Viewing page 4 of 5 pages.
Measurements were made at 4 output power levels: 65W, 90W, 150W and 300W . The PSU was allowed to run for ~10 minutes at each power level before measurements were recorded. The room temperature was 24C. Two samples were tested in exactly the same way. The difference between them was less than 1% for any measurement. Shown here are the measurements from one of the two -- neither had any significant superiority.
1. EFFICIENCY was excellent. Even at the low 65W load level where most PSUs fail to reach even 65%, the Super Tornado started with an amazing 75.6%. The highest efficiency reached was 77% at a very low 90W and also at 150W. The efficiency of the Super Tornado betters all other PSUs tested at SPCR by a big margin -- except for the Super Silencer 400, which reached 78%.
Whether this sample can be considered to have met its published specs depends on which specs one goes by. By the more technical specification document, which states minimum efficiency as being higher than 74% at 120VAC input, it clearly has; by the marketing materials, which claim 78%, it hasn't quite made it.
As with the Super Silencer 400, the most significant result of this high efficiency is that the PSU always runs cooler than other models. This will have a real effect on noise in actual use, as the thermal fan speed controller will keep the fan spinning slower over a wider variety of loads than with a lower efficiency PSU.
2. VOLTAGE REGULATION was excellent, within -/+2% on all lines in any combination of loads tried (somewhat at random). It was often within -/+1%. The low and high voltage seen on each of the main lines is shown:
+12V: 12.02 to 12.28
+5V: 4.92 to 5.14
+3.3V: 3.3 to 3.42
It should be noted that I have no way of testing line regulation, so AC conditions are steady-state, not dynamic as it would be (potentially) in a real PC; I have no way to vary input AC voltage at this time. The AC line voltage in the lab as measured by the Kill-a-Watt power meter is usually within a couple of volts of 120V.
3. POWER FACTOR was 0.98~0.99 at all power levels, as good as can be. This and the high efficiency may help you save money on your electric bill and make you feel environmentally (morally?) superior.
4. NOISE was extremely low on startup and very low throughout the testing. In essence, it is the new champ among quiet fan-cooled PSUs.
The 120mm SuperRed fan, at the ~4.7V start fan voltage, made so little noise that I had to get within about a foot to hear anything at all with the unit sitting on the test bench. My wife, who has more sensitive hearing, picked it up at about 3' distance but found that it was very faint. With my ear right up against the unit, there was a faint trace of electronic sizzle -- this would turn into coil buzz if it got loud enough, but in both Tornado samples, the noise remained at the same super-low level at all times. It is inaudible for all practical purposes. The fan noise consisted of a faint "chugga-chugga" sound and a periodic "shhh-shhh-shhh" as if something was rubbing softly. All this was, I emphasize again, with my ear within about 6 inches of the fan.
Noise measurements were made in the live test room in the evening. With all equipment turned off in the 23C temp room, the ambient noise was measured at ~17 dBA. Measured noise did not change when the Seasonic was turned on with a 65W load 1 meter away from the sound level meter. In other words, turning the Super Tornado on did not add any noise that could be heard above the ~17 dBA background ambient noise level. There was also no change in audible noise with the PSU at 90W load. In case there is any doubt, this is extremely quiet performance under these loads, better than any other PSU tested at SPCR thus far.
At the 150W load, the fan speed seemed to turn a corner and started to rise. Left running at 150W load for about 15 minutes, the fan speed climbed for several minutes till 7.25V was reached. Then, the fan voltage dropped steadily and stabilized at about 5.5V. That is where the 22 dBA was measured. Almost all the noise consisted of the whooshing of air turbulence.
The ~17 dBA measured at low-med load is 5 dBA quieter than the Seasonic Super Silencer 400. It is quieter than the Nexus NX3000, which measures about 3 dBA higher under the same conditions. The Tornado's base noise performance approaches the start noise level of the 14dB SilenX 400 PSU. It easily beats the SilenX 400 in any real application because the Tornado remains at virtually the same noise level up to quite a high output level while the SilenX 400 noise tends to rise almost immediately after turn-on.
At full load, with the fan voltage at the full 12V, the PSU is quieter in the case than out by itself on the test bench. This is because inside the case, the fan gets placed farther away from the measuring microphone -- and the human ear. One of the side benefits of this fan / PSU configuration. Even though 37 dBA @ 1 meter is plainly audible, the noise is not at all annoying, due to the broadband pink noise the whooshing most closely resembles.
5. COOLING: The high efficiency of the Super Tornado 300 allowed it to remain cool even in the in-case thermal simulation with a 100W bulb. Table C details the results, which are self-explanatory: Along with the Super Silencer 400, it is the coolest running PSU yet encountered. In a real system, running a without any case fan is a very viable option with the Super Tornado.
Usually, using the thermal speed controlled fan in the PSU as the only heat exhaust causes the fan to speed up as the PSU heats up, thereby adding back the noise removed with the case fans. The in-case thermal simulation results suggests this is not likely to happen with the Super Silencer, at least not with loads up to ~150W. Given that most systems draw less than 150W even at 100% loads, this promises very quiet operation.
6. LOW NOISE BY INTEL'S DEFINITION?
You may be aware that Intel's ATX12V v1.3 PSU Design Guide has guidelines for low noise PSUs. Section 5.7 on page 54 of the Guide calls for a maximum sound power level in a PSU designated as low noise to emit no greater than 4.0 Bels sound power at 50% load and ambient intake air temperature of 43C.
While 4.0 Bels is not exactly super quiet, the conditions of 50% power load and 43C air temp makes this a very tough standard to meet, especially for high power PSUs. Few models even on our Recommended quiet PSU list would pass this test. Most quiet PSUs are quiet at power loads under 150W.
The Super Silencer 400 review checked to see whether it would pass; it probably would not. At the time, I conjectured that a 300W version of the Seasonic "Super" family might well pass because of the lower load level it would have to pass: 150W instead of 200W. The Super Tornado 300 might be an even better candidate because of its fan's higher airflow at lower RPMs.
It is difficult to make a precise conversion of Bels (sound power) to decibels (sound pressure level), but 4.0 Bels usually falls 32~33 dBA @ 1 meter in most environmental conditions. The 100W light bulb in the case was raised and brought closer to the intake vents to get closer to the 43C intake temp required.
Final Result: YES! In the closest match to specified conditions, at an intake temperature of 39~41C with the PSU running at 150W load, the noise measured 28 dBA @ 1 meter. Given this result, the Super Tornado would most likely pass a formal, more stringent test for the section 5.7 low noise guideline in Intel's ATX12V v1.3 PSU Design Guide.