PSU review time again! This time it is Arctic Cooling’s entry into the burtally competitive market of low wattage power supplies. This unit is aimed squarely at the (meh) Tt Smart 530w we reviewed recently. Will the Fusion 550RF stomp the Smart into the ground? Or, will it live up to its name and explode? We’ll see!
Features and Specifications
Direct from the Arctic.ac product page:
|The Fusion 550 is an extremely silent and highly efficient power supply unit. With 550 watts, it is well-equipped for professional PC systems.|
|High Efficiency This eco-friendly power supply unit protects the environment and saves energy costs for you as well. Even when the PC is switched off, the low standby power consumption of the Fusion 550 helps to save energy.|
|Quiet, Yet Powerful For the ventilation of the Fusion 550 a single 80mm PWM fan is enough to cool down the PSU – It is the high efficiency of the PSU and creative thermal management that makes this possible. Since the integrated ARCTIC F8 Pro fan comes with vibration absorption mounting frame, the noise is reduced to an absolute minimum.|
|Safe and Reliable PSU The Fusion 550 makes for a very reliable power supply unit for all midrange and high-end PC systems because of its 550 watt performance and up to 99% active PFC (Power Factor Correction rate). The overload, overvoltage and short-circuit protection safeguard the hardware as well as the user from damage.|
|Money Saver By using a powerful and efficient PSU, you save electricity cost every day. Taking the total cost of ownership into account (price of the product + average use over a certain time frame), Fusion 550 is one of the most attractive PSUs available in the market.|
When we look at this list it is worth remembering that this PSU came out some years ago, in 2009 I believe. At that point 80+ was very good and there weren’t nearly as many units at that level as there are now. At this point 80+ is a bare minimum for a PSU to be considered, rather than the “super cool” thing it used to be. In my personal opinion buying a PSU for it’s claimed Eco-Friendliness is awfully silly in the first place. The other part that leaps out at me is the “Safe and Reliable” part. Reliability is great, I’m a huge fan. I’m a big fan of safety too. Neither reliability nor safety has anything to do with power factor correction. Having OPP/OVP/SCP is very good, but that leaves out OCP and UVP. I’ll check the protections IC out later when I rip it apart. The last box I have a hard time with. In 2009 it may have been relevant, but given the huge array of 80+ bronze/silver/gold/platinum units out and about now, 80+ base efficiency is unlikely to save you money. On the other hand, 80+ platinum is unlikely to save you meaningful money as well.
|Limited Warranty||3 year years|
|Dimensions (Packaging)||550R: 265L x 190W x 110H (mm)|
|Dimensions (Product)||160L x 150W x 86H (mm)|
|Effeciency||82 – 86%|
|Fan Speed||700 – 2,000RPM by Intelligent Controller|
|Noise Level||0.08 – 0.3 Sone|
|Power Good Signal||100 – 500 ms|
|Hold Up Time||>17 ms|
|MTBF||>100,000 hours at 25 oC ambient temperature|
|Safety||CUL (Level 6) / CE / CB|
|Protection||OVP / OPP / SCP|
|Input Voltage||AC 90 – 264 Watts (Auto Range)|
|Input Current||3.0 A at 230 Vac|
|Input Frequency Range||47 – 63 Hz|
|Output Power (W)||550 Watts|
|Operating Temperature||0-50 oC|
|Operating Humidity||20% – 80%|
|Storage Temperature||-40oC – 85oC|
|Storage Humidity||10% – 95%|
|Gross Weight||2 kg|
The biggest thing that jumps out at me here is the 50 °C temperature rating. It doesn’t say whether that’s at part load or full load so I’m going to assume it’s full load. Note the 550W rating there as well as the lack of UVP/OCP.
|Output Voltage||Load Range||Regulation||Ripple & Noise||Total Output|
|+ 3.3 V||0,5 A||24 A||+ / – 5 %||50 mV||3.3 + 5 V
combined 130 W
|+ 5 V||0,3 A||24 A||+ / – 5 %||50 mV|
|+ 12 V1||1 A||17 A||+ / – 5 %||120 mV||408 W|
|+ 12 V2||1 A||17 A||+ / – 5 %||120 mV|
|- 12 V||0 A||0.8 A||+ / – 10 %||120 mV||9.6 W|
|+ 5 V sb||0 A||2.5 A||+ / – 5 %||50 mV||12.5 W|
|12 V Case Fan||2 x 0.15 A|
This chart claims two 12 V rails with a combined 34 A between them. Add in the 130 W for the 5 V and 3.3 V rails and we have 538 W. Add in the -12 V and 5 VSB and we have 560 W, for whatever that is worth. The regulation and ripple specs are simply the ATX specification maximums.
