TK Lab Blog ↗
O044 / UNIT 47G-01 / Open bench / 52mm

INDIVIDUAL TEST RECORD · O044 · UNIT 47G-01

FAN
Single fan
CPU PROFILEStock settings

The cooler and each fan link to separate profiles. The air duct is a test accessory, and the result belongs to the complete configuration above.

R23 ScoreN/ANo ranking data
R23 Sustained Power160.7WOpen bench · #64 / 76
AIDA64 Heat Dissipation160.7WOpen bench · #56 / 75
NoisePendingAdded after controlled sound testing
MEASURED PERFORMANCE PROFILECompared within Open bench
  1. 01
    R23 SustainedLast-five-minute average power
    160.7 W#64 of 76 · P17
  2. 02
    AIDA64 CoolingLast-five-minute heat dissipation
    160.7 W#56 of 75 · P27
  3. 03
    Cooling per HeightAIDA64 cooling divided by total height
    3.09 W/mm#14 of 75 · P83

Farther from the center means a higher relative result within the same test environment. Only measured metrics with at least three comparable records are included; missing values are omitted, and acoustics stays out until enough formal dBA data exists.

TEST CONDITIONSConditions saved per record
Test ID
O044
Cooler ID
47G-01
Test time
2026-07-04, 13:22:00 (UTC+8)
Height
52 mm
Environment
Open bench
Installation note
The VRM heatsink was removed for this installation to avoid motherboard interference.
CPU operating profile
Stock settings
Fan
Fan adapter
XTIA 92 mm-to-120 mm fan adapter (for AXP90)
CPU
Ryzen 9 9950X3D
Ambient
28°C
PWM
100%
Thermal paste
ARCTIC MX-4

S5M COMPONENT TEMPERATURES

Component temperatures

Averages from the final five minutes of AIDA64, with maxima shown separately. These values reveal overhanging airflow and in-case heat soak and are not folded into the CPU cooling rank.

VDD VRM71.5°CMax 78°C
Motherboard39.5°CFinal-five-minute average
DIMM37.2°CMax 39.2°C
Chipset71.4°CFinal-five-minute average
Still warming at the endMaximum late-run sensor slope was 5.13°C/min. The direction of change is useful, but exact deltas remain provisional.
Test-condition noteThe heat-pipe ends of this cooler interfere with the test motherboard's VRM heatsink, so the VRM heatsink was removed for these runs. The VRM and nearby components therefore lack their normal passive cooling; the reported temperatures may be higher and are for observation only, not for direct comparison with tests that retained the VRM heatsink.
Full component-sensor data and conditions
SoC VRM Avg
52.6°C
MISC VRM Avg
60.8°C
SSD Avg
42.9°C
CPU_FAN Avg
2,068.8 RPM
Sampling window
Final five minutes of AIDA64
GPU load
No GPU stress load
Panel / duct
Documented

INSTALLATION PHOTOS

Installation photos

3 media items
01 / 03
IMAGE

AXP-90/R Full Copper Graphene with a SilverStone HA1220H12SA-Z fan, installed using an XTIA 92 mm-to-120 mm fan adapter for AXP90.

Media IDM220Captured2026-07-04 13:23LicenseCC BY 4.0CreditTK LabView O044 →View original →