The WD650K is an affordable PSU made by one of China’s largest PSU OEMS, Huntkey. This PSU is popular in the Chinese market but not so popular in markets outside Asia. I was able to spot it on Newegg at the crazy price of $188! Read the full review to learn more about this PSU and check what is available in the Asian market.
The Huntkey WD650K won’t be included in my best ATX v3.x & PCIe 5.x PSU picks article.
China’s market is full of PSU products, many of which are not available in the rest of the world. One of these products is the Huntkey WD650K, which, according to its maker, is 80 PLUS Gold certified. Back in the day, Huntkey planned to have a strong presence outside of China, but after a while, it dropped these plans and focused entirely on the Asian market. This is a shame because this OEM had some innovative designs and is currently a large PSU and mostly power adapters manufacturer, probably one of the largest in China.
This PSU is 80 PLUS Gold certified, but in the Cybenetics scheme, it scores a Silver rating at 115V, and doesn’t even get a rating at 230V because of the increased vampire power. Its noise certification at 115/230V is Standard++, meaning that its average noise output is within 30-35 dBA.
Look at our best ATX v3.1 PSU guide before buying your next PSU.
ATX v3.1 Mid-Capacity PSU Reviews:
- 1st Player NGDP 750W PSU Review
- ASUS TUF Gaming 750W Bronze PSU Review
- be quiet! Power Zone 2 750W ATX v3.1 PSU Review
- Cooler Master GX III 750W ATX v3.0 PSU Review
- DeepCool PN750D ATX v3.1 PSU Review
- Enermax Revolution D.F.12 750W PSU Review
- Sharkoon Rebel P20 750W ATX v3.1 PSU Review
- Thermaltake Toughpower PF3 750W PSU Review
- Thermaltake Toughpower GF A3 750W PSU Review
- Thermaltake Toughpower GF3 ARGB 750W PSU Review
- Thermaltake Toughpower SFX 750W Gold PSU Review
- Thermaltake Toughpower GF3 750W ATX v3.0 PSU Review
- XPG Core Reactor II VE 650W PSU Review
- XPG KYBER 750W ATX v3.1 PSU Review
- XPG Core Reactor II VE 750W PSU Review
- FSP VITA GM 750W ATX v3.1 PSU Review
- Manufacturer (OEM): Huntkey
- Max Power: 650W
- Cybenetics Efficiency: [115V] Cybenetics Silver (87.245%)
- Noise: Cybenetics Standard++ (30-35 dB[A])
- Compliance: EPS 2.92
- Operating Temperature (Continuous Full Load): 0 – 50 °C
- Alternative Low Power Mode support: Yes
- Power 12V combined: 650W
- Number of 12V rails: 1
- Power 5V + 3.3v: 100W
- Power 5VSB: 12.5W
- Cooling: 120mm Sleeve Bearing Fan (AS12025H12)
- Semi-Passive Operation: ✗
- Modular Design: Yes (Fully)
- High Power Connectors: 2x EPS (1x cable), 2x PCIe 6+2 pin (1x cable)
- Peripheral Connectors: 4x SATA (2x cables), 3x 4-pin Molex (2x cables)
- ATX Cable Length: 530mm
- EPS Cable Length: 650+50mm
- 6+2 pin PCIe Cable Length: 500+80mm
- Distance between SATA/4-pin Molex: 120mm
- In-cable capacitors: No
- Dimensions (W x H x D): 150 mm x 85 mm x 150mm
- Weight: 1.45 kg (3.2 lb)
- Warranty: 3 years
- Street Price in China (excluding VAT): around $50
Power Specifications
| Rail | 3.3V | 5V | 12V | 5VSB | -12V | |
| Max. Power | Amps | 15 | 15 | 50 | 2.5 | 0.3 |
| Watts | 100 | 600 | 12.5 | 3.6 | ||
| Total Max. Power (W) | 650 | |||||
It appears the unit failed to meet the Cybenetics Gold level solely due to 5VSB efficiency. This raises concerns about the potential for misleading conclusions, particularly when that parameter is not included in 80 PLUS certification.
Additionally, the loading values used in your test differ significantly from those defined by the 80 PLUS program. While 80 PLUS includes the -12V rail at 300mA, your data shows discrepancies of 1–2A on the 12V, 3.3V, and 5V rails. These are not trivial differences—they are large enough to prevent any fair or technically valid comparison between results. If the loading methodology is not consistent, any efficiency conclusions drawn across programs are fundamentally flawed.
In the interest of transparency and data integrity, I would be open to independently testing this unit to conduct an inter-laboratory comparison. This could help validate methodologies and clarify any discrepancies in the results.
