Skip to main content
Category

Data Centers

Turning Cooling Waste Into Compute: Insulation as a Key Lever in Maximizing Power Usage Effectiveness (PUE)

By Chilled Water Pipe, Data Centers, Industrial Insulation, Insulation Solutions

In modern data centers, efficiency is measured in more than uptime. It’s evaluated not by the energy used to support the facility itself, but by how effectively and efficiently available power is converted into compute. This ratio is called Power Usage Effectiveness (PUE), and PUE has become the standard metric for evaluating the efficiency of data centers.

When zeroing in on PUE measurements, the industry tends to focus on cooling technologies, airflow management, and hardware optimization. It is true that these factors affect the PUE calculus; however, one critical factor is often overlooked: insulation. Specifically, insulation applied to cooling infrastructure: pipes, tanks, and distribution systems. The right insulation can directly impact how much energy is lost before it ever reaches the server rack.

Dragon Jacket Insulation manufactures data center insulation systems that offer unparalleled thermal and dimensional stability and performance that translate directly into reducing that loss,  providing your facility more usable compute per unit of power consumed. 

Understanding PUE and Where Energy Is Lost

PUE is defined as:  

Total Facility Energy Total IT Equipment Energy

In data centers, the ideal goal is to move this ratio as close to 1.0 as possible, meaning nearly all energy consumed is used for computing, not support systems.

Cooling infrastructure is one of the largest contributors to non-IT energy consumption, so improving this ratio depends on how efficiently you can cool the vast numbers of servers, storage devices, networking equipment, and security hardware so that they provide continuous operation. Moreover, inefficiencies in these cooling systems don’t just occur at the chiller or air handler; they occur throughout the system, including:

  • Chilled water distribution piping
  • Chilled water buffer tanks
  • Chiller Distribution Units (CDUs)
  • Heat exchange systems
  • Backup cooling loops

In vast acres-wide campuses, the cooling infrastructure can be massive. Without proper insulation, these systems experience heat gain, forcing cooling equipment to work harder and to consume more energy to maintain target temperatures.

The Hidden Inefficiency: Heat Gain in Cooling Systems

In chilled water systems, every degree of unwanted heat gain translates into:

  • Increased load on chillers
  • Higher energy consumption
  • Reduced cooling capacity available for IT equipment

The energy lost to inefficient cooling not only negatively impacts PUE, but increases the risk that equipment is not being properly cooled, which can compromise equipment performance. The cost of this inefficiency is especially relevant in large-scale or distributed data centers where the cooling infrastructure can span miles of piping runs. Common causes of this inefficiency include:

  • Inadequate or degraded insulation
  • Gaps around fittings and valves
  • Moisture intrusion reducing thermal resistance
  • Inconsistent installation quality

These issues create a system where energy is continuously wasted instead of being allocated toward compute power.

Insulation as a Direct Lever on PUE

Improving insulation performance is one of the most direct ways to reduce cooling losses without modifying core mechanical systems. Key benefits include:

  • Minimizing Heat Gain: High-performance insulation reduces thermal transfer from the surrounding environment into chilled systems, preserving cooling capacity.
  • Maintaining Stable Supply Temperatures: Effective insulation ensures that chilled water reaches its destination at the intended temperature, reducing variability across the system.
  • Reducing Equipment Load: When less heat enters the system, chillers and pumps operate more efficiently, lowering overall energy consumption and protecting equipment from excessive strain.
  • Supporting System Balance: Uniform insulation performance across pipes, fittings, and tanks prevents localized inefficiencies that can disrupt the efficiency of the entire system.
  • Preventing PUE Drift: High performance insulation resists degradation from environmental factors, which can gradually decrease R-value and lead to PUE drift.

