Cross-Load Tests & Infrared Images
Our cross-load tests are described in detail here.
To generate the following charts, we set our loaders to auto mode through our custom-made software before trying more than 25,000 possible load combinations with the +12V, 5V and 3.3V rails. The load regulation deviations in each of the charts below were calculated by taking the nominal values of the rails (12V, 5V and 3.3V) as point zero. The ambient temperature was between at 30°C (86°F) to 32°C (89.6°F).
Load Regulation Charts
Across much of its operating range, the RM850x delivers 90-92% efficiency.
We applied half-load for 10 minutes with the PSU's top cover and cooling fan removed before taking photos with our modified FLIR E4 camera that delivers 320x240 IR resolution (76,800 pixels).
Our IR shots show that the NTC thermistor gets quite hot (close to 82°C) after barely 10 minutes. Obviously, a bypass relay was needed to keep the thermistor cool while the PSU is in use.
Moreover, filtering caps between the +12V boards look like they run hot when the fan isn't spinning, which would be the case during normal use. At least they belong to Chemi-Con's higher-quality KZH series and not the entry-level KZE family. The bulk caps also see higher temperatures than we expected, mostly because of heat dissipated by the nearby PFC FETs and boost diode.
MORE: Best Power Supplies
MORE:How We Test Power Supplies
The combes are also relatively poor quality and don't clip together brilliantly especially as there are multiples wires used in a couple of pins.
Is there a possibility to retest one of those Evga Supernova G2 units, like G2 750, in terms of the new hold-up time tests?
I read your Leadex Gold 550 review, where you find out the hold up time of that unit (and maybe the whole platform) is questionable and not really safe. So that made me think that the rest of the platform and also the Evga units are plagued by the same important mistake.
I own a G2 750 and after reading that review and some forum threads, I'm a bit worried now.
Hope you can help.
RMx have always been better than the G2 because they're quieter.
Thanks for the answer Aris,
Do you think that these units are safe though? G3 850 seems fine on your tests. Are there dramatic differences between G2 750/850 and G3 850 in terms of bulk caps etc...? You have a rough estimation maybe?
I looked into it before. The 650 G2 most likely has the problem where it drops the PWR_OK signal after the voltages go out of spec. The 550W should be fine. I'd assume the 750W is perfectly fine also. Not something to worry much about anyway.
Pretty sure those were some of the most unreliable units in history. RMA rates above 10% I believe.