Winmate DD-24AX DC-DC Module for Silent, Efficient Power

Power solutions continue to evolve in this era of ubiquitous computing. Electrodacus offers the PC enthusiast a picoPSU alternate: A silent, 88% efficient DC-DC converter module with full regulation to 130W output… at a price made for ultra SFF hobbyists. A close look the Winmate DD-24AX under the SPCR power microscope.

Product	DD-24AX V220-2 130W DC/DC power supply
Manufacturer	Winmate
Sample Supplier	Electrodacus
MSRP	US$60

This 130W computer power supply is not a standard ATX12V power supply. It is a DC/DC converter, much like the picoPSU, requiring a source of DC current to generate the various voltage lines required for a computer. The nominal required input is 24VDC, which must be provided by an AC/DC adapter. The only use for a DC/DC converter like this a few years ago would have been in a digital kiosk, signage, security, cash register or other such limited, industrial embedded computing applications. Today, with the diminishing power requirements of computing components, this modest DC/DC adapter can actually power a full-function desktop, media / file server, office computers or even a light gaming machine.

The maker of the DD-24AX 130W DC-to-DC power module is Winmate, a Taiwan-based company whose core business is developing LCD application products, digital signage solutions, and embedded systems integrating GPS, wireless, and touch screen technology. The company does not sell direct to consumers. An engineer in Canada with an interest in such power products for himself obtained a deal for 600 DD-24AX units in order to go into business offering it in the SPCR forums and on eBay for silence-focused DIY PC enthusiasts all over the world. His forum ID is Electrodacus. All of the specifications about the DD-24AX V220-2 came from Electrodacus, but they originated from Winmate.

The 130W DC/DC converter atop a 180W AC/24VDC adapter supplied by Electrodacus for use in this review.

WHY USE A DC/DC CONVERTER?

Why not stay with the tried-and-true ATX12V PSU? Here’s a summary of the advantages of using a DC/DC converter with an AC/DC adapter over a conventional ATX12V power supply. A key point is that the advantages are greatest at lower power.

High Efficiency at Lower Power

Like all machinery, every power supply achieves best efficiency over a certain range, usually at 40~80% of full capacity. With most ATX12V power supplies rated at a minimum of ~400W these days, there’s not much hope for them to achieve their maximum efficiency with systems that pull under 50W at idle and barely 100W at maximum load. The majority of mini-ITX PCs fall into this category as do many efficiency-optimized micro-ATX systems designed for use as media PCs or file servers.

Even modestly priced DC/DC converters offer very high efficiency, often 90% or higher, and AC/DC adapters of 60W~150W that achieve 90% efficiency are not uncommon. Such a combination for a PC with modest power demand can often draw less power than conventional ATX12V PSUs, even those rated for 80 Plus. Why? Because the vast majority of high effiency ATX12V PSUs are rated for 500W or higher; at 40W, many struggle to make 70% efficiency and barely reach 75% at 65W. In contrast, an efficient DC/DC converter can be used with an efficient AC/DC adapter of just the right power rating to achieve ~80% and better efficiency throughout the actual power range of the low power computer.

Electrodocus offers a comparison of AC power consumption between an SFX PSU and the DD-24AX with a 150W adapter, both driving the same low-power optimized quad-core Intel Q8400S system. The comparison favors the DD-24AX by 8~13W depending on mode. More on this later in our own testing and analysis.

Small Size for Greater Flexibility

A low power DC/DC converter is usually quite small, and because it generates little heat, there is a lot of flexibility in installation. It will go into the smallest of microATX cases, and even into very small mini-ITX cases. This is a big advantage as PC users continue to downsize.

Less Heat, No Noise

The icing on the cake with the converter + adapter approach is lower heat in the PC and zero noise from the power adapter. Half of the heat generated in the AC-to-DC converstion is outside, in the external adapter. The DC/DC converters usually don’t require a fan for cooling, and most AC/DC adapters are also passively cooled.

The cost of the converter + adapter approach is quite favorable for low power, say under 120W, but as the power requirement goes up, the price gets considerably higher. A good 200W AC/DC adapter is often quite pricey, perhaps double that of a good 400W ATX12V PSU. At >150W loads, the cooling requirement becomes greater, and if an additional fan must be used to cool the DC/DC converter, then the advantage pass quickly to the standard ATX12V PSU.

