One of the more important tests we do when analyzing any type of hardware is thermal testing.  We will analyze internal heatsink temperature in the course of this examination.  We will chart these examinations over 20 minutes to determine which power supplies are actually benefiting from ‘automatic’ fan adjustments.  Temperatures are in degrees Celsius.  The ambient temperature was 21.0C.

Power supplies that did not have automatic settings were tested on the highest setting.  Tests were conducted inside an industry standard Chenming case with no fans running. (Lower temperatures are better)

Heat over Time

 

5min

10min

15min

20min

Allied AL-A400ATX

29.5

30.6

32.8

34.0

Antec TrueControl 550

24.2

27.0

29.6

33.4

Antec TruePower 330

26.5

31.5

32.0

33.8

Enermax EG465AX-VE FCA 460W

30.5

33.1

36.0

44.1

Enermax EG651P-VE FMA 550W

31.2

34.6

37.5

45.0

Enhance ENS-0246 460W

28.7

30.8

32.7

36.0

Fortron FSP400-60PFN

30.0

31.4

32.9

33.8

Kingwin KWI-450WABK

25.7

27.4

29.6

30.4

PC Power & Cooling TurboCool 475

23.1

25.6

25.8

25.8

Sparkle FSP350-60PN

24.7

26.2

28.7

30.4

ThermalTake PurePower 480W

30.4

35.1

39.3

42.3

TTGI/SuperFlower 520SS 4Fan

26.2

28.4

29.4

30.5

TTGI/SuperFlower 420SS

28.4

30.5

31.0

32.1

TTGI/SuperFlower 350SS

27.9

29.4

31.1

33.6

TurboLink 420W

28.8

29.7

34.1

35.0

Vantec Ion 400W

23.9

24.9

25.2

26.8

Vantec Stealth 520W

27.5

29.1

30.3

32.4

Zalman ZM400A-APF

30.5

31.5

32.7

33.4

All of the power supplies performed well in this category, from the $35 Allied, to the $130 Enermax units.  The best performer was the incredibly loud TurboCool 475, which never made it over 26C. We were a little surprised the Enermax units stayed so hot during the test, but with some tweaking the rear fans could be better manipulated by the motherboard to provide more exhaust.

Features Noise Testing
Comments Locked

70 Comments

View All Comments

  • Anonymous User - Friday, August 1, 2003 - link

    1. No amp measurements were listed, a serious deficiency because without them there is no way to know how well each PS met its amp specs, and many brands are known to fall short.

    2. No overload testing results for shorts, excessive power draw, excessive temperature.

    3. Ripple is not just slow voltage variation also short term variation, such as for each AC cycle (60 Hz for the incoming AC, about 60,000 Hz for the output DC). I would have liked to see how the latter correlated with the memory noise test results.

    4. I hope you were careful when you tested the PS heatsink temperatures because some heatsinks are live with high voltage.
  • Anonymous User - Thursday, July 31, 2003 - link

    Well for the observent people who can differentiate between orange and blue, its not an issue. Also waiting a full second before clicking it reveals the location on both the bottom left and the mouse cursor. but i can see how it does get annoying.
  • Anonymous User - Thursday, July 31, 2003 - link

    yeah those in page advertisements are REALLY annoying...those have got to go. i never know when i'm going to link to another anandtech article or to an ad...i guess that's the point but it's still unacceptable
  • Anonymous User - Thursday, July 31, 2003 - link

    I would like to see the review include a Heroichi Electronic power supply, I hear they are very good but I haven't used one.
  • Anonymous User - Thursday, July 31, 2003 - link

    Some of your "In Page Advertising" links seem to be missing closing links tags or something so that it results in having a <link> in the middle or end of a sentence. Ex. "We had a lot of troubles with Vantec’s last power supply, the Stealth. We found an error in the production label<link>, which quickly led to a change in all the labeling on all Stealth power supplies."
  • KristopherKubicki - Thursday, July 31, 2003 - link

    #3 and #4, thank you for spotting these errors. I have updated and fixed them.

    Cheers,

    Kristopher
  • Anonymous User - Thursday, July 31, 2003 - link

    I hope the ripple for the PC Power & Cooling 3V wasn't 2.295.. Possibly 3.296??? 1 volt drop is unacceptable.
  • Anonymous User - Thursday, July 31, 2003 - link

    The REAL price of the pc power&coolinghttp://www.directron.com/pcpower.html
  • Anonymous User - Thursday, July 31, 2003 - link

    a little more content per page in some cases would be nice too...

    ...but good to see content on the site at all...and seemingly more regularly too...
  • Anonymous User - Thursday, July 31, 2003 - link

    woah guys, the tables need some work...

Log in

Don't have an account? Sign up now