A visit to the Enermax Lab in Hamburg – 6 PSU tested
Enermax and Lepa
For almost two years now, all sales and marketing operations for Enermax have been handled by their German office located in Hamburg. With the expansion of the office, a top-notch testing lab was put together, which under brief approximation easily exceeds the 30.000 euro mark in equipment cost alone. While the main purpose of this lab is to test retail samples that will go on sale in the European market, we were invited there to have some fun testing PSUs with their equipment.
For us, this turned out to be an excellent occasion to address the recent community requests for more in-depth power supply reviews. Today we’ll show you how a LAB501 branded review would look like, if our budget would allow us to acquire such equipment and perform the required annual calibrations.
I am sure that Enermax doesn’t need any more introductions for our readers. The company was founded in 1990 in Taiwan, boasting a 20 year history in power supply design, manufacturing and retail, but also known for their cases, peripherals, coolers and fans. We’d also like to point out the fact that Enermax, alongside Antec and Chieftec, was among the first brands of quality power supplies to be found in the Romanian market a decade ago.
Lepa on the other hand is a brand founded in 2011, with Enermax acting as the main OEM for them but also managing their European sales and marketing operations through the same Hamburg office.
From the moment I entered the Enermax office my target was clear, the testing lab. However, on my way there a computer case draw my attention, which usually doesn’t happen to me with Enermax cases.
A classical look created by the simple, straight lines and the predominantly black chromatic is combined with the contrasting white of the front grille wrapped around the frontal mask. The Ostrog case is surprisingly pleasing aesthetically but also has the features and the functionality to please almost anybody, especially considering the price tag of only 40 euros.
Enermax cases didn’t appeal too much to europeans until they establish the main European office in Hamburg. Since then, the team has grown significantly, being composed by europeans that understand the needs and know the taste of the people living in Europe. So I’m glad to say: bye bye shiny black, bye bye chrome and, especially, bye bye multicolour lights used in excess!
The lab and equipment
Now we’ve reached the interesting part for most LAB501 readers, mainly high-tech enthusiasts and hardware geeks. Despite the fact that it doesn’t occupy an entire room – as you would imagine – the testing equipment present in the Enermax lab is downright impressive, summed up in a single word: Chroma.
Chroma is the world’s largest manufacturer of Automatic Testing Equipment (ATE) for power supplies. A testing platform is formed by one or more mainframes, which houses a number of modules, set up and configured according to particular requirements (a maximum of 4 modules for the 6314 mainframe or 2 modules for the 6312 version).
I’d like to point out something here, when compared to Chroma, other solutions from SunMoon, FastAuto or TechRed tend to look like “toys” designed for testing PSU’s for fun. Chroma plays in a whole different league, but also their equipment can cost 10 times more than the previously mentioned competitors and require a very high level of knowledge and a lot of work to set up and use properly.
For our test we had at our disposal a platform composed of two Chroma 6314 Mainframes, each housing 4 (four) Chroma 63103 modules, each being capable of generating a 300W load, for a total of 2400W. 6 (six) modules are loading one rail each (3.3v, 5v, 12v3, 12v4, 12v5, 12v6), while the other two modules are loading the 12v2 rail for a maximum capacity of 600W (50A).
Above we have a newer generation Chroma Mainframe, the 6314A, outfitted with two modules. The first one is a Chroma 63102A which can simultaneously load two rails, up to 100W per channel, used for the 5vSB and the -12v rails. The second module is a massive Chroma 63106A that can put an 800W load on a single channel, being used to load the 12v1 rail. Everything can induce a staggering 3.2KW load (excluding the 5vSB and the -12v rails), which is much more than any PSU can output now or in the next 4-5 years.
On top of the mainframe tower, we find the Chroma 66202 Analyzer. This monitors the AC side of things and reports the AC Voltage, AC Amperage, AC Power Draw (in Watts) and the PFC value. Do not mistake this for the usual wattmeter we typically use, the Chroma 66202 is in a completely different precision class.
Also deserving some special attention, the connector board is an in-house design by Enermax (Chroma does not offer such equipment, only measurement and load banks). For the ripple and noise measurements we used a high-end digital oscilloscope, a Tektronix TDS 3014C, connected to the rails with two capacitors sized in accordance with the ATX12V Power Supply Design Guide.
