The Building and the Problem
1111 Broadway is a 572,000 sq ft, 24-story Class A office tower built in 1992 and refreshed in 2019. It sits directly above the 12th Street BART station in Oakland, California. After the 2019 renovation, the building earned LEED Gold certification, including a modern Siemens Desigo Building Automation System (BAS).
On paper, the building was efficient. New windows, LED lighting, re-commissioned HVAC. But like most office buildings, HVAC fan energy — driven by Variable Air Volume (VAV) systems maintaining supply air pressure across dozens of zones — was still a significant draw.
The core inefficiency: Static pressure optimization is invisible work.
🤖
Run this analysis on your building
Our AI agents use the same methodology. First query free — no credit card. Try the Agent →
In a typical VAV system, the AHU maintains a fixed discharge static pressure. If VAV boxes open wide, pressure spikes; if they close, pressure drops. The system compensates by ramping the fan motor, burning extra energy to maintain that fixed setpoint.
The opportunity: What if the setpoint itself could adapt in real time?
The Technology: Siemens Dynamic VAV Optimization (DVO)
Siemens DVO is a machine learning overlay on standard VAV control systems. It continuously optimizes:
- Supply air static pressure setpoint — lowered when demand is low, raised only when needed
- Supply air temperature setpoint — coordinated with humidity and occupancy to minimize reheat
How DVO Learns
DVO ingests real-time airflow data from each VAV box, supply and return air temperatures, ductwork static pressure, occupancy patterns, and historical HVAC performance. A neural network learns the minimum static pressure required to satisfy all zones without losing comfort — often 30-50% lower than conservative fixed setpoints.
Key Technical Features
- Green Mode: Prioritizes comfort + efficiency (typical operation)
- Defence Mode: Temporarily increases ventilation + pressure to mitigate airborne transmission
- No hardware retrofit: Runs as a software overlay on existing BAS
- Automatic tuning: The model self-corrects if a VAV box fails or a valve sticks
Benchmarked Performance: 35% Fan Power Reduction
| System Type | Baseline | With DVO | Reduction |
|---|---|---|---|
| Systems without pressure reset | 100% | 75% | 25% savings |
| Systems with basic pressure reset | 100% | 90% | 10% savings |
| Average across deployments | 100% | 65% | 35% fan power reduction |
Static pressure reduction: DVO lowers average AHU discharge pressure by ~57%.
Why the Wide Range?
- Buildings with no prior optimization see the largest gains (25-35%)
- Buildings that already had basic pressure reset show marginal gains (5-15%)
- 1111 Broadway, recently refreshed, likely sees 10-15% incremental savings
Critical transparency: Siemens has not published site-specific energy savings for 1111 Broadway. The 35% figure is the portfolio average across DVO deployments.
Implementation: Overlay, Not Retrofit
- Baseline measurement: 6 months of pre-DVO HVAC performance captured
- Model training: DVO learned 1111 Broadway's VAV dynamics
- Parallel operation: 4 weeks of optimized setpoints logged alongside existing control
- Switchover: DVO setpoints became the active control layer
- Monitoring: Continuous feedback with comfort alerts
Zero downtime. Zero hardware changes. Pure software integration.
Estimated Results at 1111 Broadway
| Metric | Estimate | Basis |
|---|---|---|
| Fan power reduction | 10-35% | Depends on existing control sophistication |
| Annual fan energy savings | 150-400 kWh/1000 sf | Typical office: fans = 40-50% of HVAC energy |
| Cost savings (annual) | $20K-$55K (estimated) | At Bay Area electricity rates |
The Bigger Picture: Defence Mode and IAQ
Beyond energy, DVO's Defence Mode can shift toward higher air change rates during periods of higher aerosol transmission risk — without permanently consuming the energy cost of high ventilation. For a Class A office tower above a major transit hub, this is meaningful.
Lessons for Practitioners
1. Pressure Optimization Is the Overlooked Lever
Most buildings treat static pressure setpoint as static. DVO proves dynamic pressure management can cut fan energy by a quarter or more.
2. "Already Optimized" Doesn't Mean "Fully Optimized"
1111 Broadway had a 2019 refresh and LEED Gold. DVO still found savings. Incremental optimization layers stack.
3. Occupancy-Aware Ventilation Is a Solved Problem
If your BMS can measure zone occupancy, DVO can modulate ventilation accordingly. Oversized ventilation for ghost zones is eliminated.
4. Green Mode vs. Defence Mode Is a New Paradigm
Optimized for energy by default, resilient mode on demand. That's a story that sells to tenants.
5. Verify Baselines Are Comparable
When a vendor claims 35% savings, ask: Was the baseline system already optimized or was it conservative/broken?
M&V Note
Data Source: Siemens DVO case studies, Smart Energy Decisions articles
Verification Status: PARTIAL
The 35% fan power reduction is verified at portfolio level. Site-specific results for 1111 Broadway are not publicly disclosed. LEED Gold status verified (not Platinum, correcting earlier claims).
Data Quality: Medium confidence for 1111 Broadway; high confidence for DVO portfolio benchmarks.
Key Takeaways
- Dynamic VAV optimization can reduce fan energy by 10-35%
- 35% fan power reduction translates to ~4-8% total building energy savings
- Implementation is a software overlay — no HVAC hardware retrofit required
- Defence Mode enables flexible IAQ strategies for health-conscious buildings
- Pressure setpoint is the invisible optimization lever most buildings ignore
- Existing optimization doesn't preclude further gains — AI layers compound
- Site-specific results vary widely — use portfolio benchmarks cautiously