NZXT has released a sequel to their excellent HALE82 power supply lineup, the HALE82 N Series. We have the 750 W flavor here today. It will get checked out, poked, prodded, abusively tested (twice), ripped to shreds, analyzed, inspected, quantified, and finally given an official Overclockers.com rating. Before we even get to the power supply we have some specifications and features to go through though. They’re hard to test, but I will be comparing them to reality as I go through the PSU later in the review.
Specifications and Features
Copied off the box this time, we have the specs:
HALE82 N Series is powered by:
- 80+ Bronze Certification. The Hale82 N Series from NZXT operates with high efficiency: at 20%, 50% , and 100% loads, efficiencies are 82%, 85%, and 82% respectively.
- Japanese Capacitors. High quality components promote longer lifespan and better reliability. Bobnova’s note: I’ll be checking this, closely.
- 120mm Two-Ball Bearing Fan. For smooth, silent rotation and optimal air intake. Bobnova’s note: Ball bearing fans are generally louder than other bearing types, but last for a very long time.
- Strong Single +12V Rail. A single +12V rail provides stability and ease of use with the ability to deliver clean currents under a heavy load. Bobnova here again, multiple rails done correctly are just as clean, just a easy to use, as well as safer in short circuit situations. They cost more though.
- Large Tower Support. Extended 8pin connector for bottom mounted and/or large cases. I’m assuming they mean cases with bottom mounted PSUs. I do like this feature.
- Keeping it Safe. The HALE82 N series offers over voltage, current, power, temperature, under voltage, and short circuit protection.
For connectors you get the following:
- CPU: 1x 4+4P.
- PCIe: 2x 8P, 2x 6+2P. This is the second unit with 8P PCIe connectors that I’ve seen. They’re rare. I prefer 6+2s.
- SATA: 12x. This is nice, more than some higher rated units.
- “Molex”: 5x, plus 1x FDD.
Power distribution is thusly:
|AC Input||100-240 V AC, 10-5 A, 50-60Hz|
That’s a nice juicy 5VSB rail, always nice to see. The 12 V rail is pretty juicy as well, while the 3.3 V and 5 V rails offer a fair amount of power as well. All around this is a juicy PSU. I think it’s time to look at the box.
Photos Part One: The Box
If the box look is familiar, it’s because it looks almost identical to the NZXT HALE82 750 W. The only difference between the boxes is the N Series lettering, the lack of the “Modular” decal, plus the HALE82 says “100% Japanese Capacitors” while the N Series says “Japanese Capacitors”. Makes me suspect they aren’t all Japanese. Like I said, we’ll see.
Photos Part Two: The PSU
The unit itself also looks very similar. I liked the look the first time, and I continue to like it now.
It’s a good looking unit, nothing flashy or over the top. Just a matte black finish, a subtle label and a white fan. The white fan is a nice touch, though it does do away with the units otherwise stealthy appearance. I expect the fan would look quite nice with an LED or two aimed at it.
A few more shots below from pseudo random angles.
Photos Part Three: Connectors and Accessories
All told a nice selection of cables, I’m happy to see four PCIe power plugs, but I’d rather see all four as 6+2 plugs. With two 8p and two 6+2p you’ll have difficulties if you want to CFX/SLI a pair of cards that both use 6p connectors. 12 SATA plugs is quite generous for a 750 W unit, and 5 Molex plugs isn’t bad either. The CPU plug is a 4+4p, which is perfect. The motherboard power is a 20+4p, slightly more annoying than a straight 24p for modern motherboards, but you can still use it in older boards. I ran into an issue earlier this month trying to fire up an older board with modern PSUs for just this reason. In a perfect world the 20+4 would clip together tightly, I’ve seen it done by one (1) company. I suspect the connector is patented.
The accessories are on the bare bones end of things. You do get some black screws and cable ties, as well as a power cord.
Load Testing, Cold
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 (APFC). For these reasons 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.