Why Fit and Consistency Matter in Data Centers

In data center environments, precision is critical. Even small inconsistencies in insulation can create measurable inefficiencies. But traditional insulation methods often introduce variability through:

  • Field-cut materials
  • Compression from jacketing deformation
  • Gaps at pipe supports
  • Inconsistent thicknesses

These problematic issues become amplified in complex systems with:

  • Dense piping networks
  • Multiple connection points
  • Limited access areas
  • Thousands of feet of piping runs

In contrast, prefabricated insulation systems address these challenges by delivering:

  • Engineered components for each system element
  • Tight, repeatable fit across the entire network
  • Consistent thermal performance from installation and for decades onward

Cooling Infrastructure: Key Applications for Dragon Jacket Prefabricated Insulation

Prefabricated insulation systems such as those offered by Dragon Jacket Insulation are ideal for applications that require consistent thermal performance and resistance to damage and degradation from the elements. Each Dragon Jacket insulation unit is an engineered monolithic clamshell component that clamps onto pipes and fittings with a custom fit, secured by a steel band.  

Eschewing traditional field-fabricated layering methods, each Dragon Jacket insulation component is designed using 3D modeling, and is composed of a rigid foam cell core completely encapsulated by a water-impermeable polyurea shell. Our insulation components are thermally and dimensionally stable, making the system ideal for critical infrastructure applications, including:

  • Chilled Water Piping: Maintaining low temperatures throughout distribution is essential for efficient heat removal from server environments.
  • Buffer Tanks: Insulation helps preserve stored cooling capacity, reducing the need for continuous chiller operation.
  • Valves and Fittings: These are common points of thermal loss; precision-fit insulation ensures continuity across the system.
  • Pipe Supports and Complex Geometry: Prefabricated solutions ensure full coverage without gaps, even in difficult-to-insulate areas.

Moisture Control: A Critical Factor in Cooling Efficiency

In chilled systems, condensation is more than a maintenance concern or irritant; it is a fundamental performance issue. Moisture intrusion can:

  • Degrade insulation effectiveness
  • Increase thermal conductivity
  • Lead to long-term system inefficiencies

Encapsulated, moisture-resistant insulation systems help maintain consistent performance by preventing water ingress and protecting thermal properties over time.

Turning Efficiency Gains Into Compute Capacity

The relationship between insulation and PUE is straightforward: 

Less heat gain → Lower cooling demand → Reduced facility energy use.

This reduction in non-IT energy consumption improves the PUE ratio, effectively shifting more power toward compute operations. In practical terms, this can result in:

  • Increased available IT load without additional power input
  • Reduced operational costs per unit of compute
  • Improved sustainability metrics

For data center operators, these gains can be achieved without major system overhauls, making insulation a high-impact, low-disruption optimization.

In short, Dragon Jacket prefabricated insulation allows data centers to operate with minimal tolerance for downtime. Our insulation systems yield tangible lifecycle benefits, including:

  • Long-term insulation performance without degradation
  • Minimal maintenance requirements
  • Easy access for system modifications
  • Consistent thermal performance over time
  • Simplified removal and reinstallation
  • Reduced need for frequent inspection or replacement

These factors contribute to sustained efficiency gains throughout the facility lifecycle, which translates directly to improved PUE.

DJI: A Practical, Direct Approach to Improving PUE

While advanced cooling technologies certainly play a major role in efficiency strategies, the simple solution of an engineered prefabricated insulation solution can provide immediate PUE gains without the need to retrofit your entire cooling infrastructure. Effective, non-degrading insulation provides a foundational layer of performance that supports every other system in the cooling chain regardless of which cooling technologies you employ.

By addressing heat gain at the source — across pipes, tanks, and distribution systems — facilities can:

  • Improve cooling efficiency
  • Reduce energy waste
  • Maximize the proportion of power used for compute

Maximizing PUE is ultimately about achieving one goal: reducing wasted energy and increasing the share of power dedicated to IT operations. Insulation, when properly engineered and installed, directly supports this objective. By minimizing heat gain, stabilizing system performance, and reducing cooling loads, high-performance insulation helps transform wasted energy into usable compute capacity.

If you are looking for ways to improve PUE in your data center, there’s a simple, straightforward method that can provide immediate results: using a more effective and efficient insulation system. Contact Dragon Jacket Insulation to explore how prefabricated insulation solutions can enhance cooling efficiency and support your facility’s PUE performance goals.

Exterior Chilled Water Pipe Insulation

By Data CentersNo Comments

Exterior Chilled Water Pipe Insulation

A framework for conserving HVAC energy in Data Centers

 

Read the full white paper here.

