NZXT HALE90 V2 1200 W Power Supply Review

At CES, NZXT unveiled the successor to their excellent HALE90 series, the HALE90 V2. Not perhaps the worlds most creative name, but it does get the point across. In theory we can look forward to even better voltage regulation, tons of power, a bright white paint-job, and general awesomeness. How will it fare in reality? We’ll see soon enough.

Features

-80 Plus Gold Certification ensures efficiency levels up to 90%
-Completely modular design with labeled back plate for easy installation and clean look
-NZXT design element infused on the exterior/interior power supply with world’s first Innovative white and black PCB design for better isolation and identification
-Patented Multiple Intelligence Ability Controller:MIA IC^TMis designed for ultimate performance, eco-friendly, and improving AC-DC power conversion
-IFCT:The intelligent fan control technology is designed for better thermal solution
-Powerful +5Vsb that supports motherboards with fast USB charging that can also power USB devices while PC is on standby
-High quality Japanese Capacitor to promote longer lifespan and better reliability surpassing 5 year warranty
-Strong single +12V rail that provides stability and ease of use with the ability to deliver clean currents under a heavy load
-SLI and CrossFire Ready
-135mm Dual ball-bearing fan for smooth, silent rotation, and optimal air intake
-Auto Range:AC main is from 90V to 264V and no need in tiny select switch.
-Active PFC
-Full protections for over voltage, current, power, temperature, under voltage, and short circuit protection
-Safety Approvals for:UL / cUL / CB / TUV / CE / FCC / CCC / GOST-R / C-tick
-Backed by a 5 Year Manufacturer’s Warranty
-Standard:Comply ATX 12V V2.31 and EPS 12V V2.92

80+ Gold is nice. Completely modular PSUs I love, very useful feature. White/black PCB design is definitely distinctive, I seriously doubt it’s any better at isolating things than other colors though. MIA IC sounds like Missing In Action IC to me, I don’t think that was the intended thought though! I’m curious to see how it does and what it is. I’m all about beefy 5 VSB rails, this one comes in at four amps. I’m really hoping there’s more than one Japanese Capacitor in this unit, I’ll be disappointed if there is only one. Single 12 V rails are easy and inexpensive, but can be a safety issue in partial-short-circuit situations. Thankfully, those are rare. SLI/CFX: I’d hope so. Ball bearing fans last a very long time, which is good. Full protections are nice to see as well, much like the last 1kw+ unit I tested I’ll be testing SCP on this unit.

Specifications

NZXT does this in single-page pictorial format, so I will too! 16 SATA connectors and 14 Molexes and two FDDs! That is a lot of connectors, even for a 1200 W unit. Very nice. Even the lower wattage units have plenty of connectors, this is nice to see. Also note the eight PCIe connectors, plenty for four GPUs. Standard GPUs anyway, if you’re planning on running crazy high-end custom PCB GPUs that want three power plugs you’ll have issues running four of ’em, but that’s OK as you’ll need more wattage anyway.

The efficiency curve shown above is only barely worthy of being called a curve, though calling it an “efficiency flatline” sounds weird. Meanwhile on the output ratings we see the 4 amp 5 VSB rail again, I still like it. The 30 amps for 5 V and 3.3 V are roughly average, as is the 150 W combined cap on them. That’s more than almost anybody needs anyway, other than people running servers with dozens and dozens of hard drives (sidenote: I’ve only seen this become an issue once, on a server with 70-80 HDDs).

Let’s look at some pictures of the unit itself.

Photos Part One: The Box

The box is quite boxish, the only thing of note really is the viewing window on the left side of the top (or front, if you prefer). In the photo it looks like a picture, but that’s actually the top of the PSU peeking through the slit there. Other than that there’s some spiel about Japanese Capacitors, efficiency, single 12 V rails and the like. Plus the 5 year warranty and some information on the <3 warranty/RMA program (which is, as a note, fantastic in theory and in real world function).

Not surprisingly, the box opens!

