or major front-end revision, rerun the relevant sequence checks before you sign off.
• 4:50
  SPEAKER_00: One more thing.
• 4:51
  SPEAKER_00: When you're talking to your controls contractor, ask about guideline 36 by name.
• 4:56
  SPEAKER_00: Ask how they verify the install.
• 4:58
  SPEAKER_00: If they hedge, that's a red flag.
• 5:01
  SPEAKER_00: Sequence drift isn't sexy.
• 5:03
  SPEAKER_00: It's not a shiny piece of equipment you can point at.
• 5:06
  SPEAKER_00: But poor implementation plus quiet degradation can create 10, 20, sometimes 30% energy waste.
• 5:13
  SPEAKER_00: Enough to erase the savings you thought you already paid for.
• 5:16
  SPEAKER_00: The research has been sitting there for years.
• 5:19
  SPEAKER_00: So audit your overrides, verify your sequences, trend your data, make it a routine, not a one off.