When it comes to discussing power banks and USB charging in general, it’s hard to overlook Anker branded products. The company was started by a group of friends working at Google, and has since put out a multitude of USB power banks, wall chargers and cables which have received recognition for their quality from a wide range of sources. In America, they hold a leading position in the USB charging market.
Because of this, many readers have been eager for me to review an Anker product to see how it really stacks up against the competition (or more likely, how its competition compares) from an unbiased technical perspective, backed up with test results. However, as it turned out, nobody was willing to donate a product to me for review and I didn’t qualify for their power user review program due to being in Australia.
Rather luckily, I met with representatives from Anker at CeBIT Australia in early May, as they were looking for distributors in the Australian region. After proving my commitment to producing high quality technical reviews, they were willing to immediately provide an Anker PowerCore+ 10050 Qualcomm Quick Charge 2.0 Power Bank (A1310) for review under the review challenge terms.
Unboxing & Features
The item comes packed inside a medium matte cardboard box which has a duotone sky-blue and black print. The front features a simple aesthetic, with the brand and model number front and center. The item sports the Qualcomm Quick Charge 2.0 logo on the front. The rear panel is used to advertise some of Anker’s other products, whereas the sides feature the approvals, model numbers and support information (including a US toll free number).
Upon opening the box, you are greeted with a thank you message on the inside lid. Neat!
Inside, the power bank is packed in a milky plastic bag in the bottom section, with a bit of cardboard isolating the rest of the included accessories.
Lifting out the cardboard, we see that even the bottom is not free of self-praising print. Included is a multilingual welcome guide, and a support details card.
Unlike many other power banks on the market, we get a sense that this product is a premium product as there is a bespoke nylon drawstring bag to store the power bank, and a charge cable with a decent length is included in the set.
The cable is part of Anker’s PowerLine series and has unique serial and model name printing on the cable. No cable thickness is specified on the cable, and while it looks a little thin, it is quite stiff for the thickness implying thicker metallic conductors.
Of course, you get the power bank itself. This particular unit is a rosy colour with sandblasted aluminium finish and etched logo. There is a power check button on the top, featuring a 10-segment LED display which is one of the most generous I have seen. The power bank weighs 238 grams.
The rear has the Qualcomm QC2.0 logo etched into it.
The power bank features one USB A output and a USB micro-B input. The specifications are printed underneath, and imply a 10050mAh/3.6v/36.18Wh capacity with 5V or 9V/2A input (QC2.0) and 5V/2.4A, 9V/2A or 12V/1.5A (QC2.0 18W) output.
The suite of features provided in Anker power banks are given in their marketing brochure. Some of the descriptions were rather interesting, which resulted in me scribbling all over it.
The first technology is PowerIQ, which helps emulate different charging protocols to ensure compatibility with picky devices. This could be quite handy, if you have particularly special devices, although some of the competitors will work with most popular devices out of the box.
The next technology is VoltageBoost, which has a very confused explanation which involves adapting the output power, and has a graph showing current. All three terms are not interchangeable! As an engineer, this really got on my nerves – what they seem to be implying is a voltage drop compensation circuit which increases the output voltage as a function of current to try and maintain a constant voltage at the device based on an assumed resistive cable loss.
Then, there is a suite of protections they term MultiProtect. They have a Qutput Current Limiter … which made me chuckle, as well as Temperature Control. They also seem to class High Energy Efficiency as a protection. I suspect, in more succinct terms, they just mean that the power bank has OCP, OTP, SCP, OVP (input and output) and OPP. The output current stabilizer feature is … well … still a mystery to me because the current really depends on the demand of the connected device, rather than the power bank.
Finally, they also include Qualcomm Quick Charge technology, which we will explore in the next section.
Qualcomm Quick Charge Technology
As this is the first power bank to come across my test rig that featured Qualcomm Quick Charge technology, it’s probably a good idea have a brief summary of how the technology works. From now on, for brevity, I will abbreviate Quick Charge as QC.