Without further blathering, we have the results!
|Load Wattages||12 V Rail||5 V Rail||3.3 V Rail||Kill-a-Watts|
This looks pretty good, the 12 V rail has 2.67% regulation, the 5 V rail has 1.59% and the 3.3 V rail has 2.44%. Nothing spectacular, but not bad at all, especially for a group regulated unit facing abusively laid out loads. Really the most interesting number is the 3 watts used with zero load. The previous record low on that test was 6.2 W, most units are somewhere in the 6-7 W range. What does this mean? Not much, really.
The intake air temperature was 15c, the exhaust air temperature with a sustained full load was 24c. The fan was easily noticeable, but only as a steady whoosh sort of noise. No annoying harmonics, ticking, or rattling.
Load Testing, Hot
I placed the PSU into the Enclosure of Excessive Warmth (it’s a cardboard box), which forces the PSU to inhale its own exhaust air. Temperature was monitored by a dual probe thermocouple thermometer made by myself. Intake air temperatures were kept in the 40-41c region. While the box is devoid of a temperature rating, the brag sheet that came with the unit says 40c. Also, I consider 40c to be the minimum reasonable rating anyway. If a unit cannot survive 40c it is not worth using in my opinion.
Only the heavy load tests are repeated hot, lacking an actual heater the box cools off at lower loads.
|Load Wattages||12 V Rail||5 V Rail||3.3 V Rail||Kill-a-Watts|
These are the kinds of results I like to see! Very slight changes to the 12 V and 3.3 V rails, a few more watts of power eaten, nothing unpleasant. With an intake temperature of 40c, the exhaust temperature was 46c. I’m impressed.
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.
This is where things get interesting, the results in the tables below are actually from a second unit that NZXT sent to me. The first unit had nearly identical regulation numbers to those of the unit above, and nearly identical ripple. Except for one set of conditions: a low to mid range load on the PSU overall AND a ~22W load on the 3.3 V rail, after a sustained heavy load with high intake ambient temperatures. In those conditions the ripple on the 3.3 V rail was off the charts at ~310 mV. I contacted NZXT about it and they had me send the unit back via standard RMA channels (which are FANTASTIC, by the way). Armed with this data and the unit, NZXT canceled the full production run for the units and dug in to figure out what was wrong. It turned out that between the engineering phase and the pre-production units, the 3.3 V smoothing inductor had been swapped out for a different inductor with (I assume) the same specs. The pre-production units were then sent out to the reviewers, including me. NZXT changed the design back to the original inductor, swapped said inductor into my unit, and sent the unit back to me. What you see in this review is that reworked unit, which is the same as the final production units will be. As this reworked unit is the final production flavor, this is the unit scoring will be based on. All of the regulation testing above was done on this second unit as well.
I think it’s time for some ripple testing, don’t you?
First we have the 12 V rail in cold conditions, scope settings, load and ripple are noted in the captions of the scope shots.
51 mV is well within the 120 mV spec. This is good. It’s not fantastic, but it’s definitely good enough.
5 V next:
48 mV on the 5 V rail is getting awfully close to the 50 mV maximum spec. It passes though.
Time for 3.3 V:
The 3.3 V rail looks great, no issues there at all.
Time to toss it into the Enclosure of Unreasonable Warmth, and see how it does at 40c!
Not much changed, though a crucial extra 7 mV of ripple on the 5 V rail pushes it over the spec. Not far over and the flavor of ripple is the transient sort, but it’s over. That’ll have to go into the cons section. The rest of the rails look just fine, with 3.3 V continuing to impress.
All told the unit does decently, though ripple under spec on the 5 V rail would definitely be preferable.
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, and doing so could very well kill you. Don’t try this at home. Don’t try this at work. Just don’t do it.
With that out of the way, please note that except where noted, this is the first PSU in these photos. The second PSU has all the same bits and the same soldering, except for the soldering on the main power leads and the 3.3 V inductor, which are hand soldered (rather nicely, I must say) to update the unit.
