Page 2:Test Setup Overview
Page 3:Voltage Regulation Test
Page 5:Ripple Voltage
Page 6:Hold-Up Time
Page 7:Inrush Current
Page 8:Transient Tests
Page 9:Cross-Load Tests
Page 10:Acoustics/Noise Tests
Page 11:PSU Testing Equipment In Detail
Page 12:PSU Reviewing Is Hard Work!
The power supply unit (PSU) is the most important part of every electronic device, including, of course, computers. It is the heart of your system, since it feeds energy to the other components. Consequently, if the PSU fails, everything else fails with it. This is the reason most experienced technicians start a failure investigation from the PSU before proceeding to the rest of the components. And, consequently, this is why you should pay extra attention to your choice of PSU and not make a decision based exclusively on price. After all, a good PSU will do its job for quite a long time, far outlasting the rest of your expensive system components.
To properly review a PSU, expensive equipment is required, and the reviewer needs to know not only how to operate it, but also have sufficient knowledge about electronics and a PSU's design. The knowledge part is especially crucial, since not even the most expensive equipment can make a good PSU review if the reviewer doesn't know how to properly use it and what tests to conduct with it.
In our reviews, we examine the PSU's performance, noise and temperature ratings, along with the build quality of the units. We also judge the PSU's individual components, cables, connectors and even product specifications and packaging while providing a performance per dollar comparison.
Test Setup Overview
Currently, we use two Chroma 6314A mainframes equipped with six 63123A loads (350W each), one 63101A load (200W) and one 63102A load (2 x 100W). This setup is among the best money can buy, providing the highest possible accuracy. Chroma loads are widely used by all PSU manufacturers and are pretty much the standard for PSU measurements. Our setup allows us to stress the +12V rail with up to 2100W, 5V up to 200W, and 5VSB / 3.3V up to 100W each. All of our equipment is controlled and monitored by a custom-made software suite that's highly sophisticated.
In addition to the Chroma loads, we also use a Chroma AC source (6530), which can deliver up to 3kW of power, a Rigol DS2072A oscilloscope, a Picoscope 3424 oscilloscope, a Picotech TC-08 thermocouple data logger, two Fluke multimeters (models 289 and 175), a Keithley 2015 THD 6.5- digit bench DMM and a Yokogawa WT210 power meter.
Our testing gear also includes a hotbox, which allows us to test a PSU at high ambient temperatures. Finally, we have three more oscilloscopes (a Rigol VS5042, a Stingray DS1M12 and a second Picoscope 3424), and a Class 1 Brüel & Kjaer 2250-L G4 Sound Analyzer, which is equipped with a type 4189 microphone that features a 16.6-140 dB(A)-weighted dynamic range.
The latest addition to our testing equipment is a Flir E4 infrared camera, which through some firmware modifications (many thanks to the fine folks at EEVblog's forums for this) now delivers a resolution of 320x240 pixels. Finally, we have several soldering and desoldering stations that we use during the dismantling process of every PSU we test. Test results are one thing, while checking out the build quality of a PSU is another.
Finally, if we encounter any unusual results during the testing process, we examine the internals of a PSU to find out what is causing the issues.
Voltage Regulation Test
A PSU should be able to keep all of its rails within some predefined voltage ranges at all cases/loads. In Table 1, you will find these ranges (following the ATX v. 2.4 specification).
DC Output Voltage Regulation
Here's an example of a voltage regulation test results chart you'll find in each of our PSU reviews:
The 80 PLUS certification measures efficiency at 20-, 50- and 100-percent load of the PSU's max-rated capacity up to the Gold efficiency certifications. For the Platinum and Titanium levels, they also measure efficiency with 10 percent of the PSU's max-rated capacity load.