 

Contents:

Executive Summary 3

Why Exterior Pipe Insulation Matters in Data Centers 3

Standards-Based Methodology 3

How Exterior Insulation Systems Degrade 4

Fibrous Insulation (Fiberglass, Mineral Wool) 4

Rigid Insulation (Calcium Silicate, Perlite, Phenolic) 4

Flexible Elastomeric Foam (NBR/EPDM-Based) 5

Estimated Energy Impact Ranges 5

How Dragon Jacket Insulation Differs 6

Key Differentiators 6

Final Takeaway 6

References 7

Appendix A — Energy Impact Sensitivity Analysis 8

Exterior Chilled Water Insulation in Data Centers 8

Purpose 8

Industry-Accepted Sensitivity Ranges 8

HVAC Energy Envelope for a 5-MW IT Facility 8

Conserved HVAC Energy by Insulation System 8

Conserved HVAC Energy — Northern Virginia & Texas 9

Interpretation 9

Appendix B — Installation Labor Methodology (Touch Labor Basis) 9

Estimating Framework 9

Purpose 9

Touch-Labor Definition 9

Why Touch Labor Is the Correct Comparison 10

Appendix C — Touch-Labor Installation Comparison 10

NoVA & Texas Case Studies 10

Touch-Labor Units (Published) 10

Case-Study Installation Scope (Both Locations) 10

Touch-Labor Totals 11

Touch-Only Duration (Single 2-Person Crew) 11

Theoretical Timeline Reduction (Touch-Only) 11

Final Integrated Conclusion (Appendices A–C) 12

 

 

 

 

 

Executive Summary

Exterior chilled water (CHW) piping in data centers located on rooftops or in mechanical yards represents a persistent and often under-quantified source of avoidable HVAC energy consumption. While pipe insulation is universally specified, real-world degradation mechanisms cause installed systems to underperform laboratory assumptions over time.

This paper outlines:

  • A standards-based thought process for evaluating exterior pipe insulation performance
  • Material-specific degradation mechanisms for fibrous, rigid, and elastomeric insulation systems
  • Industry-consistent estimates of conserved HVAC energy achievable by mitigating insulation degradation
  • How Dragon Jacket Insulation’s composite system is engineered to address exposure conditions that conventional insulation systems struggle to manage

The goal is to evaluate HVAC energy consumption in exterior CHW piping using assumptions that align with accepted engineering practice and industry norms.

Why Exterior Pipe Insulation Matters in Data Centers

In most data centers, cooling energy represents a significant portion of total facility energy use. Guidance from ASHRAE TC 9.9 and DOE publications consistently show HVAC energy commonly representing ~25–40% of total site energy [5], [6], depending on climate, architecture, and efficiency.

Exterior chilled water piping contributes to HVAC energy consumption through:

  • Sensible heat gain into the chilled water loop
  • Increased chiller and compressor energy
  • Additional pumping energy due to elevated return temperatures
  • Control instability in optimized facilities

While pipe losses are a subset of HVAC energy, degradation of insulation performance can materially increase these losses over time.

Standards-Based Methodology

The analytical foundation for this discussion is established via:

  • ASHRAE Handbook – Fundamentals, Chapter 23 (Insulation for Mechanical Systems)
    Governs accepted methods for estimating heat gain into insulated piping, explicitly recognizing that installed performance differs from laboratory values due to joints, aging, and exposure [1].
  • ASTM C680
    Provides standard practice for estimating heat gain or loss from insulated piping systems [2] using effective surface conditions.
  • ASHRAE Standard 90.1
    Establishes minimum insulation thickness for chilled water piping systems [3].
  • DOE/NREL Data Center Guidance
    Supports translating avoided thermal load into avoided electrical consumption using plant efficiency (COP or kW/ton) assumptions [5], [7].
  • ISO 2241

Provides internationally recognized calculation rules for thermal insulation systems [4].

This paper uses these references to justify ranges, not point values, consistent with accepted engineering judgment.

