Topower 420, 470 and 520 watt ATX power supplies

Review date: 7 August 2001.
Last modified 03-Dec-2011.


The Power Supply Unit (PSU) is far from the most expensive component in most modern PCs.

Most no-brand clone PCs use the PSU that came with their case. Stock PSUs in half-decent mini-tower cases are likely to have about a 250 watt rating - not many small cases come with a 300 watt PSU as standard.

By itself, one of these yum cha PSUs will commonly have a retail price of less than $US15.

If you're now thinking that it doesn't seem too likely that a $US13 PSU will be the most carefully designed and manufactured electronic device in human history, you're on the right track.

If your PSU's no good, it can cause amazingly diverse problems. Underpowered or otherwise defective PSUs don't necessarily have the decency to just go pop and stop working, you see. They may do that - or they may decide to go out in company, and pass mains voltage through to the motherboard. But they're more likely to just suffer voltage sag.

Voltage sag is your own little private brownout inside your computer case.

The three primary output "rails" from a modern PSU - the only outputs that have to supply a lot of current - are the 12V, 5V and 3.3V lines. Overloading one of them can make all of them sag, depending on the PSU design; lousy PSUs may also deliver too high a voltage under some circumstances, or have a lot of "ripple" - a regularly varying output voltage.

Different computer components have different tolerances for low input voltage and other supply oddities, and will foul up in different ways when underpowered. Most of these ways are unlikely to give you any obvious indication that it's the PSU at fault. If you don't know the PC service-person's mantra - "if the problem's bizarre, try another PSU" - you can waste days swapping boards and changing drivers and fooling with your RAM configuration, and still have a computer that flakes out whenever, for instance, all of the drives are accessed at once.

So that's one reason to buy a more expensive, beefier PSU.

Another reason has only arrived quite recently.


Topower TOP-420P4

Topower's high power PSUs have hefty 420, 470 and 520 watt ratings, the details of which I'll go into in a moment. Let's ogle them a bit first.

These PSUs are not just anonymous grey boxes. They're pretty boys.

They're a grey-gold colour, with a shiny lacquer coat...

Zooty gold grille

...and fans, covered by gold-tinted finger guards.

The original TOP-420P4 PSU I photographed for this review had two translucent fans installed; current models are coming with plain black fans. This may change again, though.

The fans are thermostatically controlled; they only spin at full speed when the PSU's hot. Power the thing up on the bench and it makes practically no noise at all.

Inside the PSU

Take the lid off the TOP-420P4 and there's a reassuringly dense component forest, with big fat heat sinks on the power semiconductors. Cheapo PSUs are light, because they contain a lot of air. The TOP-420P4 is not.

The Topower's nine-bladed see-through fans aren't massively powerful, but since there's two of them, they don't need to be. And they certainly look groovy, with their brushless motor coils visible through the hub.

Component detail

Here. Have a pointless glamour shot full of gold-anodised aluminium.

The Topower's appearance may seem a tad familiar to high power PSU aficionados. That's because it bears a more than passing resemblance to Enermax's products. I review their 550 watt PSU here.

See-through fans and gold trim are very Enermax. The Topower PSU also has a black webbing tube around its main ATX power connector cable bundle; that's an Enermax feature, too.

One Enermax feature the Topower PSU lacks, though, is the big price tag. Aus PC Market here in Australia stocked both products (when this review was young; it isn't any more, and the PSU range has moved on; the pricing info in this review is now historical) and you're looking at $AU231, delivered, for a 430 watt Enermax. The 420 watt Topower is, as of the my most recent update of this review, selling for $AU176 delivered. The 470 and 520 watt Topowers are $AU192.50 and $AU214.50 delivered.

If you're not building a show-pony computer, you can get the guts of the 420 watt Topower in a plain grey metal single-fan box for a mere $AU148.50, delivered, as I write this. Less pose value, but a lot of amps for your dollars. This version's only got seven four-pin power connectors, though, as opposed to the ten in the fancier version (see below).

Moving the juice

Connector spaghetti

Welcome to ATX Spaghetti Junction.

The three gold-box twin-fan Topower PSUs all have, on top of the normal ATX power plug, ten of the larger "Molex" drive power connectors, two floppy drive power connectors, and ATX12V and AUX connectors, too, for motherboards that need them. That's more drive power connectors than the 350 and 430W Enermax PSUs, and the same number as the 550W Enermax, which has 11 Molex connectors, but only one floppy connector.

So there are enough plugs to cover pretty much any PC configuration, including stacked tower cases. Is there enough power, though?

Well, almost certainly, yes.

But to tell exactly what a PSU can really deliver, it helps to spend a few minutes in Unethical PSU Marketing 101.

Here's how to make overly optimistic power supply specifications. It's really simple.

First, power the thing up. You can make an ATX power supply that isn't connected to a motherboard turn on by grounding pin number 14 on the big motherboard power connector. It's easy to spot that pin, because it's the only one with a green wire going to it.

PSU plus paper clip

Use any handy bit of wire - like the paper clip in this picture - to connect pin 14 to any ground contact. The ground contacts are the ones with the black wires going to them. Presto, the PSU will turn on.