First of all:
Cybenetics is ISO 17025 AND ISO 17065 compliant, so it already conducts essential inter-laboratory comparisons.
Cybenetics DOES have a lab on its own and doesn’t use outside labs.
Cybenetics reports are ISO 17065 and ISO 17025 compliant. 80 PLUS reports don’t even mention the equipment used!
-12V rail isn’t use for many years now and doesn’t play any significant role.
Why should someone follow 80 PLUS load values? Also, while 80 PLUS uses only 3 load levels, Cybenetics uses more than 1450 different ones.
Cybenetics, in case you are confused, has its own independent rating system. It doesn’t have anything to do with the flawed 80 PLUS rating system, which anyone can easily fool.
About this specific PSU:
It passes ErP only because Cybenetics allows a 5% margin of efficiency.
It doesn’t get a 230V rating because of the increased vampire power.
It doesn’t get an ATX v3.1 compliance because of the short hold-up time.
No ALPM support. T3 >150ms
Also, who are you to independently test the unit and provide inter-laboratory comparisons? Do you own/have an ISO 17025-compliant lab? Moreover, let’s entertain the idea that you do have an ISO 17025 lab. This is NOT enough because you have to use this system to run inter-laboratory comparisons.
https://www.eptis.bam.de/eptis/Security/login
Please do your homework before you try to challenge Cybenetics. I am not saying that they are perfect because nobody is, but they at least know their work WELL!
Aris,
Thank you for the detailed reply. My intent was not to challenge Cybenetics’ credibility but to highlight key differences in methodologies that can cause confusion when cross-referencing results between rating systems, especially among consumers or reviewers who may not fully understand the distinctions.
I fully recognize that Cybenetics and 80 PLUS operate independently with different scopes, metrics, and testing protocols. That said, when both certifications are presented side-by-side without clarifying those fundamental differences, such as the inclusion of 5VSB in one and not the other, it opens the door to misleading comparisons, particularly in ISO/IEC 17025 contexts where measurement traceability and equivalence are key.
To clarify: 80 PLUS conducts testing exclusively through ISO/IEC 17025-accredited laboratory. Our equipment used is tracked and documented internally as part of our compliance framework, even if not publicly listed on test reports. The point about load values isn’t a demand that Cybenetics follow 80 PLUS protocols, it’s a recognition that differing load conditions (especially when varying by amps across key rails) will inevitably yield different results, making side-by-side comparisons non-equivalent without proper context.
My offer to conduct independent testing was not intended to diminish Cybenetics’ capabilities, but rather to initiate a collaborative effort to improve transparency and understanding between our approaches. I welcome the opportunity for a joint inter-laboratory comparison between the 80 PLUS lab and Cybenetics, with the goal of identifying whether differences in reported results stem from test methodology, certification criteria, or actual product performance.
I believe this type of collaboration would strengthen trust across the industry and ensure that consumers and manufacturers alike benefit from more informed, technically grounded assessments.
Looking forward to your thoughts and the possibility of working together.
Best regards,
Peyton Sizemore
80 PLUS Program
Dear Peyton,
Thank you for the detailed response. My suspicion that you belong to the 80 PLUS program is confirmed now 🙂 It is nice to know you!
Why do you want to discuss this on a review forum and not officially through the proper channels? HWbusters just gets data from Cybenetics, as do many other sites and PSU tier lists. I believe a possible collaboration should be discussed officially and through an official channel, not through a forum post.
About this, you mention:
“I welcome the opportunity for a joint inter-laboratory comparison between the 80 PLUS lab and Cybenetics, with the goal of identifying whether differences in reported results stem from test methodology, certification criteria, or actual product performance.”
My skepticism is that there can be no direct comparison between Cybenetics methodology and 80 PLUS. One uses >1450 different load combinations to derive the results, among others, while the other program 3-4 different load levels. The differences are already highlighted in many reviews, white papers, etc. These two are entirely different systems. There is just no comparison between them, so you cannot compare two entirely different things. PT tests, we already conduct with other labs through the EPTIS system.
For anything further you could connect with Cybenetics on their official email. Info at Cybenetics.com
Thank you for the friendly conversation! I wish you the best of success. After all, our goals are identical: a greener environment with less energy wasted!
Dr. Aris Bitziopoulos
Aris i just curious about new GAMEMAX LIONCORE PSU that certified Cybenetics, on TOMSHARDWARE COMPUTEX article they said they insured Lifetime warranty, i hope you just complete the review as soon as possible https://www.techpowerup.com/336774/gamemax-unveils-latest-innovations-at-computex-2025
hm lifetime warranty is a bad idea for any PSU, not only for these specific ones.