The packing, as we can see, involves full foam protection for the PSU. Nice! On the left we have a bright white nylon bag full of cables (and cables, and more cables), under that we find the manual as well as a Quality Control tag that shows that the PSU got tested at various stations in the factory. Under the manual is a false bottom, and under that we find the accessories and power cable. The power cable uses a C19 plug, to use that plug the cable needs to use at least 14 gauge wire. It’s 16 amps, as opposed to the normal sort that can use as thin as 18 gauge wire, awfully thin for a 1200 W PSU that’ll be drawing >1300 W at full load. We also get some black screws and a few “zip” type cable ties.

Photos Part Two: The PSU Itself

As you can see the PSU’s styling on top and bottom is fairly laid back, it’s an off-white finish that I think looks quite good, especially combined with the white label and white fan. The modular output end has some fairly faint labels on the plugs, I’d prefer brighter personally. Note the extra three pin plug to the left of the ATX24P connector, that’s for the extra voltage sense wires. Also note that the upper and lower 8P plugs are for CPU power cables, while the eight 8P through the middle are all for PCIe cables. On the right we have plugs for the SATA and Molex cables. On the exhaust end of things we have the C20 power receptacle (C19 is the female cord end, C20 is the male PSU end) as well as a power switch. See the black styling bit that curves around? More on that after the side pictures.

Those black bits are a semi-flexible plastic material, NOT paint! They also clip in and can be removed, NZXT could, in theory, make a few different color kits. I’d like to see them do that, and I think case modders would too.

Lastly as usual, a few glamor shots.

The center of the fan grill is black, but very shiny as you can see. It has NZXT embossed on it.

Photos Part Three: Cables

The ATX24P cable is interesting enough (to me, anyway) to get its own little photo section, here it is:

As can be seen above, the motherboard end has doubled up wires on 5 V (red) and 12 V (yellow) to go along with the normal brown 3.3 V-sense wire (brown, far left bottom in the lower left picture). The PSU end has a small three pin connector that clips in next to the main connector. If it is left unplugged the PSU functions fine, but regulation is not as good. I did some fairly abusive testing (that’s my job, after all) and found that the PSU will power through that plug being disconnected with a load on it, as well as being plugged in. Do NOT do that, though. I can’t imagine it’s a good idea, and your job isn’t testing PSUs to destruction (I assume, anyway).

Now if you look at the cable pictures you’ll note that there are four SATA cables and four Molex/FDD cables. If you scroll up further you’ll note that there are six ports for said cables. You don’t get to use all 16 SATA plugs and 14 Molex plugs at the same time, sorry. Lots? Yes. All? No. You do get to use both CPU power cables and all eight PCIe cables though. The cables are all of the ribbon sort, not as pretty as sleeved cables, but much easier to fit through small areas for cable management. I prefer them to sleeved cables as I like functionality better than looks. They are the stiff sort rather than the really soft sort, my feelings are mixed on that one.

Let’s test this thing!

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.

Awkwardly, the first unit I received was DOA. The 5 VSB rail worked perfectly, but nothing else so much as twitched. The unit was functioning when it left NZXT according to the ride along test tag, somewhere during shipping someone Ace Ventura’d it one too many times I guess. On the plus side when a unit has issues, manufacturers have the option of sending a second unit, which NZXT did. Below are the results from that second unit.

12 V came in with 0.98% regulation, just under the magical 1% mark for excellent regulation. 5 V came in a 1.18%, just over that magic number. 3.3 V came in at 1.5%, still very good but short of excellent. Averaged out we get 1.23% regulation, very very good, if just shy of 1%. Pretty impressive really, especially given that all three rails are essentially maxed out here, and running at 42 °C intake temperatures.

The fan boosts itself during startup to make sure it starts turning as well as to let you know it’s alive, then it settles down to a very quiet speed for low load operations. There is a slight bearing hiss at low speed. As the unit heats up the fan speeds up, even at full load in a cold ambient it isn’t bad. I’ll call it average for a 1.2 kW PSU. At 42 °C and full load the fan is far from quiet, this is what I’d expect from a 1.2kW PSU’s fan.

Out of spite I’ve decided to start testing SCP in my reviews. This review is no exception! I was able to do a fair amount of arc welding using a single PCIe cable without the unit shutting down. On the other hand shorting the CPU power cable shut the unit down immediately. It’s better than usual for a single rail unit, but nowhere near as good as a multi-rail unit would be in this test.

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.