PACKAGING & SPECIFICATIONS

The DC/DC board came in a small, well-packed, plain cardboard carton with standard shipping accessories. The black cable is an input extension.

Winmate DD-24AX V220-2 General Specifications
ITEM	Our Comment
Nominal Input: 24VDC	18V~28V allowable, 8A~5A
Total Output: 130WDC	Enough for lots of systems
Ripple And Noise 5V: Less than 100mV 3.3V: Less than 66mV 5Vsb:100mV 12V: 120mV -12V: 200mV	Seems high for some lines. Normal ATX12V requirements are 50mV for the 5V & 3.3V lines, and 120mV for the 12V line.
Line Regulation Less than 2.5% at rated load with 10% change in input voltage	Pretty good.
Protection from short circuits (SCP) and over current (OCP)	Good, but not quite as complete as some ATX12V power supplies.
Efficiency: 88%	Excellent… but at what loads?
RoHS Compliant. (Lead free)	Required for sale in the EU.
Weight: 114 grams Size: 154 x 45 x 25.6 mm	It’s very small and light.

PRODUCT DETAILS

Eletrodocus took some good, high detail photos, so many of them are used here. The overall impression is of a good quality 4 layer board, nicely assembled with good quality parts.

The PCB is laid out nicely, with 105°C rated capacitors throughout. Most are Elcon brand; the three small ones on the left are S.Y. The trace side shows good, clean soldering. This view shows the different sections of the board by function. The output connectors are grouped in a row, starting on the right with 20-pin ATX, a standard 4-pin Molex, and a 2×2 Aux12V. The 2×2 connector on the left is an input connector, an alternate to the round 4-pin input shown below; its wiring configuration is the same as any standard 2x12V Aux12V output. It appears that the top two pins are positive, and the bottom two pins are negative. Each pair is wired in parallel. The 20-pin ATX output cable also provides a SATA power connector, legacy 4-pin Molex, and a 2×2 AUX12V connector. Using separate cables off the other output connectors might mean less voltage drop, but with only 130W in total, the current in the cables would be very modest, so the advantage is probably not significant. The ATX cable is only 6" long, but since the board is so small, it should not be difficult to position close to the motherboard in most cases. Wiring block diagram shows that the 5V standby line is live even when the computer is turned off — as it should be.

TESTING

For a fuller understanding of ATX power supplies, please read
the reference article Power
Supply Fundamentals. Those who seek source materials
can find Intel’s various PSU design guides at Form
Factors. For a complete rundown of testing equipment and procedures, please
refer to SPCR’s
PSU Test Platform V4.1. The testing system is a close simulation of
a moderate airflow mid-tower PC optimized for low noise.

Acoustic measurements are now performed in our anechoic chamber with ambient level of 11 dBA or lower, with a PC-based spectrum analyzer comprised of SpectraPLUS software with ACO Pacific microphone and M-Audio digital audio interfaces. In this review. no acoustic testing was necessary; the product under review is silent.

The usual practice of putting the PSU in the airflow path of the heat generated by the "components" in the simulated PC case had to be abandoned for this PSU test. The DC/DC converter was plugged into the PSU tester as shown below, and the power adapter was placed on the floor.

Winmate DD-24AX V220-2 ready for testing.

Testing a DC/DC converter is not like testing a standard ATX12V power supply. It is only half of the power supply, the other half being the AC/DC power adapter used with the converter. Changing the power adapter changes the end results.

Hence, testing was done twice, with two different AC/DC adapters. One was a no-name 24VDC 180W unit supplied by Electrodacus for use in this review. It was simply the only unit he had on hand at the time, not an item offered, sold or recommended. The second is a 120W FSP unit that happened to be in the lab, model FSP120-AAB. The FSP Group web site did not have complete specifications but the number of approvals and certifications on the unit are evidence of higher quality. A web search turned up this model (and slight variants with different output plugs) priced at $30~55, including some items described as "refurbished".

Keep in mind that if you want to achieve the maximum power output capability of the Winmate DD-24AX, the power adapter must supply at least 150W output, preferably higher, to allow some headroom. The Windmate will lose 12% of the input power — 88% efficiency, remember? Hence, in theory, the 120W FSP120-AAB is safe only for ~106W output from the Winmate.

