If you haven’t already guessed it, testing power banks gets a little tiring after a while. But it doesn’t matter, because I might as well do it while I still have some motivation to keep going – this one’s thanks to an anonymous donor.
This one comes in a hard plastic shell, and is packaged without a brand. The product, however, says Power Bank Power on the go!, so we’ll use that to refer to the product from now on. The box itself has a 6000mAh sticker on it, so I’ll assume that’s the true intended capacity. As usual, the blurb is pretty generic with a spelling mistake scattered in, as per usual.
The power bank comes with a user manual leaflet, and a charge only micro USB B cable. The power bank is mostly glossy black, with a mirrored front insert and a chromed-band around the main body. There are semi-transparent clearings in the mirrored finish that let four blue LEDs shine through to indicate charge status. These LEDs continue to indicate during both discharge and recharge.
The rear of the power bank lists the capacity as 5200mAh (probably the regular capacity of the model when initially launched), and an input current of 1A, and an output current of 1A. As with most power banks of a similar appearance, this is a single port unit most suited for smartphones and less suitable for hefty devices such as tablets.
Functionally, the device also includes a 5mm LED emergency torch, which isn’t very bright, but can be toggled into SOS mode as well through long-presses of the power button.
Despite appearing similar to some power banks already reviewed, this power bank is built pretty tough. It didn’t want to come apart without utter and total destruction.
The reason is obvious – the batteries have been covered with liquid nails adhesive and that stuff sticks pretty damn well. Already visible are the two green Samsung ICR18650-30B cells, which appear to be genuine and form the basis of the power bank. The main PCB is visible, containing a Holtek microcontroller (as most power banks do), and a dedicated charge controller chip. The LEDs are housed on an auxiliary board which nestles in-between the two 18650 Li-Ion cells.
The heatshrink of the cells had cracked under the force of attempting to separate the two halves of the shell of this power bank. Visible, just barely, through the heatshrink are the codes printed on the cell canister themselves (near the top positive terminal of the cell).
A closer look at the top PCB shows that the positive battery connection is made by tab, soldered directly to the PCB. There is adhesive green cardboard insulation on the cell tops. The LEDs are on an auxiliary PCB which is soldered to the main one. The main PCB is dated 24th August 2012.
Looking at it from another angle reveals the code TP580, which, from a quick Google, seems to have some relation with Shenzhen Scud Electronic Co. Ltd although I am uncertain about this.
More prodding and prying was required to get the cells to release from the rear shell. In fact, that’s why the bottom negative terminals of the battery got slightly scratched. They were insulated with regular plain masking tape.
A copious amount of liquid nails was used to adhere the batteries. On the plus side, this means the power bank doesn’t have any form of rattling or noise when shaken, but it does make for difficult repair and additional weight.
A closer look at the main PCB …
… reveals a rather interesting strategy of partially heat-shrinking the LED to try and improve the output – but with dark coloured heatshrink. I don’t get it.
The PCB is marked with HCX-H107A, which is a code likely belonging to BesTec Power, a manufacturer of such PCBs. However, this product was not found on their site, although it bears a striking resemblance to the HCX-H085 down to the layout and type of inductor and capacitor employed. PCBs from BesTec were seen in the Lipstick LED review.
Speaking of which, this unit uses a single-wound open inductor, which is not very desirable due to higher losses at high frequencies (i.e. lower efficiency). The regular 8205 MOSFETs make an appearance, as does the SS34 Schottky 3A rectifier diode.
This power bank differs from the others in employing an electrolytic capacitor for bulk power storage. I hope this improves the quality of the power output, however, it doesn’t seem like this is a large or quality part, being branded United and bent-over tack-soldered onto the board.
Just as a check, the tops of the batteries are verified to have white caps, and consistent shapes.
The negative terminals are cleanly done too, except for my screwdriver-gouging attempts to get this thing apart. Mind you, you should not pry too hard as deforming the batteries can result in venting with flame (aka explosion).
As usual, the new rig was used and a similar methodology to previous reviews was employed. The capacity results are as follows:
|Load (mA)||Run||Capacity (mAh)|
|Load (mA)||Run||Capacity (mAh)|
At 500mA, the effective capacity was 4600mAh. At 1A, the effective capacity was 4274mAh. Knowing that the capacity of the cells is 3000mAh each for a total of 6000mAh, the conversion efficiency is 76.7% and 71.2% at 500mA and 1A loads respectively. This is not a particularly high efficiency.
The charge termination reliability is excellent with relatively small and consistent capacity ranges reported throughout the five runs at each load.
Through running the discharge test, I wasn’t too careful and didn’t monitor the discharge too carefully, and thus accidentally found out that this power bank has an overdischarge protection but the threshold is set way too low.
The over-discharge protection doesn’t actually disconnect the cells from the output until the cell voltage falls to 2v under load. Generally the cells should be disconnected closer to 2.75v to avoid permanent cell damage and formation of copper shunts within the cell that can jeopardize future safety. As a result, this power bank is deemed not to have an effective end of discharge cut-off, although smart loads often will see the voltage from from 5v to 3v and refuse to draw any further power from the power bank.
Ignoring this, the power bank conversion circuitry had excellent regulation of the average voltage through both 500mA and 1A loads, hugging the 5v line fairly closely (considering resistive contributions to measurement error), and remaining stable.
The ripple measurements are as follows:
At 500mA, the ripple averaged 354.6mV peak to peak, which is higher than most stock chargers, but would still lie within the absolute limits of the USB specification (as the output will be between 4.75v to 5.25v). The frequency of oscillation is a relatively low 312.9kHz.
At 1A, the ripple intensifies to 627.2mV peak to peak which lies outside the USB specification. It’s not likely to be immediately fatal to devices and is not the worst experienced, but may cause stress to your devices.
Alas, it seems the electrolytic capacitor might not be big enough and may not be able to react fast enough to the fast-rise switching noise. Maybe increasing the capacitance with a quality, low ESR capacitor paralleled with a ceramic/polyester capacitor would improve this dramatically.
Despite having a set of genuine high capacity cells, I feel that the conversion efficiency could be quite a bit better (part of this may be due to the LEDs remaining on during discharge). Likewise, while the regulation is good, the ripple and lack of properly functioning over-discharge protection is concerning for the longevity of the power bank as a whole. The construction is physically solid, although difficult to repair.