For the first set of pictures, at zero load, the scope is set to 10 µs / 10 mV.

Is this a record setting lack of ripple for an Overclockers.com PSU test? Yes, yes it is. Very, very, very nice. If it can do something close to this at full load I’ll be very impressed.

Full load time, scope at 5 ms / 10 mV now.

I’m very impressed. This unit continues on its run to be the cleanest power output in Overclockers.com PSU history. No mean feat on this test equipment!

How about with some rather high intake air temperatures? Scope is still at 5 ms / 10 mV.

22 mV or less on all three main rails in all conditions. That’s pretty epic. I’m very impressed. Even with the scope dialed down into the microseconds the ripple picture stayed clear, there is an impressive (and rare) lack of meaningful transients here. NZXT did a very, very good job design wise!

Dissection

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.

The saying goes “the proof is in the pudding”. I don’t expect to find any pudding inside here, nor do I generally look for answers in pudding, but I am expecting some answers from inside this power supply! I want to know who the OEM is, and how they did this!

The first thing that greets us is, as always, the fan and some overview shots.

I’m going to end the suspense and say that this is a FSP based unit, though not a platform I recognize. The FSP roots are evidenced by the proprietary FSP APFC/PWM and DC-DC chips, not subtle.

When I popped the PCB out, I discovered something that was semi-trapped under the protective plastic shield that lives under the PCB, here’re pictures of it:

This is an issue folks. Down where it was stuck it obviously caused no problems, but should it come free and rattle around the plastic shield and start shorting things on the PCB, things could go very very poorly rather abruptly. I believe that this solderball resulted from the receptacle soldering process, as there is some evidence of hasty work there in the form of slightly melted plastic on the switch and some traces of solder on the plastic receptacle housing and one Y cap. This poses a rather serious issue, as normally I would stop and ask the company if they want to send a second unit for review. That gives them a chance to prove it a fluke. The issue of course is that this is the second unit, the first was DOA. As such, I have to score on this, and it’s a potentially hazardous issue.

We’ll move on for the moment and check out the transient filter next.

The transient filter is very complete, it has four Y caps, two X caps, two inductors, a fuse and a TVS Diode. It also has a little IC that discharges the X caps when the input voltage goes away, this replaces the standard bleeder resistor and gains a bit in efficiency.

After the transient filter bits comes the APFC section:

Things kick off with a pair of GBJ25L06 (25 A, 600 V) rectifiers, followed closely by three 6R125C6 (30-19 A, 650 V) MOSFETs. Doing the control duties for both the APFC section and the Primary MOSFETs is a proprietary (FSP?)6600 IC, no data is known about this IC outside of FSP. Usually these are marked FSP6600, the change makes me wonder if this is a new generation. APFC boost diode duties are done by a pair of STTH02R06FP diodes, about which I can find nothing at all. There is a thermistor for inrush current control, along with a relay that shorts it out once the unit has started operating. There’s a strip of copper wrapped around the APFC inductor, it’s attached to a wire that is attached to a ground trace on the PCB. My best guess to its function is that it helps keep the electromagnetic hell that is the APFC inductor better contained than it otherwise would be.

On the primary side of things, we have more pictures!

Three 17N80C3 (17-11 A, 800 V) MOSFETs wired in parallel do the primary switching, while a single 3N80C ( 3-1.83 A, 800 V) MOSFET is involved as well, possibly resetting the primary MOSFETs. I’m not really sure on that one.

On the secondary side we have four 4x 023NE7N (120 A, 75 V) MOSFETs doing synchronous rectification duties for the 12 V rail.

Author Update [Feb. 5, 2013 – 10:53 p.m. EST]: The 5 V and 3.3 V rails are not generated with the DC-DC setup I expected! They’re actually a rather interesting AC-DC design. The following paragraph and several photo captions are updated to reflect this.

