by Oleg Artamonov
01/28/2005 | 05:24 PM
This time around we got four power supply units (PSUs) of the high-end category (at least their manufacturers claimed them to be such): one model from each Antec and BeQuiet, and two models from OCZ Technology. But here we are not as much interested in the positioning of the PSUs or their retail price as in their design features. Three of these four units have dedicated voltage regulation of each of the main power rails (+5v, +12v, and +3.3v), which is a very rare thing indeed in today’s computer PSUs. You can only meet it in top-end models, and it is the first time we have such units in our test labs.
So I am the more eager to compare their characteristics with classic-design units (namely, the stability of the output voltages, since the rest of the characteristics of a PSU don’t depend on the additional voltage regulators).
The first part of this review is dedicated to the theory of the problem – the problem of stability of voltages and its possible solutions and the explanation why the developers prefer the design we will see in the tested PSUs.
As I said in my previous article called X-bit Presents: Power Supply Units Testing Methodology, we use to test computer power supplies, one of the problems of each computer PSU is the lack of independent regulation of the output voltages. In other words, if the load on one of the PSU’s outputs changes, the output voltages on the other power rails change, too. This comes as there’s physically just one regulator inside the PSU, and this regulator works basing to some averaged load on the output rails of the unit.
This problem has become the more urgent today now that the ATX12V 2.0 standard has been accepted in which the main load has shifted finally from the +5v rail to the +12v. That is, you won’t have any problems with a power supply that strictly complies with the ATX12V 2.0 standard except that this power supply won’t work with powerful computers of old production dates that put a heavy load on the +5v rail. While Intel’s platforms have mostly relied on the +12v for a few years already, many mainboards for AMD’s Socket A processors have powered up the processor from the +5v until quite recently, and top-end processors have put a very high load on this rail.
If the manufacturer wants to create a universal power supply, i.e. equally suitable for both +12v-oriented systems (computers with Intel’s processors and AMD’s Athlon 64 belong here) and for older +5v-oriented computers, the output voltages must be independently regulated. Otherwise, the PWM controller of the PSU has to be predisposed to a high load on either the +5v or the +12v power rail. Yes, it is possible to find a compromise between these two requirements, but it would rather mean that the PSU would regulate both voltages equally badly rather than equally well. Of course, bad regulation doesn’t satisfy the user as well as the manufacturer. So, again, independent regulation of the voltages is a must.
A straightforward solution would be to install several (at least two) PWM controllers into the power supply, but you wouldn’t build a PSU of a reasonable size and cost this way, since you have to put at least two pairs of power transistors and at least two power transformers… Thus, the classic one-PWM-controller design still holds – but with additional regulation of the output voltages!
A linear regulator, one that dissipates the excessive power on the control element (transistor), would be the simplest device for this additional regulation. Alas, but the efficiency factor of linear regulators is very low – the very principle of their operation implies that there must be that surplus power. In other words, the input voltage of such a regulator is always higher than the output one, and the power corresponding to the voltage difference is dissipated as heat. Best samples of modern linear regulators are capable of working at an input-output voltage difference of 0.5 volts. It means that each such regulator in a PSU would dissipate over 20 watts of heats under a high load. These linear regulators would require additional and quite large heatsinks and forced air cooling, and thus they are not acceptable.
It would also be problematic to add a dedicated switching regulator to each of the output rails. Although, placed at the output, it wouldn’t have a high-frequency part and related problems as the main regulator has, but each switching regulator outputs pulsating (rectangular, to be exact) voltage, so a smoothing LC filter has to be put after it. One LC filter already stands at the output of the main regulator, but it’s impossible to put additional regulators before it: they would receive the pulsating voltage from the main regulator. The auxiliary regulator’s own pulsation being superimposed on it, there will be low-frequency, practically non-filterable fluctuations at the output. And if we attach the auxiliary regulator after the main regulator’s LC filter, there will still be rectangular pulsations requiring a second filter. Thus, the use of auxiliary switching regulators would call either for doubling the number of filters at the PSU’s output (and they take up quite an amount of space) or for complicating the already sophisticated circuit with synchronization of the auxiliary regulators with the main one.