Mainboard 20 + 4 Pin
Mainboard 4 Pin 12 V
PCIe 6 Pin
PCIe 6 + 2 Pin
Case Fan Connector
This is a 550RF, what the F stands for I don’t know. We get a fair number of connectors here.
So far the specs are looking pretty decent. 408 W is lower than I’d like to see for 12 V output on a 550 W unit, but I can let that slide.
Photos Part One: The Box
When I say “box”, I actually mean “clear plastic blister pack”, because that is what Arctic has decided to ship these units in. It’s an odd choice to say the least, there isn’t a lot of padding here (read: none), but if it works it works.
If you can’t read what is in the second box in the right column, click the image. Now look at the PSU name. Now become offended that this is being sold, essentially, as a 550 W unit when it is really a 500 W unit. Even if you don’t become offended that’s OK too, I’ve got it pretty well covered.
I don’t have a problem with peak ratings being listed, but I have a major problem with PSUs being sold for their peak rating. Especially when that rating is for one (1) second. That serves absolutely no purpose, other than marketing. So now we’re looking at a 500 W PSU, which does explain the 408 W of 12 V we get I suppose. I am not pleased with Arctic’s marketing department right now. For bonus points check the charts/graphs out on the rear panel, I wasn’t able to make much sense of them personally. I think they’re labeled incorrectly.
I’ll drag the PSU out, hopefully it’ll be less offensive than its packaging is!
Photos Part Two: The PSU
The fan sticking out the back is rather odd. The fan power cable that heads out six inches and then goes back into the cable hole is pretty terrible. Other than that it’s a pretty decent looking unit though. The 80+ sticker is a bit much for me, around it you can see some scuffing from the plastic packaging. Let’s look at the spec label a bit more closely and see what it has to say about the 500 W / 550 W question.
This info is similar to the box, we again see two 12 V rails, and again we see that the actual continuous output is 500 W. I would vastly prefer that Arctic market this as the 500 W unit that it is rather than the 550 W unit that it isn’t. Not that anybody asked me. That UL number traces to absolutely nothing, the UL look up page says: Nope! Typo on the box or creative marketing? I don’t know. It’s not exactly promising either way though.
For cables, here’s what we’ve got:
This is a mixed bag. The wall cord is decently long, you get a sticker and four black screws too. You get four PCIe power plugs, two 6p and two 6+2p. What you’re going to do with two GPUs that require 6p+8p on a 408 W 12 V PSU is beyond me, but you have options at least. The Molex connectors are all on one cable, if you’re using a full ATX case you’re going to have issues connecting both the optical drive and the HDD. I suspect this PSU is aimed at mATX. The Molex connectors and FDD are all on another single cable. The motherboard power connector is a 20+4p type, though the 4p is very well attached. Getting it off could be entertaining. For CPU power we have a single 4p connector, I find this disappointing, though it’ll probably be fine. The real issue is that the motherboard and CPU cables are quite short. You have no hope of using this PSU in a medium or full ATX case that mounts the PSU at the bottom. Not without extensions anyway. Lastly, you get two fan plugs (rated for 0.15 A) that are controlled at the same voltage the PSU fan is. It’s a feature you don’t really see these days, I like the option though I’ve never actually used it.
Lastly, a few not at all randomly angled shots:
Now that we’ve seen it, I’ll test it!
Testing Part One: Regulation
A decent load test of a PSU requires a decent load. Contrary to what some may believe, that means you need a known load that can fully stress the PSU. Computer hardware does not cut it. Worse if the PSU fails during testing it might take out the computer hardware anyway. Commercial load testers cost a lot of money. I do not have a lot of money, so I built my own with juicy power resistors and a Toyota cylinder head. It works great. I’ll be using it to load this thing down fairly severely and will check voltages and ripple (more on that later) at various points. The down side to my tester is that the loads it can put on PSUs are fairly coarse, they go in increments of 48 W for 12 V, 50 W for 5 V and 22 W for 3.3V. Those wattages assume the PSU is putting out exactly the official rail voltage, a PSU putting out 12.24 V rather than 12 V will be at 49.9 W per step rather than 48 W. I file that under the “tough beans” category as I figure if a percent or two of load makes that much of a difference the PSU manufacturer should have hit the voltage regulation more squarely. It does make calculating efficiency difficult at best. however, given that the input power is read via a Kill-a-watt, the efficiency numbers are dubious to begin with. Kill-a-watts not being known for extreme accuracy on things with automatic power factor correction. For this reason, I am not listing the efficiency.