Controlling each rail is a FSP6601 controller. Two extra cables come out of the main transformer and connect to the 5 V / 3.3 V bits. One cable gets split on the PCB and is fed to one 2R640 (100 A, 40 V) MOSFET for each rail and then to the filter inductor input side. The other cable goes through a single 2R640 (100 A, 40 V), then is split and sent to one 036N04 (90 A, 40 V) MOSFET per rail, after which comes that same inductor input side connection. If this doesn’t make a whole lot of sense, here’s an alternative explanation: The transformer has a secondary AC output, the 5 V and 3.3 V rails are both generated off that one output with a hybrid synchronous rectification / buck converter type setup.  If that’s not odd enough, there is a 2R030 MOSFET (no data) that connects the 5 V rail to the 5 VSB rail. Why? I have no idea. All together it’s something I haven’t seen before, but I’ve gathered that FSP uses roughly this setup on many of the FSP Aurum units.

All six of the MOSFETs have thick thermal pads connecting them to the case through a hole in the plastic undershield:

As you can probably tell from the pictures, all the capacitors are Japanese. They’re all Nippon Chemi-con, even. The protections board has a little bit of hand soldering up at the top where the thermistor on the secondary heatsink attaches for fan control and OTP, the brown gunk you can see is leftover flux from that. Not an issue. The protections IC itself is a WT7527 unit. It supports all the usual functions, including two 12 V rails (though that is not used on this unit of course).

I can’t find a MIA IC, it must really be MIA! Or they’re talking about the 6600 chip, more likely.

The soldering on the PCBs is all good quality, though the amount of solder is somewhat random. There are a few joints that have way more than needed (a possible solder ball source), and some that just barely have enough to be called good joints. I can’t find any meaningful pattern to it.

The modular connector PCB has a variety of plates to get energy from point A to point B, as well as a pair of brackets on the ends that serve as extra ground paths as well as holding the PCB steady. The screws holding them to the main PCB lack threadlocker. They are tight and they have lockwashers, but I’d much prefer to see some threadlocking compound as well.

The PCB UL number is E87711, which traces to CENTRAL TECH INVESTMENT LTD. I’ve never heard of them before. The case’s UL number is E190414, which traces to FSP. NZXT was involved in the design of this unit, but FSP does the actual manufacturing.

Final Words and Conclusion

NZXT and FSP hit on a pretty amazing design here! Design wise I’m very impressed.

The regulation is very very good, just shy of the 1% excellent mark.

The ripple control is fantastic, the best I’ve seen on my equipment. They did a very good job here.

The cable selection is generous, and the flat ribbon style cables make cable management very easy. Some people don’t like how they look, I don’t have any issue with them myself.

The styling on the PSU I like a lot, the plastic/rubber inserts on the sides rather than the usual stickers is something I really like. I hope NZXT puts out a few options for them, I’d love to see that.

I like that this unit uses a C19 style power cord, no risk of a user using an 18 gauge cord off an old unit and setting fire to the thing.

The quality control process looks great on paper, with seven tests the PSU has to go through and a test sheet that comes with the unit. Unfortunately both units I got had QC issues. The first was dead of course, NZXT gets a mulligan on that one. The second unit had a 2.29mm solder ball loose in it. This is not OK. If I’d gotten quite unlucky the solder ball could have bridged a couple pins on a MOSFET (Gate and Drain for instance) and blown it and the MOSFET to pieces. I’m very disappointed to see this, as the unit is otherwise fantastic!

At a MSRP of $270 this isn’t the cheapest 80+ Gold 1200 W PSU around, though it does match the current price of Corsair’s offering for this category exactly. I’d say that given the performance and the fact that it’s a brand new design it’s priced about right.

To summarize, there are some impressive pros:

• Very good voltage regulation.
• Excellent ripple control.
• Plenty of cables.
• C19 power cord, very safe.
• Looks great. Potential for panel kits in the future.

Unfortunately there are some cons too:

• SCP is as dubious as it usually is on >1kw single rail units.
• 2.29mm solder ball loose inside the case. This is a serious issue. Quality control needs some work.

At the bottom line this is a fantastic unit that is hampered by dubious quality control at the FSP factory. With better QC this unit would be a slam dunk for an approved badge. Unfortunately as it stands I cannot give a approved label to this unit, as impressed as I am by it. I expect that I have simply been hit by a double order of bad luck, but I can only review what I have in front of me, and what I have in front of me I cannot approve due to QC issues. The design and performance is far from “Meh” worth, leaving us with an awkward No-Rating final rating.

— Ed Smith / Bobnova