I've seen reviews where the psu is using a Protechnic fan instead of a Yate Loon.

It probably depends on where you buy the PSU and whether it's a retail/bulk pack ?

That's news to me. Of the dozen or so FSP & Sparkle PSUs I've examined, none has ever sported a Protechnic fan.

I am sure it doesn't depend on where you buy the PSU or whether it's retail/bulk. It costs $$ to make changes in production lines and maintain inventory of slight model variations. Nobody like this -- not the mfg, the distr, or the retailers. Not to mention the customers. And a different fan is a pretty major change.

the Bluestorm AX500 reviewed here uses a Protechnic fan.

Moreover, I've seen a review of a bulk(?) FSPxxx-60THN (Bluestorm equivalent without the bling factor) that shows a Yate Loon fan, even though my own retail Bluestorm comes with a Protechnic.

Not too long ago, a store here sold the FSP300-60BT (old model, not even ATX v1.3). Some samples had Yate Loon, the others Protechnic. Wonder what's the rationale behind using different fans.

EDIT:

fan that came with my Bluestorm
http://www.pix8.net/pro/pic/6176dTHj1/430283.jpg

OMG, soorry, you're totally right! That's it, no more unplanned postings here from me.

No idea why they would do this -- unless they were making a transition from one supplier to another, using up existing invetory or parts...

I have read several reviews of this psu (french, chinese, english) and it's the first time I see a ball bearing yate loon inside.
If fortron has changed his supplier, it's probably a smart move since the previous fan did not seem good.

In france, this fortron can be found at a very low price considering its quality, less than 60$.
It's been called the "poor man's seasonic"

Some explanation for Green PS layout. It's very similar to passive FSP Zen: http://www.xbitlabs.com/articles/other/ ... sus_2.html. There are lots of diodes on the high and low voltage side, to spread the heat better. The 3,3 V rail is so stable because it is regulated separately (Zen has saparete 5V and 12V regulator too). What were the specifications fo the main capacitator?

Looks awesome, I love the simplicity. Hopefully some US vendors will pick these up, would be really good PSUs, especially with bigger heatsinks and/or a fan swap.

Page 4, bottom of first table:


It's a very unusual PSU indeed. Do you think the noise at high loads is problem with the fan controller or with the cooling? It would be interesting to see just how hot those heatsinks actually get.

The table footnote has been fixed -- it comes from the people who test PSUs for 80 Plus. It's the same test protocol used by Intel.

As to the 2nd question, the issue is mostly the fan controller. It ramps the fan up pretty fast and has too little hysterisis. What whould happen to cooling if the fan controller was slower to respond & didn't spin the fan as fast -- well, for sure, the internal temp would go up, but who knows by how much.

Yeah, at last my PSU got reviewed! Great review, though it's a shame the efficiency isn't quite up there with the Seasonics. The fan that came with mine about 4-5mths ago (from France, delivered to the UK) was definately not a Yate Loon. Can't remember what it was (threw it away when I moved to NZ), but it clicked at low speeds. There's a thread in the PSU forum where a couple of other people found the same, so perhaps they listened?

Anyway, swapped the stock fan out with a YL D12SL which made a huge difference - no wind turbulence noise, no discernable motor noise and the air coming out the back is nice and cool (low power SktA system though). I tried a D12SM, but it made noticeable wind noise and seemed to be pushing more air than necccessary.

Been very happy with my "nerdy" PSU, even though I can hardly claim to be pushing it hard with my system

I dunno, for some reason I kind of think it's fan-related. The 430w Yate Loon S12 reviewed had a very similar voltage curve <250w, and was only a few tenths of a volt off at most of the load levels except for 150w. However, at almost identical fan voltages, the Fortron is as much as 8dbA louder than the S12.

I know there are many factors (ambient noise, sample variations, different internal airflow characteristics of the S12 and FSP), but it makes you wonder how this power supply would do with a medium or low-speed Yate Loon sleeve.

EDIT: Very interesting Matt, thanks for the info.

see ^ no SPL values though, which is what I guess you were aiming at?

EDIT: no worries

In future reviews is there any chance of getting noise and efficiency levels at different ambient temperatures, say 20c, 25c, 30c, 35c so we can see how Power Supply’s act outside of an air conditioned office?

This is how the tester currently functions (kind of). Heat from the load resistors are exhausted into the tester. This mimics the thermal conditions in a real-world ATX tower, as intake temperature increases with load.

IMHO, it would be silly to test a power supply at a constant temperature like you suggest, because this is not how the PSU would be loaded in an actual computer (although there are a few exceptions, such as the P180 w/a separate chamber for the PSU).

No, there's no way we can afford that much time. A PSU test session already costs a whole day. There is also no way we can control the ambient temp of the room accurately, anyway; this is not an air conditioned room.

We've tried to stay as close to 20~22C as possible for all the PSU reviews, avoiding hotter days for testing and turning on the forced air gas central heating prior to testing on colder days.