How Exterior Insulation Systems Degrade

Fibrous Insulation (Fiberglass, Mineral Wool)

Primary degradation mechanisms

  • Moisture absorption and retention
  • Loss of vapor barrier and jacketing integrity
  • Wind washing and convective bypass
  • Compression and mechanical damage

Industry-accepted performance impact

  • Wet or compromised fibrous insulation can lose ~20–40% of effective R-value consistent with ASHRAE guidance regarding moisture sensitivity of fibrous materials [1].
  • Thermal conductivity may increase by 1.5–3× in severe moisture exposure

Estimated energy impact

  • ~2–6% of HVAC energy
  • ~0.6–1.8% of total site energy (assuming HVAC ≈30% of site energy)

Fibrous systems are well understood and widely specified, but are highly sensitive to moisture and jacketing integrity in exterior service.

Rigid Insulation (Calcium Silicate, Perlite, Phenolic)

Primary degradation mechanisms

  • Joint separation and discontinuities
  • Thermal cycling and microcracking
  • Jacketing failure and radiation effects
  • Aging and k-value drift (notably for phenolic)

These materials are often described as “moisture tolerant,” but still experience effective heat-gain increases due to non-uniform contact and exposure effects. These performance effects are consistent with installed-condition considerations discussed in ASHRAE fundamentals [1].

Industry-accepted performance impact

  • ~10–35% increase in effective heat gain over design assumptions over service life

Defensible energy impact

  • ~1–5% of HVAC energy
  • ~0.3–1.5% of total site energy

Rigid systems typically perform more consistently than fibrous systems outdoors, but still rely heavily on workmanship and long-term jacketing integrity.

Flexible Elastomeric Foam (NBR/EPDM-Based)

Primary degradation mechanisms

  • UV and ozone attack
  • Seam and adhesive failure
  • Compression set at supports
  • Thermal aging and surface cracking

Unlike fibrous systems, elastomeric foams do not usually fail through bulk moisture saturation but are highly dependent on UV protection and detailing. Exposure of polymeric materials is addressed in ASTM G154 [9].

Industry-accepted performance impact

  • Well-protected systems: ~5–15% heat-gain increase
  • Typical rooftop installations: ~10–25%
  • Poorly protected systems: ~25–50%

Estimated energy impact

  • ~1–4% of HVAC energy
  • ~0.3–1.2% of total site energy

A practical constraint is availability: most elastomeric pipe insulation systems are limited to ~12–14 in nominal pipe size, making them difficult to deploy on large data-center CHW mains without extensive field fabrication.

Estimated Energy Impact Ranges

Insulation type Typical degradation driver % HVAC energy % total site energy
Fibrous Moisture + air movement ~2–6% ~0.6–1.8%
Rigid (CalSil, Perlite, Phenolic) Joints, aging, jacketing ~1–5% ~0.3–1.5%
Elastomeric UV, seams, compression ~1–4% ~0.3–1.2%

These values are intentionally conservative and derived using ASHRAE and ISO heat-transfer calculation frameworks applied to representative exterior piping conditions [1], [4], [10].

How Dragon Jacket Insulation Differs

Dragon Jacket Insulation was developed specifically to address the failure modes common to exterior piping, rather than optimizing only laboratory thermal performance.

Key Differentiators

Composite System Design

  • Prefabrication process ensures proper fitment
  • Closed-cell insulation core for stable thermal resistance
  • Fully encapsulated polyurea outer jacket providing monolithic protection

Exterior-Specific Engineering

  • UV-stable, impact-resistant exterior surface
  • Near-zero water absorption (ASTM C272) [8]
  • Eliminates wind washing and convective bypass
  • Protects joints, fittings, and irregular geometry consistently

Performance Implication

Rather than relying on “best-case installed conditions,” Dragon Jacket is designed so that installed performance closely matches long-term field performance, even in harsh environments.

From an energy perspective, the value proposition is not higher initial R-value alone, but avoiding the 10–40% performance erosion commonly assumed for conventional systems over time.

Final Takeaway

From an engineering standpoint:

  • Exterior chilled water pipe insulation degradation typically represents ~1–6% of HVAC energy depending on material and exposure.
  • This corresponds to ~0.3–2% of total site energy in many data centers.
  • These losses are persistent, continuous, and often invisible in short-term monitoring.
  • Systems engineered specifically for exterior exposure offer a credible pathway to conserved HVAC energy without relying on aggressive assumptions.

The opportunity is not theoretical; it is grounded in standards, field experience, and conservative engineering judgment.

 

 

Appendix A — Energy Impact Sensitivity Analysis

Exterior Chilled Water Insulation in Data Centers

Purpose

This appendix bounds the HVAC energy conservation potential associated with exterior chilled-water (CHW) insulation by evaluating sensitivity to:

  • Power Usage Effectiveness (PUE)
  • HVAC share of total site energy
  • Insulation system performance drift in exterior service

The objective is to present realistic ranges, not point claims.

Industry-Accepted Sensitivity Ranges

Based on ASHRAE TC 9.9 guidance and industry surveys [6], [7]:

  • PUE range: 1.25 (highly efficient) to 1.60 (less optimized / hotter climates)
  • HVAC share of site energy: 25% to 40%

These bounds capture the majority of operating data centers without overstating losses.

HVAC Energy Envelope for a 5-MW IT Facility

Annual IT energy:
5 MW × 8,760 hr = 43,800 MWh/yr

Scenario PUE HVAC % Total Site Energy (MWh/yr) HVAC Energy (MWh/yr)
Low bound 1.25 25% 54,750 13,690
Mid case 1.35 30% 59,130 17,740
High bound 1.60 40% 70,080 28,030

These ranges align with DOE and ASHRAE published guidance for typical data center operations [5], [6], [7].

Conserved HVAC Energy by Insulation System

(Exterior degradation avoidance only)

These percentages represent avoidable HVAC energy attributable to insulation performance degradation using ASHRAE and ISO-based heat gain calculation methods [1], [4].

Insulation system Avoidable HVAC energy (% of HVAC)
Fibrous (exterior service) 2–6%
Rigid (CalSil / perlite / phenolic) 1–5%
Elastomeric foam 1–4%
Dragon Jacket Insulation 0.5–2%

Conserved HVAC Energy — Northern Virginia & Texas

Northern Virginia (mid case HVAC ≈ 17,740 MWh/yr)

System Conserved HVAC energy (MWh/yr) Avg kW
Fibrous 355–1,064 41–122
Rigid 177–887 20–101
Elastomeric 177–710 20–81
Dragon Jacket 89–355 10–41

Texas (mid case HVAC ≈ 22,230 MWh/yr)

System Conserved HVAC energy (MWh/yr) Avg kW
Fibrous 445–1,334 51–152
Rigid 222–1,111 25–127
Elastomeric 222–889 25–102
Dragon Jacket 111–445 13–51

Interpretation

  • Exterior insulation degradation represents tens to low hundreds of kW of continuous HVAC load in a 5-MW data center.
  • Systems engineered for exterior exposure reduce both magnitude and uncertainty of long-term losses.
  • Dragon Jacket’s value is primarily loss prevention, not reclaiming already-lost energy.

Appendix B — Installation Labor Methodology (Touch Labor Basis)

Estimating Framework

Purpose

This appendix defines the labor accounting framework used for schedule comparison between insulation systems.

All labor values presented are touch labor only, to maintain consistency across materials.

Touch-Labor Definition

Included

  • Hands-on installation of insulation system components
  • Fit, close, fasten, band, seal
  • Straight sections, support covers, fitting covers

Excluded

  • Mobilization between work areas
  • Lift repositioning and walking time
  • Weather, access delays, escorts
  • Staging logistics
  • Rework due to congestion or damage

This boundary mirrors how base labor units are defined prior to job-factor adjustments in standard estimating practice. Standard mechanical estimating practice follows this separation of base labor and job factors [10].

Why Touch Labor Is the Correct Comparison

  • Mobilization and access penalties vary by site, not by insulation system
  • Touch labor isolates system complexity and step count
  • Allows transparent scaling and later application of job-specific factors

Appendix C — Touch-Labor Installation Comparison

NoVA & Texas Case Studies

Touch-Labor Units (Published)

Straight Pipe Touch Labor (MH/LF)

System 10″ 16″
Fibrous + jacketing 0.05–0.08 0.06–0.10
Rigid + jacketing 0.06–0.09 0.07–0.11
Elastomeric + UV 0.04–0.07 0.05–0.09
Dragon Jacket 0.03 0.03

Fitting Touch Labor (MH/each)

System MH/part
Fibrous + jacketing 0.08–0.18
Rigid + jacketing 0.10–0.22
Elastomeric + UV 0.08–0.18
Dragon Jacket 0.05

Dragon Jacket fittings install nearly as quickly as straight sections (~3 minutes per part), eliminating the traditional fitting labor penalty.