Simply put, if a PSU has an 80 PLUS certification, then it must have the equivalent efficiency required by the corresponding certification. However, 80 PLUS measures at a mere 23 °C (73.4 °F) ambient, whereas we measure efficiency at a higher ambient temperature. This means that, in many cases, a PSU that is certified to a certain efficiency category fails to deliver the same efficiency at higher temperatures in our tests.
Also, many PSUs are tuned to deliver high efficiency at or above a specified load percentage, usually the minimum that 80 PLUS measures at the corresponding efficiency level. But at loads lighter than that, their efficiency is pretty low. Since many of us run our systems for long periods at low-energy consumption modes, efficiency at light loads can be highly important. So it's wise to pay special attention to our light-load test results.
In our reviews, we measure efficiency at four different light loads: 20, 40, 60 and 100W.
The ATX specification also states that the efficiency of the 5VSB rail should be measured, too. In the table below, you will find the minimum 5VSB efficiency levels that the ATX specification recommends.
Recommended System DC And AC Power Consumption
|≤0.225W||< 0.5W to meet 2013 ErP Lot 6 requirement (100V~240V)|
|≤0.45W||< 1W to meet ErP Lot 6 requirement (100V~240V)|
|≤2.75W||< 5W to meet 2014 ErP Lot 3 requirement (100V~240V)|
Testing in Standby Mode
In 2010, the European Union released a guideline on Energy Related Products (ErP Lot 6), which states that every electronic device should have below 1W power consumption in standby mode. In 2013, this limit was further reduced to 0.5W. The same year, the EU also released the ErP Lot 3 guideline for computers and computer servers.
This is why we measure the consumption of a PSU in standby mode, which is something that would be difficult without our monitoring software since the readings at such low consumption levels have significant fluctuations. We have to average them over a significant period of time to provide enough accuracy.
Ripple represents the AC fluctuations (periodic) and noise (random) found in the DC rails of a PSU. Ripple significantly decreases the life span of capacitors since it increases their temperature; a 10 °C increase can cut into a capacitor's life span by 50 percent. Ripple also plays an important role in overall system stability, especially when it is overclocked.
The ripple limits, according to the ATX specification, are 120mV for the +12V and -12V rails, and 50mV for the remaining rails (5V, 3.3V and 5VSB). Nonetheless, in modern PSUs, we expect to find much lower ripple. It should be just a small fraction in high-end platforms with quality components and the proper amount of filtering capacitors. Below, you will find a schematic that analyzes a ripple waveform.
In the above schematic, four AC components can be identified:
- Low-frequency ripple associated with AC mains frequency.
- High-frequency ripple due to PWM of the main switches.
- Switching noise that has the same frequency with switching PWM.
- Non-periodic random noise that is not related to any of the above.
Here are a few examples of the typical ripple test results you'll find in our PSU reviews:
Hold-up time represents the amount of time, usually measured in milliseconds, that a PSU can maintain output regulations as defined by the ATX specification without input power. Put simply, hold-up time is the amount of time that the system can continue to run without shutting down or rebooting during a power interruption.
The ATX specification sets the minimum hold-up time to 16ms with the maximum continuous output load. In many cases, manufacturers use smaller capacitors in the APFC converter, resulting in a measurement of less than 16ms. Manufacturers do this mostly to cut production costs, as these capacitors are expensive. The smaller bulk capacitors also improve efficiency by a little bit.
Measuring the hold-up time is a dangerous procedure since you have to connect an oscilloscope to the mains grid. Unless you are taking the right precautions, you never want to do this; it's dangerous and you could harm yourself and your equipment!
Inrush current, or switch-on surge, refers to the maximum, instantaneous input current drawn by an electrical device when it is first turned on. Because of the charging current of the APFC capacitor(s), PSUs produce significant inrush current as soon as they are turned on. A large enough inrush current can cause the tripping of circuit breakers and fuses, and may also damage switches, relays and bridge rectifiers. As a result, the lower the inrush current of a PSU right as it is turned on, the better.
