Topower is a traditional OEM manufacturer, being the real manufacturer behind units from brands like Tagan and OCZ (not all power supplies from OCZ are manufactured by Topower, though). They also sell power supplies using their own brand and today we are going to review their PowerBird 900 W unit, which uses a full modular cabling system and is based on a two-transformer design. Let’s see whether this is a good unit or not.
PowerBird 900 W is a long power supply, being 6 57/64” (17.5 cm) deep, using a 120 mm fan on its bottom and featuring active PFC, of course. But what is really different about this power supply is its fully modular cabling system. Usually when manufacturers say their products have a modular cabling system, the main motherboard cable and usually the ATX12V/EPS12V cable come directly from inside the power supply, not using the modular system. This does not happen with PowerBird 900 W, where even the main motherboard cable is attached to the modular system.
All cables use a nylon protection and all wires are 18 AWG, except the wires on the main motherboard cable, which are thicker (16 AWG).
The modular system has 11 connectors and this power supply comes with 12 cables:
- Main motherboard cable, with a 20/24-pin connector.
- EPS12V cable.
- One cable with two ATX12V connectors that together form one EPS12V connector.
- Two cables with one six-pin auxiliary video card power connector each.
- Two cables with one six/eight-pin auxiliary video card power connector each.
- One cable with one six-pin and one six/eight pin auxiliary video card power connectors.
- Two cables with four SATA power connectors each.
- Two cables with three standard peripheral power plugs and one floppy disk drive power connector each.
All cables are relatively long, with 21 21/32” (55 cm) between the end that goes into the modular cabling system and the first connector on the cable. On the cables with more than one connector, there is 5 29/32” (15 cm) between the connectors.
Having a total of six power connectors for video cards, this unit supports three-way SLI. Two of these connectors, however, are available on the same cable. This usually leads to voltage drop at the connector when the unit is fully loaded and because of that we prefer to see all connectors using individual cables.
Now let’s take an in-depth look inside this power supply.
[nextpage title=”A Look Inside The PowerBird 900 W”]
We decided to disassemble this power supply to see what it looks like inside, how it is designed, and what components are used. Please read our Anatomy of Switching Power Supplies tutorial to understand how a power supply works and to compare this power supply to others.
This page will be an overview, and then in the following pages we will discuss in detail the quality and ratings of the components used.
[nextpage title=”Transient Filtering Stage”]
As we have mentioned in other articles and reviews, the first place we look when opening a power supply for a hint about its quality, is its filtering stage. The recommended components for this stage are two ferrite coils, two ceramic capacitors (Y capacitors, usually blue), one metalized polyester capacitor (X capacitor), and one MOV (Metal-Oxide Varistor). Very low-end power supplies use fewer components, usually removing the MOV and the first coil.
On this power supply this stage is flawless. On the main AC connector (Figure 8) it has already all the required filtering components and more: four Y capacitors, one X capacitor and two ferrite coils. On the main printed circuit board it has one X capacitor, two Y capacitors and one ferrite coil, plus another X capacitor after the rectifying bridge. There are two MOV’s located after the rectifying bridge behind the two big electrolytic capacitors (not shown in Figure 9) and not before as usual.
In the next page we will have a more detailed discussion about the components used in the Topower PowerBird 900 W.
[nextpage title=”Primary Analysis”]
On this page we will take an in-depth look at the primary stage of PowerBird 900 W. For a better understanding, please read our Anatomy of Switching Power Supplies tutorial.
This power supply uses one GBJ2506 rectifying bridge in its primary, which can deliver up to 25 A at 100°. This component is clearly overspec’ed: at 115 V this unit would be able to pull up to 2,875 W from the power grid; assuming 80% efficiency, the bridge would allow this unit to deliver up to 2,300 W without burning this component. Of course we are only talking about this component and the real li
mit will depend on all other components from the power supply.
On the active PFC circuit two SPW35N60C3 power MOSFET transistors are used, each one capable of delivering up to 34.6 A at 25° C or 21.9 A at 100° C in continuous mode (note the difference temperature makes) or 103.8 A in pulse mode at 25° C.
This power supply uses two electrolytic capacitors to filter the output from the active PFC circuit. The use of more than one capacitor here has absolute nothing to do with the “quality” of the power supply, as laypersons may assume (including people without the proper background in electronics doing power supply reviews around the web). Instead of using one big capacitor, manufacturers may choose to use two or more smaller components that will give the same total capacitance, in order to better accommodate space on the printed circuit board, as two or more capacitors with small capacitance are physically smaller than one capacitor with the same total capacitance. On PowerBird 900 W two 1200 µF x 200 V capacitors are used in series; this is equivalent of one 600 µF x 400 V capacitor.
These capacitors are from Toshin Kogyo (TK) and are labeled at 105° C. Even though this is a Japanese vendor, they sell rebranded OST (i.e., Taiwanese) capacitors.
In the switching section, another two SPW35N60C3 power MOSFET transistors are used on the traditional two-transistor forward configuration. The specs for these transistors we’ve already published above. These two transistors drive the two available transformers.
This power supply uses the omnipresent CM6800 active PFC/PWM combo controller.
Now let’s take a look at the secondary of this power supply.[nextpage title=”Secondary Analysis”]
As mentioned, this power supply has two transformers, both driven by the same switching transistors. The first transformer, labeled T3 on the printed circuit board (and T122F on its top) is in charge of the +3.3 V and the +12 V outputs, while the second transformer, labeled T4 on the printed circuit board (and T121F on its top) is in charge of the +5 V and +12 V outputs. So while +3.3 V and +5 V outputs are produced by different transformers, the +12 V is produced by the two transformers, each one connected to a different transistor (see below).
The maximum theoretical current each line can deliver is given by the formula I / (1 – D), where D is the duty cycle used and I is the maximum current supported by the rectifying diode. Just as an exercise, we can assume a typical duty cycle of 30%.
The +12 V rectification is done with a synchronous design. On synchronous design the diodes are replaced by MOSFET transistors. On this power supply the direct rectification is done by two IRFB3206 MOSFETs (210 A at 25° C or 150 A at 100° C each in continuous mode or 840 A at 25° C in pulse mode each) – each one connected to a different transformer – while the “freewheeling” portion is done by three KCQ60A06 Schottky rectifiers (60 A at 69° C each) connected in parallel. For our math we need to consider the portion with the lower current limit, which is probably the rectification portion. From our understanding the transformers and thus the transistors work out of phase (i.e., switching at opposite times), and thus they are not turned on at the same time. So we have a maximum theoretical current of 214 A (150 A / 0.70) at 100° C, which equals to 2,571 W. That is what we call going overboard with overspec’ing!
The +5 V output is produced by two STPS60L45CW Schottky rectifiers connected in parallel, each one capable of delivering up to 60 A (30 A per internal diode at 135° C). This gives us a maximum theoretical current of 86 A or 429 W for the +5 V output.
The +3.3 V output is produced by one STPS60L30CW, which is capable of delivering up to 60 A (30 A per internal diode at 130° C). This gives us a maximum theoretical current of 43 A or 141 W for the +3.3 V output.
Electrolytic capacitors from the secondary are manufactured by Teapo and Hermei and labeled at 105° C.
[nextpage title=”Power Distribution”]
In Figure 15, you can see the power supply label containing all the power specs.
This power supply has six virtual rails, distributed like this:
- +12V1: Main motherboard cable.
- +12V2: The two EPS12V/ATX12V connectors.
- +12V3: Two of the four connectors for SATA and peripheral power cables.
- +12V4: Two of the four connectors for SATA and peripheral power cables.
- +12V5: The two light blue connectors for video cards.
- +12V6: The two dark blue connectors for video cards.
Each rail is clearly indicated on
the modular cabling system. The fifth video card power cable (which has two connectors) can be installed on either +12V3 or +12V4.
Power distribution is ok, although it would be better to see all video card cables on separated rails.
Now let’s see if this power supply can really deliver 900 W.[nextpage title=”Load Tests”]
We conducted several tests with this power supply, as described in the article Hardware Secrets Power Supply Test Methodology.
