Apevia Warlock Power 900W Power Supply

REVIEW INFORMATION
SUPPLIED BY: Apevia
PRODUCT: Apevia Warlock Power 900W
PROD LINK: Apevia’s current offerings.
PRICE: $169.99 @ Directron
Price is at time of testing!

It’s that time again, ladies and gentlemen. We’re going to roast us a Warlock. Pay no attention to that man behind the SunMoon, for he is the great and powerful Wolf. Sorry guys, it was either Oz or Harry Potter jokes for this one. Dumbledore!

Speaking of the SunMoon, just as a reminder, I use an SM-268 automated test environment to check out how accurate the power claims on these units are. The 268 has the ability to program in loads in up to five steps, which I am setting for measured increments between 20% and 100% of the full rated power of our Warlock. This will show you how the unit’s guts hold up to steadily increasing loads. Then, I’ll crossload it towards 3.3V and 5V, and then again towards 12V to see how the unit copes with unbalanced loading.

After that, I’ll put the Warlock into the oven and see if Hansel and Gretel can get away in time while I do it all over again to see how the Warlock can take the heat.

Warlock Power 900W – Cold Load Tests
Test # +3.3V +5V +12V DC Watts/
AC Watts
Eff. Intake/
Exhaust
Progressive Load Tests
1 4A 4A 12A 195W/
228W
86% 21°C/
29°C
3.44V 5.17V 12.14V
2 7A 7A 25A 376W/
434W
87% 21°C/
33°C
3.38V 5.12V 12.10V
3 11A 11A 37A 551W/
648W
85% 22°C/
37°C
3.34V 5.08V 12.06V
4 14A 14A 50A 727W/
894W
81% 24°C/
42°C
3.30V 5.04V 12.03V
5 18A 18A 62A 889W/
1130W
79% 25°C/
59°C
3.06V 4.65V 12.01V
CL1 18A 20A 2A 188W/
242W
78% 24°C/
37°C
3.07V 4.65V 12.14V
CL2 1A 1A 62A 762W/
909W
84% 25°C/
43°C
3.41V 5.14V 12.01V

Goodness gracious. I’m seeing some impressive numbers up there from an Apevia unit, not the least of which is the astounding efficiency in test two. Indeed, the test results were so high I thought my Kill-A-Watt was borked. But no, I hit test four and suddenly the efficiency numbers dropped like a stone. Going by these numbers alone, methinks we have an overrated unit here.

But, as if that weren’t enough, take a peek at the voltage readings. In test one, the 3.3V rail is almost out of spec on the high side. In test five, the 3.3V rail is out of spec on the low side. That’s terrible regulation. The 5V also goes out of spec on the low side. The 12V rail is the shining beacon out of all this, showing no major fluctuation in any test. The crossload tests also have some interesting tales to tell, where in test CL1 the unit again drops below spec on the 3.3V/5V rails and is nearly kissing the high end of the spec on test CL2. Atrocious. I must say that for what appears to be an independently regulated unit, we are getting some stunningly bad numbers from it.

And then, we have the temperature readings. Look at that. On test five, exhaust temps are a full 34 degrees above intake temps. All these facts combine to make me completely sure that this unit is overrated. Sigh. I was really hoping for something good from Apevia. And indeed, had this unit been allowed to stop at test four, it would have been a good unit, if a little lackluster in voltage regulation. Perhaps even worthy of 80+ certification. The 135mm fan? It was howling away by test three already.

Well, this is really just a formality at this point, but let’s just see what the hot box has to say about the Warlock.

Warlock Power 900W – Hot Load Tests
Test # +3.3V +5V +12V DC Watts/
AC Watts
Eff. Intake/
Exhaust
Progressive Load Tests
1 4A 4A 12A 195W/
228W
86% 24°C/
34°C
3.43V 5.17V 12.11V
2 7A 7A 25A 375W/
434W
86% 28°C/
42°C
3.40V 5.13V 12.06V
3 11A 11A 37A 549W/
660W
83% 33°C/
54°C
3.40V 5.09V 12.02V
4 14A 14A 50A 724W/
895W
81% 38°C/
71°C
3.35V 5.06V 11.98V
5 18A 18A 62A
FAIL FAIL FAIL
CL1 18A 20A 2A 174W/
230W
76% 32°C/
44°C
2.88V 4.31V 12.13V
CL2 1A 1A 62A 758W/
923W
82% 37°C/
68°C
3.42V 5.15V 11.93V

I must confess that after the first few reviews I did with my revised hot box, which uses convective and forced air heating from the SunMoon, I tweaked it just a bit so it wouldn’t quite get as hot anymore. It didn’t matter for the Warlock Power. Again, it did well up until test four. And then it failed test five. Repeatedly. It would be humming along fine in test four, I would hit the button for test five, and then after a few seconds all the voltage readings would begin to drop. By the time a minute was up, the power supply went into seizures, making a tick-tick-tick noise as the LED fan pulsed on and off. Why, here’s the true source of Harry Potter’s mysterious ticking noise – it was the Warlock all along.

And the strange thing was, every time the unit did this, I was able to shut down and restart the unit in test four and it would work perfectly until I hit test five again. I tried four times to get the unit to complete test five, and it failed every single time. At no time did the protection circuitry step in and shut the unit down, which is inexcusable for me. Especially since exhaust temps went past 72 degrees the moment test five started. The unit brags about having overtemp protection, but I’ll be darned if I ever saw it work. At any rate, overtemp protection is the least of my worries now, what with test CL1 giving me new records of suck on the 3.3V and 5V readings. Forget about overtemp, where’s the overcurrent protection in all this?

Awful performance here from the Warlock. But, again, if we could only have stopped at test four…

Warlock Power 900W – Oscilloscope
Test # +3.3V +5V +12V
1
2
3
4
5
CL1
CL2

Now I’m even more depressed. The rails actually look pretty clean until tests five and CL1 where the 3.3V promptly goes out of spec, hitting an abominable 80mV at one point. The ATX spec is 50mV. I’m sorry, Apevia, but my princess is in another castle.