We also noticed that the lower the input voltage, the lower the inrush current. Thus, when operating with 115 VAC, a PSU has much smaller inrush current compared with 230 VAC input. Both our Chroma AC source and the Yokogawa power analyzer have the ability to measure inrush current.
In the real world, a PSU is always working with loads that change, depending on whether the CPU or graphics cards are busy. So, a decent PSU review should contain some tests with dynamic or transient loads. We run two different transient tests:
- While the PSU is working at a 20 percent load state, a transient load is applied to the PSU for 200ms (10A at +12V, 5A at 5V, 5A at 3.3V and 0.5 A at 5VSB).
- While working at 50 percent load, the PSU is hit by the same transient load.
In both tests, we measure the voltage drops that the sudden load change causes. The voltages should remain within the ATX specification's regulation limits.
We also conduct three transient tests where we measure the response of the PSU in its simpler turn-on phase. In the first test, we turn off the PSU, dial a full load at 5VSB, and then switch on the PSU. In the second test, while the PSU is in standby, we dial the maximum load that +12V can handle and start the PSU. In the last test, while the PSU is completely switched off (we cut off power or switch off the PSU's on/off switch), we dial the maximum load that +12V can handle, then switch on the PSU and restore power.
The ATX specification states that recorded spikes on all rails should not exceed 10 percent of their nominal values (+10 percent for 12V is 13.2V and for 5V is 5.5V).
Cross-load tests are the real deal, since they show a PSU's performance throughout its entire operating range. For the generation of the corresponding charts, we test at 28 to 30 °C (82.4 to 86 °F) ambient, and we set our loaders to auto mode through our custom-made software before trying more than a thousand possible load combinations with the +12V, 5V and 3.3V rails. The provided charts show load/voltage regulation along with ripple, efficiency and noise levels.
It is hard to conduct proper acoustics measurements on a PSU since you have to isolate it from the loaders; they are way too noisy and will interfere with the readings. However, if you move the loaders far away from the test PSU, you need long connection cables, which incur increased resistance. Power losses become significant in that case.
Thankfully, we found a way around these issues, and with the help of our custom-made anechoic chamber, we are able to conduct accurate sound measurements using a Class 1 Brüel & Kjaer 2250-L G4 Sound Analyzer.
We measure the fan's noise from one meter away, inside the anechoic chamber, whose internals are completely covered in specialized soundproofing material (be quiet! Noise Absorber kit). Background noise inside the anechoic chamber is below 18 dB(A) during testing, and the results are obtained with the PSU operating at a usual 35 to 45 °C (95 to 113 °F) ambient range.
PSU Testing Equipment In Detail
Chroma Electronic Loads
The electronic loads are the most essential, and the second-most expensive component in our lab. A load tester simulates the load (static or dynamic) and gives us the capability to stress a PSU to its limits.
As mentioned previously, we use a setup consisting of two Chroma 6314A mainframes, equipped with six 63123A loads (350W each), one 63101A load (200W) and another 63102A load (2 x 100W). The above loads allow us to stress even the best PSUs available on the market today, and can handle PSUs with even more than 2kW capacity.
The electronic loads would be useless if we didn't have the proper testing fixture at which the PSU's connectors are connected. In other words, this testing fixture is the bridge that links the PSU being tested with the electronic loads.
Monitor & Control Program
In order to monitor and control our Chroma loads, along with the rest of the equipment (oscilloscopes, temperature loggers, data loggers, power analyzer, hotbox heating elements etc.), we developed a software suite which can also record and analyze all output data.
The development of this program started in early 2010 and is still an ongoing process since new features are added on a regular basis, while old ones are improved. The software suite provides even more capabilities to the Chroma loads that we use, making the PSU testing procedure much easier and accurate at the same time. Another advantage of this software is that it can easily adapt to various types of electronic loads and testing equipment. So if, in the future, we decide to move to another testing platform, the transition process will be almost painless.
