Written by SIMA Intelligent Buildings
Most chilled water plants use variable speed pumps with a constant differential pressure (DP) setpoint. Many use an improved DP Reset approach. On paper, this sounds efficient. In practice, pump efficiency depends entirely on where the operating point lands on the pump curve—and that point moves more than most people realize.
To understand why, let’s walk through a typical simplified chilled water plant configuration and observe how the pump operating point shifts under different conditions.
Chiller Plant Configuration (Simplified Baseline Scenario)
Assume the following common configuration:
- Variable speed chilled water pump
- DP sensor located at the main supply/return headers
- Constant DP setpoint
- Two-way control valves at coils
- A minimum flow bypass valve between supply and return, modulating to maintain chiller’s minimum flow
We will analyze pump behavior strictly from a hydronic and pump-curve perspective.
Case 1 — Fully Loaded Building
All valves open, bypass closed
At or near peak load:
- Terminal and AHU valves are predominantly open
- The bypass valve is fully closed as load flow is higher than chiller’s minimum flow
- Flow demand is driven by the building, not by protection constraints
Hydraulically:
- The system has a single dominant system curve
- The pump operates at the intersection of: The full-speed pump curve The full-load system curve
📌 This is one fixed operating point on the pump curve:
- High flow
- Moderate head
- Often reasonably close to the Best Efficiency Point (BEP)
This operating point becomes the reference point for all other conditions.
Chart 1 Pump curve showing full-load operating point with all valves open and bypass closed
Case 2 — Partial Load
Valves partially closed, bypass still closed
As building load decreases:
- Control valves begin to modulate closed
- System hydraulic resistance increases
- The bypass valve remains closed (chiller’s minimum flow not yet reached)
The pump control system responds by:
- Reducing pump speed to maintain the constant DP setpoint
From the pump’s perspective:
- Head remains approximately constant
- Flow decreases
📌 On the pump curve:
- The operating point moves horizontally to the left
- It now intersects: A lower-speed pump curve A new system curve representing higher resistance
Important clarification:
The operating point is not sliding along the original system curve. Each valve position creates a new system curve.
Chart 2 Operating point shifting horizontally to the left across pump curves under constant DP, bypass closed
Case 3 — Minimum Flow Condition
Valves partially closed, bypass modulating open
At sufficiently low load:
- Building demand falls below minimum allowable flow
- The bypass valve begins to open
- A low-resistance parallel path is introduced
Hydraulically:
- Effective system resistance drops as bypass valve modulate open
- Flow is now governed by minimum flow requirements, not building demand
Under constant DP control:
- Head is constrained by the DP setpoint
- Flow is constrained by the bypass logic
📌 On the pump curve:
- The operating point moves primarily vertically
- Up or down across different pump speed curves
- At approximately constant flow
This is a critical distinction:
Once minimum flow is enforced, the pump is no longer responding to load. It is responding to a protection constraint.
This vertical movement often happens at very low efficiency contours since minimum chiller flow is typically between 40-50% of nominal flow. Hence the operating point is located significantly to the left of the chart.
Chart 3 Operating point moving vertically under constant DP once minimum flow bypass is active
Introducing DP Reset
Reducing the operating envelope
Now consider a DP reset strategy, where the DP setpoint is reduced as load decreases.
With DP reset:
- Artificial head is reduced
- Valve throttling is minimized
- The pump and system curves both shift
- All partial load operating points are obtained at a lower break horse power thus reducing electrical consumption
This is why DP reset often improves chiller plant efficiency.
Chart 4 Operating points under DP reset at lower BHP
The Critical Assumption Behind “DP Reset Works”
DP reset performs best only when three conditions are met:
- The full-load operating point is near the center of the pump’s BEP.
- The distribution system is hydraulically well balanced, allowing a wide DP reset range.
- The chiller’s minimum flow is relatively close to the system’s nominal flow.
When these conditions exist, DP reset can keep most operating points within high-efficiency regions, resulting in stable performance and consistent energy savings.
In practice, these assumptions are rarely true and even more rarely verified.
Key takeaway
DP reset narrows the operating envelope, but it does not guarantee high efficiency. It preserves efficiency only when the pump and system are properly aligned to begin with.
Pump efficiency is not determined by:
- VFDs
- Constant DP
- Or DP reset alone
It is determined by:
- Where the operating point sits on the efficiency map
- And how that point moves as system conditions change
Why this matters
Two chilled water plants can:
- Use the same pumps
- Use the same DP reset strategy
- Serve similar buildings
And still exhibit very different pumping energy performance.
The difference is not the sequence—it is where the system operates on the pump efficiency map.
What’s next in the series
In the next articles, we will explore:
- Cooling tower efficiency
- Chiller efficiency
- And ultimately, How Ai can optimize a system with so many moving variables
By the end of the series, it will be clear why continuously tracking operating points across multiple interacting curves is better suited for AI-driven optimization than for static, rule-based control logic.
Written by SIMA Intelligent Buildings
Articles in this Series:
