I'm also curious to learn the internal differences between the 400w & 460w. The following makes clear what my guess would be if I were to make one.

I'm of the opinion that there isn't a sufficient market for both the 400w and 460w, particularly given the current price difference (around $20-30 at the moment). The 60w differential is unlikely to support the price difference because anyone pushing 400w, let alone filling that small gap between 400 & 460w, is likely to error on the side of caution and purchase a fan-based PSU with a higher output rating. Besides, the overload tests [url]=http://www.hardwaresecrets.com/article/Seasonic-X-400-Fanless-Power-Supply-Review/1073/8]conducted by Hardware Secrets[/url] indicates the 400w will perfectly serve the market of people needing 460w (of course, that test isn't definitive and doesn't take into potential long term consequences).

Perhaps recognizing this, Seasonic neutered the 400w package by including only a single PCI-E cable (being the 2nd of only 2 differences between the products). It's hard to imagine a common setup which draws close to 250, let alone 400w, without a video card that uses 2 PCI-E cables so the exclusion of a 2nd PCI-E cable is clearly marketing related. This is further confirmed by the present marketing of the 460w: "Coming with two PCI-E connectors, the SeaSonic X series SS-460FL Active PFC F3 allows you to build up a SLI or CrossFire gaming system." Most SLI or Crossfire setups appropriate for a 460w would also work within the confines of 400w. Again, that 60w gap simply does not open up a whole new set of configurations. There is no justification for including 2 cables with the 460w and only 1 with the 400 except for marketing purposes. The Seasonic reps failure to affirmatively address the issue further suggests this (2nd post from the top). While it may be true that most 400w units don't have dual PCI-E plugs, most 400w units are not priced at these levels and a 2nd $.50 PCI-E cable would not have altered the economics of this item.

Ultimately, marketing alone will be insufficient to maintain both the 400w & 460w given the size of the market. Most users purchasing this unit will simply opt for the cheaper unit and use a 2xmolex to PCI-E adapter if necessary. They'll be happy with a great PSU but leave some goodwill behind for feeling nickel and dimed. Consequently, the price differential will shrink between the lines removing any incentive of Seasonic & retailers to offer both units. If Seasonic develops a 500w fanless unit, they'll drop the 460w unit. If not, we'll have the 460w unit. Perhaps OEM's will find sufficient value in the difference in order to sustain both units but it's unlikely.

That being said, thanks Mike, Newegg &/or Seasonic for the promo discount. A discount is a discount for those who must buy the 460w right this moment (there's no retail price competition at this time b/c so few have the units....we'll inevitably see a drop in prices of both units as well as a decreasing price differential b/w the 460 and 400). And, thank you Seasonic for what appears to be a great PSU and filling the void of a relatively niche market. Alas, can't you just sell us a PCI-E cable for $5?

Search terms: SS-400fl x-400 x-460 x400 x460 ss-460fl PCIE video card

I've got to assume that is an availability issue. Over time they'll become more common and other retailers will get them in stock.

Why bother with a fanless power supply if you are going to use a 200W GPU? Between the GPU fan and the case airflow - you are not going to have a silent machine. Might as well use a regular quiet power supply.

But - if you insist, there is a 460W model to go with the 400W model.

Still waiting for local supplier NCIX to get the 400s and 460s in stock. Even with the promo I think I'm better off waiting until the locals have them. Save the shipping and if they are available in more places it should help drive down the prices.

scan.co.uk finally have this in stock £115 probably still buy it though, quest for silence

Uh?

What are the chances
I meant.... due to MY quest for silence I will probably still buy this.

Darn, nice review, brill power supply as hoped but 4x price that I paid for SS S12 II 330W (!)

Every focus is diverging to lower cost, lower power/small form factor/mobile, as quick as possible, so why don't SS make a 150-250W as such for that market? It's been a while since I last checked their OEM offerings. Sure, most of this market and its $$$ is in the OEM space but if its out there, I'd probably be able to snag it from work orders. Intel, AMD, nV and the rest of ARM licensees are really fighting hard for the 10-50W system TDP, industrial, embedded, all-in-one, and smaller form factor space. Client side, AMD Zacate is vying to push nettop/thin client Atom/Pentium/Celeron/CULV out of this space so now we'll for once have some competition. Many of those industrial and embedded designs these chips target are mini-ITX requiring efficient and small form factor low power band power supplies. Interesting times.

KTE, if you go and read all the PSU reviews for the past 5 years on SPCR you will see a clear trend towards the quietest PSUs with fans being above 400W with the only one below 500W being the Nexus 430W value model (that still isn't cheap compared to a S12).

As to cost the fanless PSUs have components similar or the same as units that are rated hundreds of watts higher when a fan is involved.

So yes you can get a cheap 330W unit but it won't be as quiet as a more expensive unit.

And yes they could sell you a 300W fanless instead of the 400W fanless but it'd cost them the same to make it so they have no incentive to lower the price. In fact if you look on newegg right now the 430W and 520W S12 have the same price. There is absolutely no discount at all for the lower wattage version.

I hope you can see how these factors push us (consumers and suppliers) toward a middle ground of 400 to 700 watt units as the best value.

Yep, your points are true realities. I remember arguing the required power trends and power band offerings on Jons forum back in 2007 along with Mike

But there is one critical aspect seesawing all this: certain parts aren't manufactured due to high costs only when the demand is small - the niche market is exactly that, a niche market with low volume. MFGs don't waste time on low demand markets due to the associated risks and returns. The second industry focus as a whole shifts to a trend within these very niche markets, it bcomes a cash cow - competition and resources put to development sky rocket, volume rises proportionally, and production costs fall greatly to make such otherwise niche products viable for mass manufacturing and mainstream retail. In the quest for lower costs, better returns and competitive leadership, lead players improve and streamline the production, technology and workings even further. Over the developement lifecycle, the initial high costs are amortized and they end up with high revenues. You can see this working in every technological field, like very expensive process technology development or mobile AMOLEDs. ARM licensees, Atom, Bobcat, Tegra, all followed the same demand shifts. The products they create for us consumers are mainly dictated by demand after initial market tests.

As a consumer who left the HPC/MPU markets to the LP markets in mid 2008, I seek that day where low power band fanless PSUs will be a "mainstream thing". What's difficult or hard to see happening now is only limited by the constraints of the modern day market - difficult because the industry hasn't collaborated to tackle it together. When the demand and returns are there, they will.

This really seems like the kind of PSU I've been looking for.

Basically, my current build is based on a Silverstone SG03B case, and I'm trying to get it as silent as I can (long-winded retrospective build log and ramblings about the PSU etc can be found in HardOCP's SFF forum in the odd chance that anyone's interested).

Right now the PSU I have, Nexus Value 430, is the most noisy component in my system. This is largely due to the fact that the SG03B with low fan speeds isn't the coolest enclosure there is, and thus the Nexus's fan ramps up even when idle, let alone under load - case air temp under moderate load is slightly over 40 degrees Celcius.

As I have my CPU cooler pushing air out through the PSU, this also makes the PSU warmer than it would be, something that's not so good for a cheap PSU like the Nexus, and I'm not fully comfortable using it this way. Combined with the noise, I've been drooling at having an X-460 instead (since the X-400 lacks the second PCI-e cable my GTX460 needs). Even if the CPU cooler would be pushing 50+ degrees warmth of air into it, the X-460 would probably just push through without any problems based on the stress tests I've seen.

Problem is, if this change would end up with too high temps - though it doesn't seem likely - I might have to try a different case. Perhaps the NSK3480 with its thermally separate PSU area. Would involve paint work to get rid of the silver front plate, ack. But that aside, based on the torture test, I'm wondering if the X-460 would be ok in the PSU section with no airflow? Warm air could freely exhaust through the top vents, though. Thing is that the X-560 would be cheaper, and probably better suited for the NSK3480, but in the SG03 it'd ramp up just like Nexus does... so do you reckon the X-460 would work fine in the NSK3480 too, or am I fracked in the semi-unlikely event that I get the X-460 for the SG03 and then go for NSK3480 instead?

Based on all the praise I'm inclined to think it'd work, but I'm so often let down in my expectations

Is anyone besides Newegg going to carry the X-400 and X-460? Their shipping prices and tax (CA) just kill it for me.

Have your tried flipping the PSU around so its fan is pulling in air from the outside of the case? Only 3 of the mounting screws for the PSU might engage, but it could be worth a try.

That should work fine with the SG03 as it is intended for both ways round. I had been thinking about this possibility of this too.

I has also thought of the NSK3480 as the PSU chamber has a top vent which would fit the vented side of the X400 fine and an optical drive and a low power hard disk should be able to sit in that chamber with it. Then you only need a single 120mm fan behind the CPU with a large passive heatsink to be enough to cool most hardware configurations.

I cited your post to Seasonic... and here's their reply:

I'm not sure where you are located but NCIX has the 400 and 460 listed. They show stock in the main Richmond warehouse. Newegg often does deals where they pay the shipping so that might be worth keeping an eye on.

Nice PSU, if a bit expensive.

