Here we have Cooler Master’s latest PSU. No longer content with mere Silver or Gold ratings, this unit is rated as 80+ Platinum. The 80+ Platinum badge certifies that the PSU is disgustingly efficient, and contributes a bare minimum to global warming. Or at the very least, it’s 89% efficient at full load, 92% at 50% load, and 90% at 20% load. Assuming of course that you’re using 110 VAC input power. I’m going to make that assumption, because as an American I’m legally bound to make as many asinine assumptions as possible. Before this degrades any further, let’s look at the features and specifications for this monster.
These come direct from the Cooler Master product page. The commentary is my own, and can be found in italics.
The Silent Pro Platinum PSUs are fully compliant with the 80-PLUS Platinum efficiency requirement with peak efficiency up to 94%. It features modular cable design integrated with dual 7V fan ports as well as equipped a 135mm ultra silent hydraulic bearing fan, with +12V single rail outputs efficiency that can reach up to 984W output at 82A, enough to power up most high end VGAs and CPUs at greatly quiet operation. No doubt it is specially designed for hardcore gamers who desire extremely high efficiency and powerful gaming products.
You heard it here first folks, Cooler Master isn’t sure who it was designed for, and is making an assumption. No doubt someone in marketing now regrets that line. On the other hand, a dedicated 7v rail is a pretty cool idea. The rest of that blurb is pretty standard stuff for a unit this size and efficiency. Do note that being a single rail won’t help anything compared to multiple rails, and in certain very specific circumstances can make a bad problem (a short circuit) worse.
- 80 Plus Platinum
80 Plus Platinum Certified: 90%, 92% and 89% Efficiency @ 20%, 50% and 100% Load
- Eco-friendly design: Erp Lot6 Ready!
A maximum 5Vsb current draw target in S5 off-mode of 0.1A for its motherboards and needed to ensure the system will consume < 1W in its off-mode.
This I rather like. It means that the PSU draws roughly nothing when turned off. If your computer is also Erp Lot6 compliant and you aren’t using your USB ports for charging an iThing, the whole shebang should draw less than 1 W. Once it’s running of course things are different.
- Powerful Single +12V Output
Tackling challenges from the most power-intensive peripherals head on, this solitary 12 V power rail provides ample power and strong resistance to any overloading with up to 984W output at 82A.
See previous about single rails not being helpful.
- Silent Operation
Silent Operation 135 mm Hydraulic Dynamic Bearing Fan for quieter operation. Silent, or quieter?
Make up your mind Cooler Master. If “Quieter”, quieter than what?
- Japanese Capacitors
High quality capacitors protecting equipment from capacitor leakage problem, also increasing the life-span of your equipment, especially at high temperatures.
Ahh but how many? I’ve seen this claim used on units with only two Japanese caps in ’em. I’ll be counting.
- Supports High-end Hardware
Supports 4-Way SLI / CrossFire + Dual CPUs.
Able to run toughest rig with 4-way GPUs and dual CPUs without any hassle.
This is…. not true. Two 130 W CPUs and four GTX 480s would give you a total draw of ~1460 W before overclocking. This is a 1000 W unit, it can power a lot, but not the toughest rig. If you use modern cards like the GTX680, it still can’t do two CPUs and four GPUs (1060w, this time. Closer). While it may be able to run said rig it, would involve running the PSU over it’s rating; not a good idea at all.
- Double-Layer EMI Filter Dual protection from electromagnetic interference, thus restraining noise and interference for greater protection for you and your connections.
Normal EMI filters are four Y caps, two X caps, two inductors. We’ll see if that’s what CM is talking about, or if we have a beefier filter.
- Intel Compliant ATX12V v 2.3
- Active PFC+ PWM Combo Controller. Integrating this active controller will increase the efficiency of any computer, making PFC higher than 99%.
Sort of… but not really. Unless you’re in a part of the world that bills on power factor and watts used, your power bill will be no different with an APFC PSU than a passive PFC PSU of the same output efficiency. Nor is there a benefit to putting the PWM controller in the same IC as the APFC controller. Not that I’m aware of, anyway.
- And More
Soft-start circuitry to limit inrush and thermistor currents.
Excellent. And required by the specifications it says it adheres to.
High reliability with an MTBF of at least 100,000 hours.
That’s 11 years straight.
Worldwide Five Year Warranty and free live-chat customer support.
Despite claiming it’ll run 11 years straight on average, they only give you a five year warranty. Five years is on the higher end of standard for this sort of unit though, I can’t harass them too much for this.
My overall rating of this unit’s Marketing Honesty: Dubious.
|Type||Intel Form Factor ATX 12V V2.3|
|Dimension||150 x 180 x 86mm (5.9 x 7 x 3.4 inch)|
|Input Voltage||90-264Vac (Auto Range)|
|Input Current||6 – 12A|
|Input Frequency Range||47 – 63Hz|
|PFC||Active PFC (>0.95)|
|Power Good Signal||100-500ms|
|Hold Up Time||>17ms|
|Operation Temperature||0~40°C (Nominal Input Voltage)|
|Regulatory||TUV / CE / UL / FCC / BSMI / GOST / C-tick / KC / CCC|
|Fan||Silent HDB 135mm|
|Certification||80 Plus Platinum|
|Connector||M/B 24 Pin Connector x 1
CPU 4+4 Pin x 2
PCI-e 6+2 Pin x 6
SATA x 12
4 Pin Peripheral x 5
4 Pin Floppy x 1
3 Pin fan Cable x 2
Pretty standard stuff for a top end PSU. The temperature rating is 40c which is as low and I’m willing to let pass. All the other specs (other than wattage) are copied straight out of the ATX specification guide.
Let’s take a look at the unit itself! That should be more interesting than the specs list.
Photos Part One: The Box
Not a lightweight box, this. Not the heaviest I’ve had (that’d be the NEX1500, that thing weighed a ton), but definitely not light. It’s also purple.
The box isn’t packed with labels, stickers, and bragging, I like that. It’s either black, shiny silver, or purple, depending on the side. I suspect Cooler Master marketing couldn’t make up their minds and what we’re seeing is a compromise, that’s what it looks like anyway. Sort of goofy looking really, but what matters is inside. Let’s see what’s inside.
Inside the box we find another box! This battle, the black and silver supporters seem to have won; no sign of purple to be found.
Opening this box, we find… another box! Sort of, anyway. What we actually find is a block of hard dense foam. Check it out:
This is far and away the most intense foam packaging I’ve run into, it almost seems too hard. It seems to be more intrusion protection than G load mitigation. I like it though.
Photos Part Two: The PSU
The PSU itself is inside a sack inside the sack, Cooler Master is really into the recursive packaging here. Once de-sacked², this is what we have:
We have ourselves an Enhance unit here, easily identifiable by the “T” shaped primary heatsink. No other manufacturer I’m aware of uses those. It’s a pretty nice looking unit, black and silver won the day here too. The modular connectors have a diagram showing which power pin is where. On one hand I like information, on the other hand I can see no good reason to list that. The fan connectors lack the pinmaps. The side labels are applied so that if the PSU mounts in the rear of the case, the label will always by right side up. Of course, that means it’ll always be upside down if your PSU mounts in the front of the case. The case label has a UL number on the sticker that points to Cooler Master, for whatever that is worth to you.
My overall rating of the PSU form/mechanical function: Good.
Photos Part Three: Cables
For cables, we have a selection of hard wire cables:
ATX24P (not 20+4, mind you), two 4+4P CPU power connectors, and one cable with two 6+2P PCIe power cables. I can’t say I like this layout much. Having two hardwired CPU power cables is just silly, as most people can’t use two. Having one hardwired is fine, having another modular is great. The PCIe cables I’d prefer either with sleeving between connectors, or ideally with two cables. Not that anybody asked me.
The modular cables, as well as the accessories, are inside a sack. Here is that sack:
We get a power cable, two fan cables that will only fit Thermaltake Frio Extreme PWM fans due to how the cutouts are arranged (see picture below). There are also four screws, two PCIe cables with two connectors each, three SATA cables with four plugs each. Finally there are two Molex cables, one with three plugs and one with two Molex plugs and a FDD power plug. This is a pretty decent selection, though the fan cables are going to be useless for the vast majority of people.
See what I mean about the slots? It doesn’t work well. You can jam it in, but the tab is constantly trying to push it back out. Given vibration, I expect it will succeed at the worst possible time. You could cut the tab off the cable; I suppose that would work. Still, Cooler Master messed up here.
My overall rating of the cables: Decent.
Testing Part One: Regulation Testing
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 are not 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 10 °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.
Let’s skip straight to the testing here, that wall of text is large enough without me adding more.
|Wattages (total)||12 V Rail||5 V Rail||3.3 V Rail||Kill-a-Watts||In/Out Air Temps|
|HOT results below:|
What do we see? We see some rather nice regulation numbers, that’s what. 12 V is looking nice with 1.06% regulation. The 5 V rail comes in at 0.4%, while the 3.3 V “only” manages 1.2%. That averages out to 0.89% regulation, very impressive. The fan didn’t really do much of anything until the unit was at nearly full load when it spun up slightly. At full load, with a cold intake air (the temperatures marked by *), the fan itself was quiet, but it’s also a bit out of balance and the PSU resonated with the surface it was sitting on, making a rather annoying hum. With warmer intake temperatures (even just 20c), the fan sped up further and the hum went away. At full speed the fan isn’t quiet, but it isn’t obnoxious either. Whether it will hum in your case or not depends greatly on the case itself, and the fan speed it ends up running at. I expect that it won’t in most cases/situations.