The connector sockets are situated on a thick heavy duty PCB which can sustain very high loads without negatively impacting the measured values. The Chroma modules display the current in amps (measured by a sensor inserted between the load and the PSU) as well as the voltage at the connecting terminal.
Chroma recommends placing the measuring probes together with the load wires, at the edge of the PCB. However, this method can’t offer a truly precise reading because both the PCB and the connectors generate resistance. For more precise measurements, we positioned the probes directly into the power supply connectors.
When it comes to precision measurements, Chroma goes beyond the typical high-end voltmeters, going below the mV range. More so, to maintain the highest level of precision, both the oscilloscope and the Chroma modules are calibrated annually by a specialized company.
High-end equipment and high precision measurements are not going to do you any good in the absence of an equally good methodology, one which can identify and single out both the strengths and weaknesses of each tested unit. Personally, I find none of the methodologies used by the dedicated PSU review sites 100% relevant and representative for the end user (no offense). So we had to come up with the following 7 stage load pattern for our tests:
- 10% – low load scenario, would qualify for idle values on high output units (think over 1KW), entry level units tend to display very poor low load efficiency.
- 20% si 50% – standard ECOVA 80+ certification load.
- 40%, 60% and80% – monitoring the relevant parameters as the load increases.
- 100% – full rated output, relevant for obvious reasons, for ECOVA 80+ certification as well
- Overload – an ignored aspect by most PSU reviewers, but a very important one for us. The PSU is pushed to the point the protection is triggered (OCP, OPP, etc) or the electrical parameters go out of spec (voltage or ripple noise below ATX specifications).
At each of these loads, we’ll provide current and voltage measurements together with ripple noise on four rails (two 12v rails and both minor ones, 5v and 3.3v) on the DC side. On the AC side we’ll measure the PFC and the power drawn, which will allow us to precisely calculate efficiency. For reference, the ATX limits for ripple noise are as follows: 50mV for the minor rails (3.3v and 5v) and 120mV for the 12v rails.
If you are suspicious in any way, I can tell you that all the units are retail versions, picked from the warehouse and opened in front of me. This methodology is conceived by yours truly with no input or directions meant to favour the tested units in any way.
Another typically ignored aspect is the load applied for each specific rail, which shouldn’t be random or judging strictly by the PSU label. More specifically, a 20A rating for 12v1 means nothing if that specific rail only feeds the 20+4pin ATX connector, which in turn only has two 12v wires each supporting up to 7A each. So that’s 14A total, 15A if you take into account the tolerance margins, far from the 20A rating on the label. Applying 20A on this connector will only overload the wires and the resulting voltage drops shouldn’t surprise anyone, and besides that, there’s no way on this earth that the motherboard will ever draw 20A (240W) through the 24-pin ATX connector.
That being said, it’s important to know the connector-rail arrangement before connecting them to the test board, as well as the number of wires which will support the loads so we can roughly calculate the load on each one. Today we’ll test 6 power supplies from Enermax and before moving on we’ll present the connector-rail arrangements for each one:
Enermax NAXN 450W
Enermax Revolution87+ 850W
Triathlor ECO 650W
Enermax Revolution X’t 530W
Enermax Platimax 1350W
NAXN 450W: Introduction
We start our testing session with the NAXN 450W, a representative model of Enermax’s entry-level lineup. This PSU is on the market for some time already and it is in fact named NAXN 82PLUS, because of its 80PLUS Bronze certification.
The packaging doesn’t impress in any way, but the PSU itself comes with two interesting features for an entry-level product. HeatGuard (the fan spins for an extra 30-60 seconds after system shutdown, to cool the PSU components but also to evacuate the heat from the system itself) as well as long, fully sleeved cables.
Some of you may already know, but I’ll mention anyway that the NAXN lineup isn’t manufactured by Enermax itself, but by an OEM manufacturer. This is a less important aspect, what truly matters is the actual performance of the PSU, analyzed on the following page.
NAXN 450W: Results
First off we notice the very good efficiency this unit displays at medium loads (40%-60%), above 86%, while at full load the efficiency is still above 84%. The only scenario in which the efficiency drops below 80% is at 10% load, but let’s be honest, it’s tough nowadays to build a system that would only take 45W, even at idle.
With the NAXN, we have two overload scenarios. The first one loads the PSU to 527.93W, meaning 18% above its rating. In this case, the 12v line didn’t drop below 11.86v and the ripple values were good (far from the ATX limits), while the measured efficiency was 82.7%. Truly impressive behaviour for an entry-level power supply one might say.