The ATX spec says that voltage regulation must be within 5% of the rail’s official designation, regardless of load. It doesn’t actually mention that the PSU shouldn’t explode, though I expect they figured it was implied. Exploding is a failure in my book regardless.
It is also worth knowing that I will be testing this PSU at both outdoor ambient temperatures (typically between 15 °C and 20 °C here this time of year) as well as in the Enclosure of Unreasonable Warmth. TEUW is a precision engineered enclosure that I use to route the exhaust air from the PSU right back into the intake fan, it is adjustable to hold the intake air temperature at (almost) any level I want it. This way I can test the PSU’s response to hot conditions as well as cold conditions. For the hot testing I will be running the intake temp as close to the unit’s maximum rated temperature as possible. TEUW, in case you’re curious, is a cardboard box.
Given the 0-50 °C operating range on the box, TEUW will be going all the way to 50 °C. If you curious about what TEUW is, click this link and scroll down to the end of the ripple testing on that page. In short, it’s a box and it gets hot inside. I’m not calling it a hotbox because at least half the other PSU reviewers out there call theirs hotboxes, and mine is different.
With that out of the way, here are the results:
|Wattages (total)||12 V Rail||5 V Rail||3.3 V Rail||Kill-a-Watts||Temperature intake/exhaust|
|TEUW 50c intake testing:|
What do we have here? Well it’s a group regulated unit, but that isn’t exactly a surprise. It doesn’t appreciate crossloads much at all, but that isn’t a surprise either. It does stay within spec voltage wise during the crossloads, and while the 3.3 V rail creeps upwards with load it stays within spec as well. However, it’s down to 25 mV between it and the allowed maximum once it’s in The Enclosure of Unpleasant Temperature. All told I’m actually kind of surprised, the fan stayed very quiet all the way through the first four load levels, became just barely noticeable during test five, and even at full blast in tests six and seven it was one of the least offensive full load PSU fans I’ve heard. Very nice, Arctic! (But please, run the power wire inside the case next time!)
Regulation on the 12 V rail (including, and then not including, the crossload) was 6.22% / 2%, the 5 V rail was 5% / 2%, and the 3.3 V rail was 2% any way you look at it. All in all that’s really not bad. The crossload results aren’t good exactly, but it stayed in spec and didn’t shut down so that’s a plus. Outside the crossload we have 2% across the board, while that would be “meh” for a $200 unit it’s not bad for an entry level unit.
All of this includes the fact that at the 0.375 ohm resistance on the 12 V rail in test 6 and 12.23 V on the 12 V, the 12 V rail load was ~399 W, so we were around the 515-520 W mark for the whole unit. Like I said in the block quote, units had best watch their regulation when it comes to my test rig. It punishes voltage regulation that swings high!
The bottom line for the regulation section: It passes.
Testing Part Two: Ripple
Ripple is fluctuation of the PSU’s output voltage caused by a variety of factors. It is pretty much impossible to have zero ripple in a SMPS computer power supply because of how a SMPS works, so the question is how much ripple is there? In the regulation testing phase we found out how the PSU does at keeping the average voltage at a set level, now we’re going to see what that voltage is doing on really short time frames. The ATX spec says that the 12 V rail cannot have more than 120 mV peak to peak ripple, the 5 V and 3.3 V rails need to stay under 50 mV.
If that isn’t complicated enough for you, there are three forms of ripple to keep track of as well. Long-term ripple from the PSU’s controller adjusting the output voltage and over/undershooting, correcting, overshooting, etc. Medium-term ripple from the voltage controller charging and discharging the inductor(s) and capacitor(s) that make up the VRM, and very short-term ripple caused by the switching itself. The first and second forms are the most important, if they are out of spec it can cause instability at best or damage in extreme situations. The very short-term (I call it transient ripple) flavor is less crucial, excessive amounts can still cause issues though it takes more of it to do so. The ATX spec does not differentiate, as far as the spec goes 121 mV of transient ripple is just as much of a failure as 121 mV of medium or long term ripple.
I test ripple in a few difference ways, first I test it during the cold load testing. It is tested at zero load and maximum load first. During the hot load testing I test the ripple at maximum load again. I have recently started testing ripple at fairly random loads with the unit still hot, it’s a bit unorthodox (a bit? maybe a lot) but has found issues in the past that did not show up with other test methods.