But I'm disappointed that the review wasn't done in a real case with top and side covered. The top open scenario is not realistic at all. You could say that the 400W constant power compensates for that, but I'm guessing that in a closed case at something like 300W the PSU could get hotter than with the open setup at 400W. And 300W is easily achievable with some PCs. Think of a media creating setup, where you'd leave the video/3d graphics to render for several hours on both an overclocked multicore CPU and a powerful graphics card (which also contribute to the heat inside the case). All this on a hot summer day (30+ C) and you have a possibly more intense heat torture that's also somewhat realistic.

But they have a hotbox underneath pumping hot air into the PSU as well. Either way, the internal parts are way overspec'ed (it's really more like a 700W unit), and they even had to use a hairdryer to trip the OCP; they did it several times, and it still worked fine.

This was worrying to me, too, but I found other reviews with different "cooking" setups and the same awesome results. Take a look here (testing setup described here) and here.

Yeah, I've been reading those. The first one looks good. The second one less so (40 C with positive pressure pushing air through the PSU? that's actually probably better than an average low noise PC case with zero pressure and a few hot components inside).
But indeed the PSU seems to handle all those situations just fine. The good thing is that the PSU itself doesn't seem to produce much heat on its own, so if you don't heat it up with other components too much it should work even with no airflow in a closed case.

Hardly a surprise, is it? This is purely a function of efficiency. At 90% efficiency, it turns 10% of the incoming AC power into heat. So if your system demands 100W, the heat in the PSU is 11W; at 200W, it produces 22W heat.

With a traditional PSU setup (X-400 vents facing down due to no vents in the top of the case), and a case with a front and a back fan, I wonder if there then would be any positive or negative pressure to give airflow through the PSU. So my question is, what would be the best orientation for the tower CPU heatsync fan? To have it blow back towards the case exhaust fan, or to have it blow up towards the PSU hoping to push some air through the PSU?

You just have to experiment, probably depends on the details of your setup & components.

All the Seasonic X-series use the PSU casing as an extended heatsink. So like the fanned units the X-400 has some output components on the reverse side of the circuit board which use the baseplate for cooling. If you used the X-400 as a top mounted PSU this baseplate is directly above the exhaust fan. So having the CPU fan blow towards the exhaust fan, plus the effect of the exhast fan itself should give some air circulation over the baseplate which ought to be enough to cool it. In this configuration there is evidence from users that the X-400 will intake air through its rear grill which having cooled the PSU will then pass into the case. I think it is unlikely that this airflow will impact on other components as the combined CPU/exhaust fan airflow should be enough to extract it.

On the face of it blowing air from the CPU fan directly towards the baseplate of the X-400 rather than towards the exhaust fan is more likely to make the baseplate hotter rather than cooler. Assuming that it did reverse the airflow though the PSU it would be replacing cooler air from the outside with much warmer air, again potentially worsening rather than improving the cooling of the PSU.

Not really. None of the internal heatsinks are actually coupled to the casing. This is not to say there is no conduction of heat to the casing, but it is not direct, it has to pass through the air first. Different from almost all other fanless PSUs (and probably the only ones that are efficient enough to get away with it.) This is why the X fanless models are so much lighter than all other fanless ATX PSUs.

See this page of the JonnyGuru review of the X-400
http://www.jonnyguru.com/modules.php?name=NDReviews&op=Story4&reid=200

Quote "....The underside of the main PCB. See those four MOSFETs in the middle? Those are for the 12V output and are heatsinked both to the case and to heatsinks on the other side of the PCB....."

The Seasonic X-series derivative, the Corsair AX series, use the same arrangement. There is a very clear image of it on this page of another of JGs reviews, this time of the Corsair AX850
http://www.jonnyguru.com/modules.php?name=NDReviews&op=Story4&reid=197

And again to quote JG "....Typically, one would use diodes to rectify voltages, and these diodes would be mounted to the secondary heatsink. But this unit uses a set of four Infineon 031N06L MOSFET's. Add to this that they are mounted to the bottom of the PCB! Add to this that the heat from the MOSFET's package is dissipated by the power supply housing!...."

lodestar --

Fair enough, I stand corrected -- I never did pull the entire PCB out. However, keep in mind that what's been done is not for the primary. I have played with thermal conduction pads and while they certainly help conduct heat, it is nothing like clamping/screwing down a transistor to a heatsink. I would guess it only works here because the amount of power those MOSFETs handle is small.

Trying to push air in an enclosed box, like a computer case, over anything over very small distances, or through a duct, is like trying to push a rope. It's not like water coming out a hose nozzle. Air flow will be from high pressure areas to low pressure areas.

Now I have one configuration built, so I have the question: How do I know which way is working better? How can I know the temperature the PSU is running at without any special measuring tools?

I work at Intel (high relevance? no...little relevance? sure why not ), and I can honestly say after reading this review that this here product is simply an engineering marvel.

Bravo.