Review, Teardown: Lipstick Battery LED 3000mAh Silver Power Bank

The power bank testing and reviewing continues, courtesy of the anonymous donor. Testing and reviewing is a fairly involved process, easily taking up two weeks to complete. It is hoped that by testing these “anonymously” supplied power banks, we would be able to illustrate the pitfalls of commonly available low-cost power banks, and illustrate their merits and deficiencies.

In this post, we will be looking at the Lipstick Battery LED, which claims to be a 3000mAh portable power bank with a high powered LED torch, in the shape of a cylinder. It’s probably a bit bigger than a lipstick, but what would I know?

DSC_6612The unit itself comes in a clear plastic box packaging, with the unit itself a silver aluminium colour. There is a misspelling of the word ‘compatible’ on the front, and the capacity is indicated by a reflective, iridescent sticker on the front. No manufacturer is listed on the packaging itself, and it is likely this product is supplied for re-branding by others.

DSC_6613The rear of the box is no more helpful in regards to the manufacturer. However, that being said, the rear specification table seems to suggest that the unit also comes with 2,200mAh and 2,600mAh capacities. It is likely to contradict the label on the front due to specification changes by the supplier as they pick and choose components to build their power banks with based on demand.

The input is listed as only 500mA, which means that the charging rate is fairly low but compliant with regular USB. This may mean a longer charging time than otherwise practical, as rates of up to 2A on microUSB are commonly used by tablets. The output rate is up to 1000mA, so it should be suitable for most phones but not as ideal for tablets which tend to charge best at 2000mA.


The power bank itself is a little larger than an 18650 cell, and it is provided with a short basic microUSB lead that is only wired for charging. No operation manual appears to have been supplied.


At the top, a single USB A female port is provided for plugging in your charging lead – whatever that may be. The microUSB B female port below is used for charging, and the clear button above allows you to control the power bank’s operation. The status is indicated by LEDs which shine through the clear button.


The other end of the tube has the LED and lens assembly, for the torch. It’s a fairly strong LED beam, while slightly blue (possibly overdriven), it seems to compare well with my 2W CREE LED torch. It’s not a wimpy 5mm LED that’s next to useless, that’s for sure.

Operating the device is rather interesting. Short pressing the button once causes the LED to light up and indicate the battery status, and activates the power bank for usage. The LED shines green during charging for high charge, blue for medium, red for low and blinks red when critically low. During charging, the LED blinks red when lowly charged, blue when mediumly charged and green for nearly completely charged. It is lit solid green to indicate fully charged. It seems that if left idle without load on the USB connector, the power bank turns off automatically.

To turn on the LED torch, you must long press on the top power button, which will illuminate the LED. There is several modes available, which can be toggled by double-clicking the switch while the LED is on – you can toggle between full brightness, PWM “half brightness”, slow flash and fast flash. To turn the LED torch off, a long press is required.


To open this power bank, all that’s required is to unscrew one of the threaded end-caps. Unscrewing the top one allows for the plastic fascia to be removed, and the whole assembly to be slid out of the tube.


The power management and conversion PCB is at the top with the ports. It is constructed by having two joined double-sided PCBs, in order to maximise the use of the space. It appears that the whole thing was covered by blue heatshrink which might have been torn apart by the anonymous donor before I got my hands on it – yours might look different for this reason!

The heatshrinking is necessary, as the aluminium tube itself is conductive, and thus a potential for short circuiting. It seems that tabs are used to connect the battery as well as the LED at the other end of the tube, which is why you might see more of them than you expected!


The top PCB is marked with HCX-H100BV2, dated 2012-6-26. From this, I have determined through a Google search that the manufacturer is likely to be BesTech Power of Shenzhen, China. The full specifications are available at the link, and it seems to claim to have overcharge and overdischarge protection and facilities to accomodate a 1W LED. Unfortunately, as they are only the supplier of the PCBA itself, someone else may have done the assembly of the power bank.

The top of the PCB has the SS34 3A Schottky freewheeling diode, and a few assorted passives.


Sandwiched inbetween the two PCBs, we can see a very short USB A connector, an open inductor, and an electrolytic capacitor for smoothing the output. Due to space constraints, there might be some tradeoffs with the values employed.


The bottom of the bottom PCB (marked HCX-H100CV2) has the button, indicator LED and a Holtek HT46R064B one-time-programmable 8-bit microcontroller. I would hazard a guess that the majority of the power banks on the market from China are based around a similar chip, but instead, have the markings ground off. There was another power bank I had with the same uC, although the software that runs on it can be equally important for its performance.


The bottom of the assembly has a lens which is made of plastic that removes off the LED, and a PCB with the LED mounted on it. It’s likely the LED is of Chinese origin.