While the ambient room temp during testing is almost always 20~24C, the temp faced by the PSU is far higher. The INTAKE TEMP shown in the charts is reflective of the operating ambient temp of the PSU.

If the room ambient was always the same, the amount of heat produced in in the PSU load test box would be the same at any given output load for any PSU -- because the power/heat in the loaded resistors would be the same.

BUT, airflow (from the PSU fan) affects the intake temps, so the relationship between load and PSU intake temp in our test rig is not exactly linear, but related. Ditto the relationship between ambient temp, load and intake temp.

For example...

1) for the Green PS, the room temp was 20°C, and 30°C intake temp was reached at 150W load.

2) for the ANTEC SMARTPOWER 2.0 SP-450 review done last July, the room temp was 25°C and 31°C intake temp was reached at just 90W load.

For your purposes, seeking out PSUs whose fans ramp up least at the highest intake temps would be the thing to do. This data already exists, going back to nearly 2 years of PSU tests.

In a nutshell, the PSUs we rate highly for noise also ramp up least in high temps. Not surprising, is it?

I understand that the power supply create its own heat and gets the excess heat from the PC, what I am asking is to raise the room ambient. So at 20c room ambient the power supply temp is say 40c, at 30c room ambient the power supply temp is say 55c. And show what impact that has on the noise generated (fan speed) and efficiency (which should drop).

A room ambient of 30c-35c is a much more real life measurement to me than the winter temp of 20c.

MikeC,

Thanks for the response, and I understand the way testing is done and the time and effort it requires. You have the best powersupply reviews on the web by the way.

I guess that I always see all the reviews on the web done at 20-25c ambients which is nice, but doesnt reflect real life (in summer anyway). Keeping a PC quiet in winter is easy. But the 20c jump in ambients during summer make its a whole different game. e.g. my A64 3500+ is undervolted to [email protected] which is fine in winter, but as spring warms up I've had to up the volts to 0.95v to keep the machine stable.

Time to move, methinks.

Seriously, the highest temp ever recorded in the downstairs lab, not airconditioned, with a couple windows open, was around 27~28C.

as the data's there, why not add the difference in your ambient to Mike's ambient to the intake temp of the PSU at a given power level, then read off the dBA for the higher intake temp? eg. Mike's ambient it 25C, your's is 35C so the difference is 10C. Add that to the starting intake temp of 33C to get 43C - the PSU should then (more or less) make the amount of noise at your ambient temp as it does when the intake temp in Mike's lab is 43C. It's not likely to be particularly precise, but it'll give you a rough idea.

Ah, I see what you mean now.

I have moved recently, my summer used to be ~40c in the shade, now its only going to be ~35c. I already miss the warmer temperatures.

I believe what he wants to know is how efficiency changes with a higher ambient. Because this isn't determined solely by the temperature, he can't just extrapolate from the data, he wants to know how efficiency at a given load changes with ambient.

well, I never said it was going to be precise, did I?

It won't be able to indicate efficiency (no idea how that would vary with intake temp), but it should be able to give a rough idea of noise. Shame no-one's got the time to do the science, so people can extrapolate the published data to their particular circumstances. Then again, that would assume all PSUs vary in efficiency with temperature in exactly the same way. As that's unlikely, any estimate will be inaccurate to some degree, which leads us back to the original problem of time vs. accuracy

Looks like NewEgg has these in stock, $55 for the 300w and $75 for the 400w.

And an 8cm fan unit that appears very similar.

Like the greenpower, it also has some distinct intake holes.

I found another cousin. A micro-ATX PSU in the "GL" variety:
FSP300-60GLS
It has the same hole patern as its ATX sized cousin (see the PDF drawing and the picture).
Manufacturer spec'ed efficiency: 75% / 115V; 78% / 230V.

I can get it for only €45,- w/o S&H. Very tempting to try this out for a fileserver in a custom case am contemplating to build.

It's hard being an ecologist and liking computers at the same time, but the concept for this PSU rocks: cheap, simple and ecologically friendly.
Apart from that the fact that it doesn't weigh much can be great for LAN partyers, after the case the PSU is the heaviest component. The thinner cables might also make it a bit lighter. If you painted this thing black, put a flashy LED fan in it, made a retail package that doesn't mention the ecological stuff (apart from the fact that the paint would contradict this, many people find ecological stuff uncool), with slogans that make it seem like "the beast" and "gamer special", etc..., oh, and double the price, FSP could make a killing

I'm sure I would melt in those temps.





Pete

They actually did this (with blue paint): Fortron Epsilon 700 (FX700-GLN).

An SPCR forum thread about it.

I think those four different models (ATX 120mm fan, ATX 80mm fan, micro-ATX and a "bling" PSU) shows that Fortron is indeed prepared for the upcomming eco-laws in the EU.