To find out more details about the technology, the easiest way is to look up a datasheet. I found the ON Semiconductor NCP4371‘s datasheet to be the most comprehensive, so the following diagrams have been reproduced from the datasheet for educational purposes. This particular IC is a QC3.0 capable device, which represents the state of the art at this time. But first, a little fun –
As it turns out, the whole Qualcomm QC system is another form of charger-and-device negotiation. Whereas in previous systems, the chargers signalled to the device about their current capabilities in various ways while sticking to 5v, the QC system allows devices to request the charger change voltages between fixed voltages of 5v, 9v, 12v (Class A) and 20v (Class B) in QC2.0 or to have an adjustable output voltage between 3.6v and 12v in 0.2v steps (Class A) or 3.6v and 20v in 0.2v steps (Class B) in QC3.0 capable devices.
This system works as illustrated in the state machine diagram to the left, and has specific timing requirements.
This system improves charging speed by improving the efficiency of power transfer as resistive losses in the cable become less significant at higher voltages as a smaller current is required to deliver the same power. It also allows for better power efficiency in some cases, by allowing the device to control the input voltage in QC3.0 to reduce internal losses (e.g. in linear based chargers).
A list of Quick Charge certified devices (phones, chargers, ICs, etc) is maintained and updated by Qualcomm on a periodic basis. If your device is on the list, then you will likely benefit from noticeably faster charging from a Quick Charge capable power bank. Of note is that the Anker Powercore+ power bank is capable of QC input as well, so charging the power bank from a QC capable charger is likely to save you time.
A teardown of the product was performed after testing had completed, and sadly, it was not without some significant damage to the casing as I had absolutely no idea how to get it open. I suppose it’s a good sign, because it shows how well-fitted the parts are. I would highly recommend others do not attempt taking their product apart, or risk potential damage.
The first plan of attack was to remove the front plate, which … was actually just hiding another plate, which when unscrewed … am I really getting anywhere? I tried to pull it out with the holes, thinking it was a form of extraction hole – but that didn’t work. I tried prying it out the top, and it didn’t budge. Maybe there’s something in the bottom?
Only after a lot of fiddling around did I manage to get the whole unit to push out from the bottom to the top. The fit was very tight and the internal guts were covered by a full protective plastic shell. I suppose this is what their internal short circuit protection is and gives additional peace of mind that the internal parts are not likely to ever short out on the aluminium casing.
As promised, the power bank features three Panasonic NCR18650B cells, which are rated at 3350mAh typical, and 3250mAh minimum at 25 degrees C, which sums up to the 10050mAh on the label. The batteries are tabbed and spot welded, with the negative connected by a thick piece of wire soldered to the tab. There is no sign of pack thermistor, as with some of the competitors products, or any obvious one-time protective fuse.
The circuitry is made of a two-board set-up with the LEDs and button on a separate board. The LEDs are surrounded by a black foam to isolate each LED and prevent light spill-over.
Internally, the design of the board appears to be dated to 23rd May 2015. The top side of the board shows a large toroidal inductor, as well as an enclosed inductor to the right (likely for charging). There is probably an IC behind the large inductor, but I did not photograph this.
Rather concerningly, the board uses three electrolytic capacitors from Fcon, which are not a highly reputable company and may have suboptimal performance. One of these capacitors was forced to be surface mounted due to the design with excessive solder and adhesive silicone, leaving solder blob traces on C20 which may later break off, rattle around and short something out. It also makes the capacitor vulnerable to mechanical shock. This is not what I expected, as electrolytic capacitors have poor frequency response and are not optimal for smoothing out ripple in high-frequency switching converters.
It is heartening to see a vast array of MOSFETs on both sides of the board for switching purposes unlike some other low-cost designs. Resistors on the rear side appear to be used for current shunt measurements of both input and output. The output is produced by an Holtek Semiconductor NTMP2014-3, a certified Qualcomm QC2.0 controller.
Testing of the power bank was performed using the standard (new) test rig which all the other power banks have been tested with. Test instrumentation include the Keithley Model 2110 5.5 digit benchtop digital multimeter for current/voltage (dual measure) discharge measurements, Keysight Technologies U1461A with USB current shunt for charging profile measurements and Picoscope 2205A USB digital oscilloscope for ripple current measurements.
As I did not have any QC2.0 or QC3.0 capable devices to test the power bank with, the QC capability was not assessed. I did, however, purchase a QC2.0 charger to assess the effectiveness of QC2.0 in reducing charge time.