This is the AC/DC adapter provided by Electrodacus for use during our testing: Rated for 24V, 7.5A output (180W), with 100~240VAC, max, 2.9A input. Electrodacus was not confident of its quality; it was simply the only adapter he had available at the time. This is a 19VDC 6.32A (120W) adapter from FSP Group which happened to be in the SPCR lab. Its quality is assuredly better than the adapter above.

AC/DC Adapter Specifications
Parameter	FSP120-AAB	Generic 24VDC 180W
Input	90~264 VAC, 47~63 Hz	100~240 VAC, 50/60 Hz
Output	19VDC, 6.32A	24VDC, 7.5A
Temperature range	Operating: 0 ~40°C Storage: 20~60°C	n/a
Humidity range	Operating 20% ~ 80% RH Strorage 10% ~ 90% RH	n/a
Efficiency	> 86%	n/a

TEST RESULTS

The ambient temperature was 22°C, and the ambient noise level was 11
dBA.

TEST A RESULTS Winmate DD-24AX V220-2 with FSP120-AAB 19VDC 120W Power Adapter
DC Output Voltage (V) + Current (A)								DC Output	AC Input	Calculated Efficiency	Power Factor
+12V1		+5V		+3.3V		-12V	+5VSB
12.30	0.99	5.08	0.97	3.33	0.95	0.1	0.1	22	32	69%	1.0
12.28	1.95	5.08	1.94	3.33	0.94	0.2	0.1	40	53	75%	1.0
12.22	3.88	5.05	1.93	3.33	1.91	0.1	0.1	65	82	80%	1.0
12.16	5.80	5.05	1.93	3.31	1.89	0.2	0.2	90	113	80%	1.0
12.15	7.68	5.02	4.66	3.30	2.73	0.3	0.2	130	169	77%	1.0
Crossload (most output load on 12V line)
12.16	7.81	5.05	0.98	3.33	0.93	0.1	0.1	105	133	79%	1.0
12.18	9.68	5.05	0.98	3.33	0.94	0.1	0.1	128	164	78%	1.0
+12V Ripple on any output line (peak-to-peak): <20mV @ 130W AC power with no load: 6.3W AC power when off: 0.8W
NOTE 1: The current and voltage for -12V and +5VSB lines is not measured but based on switch settings. It is a tiny portion of the total, and errors arising from inaccuracies on these lines is <1W. NOTE 2: The 130W load must be considered a serious overload condition for the FSP power adapter, which is rated for 6.32A maximum. To produce 130W output, the adapter was producing about 7.9A. This was maintained for a little over 5 minutes.

TEST B RESULTS Winmate DD-24AX V220-2 with Generic 24VDC 180W Power Adapter
DC Output Voltage (V) + Current (A)								DC Output	AC Input	Calculated Efficiency	Power Factor
+12V1		+5V		+3.3V		-12V	+5VSB
12.27	0.97	5.08	0.97	3.33	0.95	0.1	0.1	22	33	66%	0.65
12.27	1.94	5.08	1.94	3.33	0.94	0.2	0.1	40	55	72%	0.70
12.22	3.88	5.05	1.93	3.33	1.89	0.1	0.1	65	86	76%	0.71
12.16	5.78	5.04	1.93	3.31	1.89	0.2	0.2	90	119	75%	0.74
12.14	7.68	5.02	4.66	3.30	2.73	0.3	0.2	130	178	73%	0.76
Crossload (most output load on 12V line)
12.16	7.85	5.05	0.98	3.33	0.93	0.1	0.1	105	138	76%	0.74
12.16	9.63	5.05	0.98	3.33	0.93	0.1	0.1	127	171	74%	0.76
+12V Ripple on any output line (peak-to-peak): <20mV @ 130W AC power with no load: 6.3W AC power when off: 1.3W
NOTE: The current and voltage for -12V and +5VSB lines is not measured but based on switch settings. It is a tiny portion of the total, and errors arising from inaccuracies on these lines is <1W.

Comparing the two sets of results, let’s look first at the commonalities:

1. Voltage Regulation was excellent, and quite similar with both adapters. The 12V line varied a maximum of 0.15V with either adapter, which is just 1.25%. The 5V line varied just .03V, and the 3.3V line was rock solid.

2. Ripple and noise was very low, about 16mV at the highest peak seen with either adapter on any voltage line. This is excellent.

