Temperatures, Fan Speed, and Acoustics


The high efficiency means that this power supply needs to dissipate less heat, which is icing on the cake. The temperatures on both heatsinks are so low that you could practically run this PSU fanless. Once we hit 100% load, temperatures start to increase, but we don't expect very many users to come anywhere near that point. What's interesting is that fan speed begins to ramp up faster than temperatures, and in fact the temperature graphs drop between the 80% and 100% load marks. Still, better safe than sorry. As for the exhaust temperature, it's very close to the heatsink temperatures, so the heatsinks and fan are doing a good job of getting heat out of the power supply.


As we saw in the temperature graph, fan speed begins to ramp up at around a 500W/50% load. It spins at 750 RPM up until the 50% mark, at which point fan speed ramps up linearly until the maximum 1500-1600 RPM at 100% load.


Noise levels for the first half of the graph are extremely low, but unfortunately noise increases rapidly with fan speed beyond that point. This is a real pity, since the heatsinks aren't even that hot and it certainly would be possible to run the fan at a lower speed. The starting point of 18dB(A) is not something you will even notice, and we couldn't hear any ticking noises either (something we have seen in non-PWM controlled fans). However, the final output of 29dB(A) definitely doesn't qualify as being silent. The good news is that you can use this power supply up to a load of around 800W and still have an acceptable amount of noise, which is sufficient for even triple SLI systems -- and those will create plenty of noise just from the graphics cards. Truthfully, for a unit delivering over 1100W of power, 29dB(A) is not at all something we would call a poor result.

Efficiency and PFC Conclusion
Comments Locked

49 Comments

View All Comments

  • dh003i - Saturday, December 6, 2008 - link

    This is really horrible on Enermax's part. It was said that the PSU would be available at the end of the month (then Novemeber), at the latest; here it is the end of the first week of December, and the thing still isn't to be seen anywhere. Maybe Enermax will decide to bless us by making it available before Christmas.

    When you announce the availability of a product, it damn well better be available.
  • dh003i - Monday, December 8, 2008 - link

    I e-mailed Enermax and got a respond from them saying "The revolution should be available in about two weeks." That's December 22nd, or just around Christmas. So we'll see if that ends up being the case.
  • s1ugh34d - Wednesday, November 12, 2008 - link

    Enermax PSU. My liberty 400w over spins the fan, So I pulled the fans cable out, and hooked it up to 5v, It's always quiet right around 800RPMs...

    My Q6600 dual 8800GT's runs on a 610w PCP&C silencer. There is only a few small situations were over 750w is necessary.
  • dh003i - Wednesday, November 12, 2008 - link

    So-far, this sounds great, but almost like vaporware (except Anandtech has an actual working sample). I mean, when will this thing be available? I'm building a new workstation for myself, and have held out purchasing the CoolerMaster 1200W UCP because of this new Enermax Revolution 1050W, which is modular. But I don't know how long I can wait. It'd be nice if Enermax gave us some kind of clue as to when the thing will be available.
  • Christoph Katzer - Wednesday, November 12, 2008 - link

    Enermax says that the first shipments will arrive at the branch offices in the end of this month latest.
  • gochichi - Monday, November 10, 2008 - link

    Good point, cause Hybrid SUVs are still SUVs and this guy is a 1050W powersupply... definitely an "SUV".

    This thing is really GOOD looking actually. To those of you who ask, why even write an article of a 1050W power supply? I say, this is an enthusiast's site... of course it should be covered.

    I unfortunately won't be buying this beauty, but if I were more enthusiast-like (kind of past the stage now sadly)I would certainly consider it. $330.00 is a lot of money, but it's not so bad. You figure in the things prone to become obsolete are video cards (two-cards at least for a system with this power supply yes?) Those will run $200+ each. CPU... say $300+, motherboard say $150-$200.00 and on and on. If you're going to do that sort of build you may as well slap this thing on there.

    I'm happy with my switch to the sub-$500 computer space for desktops. But it's boring, and this power supply isn't. I also own a Corolla, it doesn't mean a Lamborgini Murcielago is "irrelevant".

    Oh just in case you've been in a cave for a few years. $150.00 buys you an awesome video card. Even if you're old like me, you should really get one just for goodness' sake.

  • CEO Ballmer - Saturday, November 8, 2008 - link

    I like the specs on this!

    http://fakesteveballmer.blogspot.com">http://fakesteveballmer.blogspot.com
  • iwodo - Friday, November 7, 2008 - link

    Thanks for all the hateful comment. But as far as i read, even a CoreQX97xx with Geforce GTX 280, as stated in the Corei7 review, only uses 300W at peak. ( And Corei7 uses less then that )

    Double that, so you have a Dual Quad Core, Geforce 280GTX 280 SLI, you will still ONLY arrive to 600W at peak.

    So may be the article "Debunking the PSU Myth" never got enough people read it.

    And this may be the second dumpest question ever as someone would post.
  • MrBlastman - Friday, November 7, 2008 - link

    Yes, exploded - as in a shower of sparks, loud bang and the force was so strong it broke the plastic mounting bracket where the heat sink attaches to my CPU...

    Along with thousands of other people's Enermax PSU's exploding, I am very afraid to purchase another Enermax PSU. I used to be a stalwart supporter of them - they made great stuff. But... after hearing of others warranty'ing their PSU only to get another one, or a third one that exploded, I decided to move on to something else.

    I hope they have finally fixed these problems.
  • xaris106 - Friday, November 7, 2008 - link

    how come don't you do step load tests?
    It would be really interesting and informative to see transient responses, settling times and overshoots. Please consider it.

Log in

Don't have an account? Sign up now