Here we have the fan, and an overview.
Next up, the transient filter. As usual it starts on the receptacle and continues on the PCB.
In total we have two ferrite beads (one on the ground wire, interesting) four Y capacitors, three X capacitors, three inductors and a fuse. A fourth X capacitor lurks behind the rectifier to mop up switching noise from the diodes inside said rectifier. Notably absent is a TVS diode or MOV. Other than the lack of surge suppression, this is a good looking transient filter.
After being smoothed out into a nice clean sine wave, the AC is rectified to DC by a pair of GBU1506L units, each rated at 15 amps and 600 volts. After that it is APFC time. Below are the rectifiers and an overview of the APFC.
Not much to see here, or really not much that we can see. NZXT clearly likes their heatshrink. The controller board is in that large bit of heatshrink. The primary capacitor is a Nippon Chemi-con, rated at a refreshingly high 450 V. This capacitor is Japanese.
There is a shortage in the thermistor department, namely there isn’t one. This means that inrush currents are not limited, but the PSU is slightly more efficient. Less expensive, too. I’d prefer to see an inrush thermistor though.
The primary capacitor is filled with voltage boosted by the APFC bits, in this case that means two 6R165P MOSFETs (650 V, 21 A@20c, 13 A @100c) and a STTH8R06FP (600 V, 30 A RMS, 8 A Average) diode. Pictures were rather difficult to acquire, I was forced to use a bore scope to get the diode’s photo. It could be prettier, but you can read the text at least.
Once boosted, it’s off to the switches and the transformer. The switches are two more 6R165P MOSFETs (650 V, 21 A@20c, 13 A @100c) units, just like the two in the APFC unit. If you want to be really exact, the picture above is of one of the main switches due to the APFC MOSFETs being impossible to photograph.
On the output rectifier side of things, the 12 V rail has four M6060C (60 V, 60 A) schottkys, 5 V has three 30V30CT (30 V, 30 A) schottkys and 3.3 V has two 30V30CT (30 V, 30 A) schottkys. Some overkill, but not to the huge levels we see on gold and platinum units. The protections IC is a WT7527 chip, it supports OVP/UVP/OCP (two 12 V rails), as well as having an auxiliary monitor for other uses (OTP?).
Here are the OVP/UVP trip points, direct from the WT7527 datasheet.
On the subject of OCP, it is interesting to note that there are a pair of shunt resistors on the bottom of the PCB on the 12 V rail. It’s interesting because the datasheet says it’s not needed. Also in the following group of shots are an overview of the output side, a closeup of one of the output filter caps, and a look at the output cables.
The output filter capacitors are only partly Japanese. The foil is etched in Japan and shipped to China, where CapXon cuts, fills and assembles the capacitors. Quality wise, they could be better. China isn’t exactly known for their high quality capacitors. If they do the job that’s good enough though, and these seem to do the job. I’d still like to see Teapo or better in here though, especially when the original HALE82 series used all Nippon Chemi-Con caps. One of the two 5 VSB capacitors is another Nippon Chemi-con, so the box is right. There are Japanese Capacitors in this unit, but only two.
The cables are sleeved all the way into the unit and have a rather nice grommet protecting them from the case. Another interesting feature is a pair of variable resistors on the PCB. I’d guess they’re for final voltage adjustments, but I’m not inclined to mess with them.
Lastly we have the old 3.3 V inductor that had resonance issues at high temperature and low load (left), and the new inductor that works perfectly. Following that, an overview of the soldering (quite good).
The PCB doesn’t have a UL number, but the case says “e190414”, a number that belongs to FSP Group.
The NZXT HALE82-N 750 W is an interesting PSU, while it shares almost the entirety of its name with the previous HALE82, it doesn’t have a lot in common with it other than the looks. The two big changes are that the modular cables are gone and it’s group regulated instead of DC-DC. Both modular cables and DC-DC regulation cost more, making this a clear cost savings measure. Another change internally is from Nippon Chemi-Con capacitors throughout to mostly CapXon capacitors, again this looks to be a cost savings measure. I don’t mind cost savings if the savings are passed on to the customer.