Case-Study Installation Scope (Both Locations)

  • Total exterior CHW pipe: 2,500 LF
  • Diameter mix:
    • 40% @ 10″ → 1,000 LF
    • 40% @ 16″ → 1,000 LF
    • 20% smaller sizes (averaged)
  • Support spacing: 15 ft
    • Supports ≈ 167
  • No valves/tees/reducers (main headers)

Touch-Labor Totals

Dragon Jacket

  • Straight pipe: 2,500 × 0.03 = 75 MH
  • Supports: 167 × 0.05 = 8 MH
  • Total: ~83 MH

Elastomeric (mid-range)

  • Straight: ~163 MH
  • Supports: ~22 MH
  • Total: ~185 MH

Fibrous (mid-range)

  • Straight: ~188 MH
  • Supports: ~22 MH
  • Total: ~210 MH

Rigid (mid-range)

  • Straight: ~213 MH
  • Supports: ~27 MH
  • Total: ~240 MH

Touch-Only Duration (Single 2-Person Crew)

Assuming 16 MH/day.

System Total MH Crew-days Work-weeks
Dragon Jacket 83 5.2 ~1.0
Elastomeric 185 11.6 ~2.3
Fibrous 210 13.1 ~2.6
Rigid 240 15.0 ~3.0

Theoretical Timeline Reduction (Touch-Only)

Relative to Dragon Jacket:

  • vs Elastomeric: ~55% reduction
  • vs Fibrous: ~60% reduction
  • vs Rigid: ~65% reduction

These reductions are driven purely by system complexity and step elimination consistent with documented labor efficiency gains associated with reduction of installation steps [10].

Final Integrated Conclusion (Appendices A–C)

For a representative 5-MW data center in Northern Virginia or Texas:

  • Exterior insulation degradation represents ~1–6% of HVAC energy if unmanaged
  • Dragon Jacket reduces long-term losses to ~0.5–2% of HVAC energy
  • Touch-labor installation effort is reduced by ~55–65%
  • Typical exterior CHW installation shifts from ~2–3 weeks to ~1 week of touch labor

Dragon Jacket’s value proposition is durability-driven energy stability combined with materially shorter installation timelines, evaluated on consistent engineering boundaries.

 

 

References

[1] ASHRAE, ASHRAE Handbook—Fundamentals, Atlanta, GA: American Society of Heating, Refrigerating and Air-Conditioning Engineers, 2017, Chapter 23, “Insulation for Mechanical Systems.”

 

[2] ASTM International, ASTM C680 – Standard Practice for Estimate of the Heat Gain or Loss and the Surface Temperatures of Insulated Flat, Cylindrical, and Spherical Systems by Use of Computer Programs, West Conshohocken, PA: ASTM International.

 

[3] ASHRAE, ANSI/ASHRAE/IES Standard 90.1 – Energy Standard for Buildings Except Low-Rise Residential Buildings, Atlanta, GA: ASHRAE.

 

[4] International Organization for Standardization (ISO), ISO 12241 – Thermal Insulation for Building Equipment and Industrial Installations — Calculation Rules, Geneva, Switzerland.

 

[5] U.S. Department of Energy, Data Center Energy Best Practices Guide, Washington, DC: DOE.

 

[6] ASHRAE Technical Committee 9.9, Thermal Guidelines for Data Processing Environments, Atlanta, GA: ASHRAE.

 

[7] National Renewable Energy Laboratory (NREL), “Measuring Data Center Efficiency – Power Usage Effectiveness (PUE),” Golden, CO.

 

[8] ASTM International, ASTM C272 – Standard Test Method for Water Absorption of Core Materials for Structural Sandwich Constructions, West Conshohocken, PA.

 

[9] ASTM International, ASTM G154 – Standard Practice for Operating Fluorescent Ultraviolet (UV) Lamp Apparatus for Exposure of Nonmetallic Materials, West Conshohocken, PA.

 

[10] Johns Manville, 3E Plus® Insulation Thickness Calculation Program Documentation, Denver, CO.