Chroma 6530 AC Source
This is the most expensive part of our testing equipment, surpassing even the cost of both of our Chroma mainframes, along with their loads. Nonetheless, it fully justifies its high cost by providing us with the ability to simulate different power-line disturbance conditions. It allows us to simulate a complex mains supply waveform, if needed, while delivering a steady input voltage. It also filters most external noise from the power grid, which can seriously distort ripple measurements.
In addition, an AC source like the Chroma 6530 is capable of simulating all sorts of voltage dips, interruptions and variation waveforms showing the PSU's response under similar scenarios. Finally, this AC source can deliver up to 3kW of power, facilitating the evaluation of any PSU available on the market today (even Super Flower's 2kW unit) without the slightest problem.
Besides the Chroma 6530 AC Source, we also have the lower-end Chroma 61604 at our disposal, which provides up to 2kW of power. This AC source was pushed to its limits during the evaluation of PSUs with over 1.5kW capacity, so we replaced it with the stronger Chroma 6530. Our testing equipment also includes a variable transformer (variac), which is able to deliver up to 3kW.
A power meter provides only basic functions, whereas a power analyzer is a highly sophisticated and expensive piece of equipment able to deliver accurate power and harmonics measurements. We use a Yokogawa WT210 power analyzer to measure the exact (AC) wattage that the PSU pulls from the power grid, along with other crucial parameters like power factor and AC volts/amps. Given a known power consumption on the DC side, we can easily calculate the efficiency (DC watts/AC watts) of a PSU in real time through our testing software suite.
Every PSU reviewer needs a good power meter, or, ideally, a power analyzer, with a high sampling rate. You see, sometimes the APFC stage of a PSU can be tricky, resulting in inaccurate readings with cheap Kill-a-Watts. Unfortunately, a power analyzer is pretty expensive. But if you want to have accurate readings, especially in light loads (<100W) or demanding ones (>1000W), then you have no other choice. As backups, we have a GW Instek GPM-8212, one of the best power meters for its price, and a Prova WM-01 power analyzer. Our main instrument for measuring the electrical characteristics of the PSU is the Yokogawa WT210, which reports directly to the control/monitor program, allowing the calculation of a PSU's efficiency in real time.
To measure voltage ripple on the DC rails with static or dynamic (transient) loads, an oscilloscope is a one-way road. Back in the early days, most PSU reviewers used the limited bandwidth (250kHz) Stingray DS1M12 because it was affordable and did a pretty good job. However, for high-speed transient response tests, a higher-bandwidth oscilloscope is needed. We use a Picoscope 3424 scope for ripple- and transient-response measurements, while our hold-up tests are conducted with the help of a Rigol DS2072A scope.
Although almost all load testers are equipped with their own current/voltage meters, a good multimeter is always useful in a PSU review. It doesn't need to be a 4.5-digit one. However, it needs to be recently calibrated in order to provide accurate values.
We have a large number of multimeters at our disposal, including a high-end Fluke model with 4.5 digits (289), a mid-range one (175) and a high-end bench multimeter with 5.5-digit precision (Keithley 2015 TDH). We also have a Labjack U3-HV multifunction data acquisition (DAQ) device to obtain some real-time readings from the PSU being tested. Finally, we have a DMMCHECK Plus device, through which we are able to check the accuracy of all equipment we use for measurements.
We strongly believe that there is no point in measuring a PSU at room temperature since it will spend all its life inside a case where temperatures will be much higher. So, the most interesting test results are the ones obtained with the PSU operating at temperatures above 40 degrees Celsius (104 degrees Fahrenheit) ambient. The ATX specification states that a PSU should be able to operate at ambient temperatures of 10 to 50 °C (50 to 122 °F) at full load, with a maximum temperature change rate of 5 °C (41 °F) per 10 minutes, but no more than 10 °C (50 °F) per hour.
