Xiaomi power banks (Xiaomi) - the most complete rating of devices. It doesn't matter day or night

I welcome everyone who stopped by. The review will focus, as you probably already guessed, on the popular Xiaomi 10000mah PB and its non-standard application, namely the “mining” of high-capacity cans for flashlights and the sale of ownerless cans from a laptop. The review will include thorough testing, detailed disassembly, efficiency measurements, as well as detailed modifications in the form of replacing batteries with cheaper/unnecessary ones. If you are interested, please see cat.

Part one. Description, testing and disassembly:

So, here is the “people's” PB Xiaomi 10000mah:

TTX:
- Manufacturer - Xaiomi
- Model – Mi Power Bank 10000mah
- Housing – silver anodized aluminum
- Battery charge indicator – 4 LEDs
- Real total battery capacity - 10000mah (3 banks of 3350mah each)
- Useful PB capacity – at discharge 1.75A (6090mah), at 1A (6420mah)
- Converter efficiency – 93% (I intended on average 90%)
- Input interface (charge) – 1 microUSB port (5V 2.0A)
- Output interface (return) – 1 USB port (5V 2.1A)
- Estimated current charging time (2A) – 5 hours (in 4 hours the PB is “compacted by 85-90%”)
- End of charge/discharge voltage of built-in banks – 4.195V/2.95V
- Simultaneous operation of charge/discharge modes – yes (independent channels)
- Built-in protection – yes (against overdischarge/overcharge/overvoltage and short circuit at input and output/overcurrent at output/overheating of electronics/incorrect connection)
- Power off button – no, the power bank is always in “sleep” mode (designed to reboot the power bank in case of problems)
- Dimensions – 91mm*60mm*22mm
- Weight - 208g

Why, exactly, did I buy this PB:
A few months ago, I noticed a comment from a comrade that the Xiaomi 10000mah PB contains not 4 typical 2.6Ah banks, but 3 high-capacity Sanyo NCR18650BF batteries, 3350mah each. Having looked at the reviews, in particular great review Comrade, I was convinced that it really contains 3 good jars, and in retail the cost of one jar fluctuates around $5-6, i.e. When purchasing you need to pay at least $15. And if you take the PB, then it comes out to around $11. Moreover, in addition we get an excellent filling with an efficiency of about 93% and a stylish aluminum box, or simply an excellent capacious power bank with reduced dimensions. Agree, this PB is a very advantageous offer.

After comparing prices, I decided to buy from the Gearbest store because:
- judging by the reviews, reviews on thematic forums and the site itself, this is 100% original;
- taking into account the frequent price reduction to $12.3 (at the beginning of October for $10.99) and the use of points, it comes out to about $10. If anyone doesn’t know, for registering and uploading an avatar on GearBest you get 30 points, which is $0.6 (50 points = $1). In addition, relatively recently, there was a promotion where you could get 30 points (shown here in the “discounts” section) and in the end you can get it at a very “tasty” price, which is what I did.

A sore point for many - bought with my own money, taking into account promotional points it came out to $11 (at that time it was the cheapest price I found):

Checking for originality:

We punch, enter the 20-digit code, which is located on the package under the protective layer:


If the PB is original, then something like this will pop up:


Well, attention to detail, in non-original PBs there is no “mi” symbol in the depth of the USB connector:


The logo is present, which, coupled with the code verification, means that the PB is 100% original.

The scope of delivery includes:
- box
- Xiaomi PB 10000mah
- charging cable, length 35cm
- instructions in Chinese

The box itself (it is slightly larger than the PB itself):




Everything fits perfectly inside (I removed the protective bag and sticker):


The front end panel is protected by a sticker; there is no sticker on the bottom, but there is some helpful information by parameters:


Hidden under the sticker are battery charge indicators, input/output ports and a reset button (2 indicators are lit, capacity from 25% to 50%):


Here's a comparison with the popular single-can cylindrical PB:


The dimensions are very small (91mm*60mm*22mm), perhaps this is the most compact PB at 10000mah. You can safely put it in your pocket, it won’t stick out much:


The weight of the PB is also small, only 208g, the case separately is 52g:


The kit contains the following instructions in Chinese:


And also a very high-quality flat cable in silicone insulation (very flexible), which can handle high currents without problems. The male USB plug contacts are gold-plated to reduce losses at high currents:


Well, very useful feature simultaneous charging/discharging, i.e. By inserting, for example, a smartphone into this power supply unit for charging, and the power supply unit itself into the adapter/power supply unit, then in the end both devices will be fully charged, very convenient:

Distinctive features of PB:
- stylish, durable anodized aluminum body (does not rust, does not fade/turn yellow in the sun, has increased mechanical strength)
- ergonomic design (all controls are located on one side)
- reduced dimensions (21.4% more compact compared to the previous model),
- large capacity 10000mah with small dimensions
- high converter efficiency 93%
- high input/output current (2A/2.1A)
- simultaneous operation of charge/discharge modes (independent channels)
- compatible with most devices (automatically adapts the charging port to the type of connected cable)
- protection against overdischarge/overcharge/overvoltage and short circuit at input and output/overcurrent at output/overheating of electronic filling

What the manufacturer tells us:

Increased capacity and efficiency (by 30%) with reduced dimensions (by 21.4%):


Reliable internal protection. 9 protective layers Kevlar manufactured by Texas Instruments:


Heap various protections(protection against overdischarge/overcharge/overvoltage and short circuit at input and output/overcurrent at output/overheating of electronic filling/incorrect connection):


Minimum losses (efficiency 93%):


Various colors (but for some reason red and gold are more expensive):

- Filled container– how much is poured into the battery, taking into account the converter (usually a little more than the actual capacity, since some energy is lost in the charging module/converter). Usually it is not of particular interest, but sometimes it is very helpful, for example, when charging a battery. It immediately becomes clear how much the battery is “packed”. Usually displayed on the screen in branded chargers. In our case, you can only look using a white tester.

Features of charging and measurements of the filled capacity:

First, let's see how much it was charged:


About 5400mAh was “filled” (8550mAh is filled into a fully discharged battery), therefore, the PB was charged by a third. But here’s what confused me, and really confused me - the charge current is only 1.33A, which is not good. After a complete discharge, I tried to charge this power bank again from a 4-port Orico DCH-4U charger and an Orico cable that easily passes 2A:


Alas, the miracle did not happen. I really hoped that in this PB the 2A current charge was normally implemented, so I had already finally decided that I would redo it. But why do some users claim that PB can take 2A? The Orico DCH-4U charger delivers 2A without any problems, the problem is clearly not there:


Let's try it with the supplied cable. It can be seen that it is worse than Orico, there is a slight drawdown and, as a result, a lower current, only 1.26A. It’s a long way from the stated 2A:


Having disassembled the guts, we see that without “charging doctors” a little more than 2A goes to the banks. In the photo, the ammeter (on the left) is connected between the “plus” of the board and the “plus” of the battery. Everything is fair, as the manufacturer claims and even a little more (2.2A):


But once you install the “charging doctor”, the current begins to be limited to around 1.3-1.4A:


The situation with the white doctor is similar:


There may be some restrictions, because... the manufacturer claims a “smart” controller that adapts to different types of devices. But at first I thought that either my copy was defective or the manufacturer was being disingenuous. It turns out that no, but still some users post photos where 1.6A is poured through a white tester.

We've sorted out the filled container, now let's look at the “useful” container, i.e. given after the converter.
- Useful capacity– how much we will actually get. It is calculated after the converter. The PB usually contains a boost/buck converter, which is far from 100% efficient, so by raising the voltage from 3.7-4.2V to 5V, it loses some energy (heating, inefficient components, etc.). Buck converters have the highest efficiency, but are more expensive. Therefore, if a manufacturer claims the capacity of their PB is 10Ah, this does not mean that we will receive them all. Approximate formula for calculating useful capacity:
usable capacity = actual capacity * 3.6 / 5 * efficiency

PB discharge current 0.95A (the result is quite expected, 6420mah):


PB discharge current 1.75A (6090mah):


Let's summarize the output parameters:

Well, very good. The manufacturer claims an efficiency of 93%, let's see if this is true...