First we tested this power supply with five different load patterns, trying to pull around 20%, 40%, 60%, 80%, and 100% of its labeled maximum capacity (actual percentage used listed under “% Max Load”), watching how the reviewed unit behaved under each load. In the table below we list the load patterns we used and the results for each load.
For the 100% load test we faced a small challenge. The +12V2 input from our load tester was exclusively connected to the power supply +12V2 rail through its EPS12V connector (+12V1 input was connected at the same time to the power supply +12V1, +12V3 and +12V6 rails), and the power supply over current protection circuit (OCP) wouldn’t allow us to pull more than 30.5 A from this rail, and we wanted it to be at 33 A, with +5 V and +3.3 V set at 12 A each. The difference from what we could actually configure and what we wanted to configure was very small, though.
If you add all the power listed for each test, you may find a different value than what is posted under “Total” below. Since each output can vary slightly (e.g., the +5 V output working at +5.10 V), the actual total amount of power being delivered is slightly different than the calculated value. On the “Total” row we are using the real amount of power being delivered, as measured by our load tester.
|Input||Test 1||Test 2||Test 3||Test 4||Test 5|
|+12V1||7 A (84 W)||12 A (144 W)||19 A (228 W)||26 A (312 W)||33 A (396 W)|
|+12V2||7 A (84 W)||12 A (144 W)||19 A (228 W)||26 A (312 W)||30.5 A (366 W)|
|+5V||2 A (10 W)||6 A (30 W)||8 A (40 W)||10 A (50 W)||15 A (75 W)|
|+3.3 V||2 A (6.6 W)||6 A (19.8 W)||8 A (26.4 W)||10 A (33 W)||15 A (49.5 W)|
|+5VSB||1 A (5 W)||2 A (10 W0||2 A (10 W)||3 A (15 W)||3.5 A (17.5 W)|
|-12 V||0.5 A (6 W)||0.5 A (6 W)||0.5 A (6 W)||0.5 A (6 W)||0.5 A (6 W)|
|Total||196.3 W||354.1 W||538.6 W||726.1 W||905.9 W|
|% Max Load||21.8%||39.3%||59.8%||80.7%||100.7%|
|Room Temp.||48.3° C||48.0° C||49.4° C||46.9° C||49.8° C|
|PSU Temp.||48.5° C||48.6° C||49.7° C||47.6° C||53.0° C|
|Ripple and Noise||Pass||Pass||Pass||Pass||Pass|
|AC Power (1)||223 W||395 W||604 W||837 W||1,085 W|
|AC Power (2)||232.3 W||413.5 W||631.0 W||868 W||1,117 W|
|AC Voltage||112.3 V||109.9 V||108.1 V||105.3 V||102.3 V|
Updated 06/24/2009: We re-tested this power supply using our new GWInstek GPM-8212 power meter, which is a precision instrument and provides accuracy of 0.2% and thus presenting the correct readings for AC power and efficiency (results marked as "2" on the table above; results marked as "1" were measured with our previous power meter from Brand Electronics, which isn’t so precise as you can see). We also added the numbers for AC voltage during our tests, an important number as efficiency is directly proportional to AC voltage (the higher AC voltage is, the higher efficiency is). Also, manufacturers usually announce efficiency at 230 V, which usually inflates efficiency numbers. We added power factor (PF) numbers as well. These numbers measure the efficiency of the power supply active PFC circuit. This number should be as close to 1 as possible. Under light load (20% load, i.e., 180 W), the active PFC circuit from this unit isn’t as good as when operating under higher loads, but 0.968 is still a good number.
PowerBird 900 W achieved efficiency above 85% when we pulled between 40% and 60% from its labeled wattage (i.e., between 360 W and 540 W). Under light load (20% load, i.e., 180 W) efficiency was still high, at 84.5%. At 80% load (720 W) efficiency was at 83.7%, which is a good number. At full load efficiency dropped to 81.1%, but still above the 80% mark.
Voltage stability was another highlight from PowerBird 900 W, with all voltages inside 3% of their nominal values(i.e., voltages were closer to their nominal value than needed, as ATX spec allows voltages to be up to 5% from their nominal values, 10% for -12 V). The -12 V output, however, was at -11.12 V during test number one, -11.29 V during test number two, -11.45 V during test number three and -11.60 V during test number four. These numbers are still inside the 10% margin allowed for this output, but we always like to see outputs as close as possible from their nominal values. During test number five this output was within 3% from its nominal value.