Both of the above results can be attributed to the Winmate DD-24AX, which regulates all the output voltage lines. This is a major difference between the Winmate DD-24AX and the picoPSU: The latter simply passes the 12V line through from its input. In other words, the 12VDC line is filtered, regulated and protected by the Winmate, while the 12VDC line in the picoPSU is entirely dependent on the quality of the 12VDC adapter feeding its input. For this reason, the picoPSU cannot be considered as secure or reliable a power supply as the Winmate DD-24AX.

3. There was no significant noise from either of the adapters, with any load or plugged in without any load — the buzzing or whining that can occur under the latter condition is a serious problem with some adapters.

4. The DD-24AX made no audible noise at any load.

5. Starting with no load was no problem with either adapter, at the same 6.3W AC power draw.

6. When turned off but still plugged in, the AC power draw was low, slightly lower for the FSP.

Now, consider the differences:

7. Efficiency: By now, most SPCR readers know that this is a measure of AC-to-DC
conversion efficiency. The ATX12V Power Supply Design Guide recommends 80% efficiency
or better at all output power loads. 80% efficiency
means that to deliver 80W DC output, a PSU draws 100W AC input, and 20W is lost
as heat within the PSU. Higher efficiency is preferred for reduced energy consumption
and cooler operation.

At the super low 22W load, efficiency was quite good with either adapter. But as the load was increased, efficiency with the FSB rose quickly, reaching 75% at just 40W, and the high of 80% by 65W and staying there to about 100W. The 40W~65W efficiency performance can only be matched by Bronze (or higher) 80 Plus ATX12V PSUs. By contrast, with the generic 24VDC adapter, efficiency peaked at 76% at 65W and fell beyond that point. This is bettered by most 80 Plus approved power supplies.

Accepting the claimed 88% efficiency for the Winmate DD-24AX, the efficiency of the power adapters can be calculated:

Estimated Efficiency of Power Adapters (assuming 88% efficiency in DD-24AX V220-2)
Load	FSP120-AAB adapter	Generic 24VDC adapter
22	78%	75%
40	85%	82%
65	91%	86%
90	91%	85%
130	88%	83%

It’s obvious that the FSP is a much more efficient adapter. At 130W load, when the FSP adapter was in overload by at least 20W, the efficiency was still an estimated 88%. The droop at the extremes for both adapters may be a bit exaggerated; the DC/DC converter surely must vary in its efficiency as well, and run slightly less efficiently at very high and very low loads.

Given that FSP only rates this adapter to be better than 86% efficient, it’s possible that the 88% efficiency spec of the Wingate is conservative. It could easily be higher. In any case, since the FSP adapter is claimed to be 86% efficient and the Winmate board is claimed to be 88% efficient, the combined claimed efficiency of the two together amounts to 75%. The test results show better efficiency than this for most of the operating range, so the claim of the Winmate board is verified.

It’s also possible that the Winmate DD-24AX is more efficient with 19VDC input than with 24VDC; this cannot be determined SPCR’s existing power testing gear.

EFFICIENCY UPDATE – Oct 28/09 After some experimentation with the SPCR power supply test system, a way to measure the energy efficiency of the power adapters was established. This methodology has to be refined, but the preliminary results look accurate. The FSP120-AAB adapter was found to have about 86~89% efficiency. This suggests that the Winmate DD-24AX V220-2 has a much higher efficiency than 88%. At the high point of the curve, it appears to be at least 93% efficient, and the 88% figure can be regarded as a worst case. More on all this to come later, as a method of directly measuring the efficiency of the Winmate DC/DC adapter is being devised.

8. Power Factor is ideal when it measures 1.0. In the most
practical sense, PF is a measure of how "difficult" it is for the
electric utility to deliver the AC power into your power supply. High PF reduces
the AC current draw, which reduces stress on the electric wiring in your home
(and elsewhere up the line). It also means you can do with a smaller, cheaper
UPS backup; they are priced according to their VA (volt-ampere) rating. Power
factor with the FSP was as good as it can be in every state. With the generic 24V adapter, it was mediocre and never went above 0.76.

9. Temperature data was not put into a table because readings were made with an infrared hand-held thermometer at many points. The DC/DC board was checked for temperature after about 5 minutes at full load. With the 24VDC adapter, the single hottest point on the Winmate DD-24AX was the large coil at the power input, which reached over 65°C at full load. Most of the other components were substantially lower, varying from 30°C to about 55°C. With the FSP adapter, the temperature of the big coil was ~15°C lower. The 24VDC adapter itself ran considerably hotter, reaching nearly 50°C in some spots. The FSB, in contrast, measured under 40°C throughout testing.