The 7 V rail didn’t get as much attention in this section. It did come out to be roughly 7 V, but not being part of the ATX specs, my load tester doesn’t have any way to test it. It did run fans happily enough. I did plug it into my oscilloscope for ripple testing, but that’s in the next section.
As far as load testing goes, this unit did very well. I’m impressed. It also showed itself as being able to deal with 40c intake air at full load, for quite some time. This is something a number of Enhance units can’t do, because they tend to have overly twitchy OTP. Better than not twitchy enough, but not ideal either.
My overall rating on this unit’s regulation abilities: Excellent.
Testing Part Two: Ripple Testing
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 with zero load and cold intake temperatures, for this round the scope is set to 10 mV and 10 microseconds for the three main rails. The 7 V rail is set to 10 mV and 5 microseconds.
All three primary rails are well within spec, though not as low as I’d like to see at zero load. The 7 V rail is plagued by transients that come from the 12 V rail that feeds it. That’s the 3.3 V rail’s largest ripple source too. None of them look like they have anything close to an issue though.
Next up full load, cold. Scope is still set to 10 mV / 10 microseconds.
At full load the 12 V rail’s ripple still looks decent, not fantastic, but decent. 5 V is right at the limit, while 3.3 V is under the limit by a mere 6 mV. The 7 V rail results can be found below, scope settings are in the captions. The left photo is with the unit at full load and the 7 V rail with zero load. The right image is with two Tt 140mm 0.5 amp 12 V fans on the rail. Note that the scope settings are quite different between the two shots!
The first picture shows a regulator that isn’t overly happy, but is at least holding itself together. Lots of switching noise, mostly. The second picture, with ~4 W of fans installed, shows a rather unhappy regulator. You can see around 220 mV of ripple on a constant basis, plus the large waveforms every time one of the fans triggers its motor magnets. The grand total of 450 mV of ripple is, in a word, terrible. To be fair fans aren’t likely to care in the slightest. Anything smarter than a fan however will not appreciate this rail much. I’m not going to hold the 7 V rail against this unit too much as it isn’t in the ATX spec and it will run fans, but I’m not impressed.
Cold 12 V crossload results follow. Scope is set to 10 microseconds / 10 mV for all three.
12 V crossload results are pretty similar to the full load. The 5 V rail is a bit happier, while 3.3 V has slipped to 2 mV over spec. Not a huge issue.
Now for hot results. Intake temperatures were held to 40-41c for the following ripple testing.
For the 12 V rail, the scope is set to 10 microseconds and 20 mV. For 5 V and 3.3 V, the scope is at 10 microseconds and 10 mV.
The 12 V rail is well within spec with a decent margin, though less than I’d prefer. The 5 V and 3.3 V rails are both over, not by a ton, but over. Will I hold this against the unit? Yes. Will it prevent it from getting Approved? Probably not.
How about the crossload results with 40c intake temps? Hope it sounds good, because that’s next. Scope is at the same settings as the full load hot results.
Pretty much the same deal here, both 5 V and 3.3 V are 8 mV over spec, 12 V is well under it. Not impressed am I. Offhand I’m going to guess that the issues are largely due to noisy MOSFETs on the DC-DC boards. We’ll check them out in as much detail as possible in the dissection section.
My overall rating of this unit’s ripply suppression abilities: Dubious.
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.
When I popped the lid off the unit (after spending entirely too long trying to get the warranty sticker off, without leaving a million pieces behind mired in sticky gunk), this is what greeted me:
To kick things off, let’s look at the transient filter. It contains four X caps, four Y caps, and two inductors. Solid.
Also making an appearance are a fuse, a thermistor for inrush protection, and a relay to short it once the unit is running (and gain some efficiency), plus a TVS diode for surge protection. If you think of a TVS diode as a Super Awesome MOV, you’d be on the right track.
Next in line are the two bridge rectifiers that are bolted markings into the heatsinks, no hope on getting that spec, sorry. Past them is another X cap to clean up the diode switching noise, followed by the APFC bits.
There are two MOSFET switches in the APFC plus a single CREE diode, the switches are 5R140P (25 A @ 25c, 15 A @ 100c, 550 V) bits, while the diode is a C3D10060 (10 A, 600 V). As you’ll note in the picture below, taking clear photos of them was a lost cause. The three APFC caps are Panasonic 270uF 420 V jobs. In charge of the operation is a CM6802 PFC/PWM controller.