Moving on we took the PSU to 27% above its rating, at which point we were pulling 566.45W from it. Despite the 12v rail still hoovering at 11.85v, the ripple noise exceeded the ATX specs by a good margin, on all four measured outputs. Even though I haven’t triggered any protection, I decided to stop because it’s dangerous to run a system under such ripple.
The fan can’t be characterized as silent, but it kept a decent noise profile even when the PSU was seriously overloaded.
Revolution87+ 850W: Introduction
The Revolution87+ series is the successor of the well known Revolution85+ PSU line, which was phased out by Enermax. The Revolution85+ name doesn’t bring up any fond memories, with its modest power reserves and outrageous price tag.
So, is the Revo87+ a new platform or simply a cosmetically revised version of the old series? I’m pleased to say that the Revo87+ uses a tweaked version of the excellent Modu87+ platform, so I’m quite anxious to see the results.
The unit is modular, offers the same HeatGuard feature mentioned earlier and it’s certified for continuous operation at 50oC.
Revolution87+ 850W: Results
Can’t decide what to tell you first about Revo87+: the ultra quiet fan which became noticeable only at loads beyond 1000W, the 92% efficiency displayed at medium loads (40%-60%) or the fact that the 12v rails got stuck at ~12.1v even when delivering 38% more power than the unit’s rated output?
Yes, you read that right, the Revo87+ 850W was able to deliver 1180W without any issues whatsoever, situation in which the ripple noise was still well inside the ATX specifications (albeit the minor rails were close to the ATX limits). As if this wasn’t enough, the 84.55% efficiency displayed at 1180W was bordering the 85% efficiency required for 80+Silver certification.
For fun we calculated the largest voltage variation on the main rails, specifically the 12v4 which dropped from 12.27v (@ 10% load) to 12.08v (@ 138% load) translating into a 1.5% drop… brilliant power supply! Beyond 138% load the OPP was triggered 🙂
Lepa G1600: Introduction
The most interesting PSU present in this review is, as expected, the Lepa G1600, a 1.6KW monster boasting an 80PLUS Gold badge. As if 1600W was not enough, the specs tell us this unit is capable of 1700W continuous output and up to 93% efficiency at 230vAC. We’ll test these two claims today, since we have at our disposal equipment capable of generating more than 3.2KW load.
Despite being based on the Platimax 1500W platform, the Lepa G1600 is considerably more affordable than its brother-from-another-mother. To make this possible, Lepa made a few compromises of which we could mention the less impressive box and packaging, the slim bundle, the less expensive casing, getting rid off Enermax proprietary technologies such as the HeatGuard, using a cheaper and noisier fan, etc
Of course, what matters the most (the electronics) was kept intact. We’re dealing with the same fully modular design, 105oC rated Japanese capacitors and solid polymers in the secondary, and for the minor rails we have the DC-to-DC VRMs.
Lepa G1600: Results
We find the values displayed in the light load scenario somewhat amusing, considering at 10% load we reached 168W, a value considerably larger than the power requirements of a typical entry level system. At 20% load the unit is returning an impressive efficiency value of 91.48%, at a nominal output of 336W, more than most generic 15$ power supplies can cope with.
The only scenarios in which efficiency dropped below 90%, were the full load test (100%) and Overload test (110%). Ok, maybe 110% may seem somewhat unimpressive for an overload test, but you need to keep in mind that in absolute terms this means 1755W, or around 2000W AC drawn from the power outlet… doesn’t seem that low now, eh?
The excellent ripple results observed even at 1755W output, vouch for the quality design and excellent electronics. The largest voltage variation can be observed on the 12v1, where the rail drops from 12.27 (@ 10% load) to 11.96v (@ 110% load), meaning 2.6% which is very good considering the enormous power provided.
If the Revo87+ impressed us with its quiet operation, the G1600 can comfortably rival a vacuum cleaner in noise output when its loaded beyond 1100-1200W.
Triathlor ECO 650W
We continue the tests with Triathlor ECO 650W, a 80PLUS Bronze PSU that is part of the mainstream range from Enermax. Note that the model tested is a recent appearance on the market, being an upgrade to the original Triathlor released a while back (550W model has been tested by us here).
