G1116-0900WNA – 900W CRPS Power Supply for Enterprise Switches, Access Platforms, and Distributed Computing Nodes

The G1116-0900WNA 900W CRPS power supply is built to support enterprise switches, access-layer networking platforms, and distributed computing nodes that demand reliable mid-range power delivery. Its compact 1U CRPS-standard footprint (185 × 73.5 × 40 mm) ensures efficient use of chassis space while maintaining front-serviceable installation. This model provides a main output of 54.5V at 16.88A, complemented by a 12V standby rail used for system management and auxiliary logic loads. With a wide input range of 90–264Vac and 180–300Vdc, it adapts easily to global AC and DC power environments. Featuring PMBus 1.2 digital control, the G1116-0900WNA supports telemetry reporting, remote diagnostics, and intelligent system-level power coordination. Its Platinum-grade efficiency contributes to lower operating temperatures and improved long-term stability under continuous workloads. Thermal performance is enhanced through adaptive fan control, which balances airflow and acoustic output to suit varying load conditions. The hot-swap CRPS architecture enables quick maintenance or replacement without service disruption, making this unit well-suited for resilient network and compute deployments.

Features

CRPS-185: 185×73.5x40mm (LxWxH)

CRPS-265: 265×73.5x40mm (LxWxH)

3A Max standby output current

Voltage Tolerance : 54.5Vdc ±3%(52.8-56.1Vdc)

Hot-plug

Full Digital control

Active Power Factor Correction

Intelligent-thermal Fan Control

N+1 Redundant

Reverse Airflow Option

Applications

POE Switch

Networking

Telecom

Artificial Intelligence

Router System

Data Centers

5G Base Stations

IoT Gateways

Approvals

UL/CUL

CB

TUV/Mark

CCC/CQC

FCC

CE

Specifications

Output Power (W): 900
Length (mm): 185
Width (mm): 73.5
Height (mm): 40
Mounting Type: Hot pluggable
Minimum Output Current (A): 0
Maximum Output Current (A): 16.8
Output Voltage (V): 54.5
Minimum Output Power (W): 0
Maximum Output Power (W): 900
Minimum Input Voltage (V): 90
Maximum Input Voltage (V): 264

Model Selection Comparison Table

Model

Power Class Use Role Input Form Factor Recommended Use
G1116-0600WNA Entry Enterprise & network compute AC/DC CRPS-185

General server roles and branch workloads

G1116-0900WNA

Mid Heavier virtualization & caching AC/DC CRPS-185 Moderate I/O and multi-VM tasks
G1116-1200WNA High GPU-assisted or dense compute AC/DC CRPS-185

Inference & intensive concurrency

G1116-1600WNA

Extreme HPC or multi-accelerator nodes AC/DC CRPS-185 Dense compute and AI workloads
G1116-2000WNA Peak Continuous heavy duty AC/DC CRPS-185

Large-scale cluster/AI backbones

Deployment Scenarios

The G1116-0900WNA sits above the 600W tier within the G1116 family, offering 900W continuous output suited for infrastructure where virtual machines, storage activity, and concurrency levels are more demanding. This capacity class fits branch-level cloud nodes, regional POP compute, monitoring clusters with heavy log bandwidth, hybrid file systems, and container platforms that require consistent power delivery without transition into GPU-heavy territory. Compared to the 0600W model, it provides more watt overhead for cache synchronization, VM scaling, data handling surges, and higher I/O duty, ensuring sustained responsiveness when workloads move beyond basic utility roles.

 

Scenario

Expected Load Pattern Why 900W Tier Fits
Mid-tier VM clusters Multi-small workloads

Higher concurrency stability

Branch POP compute

Mixed user/device load Prevents voltage dip on spikes
Hybrid NAS/SAN High storage churn

Supports rebuild and parity events

SD-WAN infrastructure

Routing + monitoring Ripple handling keeps links stable
Medium observability stack Constant write throughput

Sustains logging during peaks

Distributed DC light-cores

Shared edge workloads

More growth headroom than 600W

Power Architecture & Reliability Design

The G1116-0900WNA employs a power conversion path engineered to maintain stable voltage behavior during cache-heavy read/write patterns, synchronous virtual machine scaling, and moderate data traffic flows. Its electrical design is optimized for environments where load changes are frequent but not extreme, allowing predictable power delivery during database activity, logging services, and mixed compute workloads.

 

Ripple suppression and switching control are tuned to protect storage links and PCIe pathways when burst events occur, such as commit operations, cache flushes, or network rule rebalancing. This controlled electrical response helps reduce I/O jitter and preserves system responsiveness under overlapping storage and network activity, a common pattern in virtualized and distributed compute platforms.

 

Thermal mapping and component spacing reduce localized heat accumulation, enabling consistent performance in POP racks, branch deployments, and lightly cooled environments. PMBus access provides visibility into thermal slope behavior, fan response patterns, and real utilization distribution across uptime cycles, supporting proactive lifecycle planning rather than reactive servicing. As a result, the G1116-0900WNA is well suited for 24/7 infrastructure roles that require stable operation, moderate scaling headroom, and long-term reliability across distributed environments.

Power Operating Notes

Reference Condition

Suggested Guidance
VM clusters & orchestration

Leave buffer for workload expansion

Hybrid storage paths

Balanced fans maintain thermal health
POP deployments

Review telemetry semi-annually

NAS replication windows

Stable under parity rebuild
Routing & security loads

Expect low EMI impact

Sustained audit logging

Low ripple supports long-term disks
Edge compute

Efficient cooling recommended

Scale planning

Upgrade to 1200W for GPU needs

FAQ

Q1. Where does this PSU fit best?
Mid-tier virtualization, caching servers, regional compute POPs.

 

Q2. Is continuous load reliable?
Yes — tuned for long uptime cycles with predictable thermal behavior.

 

Q3. Difference vs 600W?
More watt margin for VM count growth and heavier storage demand.

 

Q4. Redundancy ready?
Supports N+1 architecture and hot replacement planning.

 

Q5. Suitable for SD-WAN with logging?
Yes — ripple control aids network+storage workloads.

 

Q6. Good choice for observability stacks?
Handles heavy ingest with consistent power delivery.

 

Q7. Any airflow recommendations?
Maintain unrestricted intake to avoid thermal creep.

 

Q8. When is 1200W preferable?
When GPU assisted workloads or continuous high concurrency emerge.

 

Scroll to Top