And finally we have noise and ripple, which were low all the time. Below you can see the results for test number five. As we always point out, the limits are 120 mV for +12 V and 50 mV for +5 V and +3.3 V and all numbers are peak-to-peak figures.
Now let’s see if w
e could pull more than 900 W from this unit.[nextpage title=”Overload Tests”]
Before overloading power supplies we always test first if the over current protection (OCP) circuit is active and at what level it is configured.
As mentioned, we discovered that OCP was set at 31 A as we couldn’t pull more than that from +12V2.
Since we were already pulling the maximum we could using our equipment (33 A from its +12V1 input and 30.5 A from its +12V2 input), the only option for us to overload this power supply was to increase current at +5 V and +3.3 V.
The idea behind of overload tests is to see if the power supply will burn/explode and see if the protections from the power supply are working correctly. This power supply didn’t burn and when we tried to pull far more than it could deliver it would shut down, so this unit passed on this test.
|+12V1||33 A (396 W)|
|+12V2||30.5 A (366 W)|
|+5V||25 A (125 W)|
|+3.3 V||16 A (52.8 W)|
|+5VSB||3.5 A (17.5 W)|
|-12 V||0.5 A (6 W)|
|% Max Load||106.5%|
|Room Temp.||49.8° C|
|PSU Temp.||53.0° C|
|AC Power (1)||1,162 W|
|AC Power (2)||1,203 W|
|AC Voltage||100.7 V|
With PowerBird 900 W delivering 950 W efficiency dropped below the 80% mark (consider the results marked as "2", as they are the correct ones, measured with our precision power meter).
[nextpage title=”Main Specifications”]
Topower PowerBird 900 W power supply specs include:
- ATX12V 2.3
- EPS12V 2.92
- Nominal labeled power: 900 W at 40° C.
- Measured maximum power: 958.3 W at 49.8° C.
- Labeled efficiency: Minimum 80%, 87% typical
- Measured efficiency: Between 81.1% and 85.6% at 115 V (nominal, see complete results for actual voltage).
- Active PFC: Yes.
- Modular Cabling System: Yes, full.
- Motherboard Power Connectors: One 20/24-pin connector, one EPS12V and two ATX12V connectors that together form another EPS12V connector.
- Video Card Power Connectors: Three six/eight-pin connectors and three six-pin connectors.
- Peripheral Power Connectors: Six in two cables.
- Floppy Disk Drive Power Connectors: Two in two cables.
- SATA Power Connectors: Eight in two cables.
- Protections: N/A. Over current (OCP) and short-circuit (SCP) protections present and active.
- Warranty: Three years.
- More Information: https://www.topower.com
- Average price in the US*: USD 180.00.
* Researched at Newegg.com on the day we published this review.
PowerBird 900 W can really deliver its labeled power at 50° C, which is excellent. It also delivers high efficiency between 83.7% and 85.6% if you pull up to 80% of its load (i.e., up to 720 W). At full load efficiency was still above 80%, at 81.1%.
Noise and ripple were low and the main voltages were within 3% their nominal value, as promised by the manufacturer (tighter than the official standard, which states a 5% tolerance).
The fully modularized cabling system is a must for the high-end user that wants to have only the cables he (or she) will be really using, making things more organized inside the case and improving airflow.
It also supports three-way SLI directly, which is desired by its targeted audience.
This is an excellent buy for the high-end user. Although officially quoted at USD 250 it can be found for USD 180 at Newegg.com, which isn’t a bad price at all for a true 900 W power supply.
Many people are still asking why manufacturers sell high-wattage power supplies if even fully loaded PCs with three or four video cards don’t pull even close to 900 W. The answer lies on efficiency. Power supplies present their highest efficiency when delivering between 40% and 60% from their labeled power. Thus by buying a power supply labeled at double the actual power pulled by the system you are making sure that you are getting the best efficiency and this way saving on your electricity bill. For more information on this, please read our Everything You Need to Know About Power Supplies tutorial.