The 80 PLUS organization tests efficiency at only 23 °C (73.4 °F) ambient, which, in our opinion, is way too low for this purpose. Performance at high operating temperatures is what separates the good PSUs from the mediocre and bad ones. A well-built power supply should be able to output its full power continuously at up to 50 °C (122 °F), while lower-grade ones can only hang at up to 40 °C. Finally, low-quality PSUs are restricted to 25 °C (77 °F).
In order to set a standard for our PSU reviews, we decided to conduct our full load tests at 45 °C (113 °F). In case a PSU explodes or delivers a really poor performance under the above conditions, we note this in the review and subtract the corresponding performance points from its overall performance score.
To be able to test at high ambient temperatures, an environmental chamber is needed. We constructed our own hotbox with heating elements that are software-controlled and can operate in a fully automatic mode. That is, we set the desired temperature and the heating elements operate accordingly in order to keep it within a specified range.
Digital Thermometer/Temperature Logger
An accurate digital thermometer or temperature logger is essential for the measurement of the PSU's intake and exhaust temperatures. For this purpose we use a Pico TC-08 temperature logger with eight probe inputs, and a CHY 502 thermometer as a backup, with two thermocouple inputs.
Similar to the power meter and the power analyzer case, the difference between a sound meter and a sound analyzer is huge. A sound meter can only report basic sound measurements, while a sound analyzer provides a detailed look at the output noise, including frequency analysis, FFT sound and vibration, sound recording, and so on.
The American National Standards Institute (ANSI) classifies sound level meters in three different types: 0, 1 and 2. Type 0 is used in laboratories, while Type 1 is used for precision measurements in the field and Type 2 for general-purpose measurements. The better Type 2 sound meters usually have an accuracy of ±2 dB(A), while a Type 1 sound meter usually has ±1 dB(A) accuracy. IEC standards divide sound level meters into two "classes." Class 1 instruments have a wider frequency range and a tighter tolerance for error than the lower-cost Class 2 units.
Another important factor in noise measurement is how "low" a sound meter can go. For example, most inexpensive sound meters simply cannot measure anything below 30 dB(A) because of their low-quality microphones. To be able to measure below this, you need to invest in a good Type 1 or Class 1 meter with a decent mic.
Our Class 1 Brüel & Kjaer 2250-L G4 Sound Analyzer is equipped with a type 4189 microphone that allows it to measure noise down to 16.6 dB(A). To fully exploit its capabilities, we built a mini-anechoic chamber, inside which we were able to measure down to 17 dB(A) during quiet hours. However, a good sound analyzer is worthless if it's not calibrated regularly (ideally, before any sound measurement). For this reason, we use a Brüel & Kjaer type 4231 sound calibrator, which provides a calibration accuracy of ± 0.2 dB(A).
An infrared (IR) camera isn't an essential piece of equipment for a PSU review, but it can be useful if you need to check the thermal dissipation and the cooling fan's performance. We use a Flir E4 and, thanks to the ingenious folks at EEVblog, we managed to raise its resolution to 320x240 pixels.
Soldering and De-Soldering Tools
In addition to the required testing equipment, we also have the tools to fully dismantle a PSU for a build quality check. We use a Thermaltronics TMT-9000S soldering and rework station. We also have a Hakko 808 desoldering gun, an AOYUE 474A desoldering station and a Weller WSD 81 soldering station.
PSU Reviewing Is Hard Work!
In this article, we described the necessary equipment needed for a proper PSU review and outlined our testing methodology, which includes a rich variety of tests. As you have probably noticed, PSU reviewing is a complex, time consuming and expensive task. Even if you possess all of the necessary tools, you need to have the proper electronics background as well so that you can dismantle the PSUs without damaging them or harming yourself. You also need to be able to provide an accurate analysis of the PSU's platform and its components in order to judge its build quality. PSU reviewing admittedly is a hard job, but someone has to do it and, thankfully, we enjoy the work!