We’ll start, as usual, with all three main rails at zero load. Note that this unit wants at least half an amp to two amps on every rail. Tough.
For all three shots my scope is set to 10 mV / 10 ms.
Very good. Very good for a unit that wants a minimum load and isn’t getting it. The 12 V waveform is a pretty classic “I want a load and I don’t have one!” waveform, while the 5 V and especially the 3.3 V are more interesting.
I’ll throw a big load (test six, above) load on here and see how they do. For the following shots the scope is set to 10 mV and 5µs (microseconds).
Here we have some even more interesting waveforms. With the scope at 5 microseconds we see a ~83kHz switching rate, and a PSU that is starting to struggle a bit. The 12 V is especially interesting, the bands are thick because successive passes are a bit higher or a bit lower voltage than the previous passes and the scope tracks multiple passes on one screen. Based on the waveform I would say that it wouldn’t take all that much more 12 V load to get into trouble. Then again the unit is loaded to almost the maximum spec for 12 V load, so that’s okay. The 12 V ripple is actually quite good for an entry level unit at only 42 mV, that’s well within the 120 mV spec. 5 V and 3.3 V are also within spec, though 30 mV and 37 mV are rather closer to their limit of 50 mV. I call this “good” ripple control, considering the market it’s aimed at.
Next is hot testing, this is the same load but with a 50 °C intake temperature that has been sustained for a good bit of time before testing. Some units score better here as the capacitors wake up as they warm up. Others go to hell, so it’s not all sunshine and roses in hot testing. Scope settings are the same 10 mV 5 microsecond settings as the previous batch.
12 V and 5 V both got a bit worse, 12 V only by 4 mV. 5 V gained 9 mV and is now only 11 mV from the spec. 3.3 V said “Wait, what? Did something change? Nobody told me!”
All told this unit passed the hot load testing easily. I have to admit that I was surprised, when I saw the age of the unit, the price tag, the market segment, and then saw the 50 °C operating temperature my first thought was “Oh man, maybe I should go buy another fire extinguisher”. It turns out that I was far from correct on that assessment, I think this unit could probably run even hotter than it did. It certainly doesn’t seem to be struggling much temperature-wise. I don’t recommend running it hotter, but you’d have to have a pretty lousy case to have 50 °C internal case temperatures these days anyway.
Now that it has proven itself to be good or better, it’s time to rip it to shreds and see what the guts look like! That sounds evil. It’s not really evil. It’s just a bad idea if you don’t know what you’re up to. I’m pretty sure I know what I’m up to, so it’s only a dubious idea. Seriously though, read the huge red disclaimer in the next section and take it to heart. Not physically to your heart though, that could be bad for your health, along with difficult given the fact that this text, like that text, doesn’t really exist physically in the first place.
Disclaimer: Power supplies can have dangerous voltages inside them even after being unplugged, DO NOT OPEN POWER SUPPLIES. It’s just not a good idea. Opening a power supply and poking around inside could very well kill you. Don’t try this at home. Don’t try this at work. Just don’t do it.
Told you there was a disclaimer.
Popping the top off the unit doesn’t show us the fan hub for a change, it shows us a blank metal case that I’m going to skip, and the PSU innards. I’m going to show you the fan hub too, even though you don’t get a good look at it till you take the PCB out.
Experienced eyes will spot the same Seasonic platform as the Corsair VX450 and the Antec Earthwatts 500 W units. Not that there is a UL number on the PCB, there isn’t one that I can find.
We’ll start our tour with the transient filter, some of which is soldered to a PCB that is soldered to the receptacle and the rest of which is on the main PCB.
All told we have three inductors, four Y capacitors, two X capacitors, a fuse and a TVS diode. TVS diodes are like MOVs, but better. They can dump more energy and react faster. The empty hole on the right of the right picture (with an L above it) is where the power leads go into the PCB. There’s another X capacitor after the rectifier to soak up the diode switching noise.
That brings us to the APFC unit, it takes the incoming voltage (whatever it may be), rectifies it to DC and then boosts it to ~380 V and stores it in a big capacitor that you really don’t want to touch when charged.
The GBU806 rectifier is rated at 8 amps and 600 volts. In theory it could cough up 880 W for the PSU. In reality something else would almost certainly fail first. The APFC switching is done by a pair of 18N50 MOSFETs rated at 18 A @ 25c, 12 A @ 100 °C, and 500 V either way. 500 volts is lower than normal (most units use 600-650 V bits), but far from being a problem. The boost diode is a STTH8S06D rated at 8 amps and 600 volts. The capacitor is a Nippon Chemi-con part (good to see) rated at 85c. I’d prefer a higher temperature rating, but at least it’s Japanese. Also featured in the APFC bits is an inrush protection thermistor, it’s the green thing in black shrink wrap.