The packaging is not very impressive, but instead the PSU comes with two interesting features: HeatGuard (fan still rotates 30-60 seconds after shutdown to cool your PC and also the power supply’s internal components) as well as a semi-modular design which includes flat cables for better wire-management.
However I have to say that, compared to the original models from the Triathlor series which was based on Enermax’s own Revo87+ design, to get a more competitive price the new Triathlor ECO models are based on an OEM design. However, more important are the test results and that’s what we will see on the next page.
Triathlor ECO 650W Results
Firstly we notice the great efficiency displayed by the PSU, very close to obtaining the 80PLUS Silver certification, and that’s something worth mentioning for a mainstream PSU. The good efficiency is maintained when we move to the low-load scenario (10%) but also when we get to Overload (if we round the results we can see 80PLUS Bronze there also).
Triathlor ECO 650W provided stable voltages from one end to the other in our tests, the variation on the +12v rail did not exceeding 1.08% in the worst case (Overload). If we mentioned it, Overload capability is also good, not less than 27% over the power certified on the label. This means that we got 814.4W from this 650W PSU, which is not bad at all.
The ripple noise maintained very good levels, even when I have abused this PSU with loads for which it hadn’t been designed. At 100% load Triathlor ECO manages to keep a ripple level that does not even reach half of the maximum ??allowed values as specified by the ATX standard. That shows that we are dealing with high quality electronics, even if it’s not Enermax’s own design.
Revolution X’t 530W
X’t is the most recent appearance in the Revolution series, placed just under the Revolution 87+ in the current range of Enermax. Revolution X’t is targeting the mainstream market of 80PLUS Gold certified PSU designed for the price-sensitive clients.
Inside we have a platform developed by Enermax with DC-to-DC VRMs on the secondary, while the capacitors used are mostly manufactured by CapXon (exception is made by the primary capacitor being manufactured by Panasonic in Japan). The 13.9cm fan is based on the patented Twister Bearing technology, so they did not make too many compromises in terms of features and quality.
Talking about the technology implemented, Revolution X’t comes with: HeatGuard (fan still rotates 30-60 seconds after shutdown to cool your PC and also the power supply’s internal components), CordGuard (system that prevents accidental cord removal) as well as a semi-modular design which includes flat cables for better wire-management.
Revolution X’t 530W Results
Firstly it should be noted that I felt immediately the switch to a higher efficiency model and the transition from one 12cm regular fan to 14cm with Twister Bearing, the noise emitted by the Revolution X’t being significantly lower than that of the PSU tested earlier. Efficiency was measured as the standard for 80PLUS Gold, the only problem is at 50% load where the PSU missed a few hundredths to achieve the necessary 92%, but we can put that under the measurement tolerances.
In the Overload test the little 530W power supply managed to achieve 625.7W, maintaining an efficiency of over 88%… wow. If we look at supply voltage ripple noise we can observe that the results are somewhat weaker than Triathlor ECO 650W , though it’s admirable that it didn’t exceeded the ATX specifications not even in the Overload test .
Analyzing the results, is pretty easy to see that the Revolution X’t is targeting a lower segment of the market compared to the other PSU in this series: Revolution 87+. In addition to the efficiency and noise emission which are indeed lower, it does not feel like a superior PSU compared to the one presented earlier. Furthermore, I would even say that it’s inferior judging by the ripple noise or Overload capability.
The last PSU tested here is probably the most impressive of them all, the 80PLUS Platinum certified 1350W monster, Platimax 1350W.
The colour theme used by Enermax for the box, silver-based, is nothing new for 80PLUS Platinum PSUs for obvious reasons. The package itself is generously sized and is carefully divided so that it fits the PSU modular cables, power cord, Velcro straps, 4 mounting screws, user manual and two pouches to deposit unused modular cables.
The PSU keeps the same neutral colours but the glossy fan mask interrupts the monotony. The rest of the PSU housing is painted matte gray and has a rough finish that it’s used more and more as time passes by. The construction is very good, although we noticed that Enermax replaced the 14cm Twister Bearing fan used on all the other Platimax models with a 14cm fan manufactured by ADDA. The PSU is completely modular and, besides the 24-pin and EPS cables, we have ribbon cables for easier wire-management.