Now, break out your brick-sized power resistors and load the heck out of one of the output rails - the +5V rail, for instance. Measure the current as you increase the load, until the voltage sags unacceptably far below the rated voltage.

How do you tell what an unacceptable voltage sag is? Well, you could choose a nice conservative small permitted sag - say, 0.1 volts - so that your results are genuinely useful to your customers. Or you could just ignore the voltage and say that when a fuse (or some other component...) blows, that must have been the limit, right there.

OK. Now you've made a big fat amperage number for the +5V rail. If you blew up the PSU in the process, get another one, and repeat the process for +12V and +3.3V, and for the low current rails as well.

On no account, though, should you test more than one rail at a time. This is the key to the whole scam.

A big beefy PSU may be able to deliver 50 amps (say) on the 5V rail when nothing else is under load, and 25 amps (say) on the 12V rail when it's similarly all alone. But the 12V and 5V rails together may only be able to deliver, say, 350 watts between them, when they're both under load. Watts equals amps times volts.

In a real PC, all of the power rails will always be under load together.

But you're not testing what the PSU can really do - you're making pretty numbers for the sales brochure!

So test all of your rails alone, get an amperage figure for all of them, multiply that figure by the voltage of the rail it came from (the nominal voltage, not whatever the voltage had sagged to as the PSU pumped electrons through the dessert spoon you'd soldered to the circuit board), then take all of the resulting wattage figures and add 'em up. That's a wrap, folks. Ship it!

Getting real

Pretty much all PSUs, whether they're realistically rated or not, will have maximum-current quotes on the specification sheet that tell you what each rail can do when it's the only one under load. Quality PSUs, though, will also tell you what the rails can do in combination. And, I'm happy to say, that's the sort of specifications you get with these Topower PSUs.

The 420 watt TOP-420P4 has an 18 amp rating for its 12 volt rail, giving that a 216 watt rating by itself, and the maximum current ratings for the 3.3 and 5V rails are 26 and 42 amps, respectively. That's 85.8 and 210 watts, respectively, but right there on the PSU's spec sticker it tells you that these two rails put together are only rated for 220 watts. You can draw full current from either of them, but not both at once.

There's a similar maximum aggregate rating stated for all three high current rails together. The total you can draw from all three at once is 400 watts. If you want all 216 watts from the 12V rail, you'll need to be drawing not more than 184W from the 3.3 and 5V ones.

Where's the extra 20 watts in this PSU's rating come from, then? The low current -12V, -5V and 5VSB (standby) rails, which put together are rated to deliver 22 watts. Once again, their separate ratings add up to more than that, but if they're all being used, 22 watts is their aggregate rating.

The 470 and 520 watt PSUs have the same sort of power distribution. Rather than weigh this page down with piles of numbers, I'll just point you to Topower's spec sheets for the PSUs - the 470P4 page is here, and the 520P4 page is here. The 420P4 page is here.

OK. Now we've got another problem. How the heck do you use all of that power?

Warming the wires

These PSUs are unremarkable in that they use 16AWG (American Wire Gauge, a US standard cable size measure) wire for the big ATX connector, and 18AWG for everything else. The higher the gauge number, the thinner the wire.

16AWG copper wire has a resistance of only about .005 ohms per foot. 18AWG's about .006. But the actual end-to-end resistance of the connecting cables in real computers is considerably higher than you'd expect from these figures, because the connectors have resistance as well. And cable resistance can become a factor in high current, low voltage applications.

The maximum current ratings for 16 and 18AWG wire are only 11 and nine amps, respectively.

Now, if you pass ten amps through an ordinary 18AWG wire, it doesn't instantly explode in a shower of red hot copper droplets. But it does get warmer. In an average sort of situation - where it's neither sitting in an ice bath nor routed through an insulative vacuum bottle - it's likely to get warm enough that its PVC insulation will soften a bit. Which can be a very bad thing if the wire's up against a sharp metal edge, as it's likely to be in a computer case.

Another thing that happens when wire gets warm is that its resistance increases. All materials undergo resistance changes with temperature, and most materials have a positive temperature coefficient of resistance (TCR) - their resistance rises as they get warmer. High power wirewound resistors are made with low TCR wire, which can keep the change down to a fraction of one per cent over the resistor's normal operating temperature range. Normal wire, on the other hand, is made of copper, which does not have a particularly low TCR.

To prove this to myself, because I never miss an opportunity to try to set fire to something, I grabbed a one metre alligator clip lead made from wire of about 18AWG thickness. I hooked it up to my high current bench power supply through a low-value resistor that stopped my bench supply from making rude stop-shorting-me-out noises, and measured the voltage drop across the one metre lead with more than ten amps flowing.

The voltage drop started out at about 0.82 volts - which indicated about 0.06 ohms of resistance - but as the wire warmed up, the drop peaked at about 0.88V - 0.07 ohms, more than 15% higher.

Then the voltage started falling off again, because the cooling water in the jam jar containing my in-line resistor started boiling.