Keep in mind that I am not an accredited testing laboratory – merely a hobbyist who has worked to obtain the necessary test equipment in previous reviews, and one who wants to put this equipment to good use in generating knowledge for the greater internet community in good faith. I generally stand behind my test results, but slight variations in the absolute values are possible due to accuracy/calibration limitations and unit-to-unit variances especially as only one unit has been tested.
Charging Current Profile
When the power bank is first connected to the charger, the charger consumes only a small current for a few seconds as it tries to negotiate with the charger to obtain its optimal charging voltage. When operated with a regular 5v 2A charger, it can be seen that the initial current peak slightly exceeds 2A, as the power bank tries to determine just how much current is available. The charge current then settles down into the 1.6-1.8A range as it backs-off to optimize the voltage drop (some lost in the current shunt). The charging current profile has discrete steps in the tapered section, and a sawtooth appearance during constant current. The charge terminates at about 5 hours and 47 minutes using a regular charger, at a current of about 300mA. The estimated per-cell termination current is 119mA, which is a little higher than the 65mA recommended by the datasheet. As a result, it may not extract every last drop of capacity from the cell, but will avoid overcharging.
When connected to a QC2.0 capable charger, the current is very similar (likely due to losses in the shunt) but the delivered power is increased by 80% (9v versus 5v). As a result, charging terminates in just 3 hours and 54 minutes, which saves you almost two hours compared to using a regular charger. If you’re pushed for time, this difference could well be important to you.
As it utilizes virtually all the current available from the respective chargers, the charging process is pretty much optimized, making for relatively fast charging given the capacity.
Discharge Voltage Profile
The accuracy of the output voltage as a function of current loading and time is graphed above. The blue trace representing a 500mA load shows that the voltage accuracy is astoundingly good, resulting in a very stable voltage which is pretty much spot-on at 5v, varying by less than 0.02v.
When the load is increased to 1A, the voltage falls somewhat to about 4.92v, which is still very much within the acceptable territory. Some of this may be due to resistive losses in the cabling in the test rig.
Once the load is increased further to 2A, the voltage actually increases slightly to about 4.93v, which demonstrates the VoltageBoost feature in action – it must boost the voltage as a function of the current to compensate for an assumed loss in the cable, and it actually does quite well. The result is an output voltage that varies <2%, which is a very positive result.
Discharge Capacity and Conversion Efficiency
|Load (mA)||Run||Capacity (mAh)|
|Load (mA)||Run||Capacity (mAh)|
|Load (mA)||Run||Capacity (mAh)|
All power banks are sold on their cell capacity, however, the actual amount of energy that your devices can extract from the power bank also depends on the efficiency of the conversion circuitry. For consistency, the capacities have been calculated based on a nominal cell voltage of 3.7v to be comparable to the rest of the power bank reviews to date. The capacity at 500mA was 8704mAh, at 1A was 8516mAh and at 2A is 8036mA. On a 3.6v basis (as specified in the cell datasheet), the efficiency at 500mA was 89%, at 1A was 87% and at 2A was 82%.
On a whole, the range of recorded values shows variations up to 200mAh. In most cases, about 100mAh of variation can be attributed to my instrumentation, so the additional error suggests that the low-battery termination and full-charge termination may not be as consistent as it could be, resulting in capacity variations from cycle to cycle.
The efficiency figures are not the best figures I have come across, and have been bested by others, however this may be because of the energy consumption of having a 10-LED continuously lit capacity display which consumes some energy, and the increased complexity of the QC2.0 capable chipset. In light of this, the result is still quite good.
Discharge Voltage Ripple and Noise
Ripple and noise are an often neglected output power quality parameter. Ripple and noise are basically high frequency variations in the output voltage which occur mainly due to the nature of switching converters which produce discontinuous bursts of output which need to be smoothed out. These cannot be measured with multimeters which normally measure just the average value, and require the use of an oscilloscope. High ripple voltage can cause malfunction in connected devices including interference with touch screens, stress to electrical components, audio noise or even damage to the connected device.
At 500mA loading, the ripple voltage measured 309mV peak to peak with a frequency about 74.46kHz While not likely to cause immediate damage, this level of variation is somewhat higher than the 150mV of most OEM chargers, and higher than the 50mV ATX standard (which you can expect from a computers’ USB port). From the graph, it seems to show a positive excursion of about 200mV, and a negative excursion of 100mV. With an average voltage of 5v + 0.2V ripple, the peak output voltage reaches 5.2V, just below the 5.25V absolute maximum USB voltage (5%).