Disassembly of PB Xiaomi 10000mah:

All stages of disassembly in detail

This PB is quite easy to disassemble. To begin, use a thin watch screwdriver to remove the plastic covers from the ends of the device. They are held on with double-sided tape, so we tear them off little by little so as not to break them and spoil the stylish appearance (it’s better to pick them out from the inside of the ports first):




The other strip is picked out in the same way, do not remove the tape, it will still come in handy:


There are 4 screws at each end. First, unscrew the bottom, then use a flat object to “glue” it to the end plate and take it out (I stuck a flat screwdriver):


It turns out this picture:


Next, unscrew the other bar and push the entire frame outward from the negative terminals of the batteries:


It should look something like this:




The body is aluminum with stiffening ribs, and screws are screwed into them:


You can get the reset button, it is held on by latches:


To remove the board, you need to unscrew 2 more screws on the board and bend these latches:




That's it, the complete disassembly is complete:


Close-up of the board:


As you can see, there is a temperature sensor, although only one:


The board works well even without a case. During operation, it does not heat up much (the temperature on the side of the microUSB connector is 55°C when discharging with a current of 1.75A, at the end of the discharge it rises to 70°C, since the currents in the converter are about 4A, but in a closed case there will be all 80° WITH):


Reassemble in the reverse order, remove the remnants of old double-sided tape from the plastic end caps with mild solvents, such as alcohol. Acetone dissolves black strips!!! See below for disconnecting the cans, soldering new ones and assembling the PB.

Inside there are 3 high-capacity Sanyo/Panasonic cans with a capacity of 3350mah each:


Having disconnected them, we get 3 good high-capacity jars; if necessary, we clean the residues from the contacts spot welding(sandpaper/knife helps):


Can production date:
I already wrote about decoding some folk cans in my article about and testing, here is the same standard marking of panos.
Our banks have the following designations:
SANYO L
NCR18650BF 5422

NCR18650BF – model name
- the first digit is the year of manufacture: 1 – 2011, 2 – 2012, 3 – 2013, 4 – 2014, 5 – 2015, etc.
- the next digit is the month of issue: 1 – January, 2 – February, 9 – September…… X – October, Y – November, Z – December
- the third and fourth digits are the day of the month: 05, 14, 29.
Total: the production date of our Sanyo NCR18650BF 5422 cans is April 22, 2015, you see, they are quite recent.

Battery Specifications

Here is the datasheet for this battery model:


Here is a comparison of cans with their closest competitors in a comparator at a discharge current of 1A/2A/5A:






Discharge with typical current 2A to 3V:
Sanyo/Panasonic NCR18650BF 3350mah -> 2979mah (10,497Wh)
Sanyo/Panasonic NCR18650B 3400mah (basic of all varieties) -> 3104mah (10,821Wh)
Sanyo/Panasonic NCR18650BL 3350mah -> 2950mah (10,430Wh)
Lg INR18650MH1 3350mah (high current) -> 3014mah (10,685Wh)

Final table:


We run the banks on Opus BT-C3100 V2.2 with a current of 1A:


On average 3400mah per jar, not bad at all (one jar seemed to be constantly heated in the PB, even the adhesive tape got stuck, so only 3300mah). At the same time, the cost of one battery is around $4, plus excellent electronic components/housing. In retail average price these cans are $5-6, and stale ones can arrive.

Converter efficiency measurements:

A few words about the coefficient useful action(efficiency). I don’t think there’s any need to explain what this is. For those who have forgotten, roughly speaking, this is an indicator of the effectiveness of the device. The manufacturer claims the efficiency of this device is around 93%. This means that 7% of the energy is lost in the converter/heat. Let's see if this is true. The calculations are not very accurate, because... The devices are not calibrated, but they display the general picture. For the readings of the white tester, an addition of +0.05V is made, which is extinguished in the “tail” of the tester. From the review - “The indicator slightly underestimates the displayed voltage (by 2%) and current (by 3%)”

The battery is fully charged, load discharge current is 0.95A (voltage about 5.1V):


We have:


- voltage at input/bank – 4.1V
- current at the input/banks – 1.3A
Power P1=5.1*0.95=4.85W
Power P2=4.1*1.3=5.33W
Efficiency = P1/ P2 = 0.91, that is, 91% with a return of 0.95A and charged banks, not bad.