The most reasonable way out of this situation is the use of the so-called magnetic regulator or magnetic amplifier, which employs an inductive element as a controlled switch. While the main regulator generates rectangular pulses of a certain frequency, adjusting their length as necessary, the magnetic amplifier cannot generate anything, but can reduce the length of the pulses generated by the main regulator.
The picture below shows you a simplified schematic of a magnetic amplifier taken from the materials of a Unitrode Corporation seminar on the topic:
In this figure Ns is the secondary winding of the main regulator’s power transformer, on which there are rectangular voltage pulses with an amplitude of, say, 10 volts (the oscillogram of the voltage V1 in the figure), a period of 20 microseconds and a duration of 10 microseconds. Without the choke L1 and the voltage source Vc (6v) the voltage coming to the load would be 5 volts.
Suppose, the core of the choke L1 was saturated at the beginning and the choke let pass the current freely. At time t=0 the voltage on the output of the main regulator changed for negative, and the voltage on the choke is 4v (V1-Vc = 10v - 6v = 4v) and remains such during all the time of the passing of the negative pulse, i.e. 10 microseconds. This switches the choke from the saturated state into high-inductance unsaturated one in which it inhibits the current flow, as any inductance does. So, when at time t=10µsec, there appears positive +10v voltage on the output of the main regulator, the current doesn’t flow through the choke and the voltage V3 on the regulator’s output remains zero. But now the 10v voltage is applied to the choke, driving it back to saturation, but in 4 microseconds (this time depends on the length of the negative-polarity impulse and on the voltage difference on the choke: 10ms*4v/10v = 4ms). Thus, at time t3=14µsec the choke becomes saturated and starts to pass the current through, and the output voltage grows up to 10v. In 6 microseconds more, at time t4=20µsec, the voltage V1 on the output of the main regulator drops to ‑10v, so the output voltage V3 goes down to zero.
Thus, without the choke L1 the load received pulses of 10µsec duration and a period of 20 microseconds, but with this choke the period remains the same and the duration is diminished to 6 microseconds. Easy to calculate, this leads to a voltage drop from the earlier 5v to 3v. The dissipated power is small here – when unsaturated the choke doesn’t practically conduct the current; in the saturated state, the voltage drop is close to zero, so there’s no significant loss. The power consumption from the source Vc is only determined by the material of the choke’s core and the number of coils, so it can be made small irrespective of what current the regulator should give out to the load. In practice, the voltage that determines the length of the pulses on the output of the auxiliary regulator (and, accordingly, its output voltage) is not fixed, but has feedback with the regulator’s output, so the output voltage can be kept the same almost irrespective of the load.
Thus, we have a voltage regulator with a high efficiency factor (and thus not requiring forced cooling), which easily works in pair with the main regulator and takes little space. The choke L1 is rather small, and the rest of the elements are not even worth mentioning.
Until very recently such regulators were typically used on the output +3.3v rail of ATX power supplies. This voltage was derived from the 5-volt winding of the power transformer in the above-described way. Yet it is clear that nothing prevents us from regulating the rest of the voltage likewise. We can just wind the power transformer in such a way as to make its output voltages 2-3 volts above the required levels and then put a magnetic amplifier on each of its outputs that would step the voltage down as necessary. Thus, we have a two-step regulation of the PSU’s output voltages at a relatively low cost, and this PSU is more tolerant to variations in the load than ordinary units with group regulation of the voltages.
Well, considering that there are only three critical output voltages in the computer PSU, among which one (+3.3v) is already regulated with a similar circuit, the problem appears most simple. It is only necessary to regulate one more voltage with a magnetic amplifier and to send a feedback signal from the remaining voltage to the main PWM controller as it can handle the task of regulating one voltage with ease. The negative output voltages can be regulated by ordinary low-power linear regulators – load currents are small there, so they won’t heat the PSU up much.