The ripple voltage at 1A loading reduces slightly to 257mV, with the frequency increasing to 138.8kHz. This suggests the use of a variable frequency converter, and the change in ripple voltage amplitude likely reflects the frequency-dependent effects of any filtering (inductor/capacitor) elements. This is still somewhat above the 150mV of OEM chargers.
At 2A loading, the ripple voltage falls again to 181mV, with the frequency increasing to 325.5kHz. At this loading, the ripple voltage is now much closer to the level of OEM chargers.
Unfortunately, while the Anker was able to do well or excel in the other parameters, the ripple voltage performance is somewhat poor in comparison. Other competing power banks have been able to achieve ~100mV ripple levels across the board which is more likely to be kinder to end devices and avoid unintended problems. The reason for this may be due to cost – as Qualcomm QC2.0 increases the output voltage, any filtering elements (e.g. high frequency capable MLCC capacitors) have to be rated for the higher voltage which increases cost, and size. To compensate, it seems they have added bulk capacitance in the form of electrolytic capacitors, but these do not function well at higher frequencies, which would let down the ripple performance.
On the whole, the power bank had many attractive qualities. As an end user, I definitely appreciated the higher energy density cells which reduced the size of the power bank for the same amount of energy making it easier to travel with. While I did not have any particularly picky devices, the power bank was compatible with all the devices I threw at it – Android, Apple and even a Sony PS Vita. The unit had a solid feel, and the 10-segment power indicator was much more accurate and descriptive both during charging and discharging than the more common four-segment efforts by competitors. This reduced anxiety as to whether there was enough charge in the bank, and as to how long it would take to finish recharging.
One of the downsides was that the included nylon drawstring bag had lots of loose frays inside, so whenever I’d take out the power bank, it would be covered with black nylon scraps which needed to be brushed off and blown out of the USB port. Improving the quality of the case would be a nice touch, but I think it’s quite amazing that it is included in the bundle in the first place and will help prevent the body from being scratched.
The Anker PowerCore+ 10050 is a premium package, using more expensive, highly energy dense Panasonic cells to achieve the smallest size for the power. The exterior build quality of the power bank is exquisite, and it feels extremely solid. It even comes with a decent cable and nylon drawstring bag. It comes with the full suite of technologies, including Qualcomm QuickCharge 2.0 which will allow some devices to charge quicker from the power bank and allow the power bank to charge quicker from compatible chargers.
Performance wise, the power bank showed exceptionally good voltage stability, relatively good efficiency and good charging efficiency. Where the power bank faultered was in voltage ripple and noise, which had ripple voltage up to twice as much as OEM stock chargers which could cause potential issues. The internal build quality from a teardown shows some minor deficiencies in its design and construction quality.
It is currently listed on Anker’s website for US$29.99 (plus shipping) which is quite a bit less than I had expected based on the premium feel of the product, and would probably be worthy of your consideration if you absolutely require the best energy density, have very picky devices that won’t work with other power banks or need QC2.0 capability.
Thanks to Anker for providing this sample for review.
Bonus: Anker micro-USB Cable (3ft) (A7103)
After rummaging through the bag, I found a very nice inclusion – a micro-USB cable, so I thought I might as well unbox this in the review as well. This product comes from their Micro USB cables category, and is less flashy than the higher “Powerline” and “Powerline+” series cables.
The item is boxed in a matte finish colour thin cardboard box printed in duotone sky blue and black, consistent with their other products. The package touts very similar features of eco (and user) friendliness.
Inside, there is the same support information card, and the cable itself bundled with a reusable velcro tie which is a nice addition. The cable itself does not have any Anker-specific printing like the bundled Powerline lead with the power bank, and it does not have any indications as to its thickness. That being said, the cable had a thicker feel to it compared to the Powerline cable, although the stiffness was similar suggesting the conductors may be quite a similar gauge. This has a significant bearing when it comes to the speed and efficiency of charging your devices.
In general use, it performed as expected, performing indistinguishably compared with the OEM supplied cables when charging my tablets and phones. Indications made using the Ampere app and various “charger doctors” showed very similar values. The connectors felt sturdy, and the moulding was solid. As I don’t have a specific test rig to test cable and connector resistance, I can’t really say much more than that.