COMPARISONS

Our usual comparison table showing the noise level of various power supplies at loads starting from 90W just doesn’t apply here. The Winmate DD-24AX is silent, and if the adapter used with it is silent, it beats all the ATX12V PSUs acoustically except the fanless ones or those with good semi-passive operation.

A more relevant comparison is for efficiency, against other silent or near-silent power alternatives. The data in the table below is culled from previous SPCR reviews.

Comparison: Silent or near-silent PSU Efficiency in %, 20W~150W
Model	20W	40W	65W	90W	150W
Winmate DD-24AX w/ FSP120-AAP adapter	69	75	80	80	77 (130W)
picoPSU w/ 80W adapter	79	83	85	85	–
Enermax Modu82+ 625	66	78	80	81	84
Coolermaster M700W	66	78	80	83	84
Seasonic X-650	65	77	83	87	88
Seasonic M12D 850W	62	73	78	82	84
Nexus Value 430	62	72	77	81	82
Silverstone ST45NF fanless	61	72	77	81	83
Antec CP-850	55	69	74	82	83
Antec Signature 650	55	67	73	79	82

This table may not be that useful. It puts the Winmate at the top, then orders the rest by highest efficiency at 20W and 40W. The ordering does not jibe if you look at efficiency at higher power points.

The actual watts is also worth considering: 69% vs 79% at the 20W load makes you think "wow!" — but the difference is only 3.7W.

What’s clear is that the picoPSU with its little 80W adapter leads in efficiency up to 65W load. It’s even smaller than the Winmate DD-24AX, which makes it very useful for SFF and Ultra SFF systems. One reason for its high efficiency is that the 12VDC line is taken directly from the adapter, without any conversion loss within it, and the 12V line dominates the demand in most modern PCs. But this is also the picoPSU’s downside, the lack of 12V output regulation, which may be significant in areas with less than ideal AC power service or when very high precision and reliability is sought.

Price favors the Winmate slightly, at least if you’re buying from Electrodacus. Plus, if a more efficient power adapter is used, the Winmate might equal the picoPSU’s efficiency… although this would be a challenge because the Winmate always loses something in the conversion to 12VDC output, while the picoPSU does not. As with the picoPSU, the Winmate’s advantage is also its weakness.

Both the Winmate DD-24AX and the picoPSU have a huge advantage over conventional PSUs if you seek to assemble a silent, fanless, super small low-power PC. They encourage and reward creative modding or system building.

CONCLUSIONS

The Winmate DD-24AX is a high quality DC/DC adapter that meets the specifications claimed by its manufacturer. It’s also a bargain at the $30 price Electrodacus is offering., which is half the suggested retail. When combined with a suitable, efficient power adapter, the Winmate can easily power a highly capable desktop or server for home use, adding no noise and very little heat at very modest cost. The ideal use of the DD-24AX is for a system that demands very low power in idle, under 50W, with brief peaks up to 130W under load. This takes advantage of the unit’s highest efficiency range, but the 130W peak capability will allow at least a modest discrete graphics card along with a couple of hard drives.

In many ways, the biggest challenge is to find a suitably high efficiency power supply of adequate power output at a reasonable price. To that end, you are directed to Energy Star, which maintains a database of high efficiency external AC/DC adapters, with ratings for power load, output voltage, and efficiency at both 115VAC and 230VAC. As a value added service, here’s a pre-sorted compilation from that database, in MS Excel format, of all current adapters which offer 88% or higher efficiency, with output of 19~28VDC, 70~230W; ie, the best of the highest efficiency external adapters.

We’ll be scouring that list as well, for an ideal adapter to make use of our review sample in a new system, perhaps a new file/media server that can double as a lab PC or HTPC.

* * *

Thanks to Electrodacus whose efforts made possible this review and the opportunity for PC enthusiasts to obtain an unusual, high quality power product at a modest price.

Winmate DD-24AX

SPCR Recommended Award

SPCR Articles of Related Interest:
Power
Supply Fundamentals
Recommended
Power Supplies
SPCR
PSU Test Rig V.4
Tiny, Silent and Efficient: The picoPSU

Antec
Signature 650
Seasonic M12D-850W
Seasonic X650 80 Plus Gold
Antec CP-850: Unique PSU with Top Performance

* * *

Discuss
this article in the SPCR Forums.