We haven’t yet tested such units on our site – they are rather scanty in market and cost quite a big sum of money, but for our today’s tests we’ve got only one unit (BeQuiet P4-450W) designed classically, while the products from OCZ and Antec are designed as described above – with dedicated auxiliary regulators on their outputs.
But first I want to present you a small table with the basic load characteristics of the units I am about to test:
Externally the True430P power supply doesn’t differ from a majority of PSUs of the same price category from other manufacturers – a neat case of thick steel; a number of output connectors on thick wires, carefully tied up with nylon braces; two 80mm fans… I guess there’s only one unusual feature here – a connector on the external wall of the PSU that outputs +5v and +12v in case you need to power up some peripherals:
This is in fact a standard connector like those you use to power up the hard disks and optical drives. I only want to note that the restriction of 6amp refers to the total current of the output voltages. Inside the case, inline with the ground wire of the connector there is a self-healing 6amp safety device.
Here’s a surprise for you inside – instead of two traditional chokes on ring cores (one group regulation choke with three windings for +5v, +12v and -12v, and one magnetic amplifier of the +3.3v regulator) we have three chokes here, two of which have one winding and the third choke has two windings:
This is just the dedicated-regulation circuit I was talking about above. The PSU’s main regulator receives feedback only from one of the output voltages, while the remaining two are regulated by their own auxiliary regulators. The result is an excellent stability of all the voltages. The second winding on the third choke is required for the -12v voltage. This choke works like the group regulation choke in the classic PSU design, but now for one voltage only (as for the ‑5v voltage, it can be derived from the -12v by means of an ordinary linear regulator as the load currents for the negative voltages are very small). The stability of this voltage isn’t important: the negative voltages may deviate by up to 10 percent off the norm, while the positive by 5 percent only.
The active PFC device is found on a separate board that is held on poles above the main PCB.
The rest of the PSU’s assembly leaves only pleasant emotions. It’s all conscientiously made – I can’t have any gripes at all.
My measurements showed that the efficiency factor of this power supply was about 85 percent under full load, while the power factor almost equaled 1 – thanks to active PFC.
High-frequency pulsations of the output voltage under full load proved to be within the norm on the +5v rail (green line of the oscillogram) as well as on the +12v rail. Their swing was about 25-35 millivolts, while the maximum allowable values are 50 and 120 millivolts, respectively.
If the load is above 300 watts, there became apparent low-frequency pulsations on the +12v rail (100 hertz or the double frequency of the power grid here), but their swing is no more than 40 millivolts under the maximum power load, which is quite normal. On the +5v rail I observed no low-frequency pulsations at all.
The speed of the fans in the unit doesn’t vary greatly under different loads. The maximum speed isn’t high but it suffices for keeping the unit cool.
The most interesting part of our tests is the cross-load characteristics of the unit. Before showing you the results I want to say that the True340P is in fact an intermediate between older ATX12V 1.0 and newer ATX12V 2.0 units. On the one hand, it has the older 20-pin power connector for the mainboard and its maximum allowable current on the +12v rail isn’t as high as with version 2.0 PSUs, but on the other hand it allows for 20 amperes on the +12v, while older units allowed only about 15-18 amperes there.
As you see, the cross-load characteristic of this unit is close to perfect – it is only limited by its specification (i.e. by the maximum allowable wattage and currents) rather than by the deviations of the voltages. The voltages are regulated excellently if we don’t pay too much attention to random out-of-order dots. The +5v voltage doesn’t practically depend on the load at all, while the +12v and +3.3v voltages, although vary, don’t get anywhere close to the limits of their designated ranges. None of the PSUs we have tested so far in our labs have ever had such an excellent result.
So, the Antec TruePower True430P did its best to show you what the independent regulation of the voltages can do. This PSU will suit well for systems that mostly consume from the +5v rail (many Socket A systems belongs here) as well as for systems mainly powered by the +12v rail (all systems with Intel’s processors and with AMD’s Athlon 64). In any case the voltages won’t deviate beyond their standard ranges as often happens with classic PSUs with their group voltage regulation.