With a MSRP of $109.99, this unit is priced about right. When you consider that almost nothing in the computer world actually sells for its MSRP, I expect this unit will be somewhere in the $90-$100 range within a few months of release. At that price, it’s definitely a good value. What NZXT is showing us is that they actually have a decent idea of current values. As strange as it may seem, this is fairly rare. I’d like to see the price maybe $10 lower given the ripple and CapXon caps, but it’s not really overpriced, even at the full MSRP.
As far as this PSU goes from an absolute standpoint (rather than a comparison standpoint), it looks to be a good unit. The ripple is borderline on 5 V and 3.3 V at heavy loads, and slightly over spec on 5 V when heavily loaded and hot. If you can actually manage to use 100 W of 5 V I’ll be impressed.
The regulation is good, but not spectacular. Certainly no issues there, but it’s not jaw droppingly good either.
The build quality is quite good, NZXT is lavish with the heatshrink and caulk (both of which protect the unit in shipping and reduce inductor whine) and the soldering is quite good overall. There are a few long component leads, but they’re all very carefully bent in safe directions. A couple are even intentionally used to increase current handling on those parts. There are a few bits that could do with a little bit more solder, that’s all that stands between this unit and soldering I’d rate as excellent.
The CapXon capacitors are disappointing, I was hoping for more Chemi-Cons. I’ve seen a number of PSUs with CapXon recently though, and they seem to work well enough. Technically the box’s marketing is correct, there are (two…) “Japanese Capacitors”. The CapXon foil being etched in Japan isn’t enough for me to call them Japanese.
Unlike most reviews, during this review I had to use NZXT’s PSU RMA process. Wow is that easy. They send you a PDF that you print out and take along with the unit to any FedEx location, you give the printed PDF and the PSU (in box) to the FedEx people and you’re done. NZXT pays shipping both ways, you don’t even have to write in NZXT’s address. Ideally of course you won’t ever have to use the RMA service, but it’s nice to know that they care if you do end up having to use it.
The lack of a MOV/TVS Diode as well as the lack of an inrush control thermistor is disappointing. APFC bits can weather small and mid range surges pretty well, but for larger surges you really need some surge protection. That said, you should be using a surge protecting power strip anyway.
The connector selection is a mixed bag. You get a ton of SATA plugs, but of course being non-modular that means you have to stuff them somewhere. You get a decent selection of Molex plugs and a FDD plug, that’s nice. Four PCIe power plugs is perfect for a 750 W unit, but the two 8p connectors could cause you trouble if you want to use a pair of GPUs that both have 6p connectors. Depending on the GPU you may be able to simple plug in 3/4 of the 8p plug and run with it, but that is not ideal and some GPUs with a lot of shrouding will not allow it. For cards that DO use 8p plugs, having solid 8p plugs is very handy. It could be argued that a 750 W PSU is way overkill for two GPUs that use 6p plugs anyway, but being a firm believer in overkill I can’t quite buy that.
The fan does a good job of staying quiet, even at full blast it isn’t an annoying noise. A CPU/GPU combo capable of eating enough power to convince the PSU fan to spin up is almost certainly going to be louder. The box’s marketing blurb says silent, and while not at all loud, it is definitely not silent.
All told there are, as always, pros:
- Looks great, nicely sleeved cables.
- 12 SATA power plugs is a lot for a 750 W unit.
- Plenty of other power plugs, too.
- Good voltage regulation.
- Good fan control/noise level.
There are some cons, too:
- Fails ripple control on 5 V rail by 4 mV when hot.
- 8p PCIe connectors are an issue with two GPUs that want two 6p connectors.
- CapXon caps aren’t especially Japanese.
- No built in surge protection.
All told this is a solid PSU that does what it says it will do and has a decent price, it is definitely worthy of consideration in your next build!