The cell itself is marked with ICR18650-30B, from Samsung SDI. It appears to be a genuine cell at a glance, and feels of an appropriate weight. The specifications claim it to be a 3000mAh cell (consistent with the sticker), with a discharge cut-off voltage of 2.75v.

As with most other power banks, it seems there are no secondary protections (e.g. failsafe fuses in case of uC failure). The power bank was re-insulated with tape before reassembly, to ensure safety.

Performance Testing

Testing was performed on the new rig consistent with how the other power banks have been tested so far. The capacity results are as follows:

Load (mA) Run Capacity (mAh)
500 1 2120.018343
500 2 2146.134779
500 3 2167.477568
500 4 2163.364908
500 5 2180.28722
Mean 2155.456563
Range 60.26887696
StDev 23.27214587
Load (mA) Run Capacity (mAh)
1000 1 2013.519713
1000 2 1972.488427
1000 3 1939.842499
1000 4 1965.25055
1000 5 1964.322036
Mean 1971.084645
Range 73.67721434
StDev 26.73268995

It can be seen that at 500mA load, the mean effective capacity is 2155mAh, and at 1A load, the effective capacity is 1971mAh. Knowing the cell capacity should be 3000mAh, this translates to an efficiency of 71.8% and 65.7% respectively. This is a bit on the low side, but part of the reason is that this power bank lights up its LED continuously during the discharge process which consumes energy. The other part may be due to poorly made inductors (high losses) and other compromises to fit it within the form factor.

It’s curious, as the PCBA manufacturer claims ~85% efficiency. If this were true, the Samsung cell would have a capacity of 2535.8mAh at 500mA or 2318.9mAh at 1A load which would suggest the cell is a 2600mAh rather than a 3000mAh cell. Which one of the two hypotheses are true, I cannot easily tell without actually testing the batteries in isolation, but I don’t see that as being an efficient use of my time, so I’ll leave it at that.

Testing this particular power bank posed additional challenges as it seems the over-discharge protection was not functional or well matched to the cell. As a result, manual intervention was required to terminate each and every test, thus meaning I had to baby it all the way!


This is clear from the graph above which shows the power bank output voltage vs sample number (i.e. time). Once the cell is depleted, the voltage falls sharply to the battery terminal voltage of about 3v at end of discharge, but no MOSFET isolates the battery from the load. Most USB loads will recognize the low voltage and not be able to draw from the bank, however, dumb loads may become confused or continue drawing from the power bank. In that case, the voltage of the cell can fall to damaging levels.

The datasheet for the Samsung cells state that 2.75v is the maximum discharge voltage, however, the BesTec specifications claim that their over-discharge protection doesn’t kick in until 2.40v +/- 0.1v. I got a bit nervous when I did a run to deliberately see if it was functioning, I saw it reach 2.40v and it was still going, so I disconnected the load manually to avoid permanent damage to the cells.

If there is an over-discharge protection, it should be matched with the cells. If the maximum discharge voltage is 2.75v, make sure that the protection activates well in advance of this!


A cropped graph does show the voltage regulation is pretty darn excellent, with the voltages very close to 5v and consistent throughout most of the discharge period.


However, when we take a look at the ripple figures, it’s not as rosy. Again, it seems the form factor is a penalty here, with a ripple average peak-to-peak voltage of 288.1mV at 500mA. The frequency of oscillation is a relatively low 76kHz.


It increases to 320mV at 1A loads, with an oscillation frequency of about 101kHz. Considering that wall chargers are generally 150mV or less, this is a little high, although probably not immediately critically dangerous (USB specifications state voltage between 4.75v to 5.25v which is a 500mV peak-to-peak window around 5v). I would prefer it to be closer to 150mV though.


This particular style of power bank is becoming popular with users due to its small size and satisfactory capacity. It’s definitely relatively small, and some have been engineered in ways to have interchangeable 18650 cells as well (although the PCB may be different). Regardless, the performance is a little lackluster when it comes to conversion efficiency and ripple, which is unfortunate, as it is very much likeable. While it’s not critically dangerous for your gear, it’s definitely not quite as clean as the power that comes from a quality wall charger.

However, it performs very well as a torch – it’s not like the useless 5mm LEDs on many other power bank products, and I’m quite happy to use it as a rechargeable torch alone. It could definitely be handy in a bag, say for lighting your way at night or when fumbling for keys. It’s a shame the power bank segment didn’t perform a little better.

About lui_gough

I'm a bit of a nut for electronics, computing, photography, radio, satellite and other technical hobbies. Click for more about me!
This entry was posted in Electronics, Power Bank and tagged , , . Bookmark the permalink.

Error: Comment is Missing!