Although the True430P with its cross-load characteristic will suit for almost any modern computer, I want to remind you that it is an intermediate between the old and new PSU standard (such units are sometimes referred to as complying with the ATX12V 1.3 standard, but it doesn’t actually describe any units with a total wattage of above 300W), so if you’re shopping for a future-proof unit you may want to consider the NeoPower series, also from Antec. This series complies with the ATX12V 2.0 standard, but provides for a much higher load on the +5v than the new standard describes and this makes them suitable for any computer system, new or old.
The Blackline series PSUs from BeQuiet are advertised as noiseless in the first place. Added that other products from this company include quiet fans, passive heatsinks for graphics cards and sound-absorbing pads, Zalman seems to have acquired a direct competitor…
The unit comes in an attractively-looking black box which besides the PSU proper contains a manual, a power cord, mounting screws and multicolored sticker braces for laying the cables neatly. These stickers look unusual, but frankly speaking the ordinary nylon braces are much handier.
The power supply isn’t just colored black – it carries a special shiny black coating. The user manual says this coating is an additional protection against EMI, but I think its serves aesthetic purposes more than protective ones.
The dimensions and the quality of the heatsinks impress. The thickness of their base is more than the ordinary 5-6 millimeters, and they split into a lot of thin ribs above.
Such heatsinks should help to cool the PSU at low fan speeds – for how else could you achieve noiselessness? On the other hand, a big portion of the air stream from the fan on the top of the case comes right to the plastic protective plate of the PFC board screwed up to the heatsinks, rather than on the heatsinks proper, and this PFC board doesn’t seem to require any cooling.
So, the active PFC device is made on a separate board and fits exactly into the crammed space between the heatsinks and the walls of the PSU. The most interesting thing about it is the logotype of Topower Computer. To all appearances it is Topower who makes power supplies for BeQuiet.
The regulation of the voltages in the BeQuiet unit is realized along the classic guidelines, as you can see on the screenshot. There’s a big group regulation choke behind the vertical board with the PWM controller. To the left of it, there’s a humbler choke of the dedicated +3.3v voltage regulator. Leftmost is a vertical cylindrical choke which is just part of an ordinary LC filter that smoothes out the pulsation of the output voltage.
The efficiency of the P4-450W PSU proves to be lower than that of the Antec, just over 80 percent at the maximum, but the power factor, thanks to the active correction, exceeded 0.95 even at a power load of 125 watts. Under a higher load it closely approaches the ideal, i.e. 1.
The high-frequency pulsations of the output voltages are about 40 millivolts, which is not low, but fits into the acceptable limits.
The management of the fan speeds is very efficient in this PSU, and the fans themselves have low speeds. I can’t but mention two additional functions of the unit. First, it has fan connectors whose voltage changes depending on the temperature inside the PSU (this voltage is listed in the third column of the table above). Second, after the computer is shut down, the fans still work at their minimal speed for three minutes, powering from the standby +5v source, to cool the system down completely. Well, the first function loses its appeal somewhat now that there are modern mainboards capable of controlling the speeds of the attached fans depending on the chipset and CPU temperatures or manually. The second function is going to elongate the life of the PSU and computer components, but its practical effect is rather small.
In spite of the large ribs of the heatsinks, the low speed of the fans told anyway: the temperature of the diode assemblages was rather high (although far from critical). Yet, I couldn’t call this unit a noiseless one, since the top fan produced a soft buzz, quite audible in a quiet room.
Of course, I didn’t wait for any miracles from the P4-450W after I had tested the True430P, but its cross-load characteristics proved to be worse even than those of many other classic-design power supplies, not mentioning PSUs with independent voltages regulation.
The diagram shows that this power supplies doesn’t keep the output voltages as stable as they should be. It doesn’t tolerate a load misbalance towards the +5v as well as the +12v power rail, although many lower-wattage units are better with the latter. Well, partially this comes as the initially high +5v voltage went too quickly out of the acceptable range when the load on the +12v grew up. But on the other hand, the power supply cannot boast a high stability, either, when there’s a high load on the +5v and a low load on the +12v.