As expected in a high-end PSU, Enermax adopted DC-to-DC technology that involves obtaining the +3.3v and +5v voltages from the main +12v rail, using two VRMs. Being the flagship of the taiwanese manufacturer, Platimax 1350W has the most advanced technologies in terms of R&D: FMQ design, Copper-Bridge Array Transmission, CordGuard, HeatGuard and exclusive use of Japanese capacitors certified for 105 degrees C operation.
Platimax 1350W Results
One can easily see that Platimax 1350W boasts top of the line electronics. The measured ripple noise is extremely low, even when we apply two levels of Overload on the PSU: 1631W (20%) and 1799W (33%). Wait… what? Yes, you read right… Platimax 1350W could deliver 1800W without any problems, stability issue and the ripple noise were unbelievably small for such a load .
We can notice the rock-solid electronics design also by judging the 2.75% variation between the lowest and the highest +12v rail voltage. I can say that I’m really impressed by that! If Lepa G1600 delivered smoothly 1755W, Platimax 1350W exceeded all expectations and managed a clean and stable 1800W… maintaining also a 88.64% efficiency.
Efficiency obtained at 20% and 100% loads complies with the 80PLUS Platinum standard, in contrast the 50% load test attained only 92.48% efficiency instead of >94% as specified by the standard. I changed the parameters, varied the settings but I couldn’t get more, maybe it’s a problem of the test sample or maybe the 80PLUS organization measures differently. Anyway, the efficiency is enough but I felt the need to point out this issue.
Every power supply tested today impressed us in its own way:
The NAXN 450W impressed us with its very robust power reserves, but also with its very good efficiency, its cable length and design.
Although it is based on an OEM platform, Triathlor ECO 650W shows its muscles and manages to exceed 800W in the Overload test while maintaining stable voltages and unexpectedly good ripple noise for this price range. Oh, and the PSU efficiency is very close to the 80PLUS Silver standard…
Revolution87+ 850W couldn’t disappoint, being based on the Modu87+ series which is a solid, proven platform. Effortlessly delivering 1180W, 330W more than it’s rated for, while providing rock solid regulation, coupled with an ultra quiet fan at light and medium loads, certainly makes a compelling argument for a recommendation targeted at enthusiasts.
Revolution X’t 530W is not really worth the Revolution name, mainly because of the limited power reserve but also the ripple noise which is a little disturbing considering that we are not dealing with an entry level PSU. Well, we have the 80PLUS Gold efficiency and quiet operation but still, knowing the results of the Revolution87+ I expected more from the X’t.
The Lepa G1600 and Platimax 1350W are based on the most advanced platform in terms of build and electronics quality. Both PSU’s are a demonstration of power, being capable of high quality output, noise and ripple levels low enough to make most 600-700W power supplies die of spite. It’s a monster power supply intend for extreme overclocking, it won’t say no to anything, the Overload test further certifying that. Maybe a Gulftown in conjunction with a GTX 480 4-way SLI setup could push it beyond its limits, I can’t think of anything else that would make it sweat.
The tested power supplies weren’t the only ones to impress us today – visiting the Enermax lab was a very pleasant experience, the opportunity to work with such testing equipment isn’t something many people can boast about. Like with a good motherboard which isn’t undermining your overclocking efforts, allowing you to push the CPU or the memory to their limits, a stable and well calibrated testing platform makes possible a scrupulous and in-depth PSU review. While testing, I’ve noticed that the only varying parameter under a certain maintained load was the efficiency, which dropped slightly as the PSU was getting hotter. Beyond that, the voltages, loads and everything else, were extremely stable, even if the Chroma was getting hot..
If our budget would ever allow it, this is the exact setup we would use for power supply reviews at LAB501. We find other load testers (ready-made or home made) unacceptable from a result perspective, as well as limited in terms of absolute load values and ability to maintain stable loads for relevant time intervals.
Until then however, we prefer to stick to our current methodology, which has the advantage of offering relevant results from an end-user perspective, with testing being done on an actual system. Maybe the methodology presented today can tell us everything about a PSU in terms of numbers, but in the end the home user will drop the PSU in an actual system, and as we’ve seen in the past, a well though out practical methodology is much more relevant for the actual user than a synthetic methodology executed using less precise, less accurate and generally less capable equipment than the one shown by us today. Don’t get me wrong, it would be a pleasure to test each and every unit using such a setup, but since the cost is still prohibitive for us, and we find anything below it an unacceptable compromise, we’d rather not do it at all.