This brings us to the output side of things (sort of, the primary switches are technically still on the APFC half of the unit), the same CM6800G chip controls these bits too.
The primary switches are two more of the same 18N50 MOSFETs that the PFC uses. The 12 V output is rectified by a pair of 20A50CT schottky diodes rated at 30 A and 50 V each. 5 V and 3.3 V are each done by a single STPS30L30CT schottky. The 12 V is decently over-specified, the 5 V and 3.3 V label ratings are closer to the diode ratings that I’d really like, but it survived TEUW so it works well enough. The output capacitors are all OST units, OST is a taiwanese brand that I haven’t seen much from or about lately. In the capacitor rages of the early 2000s they had a bad reputation, but I haven’t seen any reports of them having issues since then. I expect they beat most if not all Chinese capacitors, which is a good start. The 12 V “rails” are in fact just a pair of filtering inductors separating the “rails”. They do not have individual OCP, and hence are not really rails at all. Added to that is the fact that the protections IC doesn’t support any form of OCP (or SCP!) in the first place. It does support OVP on all three rails and UVP on the 3.3 V and 5 V rails, though the trip points are laughably low. The OVP trip points aren’t much better, really. It’s not an overly useful protections IC from a protection standpoint. It does monitor the power signal from the motherboard, which is a crucial function.
Last for this section, the soldering!
Overall the soldering is excellent. There are a (very) few leads that aren’t far from being too long to the point of causing issues. Nowhere are there any that could actually reach something and aren’t bent in a safe direction.
That does it for the dissection!
Final Thoughts and Conclusion
I’m forced to admit that I did not come into this review with high hopes, or at least not high expectations.
The marketing bits on the product page, box and PSU did not help, at all. I really don’t like peak ratings. At all. Nor do I appreciate pretend rails much.
The actual performance was good however, across the board. It doesn’t like crossloads much, but that’s unlikely to be a problem and it does stay in spec. Ripple was similarly good, no issues there.
The 50 °C operating temperature is an honest rating, it really can do it and doesn’t care about the heat. I’m impressed on that front.
The fan is a mixed bag. It looks weird out there and the wiring job is dubious at best, but I’ve decided that I rather like how the fan looks. Performance wise the fan is excellent, it makes very little noise at most load levels and even flat out full burn it is much quieter than other units.
The cables are short, the motherboard and PSU cables are very short. It’s mATX or top mounted PSU ATX for this thing. Forget bottom mounted ATX or larger setups, it’s not happening without extensions. You do get lots of PCIe connectors. Unfortunately the layout of the SATA power connectors will also be an issue for large cases. This is simply a top mounted and/or mATX PSU. It’d be nice if it said so on the box. In my opinion Arctic should market this (honestly) as a HTPC PSU. It’d be great for that purpose.
The build quality is really quite good for a 500 W 80+ unit, the soldering is great, the components are all capable, it works well.
The packaging scuffs the PSU up, which is not surprising.
Currently I can only find this unit for sale on buy.com for a terrible price, sears.com for a horrendous price, and ebay for $60.11. There are a number of them at $60.11, so I’ll use that price for the pricing comparison. At $60 it is arguably the best 500 W PSU currently available, if you have a small case. Now another $10 buys you modern, long cabled, more efficient, units, but that’s a 16% price jump. I’ll call the value, at $60 and 500 W, decent. Not a pro, not a con. Of course, it’s not marketed as a 500 W unit, which makes the price more dubious.
I’ll summarize! Brilliant!
There are pros!
- Very quiet.
- Lots of PCIe cables for a 500 W unit.
- Comes with a sticker.
- Can deal with very hot cases.
There are cons!
- Short cable are, in fact, short.
- Despite the packaging, it’s a 500 W unit.
- The marketing is misleading at best.
- External fan wiring is messy.
- Will not work well, or at all, in a bottom mount in a large case.
Here we are at the bottom line where I apply a final judgement and slap a stamp on this unit. The unit itself is solid, I approve of it. The marketing is terrible, misleading, and generally dishonest at best. I do not approve of it. Due to this split I am going to apply a double result. The marketing deserves a fail, the PSU deserves an approved, it all averages out to “meh”, in the end.
Click the stamp for an explanation of what this means.