Posted: Tue Mar 29, 2005 6:15 pm
It's the wet weather in north west europe. They expect us to always suffer from a slight cold. Thus we should have worse hearing than the taiwanese.
Discussions about Silent Computing
https://www.silentpcreview.com/forums/
https://www.silentpcreview.com/forums/viewtopic.php?f=6&t=19497
96%Mats wrote:Looks like a winner except being second best (just 2 W more, nevermind) at the lowest output. Draws 32 W less than the Phantom at their highest load? If true, then we're really talking high efficency!
The Phantom have a 86-87 % efficency at that output and 120 V, so even better at 240 V, which should put the Zen beyond 90 %??? Since it's an overclocked system I guess the output from the 3.3, 5 and 12 V lines are different to SPCR's tests, they're probably having a higher load on the 12 V line. But let's use SPCR's measured efficency anyway:
309*0.86/277 = 0.959.... With that kind of efficency I really can consider buying a fanless PSU.
Ok, I must be wrong, or maybe Matbe...
Hold it there! This was measured on a non-PFC model at 120VAC. The EU version has Active PFC and is spec'd by Antec to be something like 4% less efficient.Mats wrote:The Phantom have a 86-87 % efficency at that output and 120 V, so even better at 240 V, which should put the Zen beyond 90 %???
Thanks Mike. Forgot about PFC. I was just waiting for someone to find out what I did wrong, I didn't really believe in it myself!MikeC wrote:Hold it there! This was measured on a non-PFC model at 120VAC. The EU version has Active PFC and is spec'd by Antec to be something like 4% less efficient.
Tested on: P4 3.0F @ 4.0 Ghz, 1.55V. ABIT Fatal1ty AA8XE, 1GB OCZ PC2-5400 EB, Radeon X850 XT, Maxtor DiamondMax 10 SATA.techPowerUp.com wrote:Test Equipment
- The PSU is connected to an APC SmartUPS which supplies clean 220V input.
- AC current is measured using a Peaktech 4010 desktop RS232 multimeter with 0.01A accuracy.
- To measure DC output voltages of the PSU we use a 20-bit data acquisition system calibrated to 10uV accuracy.
- Power Factor is measured using a generic Power Efficiency Meter.
- Heatsink temperature was measured using a HP 2801A thermometer calibrated to 0.01°C.
- Measurements for Ripple Voltage were obtained using a 30 MHz HAMEG Analog/Digital Oscilloscope.
techPowerUp.com wrote:All lines are fluctuating a bit, but nothing too serious. Considering the load we put on this small 300W PSU, it is suprising that it kept the system stable.
All lines are well within the limits set by the ATX Specification.
[...]
However, we tried a few things to break the FSP Zen but it always came out top.
techPowerUp.com wrote:What I noticed, is that even though the PSU is fanless, it emits a some electric noise. During normal operation it is not noticable, but when you put your ear right on the PSU case, you can hear it.
Also when load is high, you can hear the sound coming out of your case. What is even more disturbing is that the tone changes with the load, so you notice it much more, compared to a constant high pitched sound.
Could be...a 100% efficient PSU would produce no heat at all...mathias wrote:Maybe the heatsinks are so small because it doesn't really heat up much in any particular place due to the high efficiency?
It's a bit warm in the Athlon 64 and warm in the K7 conf; but not hot. The CPU temperature also rised 2C in both configuration, maybe because the lower RPMs of the 12cm fan. (Room temperature 21C)kjf105 wrote:Tamas: Does it seem to run hot under load?
If I mod this PSU, I would change to this fan:Green Shoes wrote:I wonder if the fan could be modded to something more to our liking, Nexus, Globe or something. If there's no coil whine it seems like there wouldn't be too many other factors to consider. The power consumption figures are promising, anyway....
While I appreciate your mini-review, the reasons for your disappointment are mostly superficial, and your own data tells a different story than your words.Tamas wrote:In my opinion this is not a good PSU, the Protechnic fan is very bad (clicking,resonating). The heatsinks are too flat, too small. But the fan contorller works well, the fan didn't ramp up too much, it stayed at low speed even in the tuning K7 configuration. But in summer maybe these small heatsinks cause much higher fan RPM's.
Not recommended.
Even better would be if Nexus asked Fortron to develop a successor for the NX-3500 based on the design of the Green Power and the fan control and fan of the NX-4090. (The NX-4090 is to new to be replaced, but the NX-3500 is an older design by now.)MikeC wrote:Too bad about the fan... but at least it should be easily replaceable.
The Hungarian Nexus dealer told me 2 month ago that a new high eff. Nexus PSU coming.Tibors wrote:Even better would be if Nexus asked Fortron to develop a successor for the NX-3500 based on the design of the Green Power and the fan control and fan of the NX-4090. (The NX-4090 is to new to be replaced, but the NX-3500 is an older design by now.)MikeC wrote:Too bad about the fan... but at least it should be easily replaceable.
Thanks your advices.MikeC wrote:While I appreciate your mini-review, the reasons for your disappointment are mostly superficial, and your own data tells a different story than your words.Tamas wrote:In my opinion this is not a good PSU, the Protechnic fan is very bad (clicking,resonating). The heatsinks are too flat, too small. But the fan contorller works well, the fan didn't ramp up too much, it stayed at low speed even in the tuning K7 configuration. But in summer maybe these small heatsinks cause much higher fan RPM's.
Not recommended.
In the 2 setups you described...
-- the Green Power model produces 12W & 18W less than the FSP 350-60PN(PF) in IDLE mode.
-- the Green Power model produces 17W & 23W less than the FSP 350-60PN(PF) at max draw in the game.
Those are NOT small differences.
WE can guesstimate that the efficiency of the 350-60PN(PF) is no better than 70% at the higer load. (Based on my review of a very similar model.) Assuming your power figures are reasonably accurate, this means it would have been producing ~58.5W of heat and delivering 136.5W (based on the 195W AC draw). The Green model would have been producing just 35.5W of heat delivering the same about of power. This gives us an approximate efficiency of 79.3%. (All this is based on many assumptions, so it is very rough)
The greatly reduced heat output is a perfectly good reason for the smaller HS, and there is absolutely nothing "bad" about using a simple circuit w/fewer parts. It is a natural cost venefit to achieving high efficiency in a PSU. Engineering, generally, is the art of doing more with less. Low parts cost /count is part of the efficiency formula. You say yourself that the exhaust air stayed cool.
Too bad about the fan... but at least it should be easily replaceable.
dimensions approximately: 14,6cm X 11,4cm, max. height 5,5cmmb2 wrote:looking at the small internal size of this (and being as i stupidly recently broke my m-ATX psu ) it makes me wonder about the potential of modding this into a M-ATX psu?
the only possible problem i can see is the 'width' of the board?.. anyone got any dimensions?
makes u wonder why no one produces more powerful m-ATX psus!.. (~280w is the most powerful IIRC.. someone correct me if u know better)
and any word on availablilty or pricing in the uk? ..i'm hoping this 'low raw materials use' idea translates to low cost too!
i'm hoping that nexus take this and mod it with monster heatsinks
hm that 14.6 is the killer like i thought.. it would be rather difficult to mod.. it would need a new case built (ie couldn't go into matX psu shell) and be rotated 90*.. but.. possibleTamas wrote:dimensions approximately: 14,6cm X 11,4cm, max. height 5,5cm