That said, the P4-450W power supply from BeQuiet is an ambiguous product. Yes, the assembly is neat, the unit itself is quiet and has good parameters, but these parameters still don’t allow it to be called a high-power and noiseless PSU. If we were to compare the specifications, the P4-450W would appear slightly better than the above-described Antec, but it is much worse in practice as the comparison of their cross-load characteristics shows. Moreover, the average retail price of the P4-450W (about 80 euros) is considerably higher than the price of the Antec True430P and is close to the price of the 470W unit from OCZ I’m going to discuss below.
OCZ Technology is widely known as a manufacturer of overclocker-friendly memory modules, but recently it has shown interest in power supplies. The two PSUs from OCZ reviewed here differ in their maximum wattage only, so I will describe the 520W model save for one case where there’s a notable difference between the two – I mean their cross-load characteristics.
The units come in nice-looking boxes inside which you can find a power cord, a pack of mounting screws, a user manual, and an adapter from the 24-pin power connector of the PSU to the 20-pin connector of older mainboards.
Like the above-described unit from BeQuiet, the OCZ model is made in a pretty, dark, polished case, and the fan is highlighted with emerald-color LEDs.
I guess you’ve noticed the three LEDs and three trimming resistors next to the power-on button. Using these resistors you can accurately set up the main output voltages of the PSU. Again, this is made possible by the dedicated regulators, since the classic power supply design only allows regulating the voltages all at once. The LEDs indicate the current value of the voltage: green means this voltage in within the norm, red means the voltage is above the norm and yellow means the voltage is below the norm.
Although this system is interesting by itself, I want to warn you against using it carelessly lest you inflict a lethal damage on your computer. By the way, you won’t improve the overclockability of your computer by raising the voltages of the PSU, since such components as the processor, memory or graphics card are supplied from their own regulators and your changing the PSU’s voltages won’t affect the voltages of these components at all.
So, you’d better not touch these controls or even remove their stickers, if you don’t feel absolutely sure about what you’re going to do. If you still think the output voltages of the PSU need correction, check them out first with a good digital voltmeter. The hardware monitoring tools of the mainboard many users are prone to rely upon are not precise measurement tools and often err. Overall, my recommendations about the manual adjustment of the voltages can be expressed like “Don’t repair it, if it already works”.
The unit is no less pleasing inside than it is on the outside. The size of the heatsinks impresses again, but their ribbing is as massive as their base, unlike with the BeQuiet model.
I can’t find any flaws here – everything is neat and nice. Well, you can demand that from a PSU of that class!
Yet another curious feature of OCZ’s power supplies – besides the manual adjustment of the voltages – is the connector with LC filters. In fact, this is a normal Molex connector with two ceramic (0.1µF each) and two electrolytic (10µF each) capacitors soldered up to it; a ferrite ring is also put on the power cable leading to the connector. You can read our report on the purpose and efficiency of this solution in our article called "Wondrous Wires" by OCZ Technology. In brief, these connectors are capable of filtering out high-frequency noise; they are overall a useful, but not very necessary addition.
When I attached the oscilloscope to the ordinary, “non-filtered” connector of the PSU, I saw a high-frequency pulsation beside the ordinary pulsation at the double frequency of the PWM controller.
The picture is noticeably smoother when the oscilloscope is attached to the filtered connector, as the high-frequency constituent of the pulsation is successfully filtered out.
Well, the swing of the pulsation is always within the norm, so there’s actually no great need for the filters. But I admit that in some cases the radio-frequency interference can become a source of noise at the output of the audio card or the TV-tuner, for example, and the filters would be of some help then.
Unlike the above-described units, the models from OCZ have neither passive nor active power factor correction, so their power factor is only about 0.7 as my measurements show. The power efficiency, on the contrary, goes as high as 88 percent – much above the results of the Antec and the BeQuiet. By the way, this is a refutation of the common opinion that active PFC increases the efficiency factor of the power supply. As a matter of fact, PFC doesn’t greatly affect the efficiency in general.