The resistor was a couple of centimetres of HB pencil lead, the carbon in which actually has a negative temperature coefficient of resistance - the hotter it gets, the lower its resistance. But the steel alligator clips holding it more than made up for any resistance drop in the resistor itself, when steam bubbles started allowing the pencil lead to glow.

I really should have an assistant called Igor.

Anyhow, the resistance gain demonstrated by my unremarkable length of unremarkable wire is realistic enough, and something that people planning to make use of high power computer PSUs should think about.

If you want 15 amps of current to flow in a 12 volt circuit, the total circuit resistance must be 0.8 ohms. That's Ohm's Law; voltage equals current times resistance.

In this situation, around 0.01 ohms of extra resistance from a warm wire won't make much difference. Even if you've got crummy connectors and more wire in the circuit, you're still not likely to see more than 0.1 ohms from a metre of wire, and that's all assuming that the entire 15 amp current is going through just one 18AWG-wired circuit, which in a PC it won't be.

When you're dealing with a computer PSU, there are lots of connections for everything, and if you avoid using just one of the power leads coming out of the PSU for all of your high powered stuff (possibly with a Christmas tree of Y-adaptors...), then you'll be able to split up the load over enough wire that heating effects should be insignificant.

But what about, say, 20 amps at 3.3 volts?

Now you're talking only 0.165 ohms of total circuit resistance. Any more than that, and the stuff at the end of the wires just won't see 3.3 volts. There's no way for a motherboard that notices its 3.3 volt line is now only 3.1V to ask the PSU to goose up the voltage a bit - it's got to take what it's given. People running overclocked Athlons may well have motherboard power draws that get up around this level.

Again, fortunately, there are multiple wire pairs involved. The standard ATX motherboard connector has three wires delivering power from the 3.3V rail. But if you're sharing 20 amps between them, that's still nearly seven amps apiece. If your motherboard uses the AUX connector - which few consumer boards do - there's another two 3.3s there.

The worst case scenario comes up if you're trying to use most of the 5V capacity of a high-spec PSU like these Topowers. 40 amps at 5 volts isn't going to happen with more than 0.125 ohms of total circuit resistance. So you're going to want very low cable resistance.

Once again, the current will be split among various wire pairs (one 5V line per standard power cable, four 5V lines on the big ATX connector, two 5Vs on the little-used AUX cable), but if it's not shared evenly, it could be quite easy to end up with enough resistance in the supply wires that one or more of your high current five volt whatever-they-ares doesn't see enough volts to work properly.

Getting real, again

Is all of this likely to be a problem in the real world?


Because, in the real world, people are not likely to buy shiny gold-grilled Ultimate Pose Value PSUs with monster power ratings because that's what they actually need.

Frankly, most of the people who buy these things would, in fact, be just fine with any quality 300W unit. There aren't many PCs around that need more than a 300 watt AOpen PSU, for instance, can deliver. That PSU's yours for $AU115.50 delivered.

The only way most people are going to draw 42 amps at 5 volts is if they slide the side of their PC's case shut without noticing that there's a power lead sticking out, and short the 5V rail to ground. Which is likely to result in a current draw of rather a lot more than 42 amps, for the brief period before the fuse blows. In that case, a higher-rated PSU will simply be able to generate a larger cloud of insulation smoke in that tenth of a second.

Even if you install a cigarette lighter in your case, you may not get very close to the 12V delivery capabilities of a high-rated supply. Automotive cigarette lighters don't seem likely to want more than ten amps; the one from my car draws less than 8A from 12V.

Apart from pure cosmetic value and the appeal of the thermostatically controlled fans, though, you can make a case for buying a high-rated supply. Since it's not likely to be working too hard in pretty much any PC, you can be reasonably certain that if you see a problem, the PSU isn't the cause. If things aren't working properly, you don't have to monitor voltages like a hawk to see if they plunge at any particular moment (if the computer hangs every time the voltage sags, your automatic voltage monitoring software will of course hang as well, and not tell you about the problem...).

Similarly, buying a PSU with a higher rating than you need means that you never have to wonder whether your new quad-fan CPU cooler might perhaps be overstraining something. You can add neon tubes and water cooling pumps and Peltier coolers with impunity. Et cetera.

So if you're looking for a pretty PSU with serious, yet plausible, current ratings, the Topower models look like an excellent alternative to the more expensive Enermaxes.

They're well priced, they look great, and despite the basic problems involved with moving lots of amps down relatively skinny wires, there genuinely is some benefit to having a PSU with great big muscles. And there's even the cheaper grey-box version, for people who don't advertise.

If you're building a humdrum business box, there's no reason to buy these things. Pay less and get a decent 300W PSU. But if your computer needs more grunt - or just needs more gold highlights - then the gold-box Topowers are a good choice.


Topower's page for the ATX12V-420P4

Topower's page for the ATX12V-470P4

Topower's page for the ATX12V-520P4

Buy one!
Aus PC Market don't sell this product any more; you can check out their current PSU range here, and you can also still try a price search at DealTime!

Give Dan some money!
(and no-one gets hurt)