Two more of the same 5R140F MOSFETs are in charge of primary switching, they feed the transformer which in turn feeds the synchronous rectifying MOSFETs:
Four 023N04N (90 A, 40 V) MOSFETs rectify the transformer’s output into 12 V, and four huge Teapo 105 °C capacitors filter it. On the left you can see the 5 V and 3.3 V DC-DC boards, here’s a closeup of them:
Each has a 86350D MOSFET (no known specs), three polymer caps (plus a Rubycon electrolytic on the main PCB), and a APW7073 controller. The controller wants to run two MOSFETs, so I’m assuming that the 86350D is a two-in-one unit. It has a couple little blotches on the side that might be part of the copper output rail through the middle of the IC. Maybe it uses magic, I don’t know. Whatever it uses it could do with another capacitor or three for ripple control. As you can see the rectifiers get a copper heatsink, they’re also electrically connected to this heatsink, which is an interesting (and rare) touch. This may give a bit more efficiency, as it’s far more conductive than a relatively thin trace on the PCB. Note the two black sleeves, those are heatshrink that hold thermistors against the heatsink. One for fan control, one for OTP I imagine. I can’t say I’m overly happy to see Teapo caps in this unit. While Teapo caps aren’t bad at all, they aren’t top notch either. Given the price we’re paying for this unit ($250), I’d like top notch caps. In charge of protections, we have a PS232S protections IC. Not one of my favorites due to its rather loose (in my opinion, anyway) OVP/UVP trip points. It does support OCP for multiple rails, as well as OTP and such.
The modular connector PCB is fairly busy. We find two big Teapo caps for the PCIe power plugs, the 7 V DC-DC bits, and the mounting standoffs. Much to my delight the standoffs have nuts and solder holding them in place. We also find four little Teapos for the modular Molex/SATA cables.
The 7 V power bits are about as impressive as their performance, not very. The little 8pin job there is the controller, a TI 5430 unit. It has the high side switch integrated into it, while the diode below that IC handles the low side. What is really lacking are decently sized capacitors, the lower two MLC caps are it for filtering (the top two filter the incoming 12 V). A single small Teapo, like the Molex and SATA cables get, would probably perk the 7 V rail right up.
Lastly, the soldering. It’s pretty good, I wouldn’t call it fantastic though. There’s a fair bit of hand soldering that could be prettier. No issues though. There are shunt resistors for measuring current coming out of the APFC circuit (on the ground trace, interestingly), as well as four wired in parallel for the 12 V rail.
That does it for our dissection, now you know as much about this unit as I do.
Final Thoughts and Conclusion
Overall, this is a good unit. It looks nice to kick things off, going with a black and silver theme.
The regulation is very good indeed. It has managed the best average of any unit I have tested so far.
Ripple control on the other hand is not very good. The 12 V rail is within spec, so that’s ok. The 5 V and 3.3 V rails are over spec by a bit when hot. Not a huge amount, but over is over. It’s not going to cause you issues, but its disappointing. The 7 V rail has enough ripple to cause issues. Don’t use it for anything but fans. Fans should be able to tolerate it as they can take a lot.
Noise wise, other than the vibration/resonance hum issue at mid fan speeds, the unit is great. If it hums like that in your case you aren’t going to be especially happy though.
Cable selection is good, the two hardwired CPU power cables are annoying, but not a showstopper by any means.
The unit takes 40c like a champ, which is nice to see. No shutdowns, no issues.
The price tag is a brutal $250. Given that Seasonic and Kingwin both make similar platinum 1kw units for $230, and XFX has one for $237, the price of $250 is too much. $20 isn’t exactly huge, but it’s quite a few burgers worth.
Given the price, I’m disappointed to find mid range Teapo capacitors inside. While their good enough for the job, I’d like to see higher quality. Amusingly, the 5 V and 3.3 V rails have higher quality caps, and ripple issues. That’s not the caps fault so much as the designers not putting enough caps on those rails.
The build quality is solid, no real issues there. The nuts+solder on the modular PCB standoffs is nice to see.
All told there are plenty of pros:
- Great regulation.
- Nice efficiency.
- Plenty of cables/connectors.
- Good build quality.
- Looks great.
- Does what it says it does.
There are some cons too though:
- Ripple over spec (not by a huge amount) on 3.3 V and 5 V rails.
- Unit shakes at certain fan speeds, causing a hum on some surfaces.
- Price is $20-$30 too high.
- Teapo caps in a top end unit is disappointing, if not an issue persay.
All told while it isn’t a perfect unit, it’s a solid one. Most of the issues are fairly minor things in specific areas. Even as it is, at $250 it’s not a bad choice, I approve of it. Please do click the approved badge and read exactly what it means. Once you’ve done that, my Approved rating will make more sense.
–Ed Smith / Bobnova