The speed of the PSU fans, contrary to my expectations, doesn’t depend heavily on the load. Yet in spite of a higher speed than with the BeQuiet P4-450W, the unit from OCZ wasn’t much noisier. There’s no total noiselessness here, too, as the light buzz of the fan can still be heard. The PSU is cooled well, even though the numbers in the table above are a bit lower than the real values – I couldn’t put the thermode in the most appropriate location when taking down the measurements.
Now let’s proceed to the single difference between the results of the two PSUs from OCZ, i.e. to their cross-load characteristics. The CLC of the OCZ-470ADJ model looks exactly like the CLC of a unit with dedicated voltage regulators should look:
Like with the Antec True430P, the CLC of this power supply is only limited by its specification, not by the deviation of the voltages out of the norm. Moreover, even the small initial deviation of the +12v voltage (the +12v diagram is all yellow because of it) can be easily corrected with the PSU’s trimming resistors, if necessary.
Alas, it’s not that well with the higher-wattage OCZ-520ADJ unit. At first it was going to draw an ideal diagram as its junior mate had done, but when the load on the +5v rail became very high and on the +12v rail very low, its regulator would lose stability, changing the output voltages in a leap. You can see that in the CLC diagram as random-shaped white areas that show where the voltages deviated beyond the norm. Anyway, this unit handles power loads of up to 200 watts inclusive on the +5v and +3.3v rails without problems.
So, OCZ Technology has debuted in the PSU market with some success. The OCZ-470ADJ is in fact an example of how a high-quality universal power supply should be made: a wide range of output currents (the unit fully complies with the ATX12V 2.0 standard, but can give out above 250 watts on its +5v and +3.3v rails); a number of output connectors, including two separate connectors with dedicated, although simple, LC filters; neat assembly; good looks; quiet operation.
The only reprimand to the other unit, the OCZ-520ADJ model, concerns its instability when the load is high on the +5v rail and low on the +12v. Still, it is rather hard to distribute the load in such a way in a real computer system, so I don’t consider this disadvantage as a serious one.
Summarizing the results of my tests I have no doubts about the best performers – they are the models from OCZ Technology. These are well-made units capable of supplying good power to any computer system existing or likely to appear in the next couple of years.
Antec’s unit isn’t inferior to OCZ’s products in the design, but it belongs to the relatively old TruePower series that doesn’t comply with the perspective ATX12V 2.0 standard. Unfortunately, we don’t yet have units from the new NeoPower series that would make a tougher competitor to OCZ’s PSUs. The reviewed Antec will suit nicely for today’s systems, but if you want your PSU to be future-proof, consider either Antec’s NeoPower line or the above-described OCZ PowerStream series.
The BeQuiet P4-450W model was a disappointment. Yes, it is a high-quality PSU with good parameters but its characteristics lack luster against the models from Antec and OCZ. The competitors have a much better stability of the output voltages, so the cross-load characteristics of the P4-450W look pale in comparison. This PSU couldn’t also meet our expectations as concerns the noise. The soft buzz of its fan makes it no less quiet than the models from OCZ, but OCZ’s PSUs have much better other parameters.
Overall, we see that the PSU manufacturers have managed to develop truly universal models that comply with all the existing standards by using the independent regulation of the output voltages. Moreover, even within the requirements of a particular standard these models surpass their competitors in the stability of the voltages, so this is not a marketing trick or beautiful words on the label, but a very useful innovation. The apprehensions that the additional regulators would reduce the efficiency of the PSU didn’t prove true as the magnetic amplifiers themselves have a good efficiency factor. In my today’s tests the classic-scheme PSU had the worst efficiency, so you should agree that the efficiency depends more on the design features of a unit rather than on the presence of an additional output regulator.
Of course, such power supplies are rather expensive today, but I wouldn’t call their price sky-high. Units of the same wattage, but made by the classic scheme, have cost the same money, if not even more, quite recently.