The battery is fully charged, load discharge current is 1.7A (voltage about 5.05V):




- input/bank voltage – 4.06V
- current at the input/banks – 2.4A
Power P1=5.05*1.7=8.6W
Power P2=4.06*2.4=9.74W
Efficiency = P1/ P2 = 0.88, that is, 88% with a return of 1.7A and charged banks, not bad.

The battery is half charged, the load is discharged with a current of 0.95A (voltage about 5.1V):


- output/tester voltage – 5.1V
- output/tester current – ​​0.95A
- voltage at input/bank – 3.72V
- current at the input/banks – 1.42A
Power P1=5.1*0.95=4.85W
Power P2=3.72*1.42=5.3W
Efficiency = P1/ P2 = 0.91, that is, 91%, not bad.

The battery is half charged, the load is discharged with a current of 1.7A (voltage about 5.05V):


- output/tester voltage – 5.05V
- output/tester current – ​​1.7A
- input/bank voltage – 3.7V
- current at the input/banks – 2.7A
Power P1=5.05*1.7=8.6W
Power P2=3.7*2.7=10W
Efficiency = P1/ P2 = 0.86, that is, 86% with an output of 1.7A, decreased a little, but still excellent.

The battery is almost discharged, load discharge current is 0.95A (voltage about 5.1V):


- output/tester voltage – 5.1V
- output/tester current – ​​0.95A

- current at the input/banks – 1.77A
Power P1=5.1*0.95=4.85W
Power P2=3.08*1.77=5.45W
Efficiency = P1/ P2 = 0.89, that is, 89% with a return of 0.95A and discharged banks, not bad.

The battery is almost discharged, load discharge current is 1.7A (voltage about 5.02V):


- output/tester voltage – 5.02V
- output/tester current – ​​1.7A
- voltage at input/bank – 3.08V
- current at the input/banks – 3.43A
Power P1=5.02*1.7=8.53W
Power P2=3.08*3.43=10.56W
Efficiency = P1/ P2 = 0.81, that is, 81% with a return of 1.7A and discharged banks, more or less.

The efficiency of the converter is on average about 90%, but if you take into account inaccurate instrument readings and losses in the measuring wires, then the converter maintains the declared 93%. This applies to energy, but practically nothing to capacity.

Now for capacity:
Since the discharge of the cans goes up to 2.95V, and they give full capacity when discharged to 2.5V (see the “batteries” section), the result is about 9300mah (3100mah per jar). Standard formula for calculating useful capacity:
Useful capacity = actual capacity * 3.6 / 5 * efficiency
We get - 9300*3.6/5*0.9=6030, we got a total of 6420mah at a discharge of 0.95A, and 6090mah at 1.7A.

Atypical uses of PB:
- “mining” cheap high-capacity Sanyo NCR18650BF 3350mah batteries at a low cost and using them in various devices ah (the market value of one can is $5-6);
- use the converter board in your DIY projects. It can simultaneously charge/discharge, has excellent performance characteristics and all kinds of protection, including against overheating, which is sometimes very important;
- nice stylish aluminum case for various crafts;
- assembling your “monster” on a large capacity (when using Li-Pol batteries) or on another electronic stuffing;
- in addition to everything, an excellent stylish 2A cable of small size (high-quality cables of this length are difficult to find, branded cables come one meter each, but using them in a purse/bag is very inconvenient)

So, let's look at the advantages and disadvantages of this PB:

Pros:
+ brand, quality guarantee (this is not a name and not a “green” company, has been operating since 2010)
+ “honest” efficiency of the DC converter of 93% (minimum energy losses in the converter)
+ high output current of 2.1A (the voltage practically does not sags)
+ high charging current of 2A, and this reduces the charging time by 2 times compared to the typical current (some competitors still have 1A for 4 banks, and this is 0.25A for a 2.6Ah bank, quiet oooooooooo)
+ simultaneous operation of charge/discharge modes (independent channels)
+ stabilized output parameters (the output voltage does not “jump” as the PB discharges)
+ the presence of all kinds of protections (against overdischarge/overcharge/overvoltage and short circuit at the input and output/excess current at the output/overheating of the electronic filling/incorrect connection)
+ stylish appearance
+ the ability to disassemble without damaging the appearance (for example, Pineng PB and some others are tightly glued together)
+ reduced dimensions and weight with the same high capacity (almost all 10Ah PBs have 4 banks, including Xiaomi 10400, instead of 3 banks)
+ low price (analogs from other companies cost about $18-20)
+ high-quality complete cable ( low drawdown under load, small size, just for PB)
+ availability of additional accessories (cases/bumpers, for example)
+ light weight and dimensions
+ steel case (a kind of mechanical protection, since many companies prefer cheap, poor plastic, which doesn’t look very good)

Cons (in general, these are not even disadvantages, but just whims):
- a primitive capacity indicator in the form of 4 LED indicators (I love devices with screens, everything is shown there clearly, for me this is a minus, for others it is not)
- one USB output port (the problem can be solved, but who carries and simultaneously uses two power-hungry devices)

Conclusion: if you are looking for a small, capacious power bank with excellent performance characteristics, this is definitely your choice. Coupled with the above advantages, I believe that this PB can rightfully be called “national”. I recommend to buy!

Part two. Replacing built-in batteries with less capacious/unnecessary ones

Reasons for modernization:

We needed high-capacity 4.2V cans for flashlights at a low cost
- I wanted to attach batteries from a laptop battery (5 good cans and 1 not very much at 2Ah)
Once upon a time, my laptop battery died. Having disassembled it, I found 6 LG LR1865SE 2000mah cans, one of which died (it had an increased self-discharge, discharged to 2V after a couple of days of sitting in a discharged state). At first, I used them in the black 4-can PB AiLi169, which appeared in my first review of . But it had 5 huge disadvantages: large dimensions, heavy weight, high quiescent current (self-discharge), artisanal appearance and a charge current of only 0.5A, so it took 24 hours to charge it, and in the end the capacity was relatively small. Therefore, I almost never took it with me and collected it, which I often used. But the charge current there was also small, only 0.5A, and the “useful” capacity left much to be desired. At first, I wanted to attach three Sanyo 2.6Ah popular cans to this Xiaomi PB, but after thinking about it, I decided to abandon this idea, because... all the trump cards of the Sanyo 2.6Ah will not manifest themselves in any way (good discharge curve + frost resistance), but the popular Samson 2.6Ah were not at hand, so the choice fell on the 2Ah Skis from the laptop. It was decided to assemble my own PB by simply rearranging 3 more or less successful banks into Xiaomi. On my own behalf, I can add the following: if we put 2000mah banks instead of 3350mah, we will get 6000mah instead of 10000mah (9300mah, since the complete ones give up the entire capacity when discharged to 2.5V), and in terms of “clean” capacity we will get about 4190mah instead of 6420mah , which is more than enough for most consumers. Formula for calculating useful capacity = actual capacity * 3.6 / 5 * efficiency

Here are our players (I recommend installing the ultra-cheap Samsung ICR18650FM 2600mah). In my case, these are LG LR1865SE 2000mah batteries from a faulty one battery laptops that were waiting in the wings in this PB AiLi169:


First, let's check the battery capacity on the Opus BT-C3100 V2.2 analyzer:


Let's take 1, 2 and 4 jars, because Their capacities are almost the same. To be on the safe side, I recommend following the following algorithm:
- discharge the original batteries (Sanyo NCR18650BF) in the Xiaomi PB
- discharge the “donor” batteries (LG, Samsung) to the same minimum potential, it is more convenient to do this in some multi-bank PB
- first turn off “plus”, then “minus”
- connect new banks in the reverse order, i.e. first "minus" and then "plus"

So, here we go (direct modification):

Just in case, we isolate the positive terminals of the batteries with electrical tape, you never know:


Next, we bend the board and use a thin blade of a knife/clock screwdriver and carefully, without poking the tip into the board, pry out the positive current lead from the batteries ( There is no need to tear off the “plus” board, I just measured the efficiency):


It’s the same with negative ones, you don’t need to rip them off the board either:




We get something like this:




Since I don’t have spot welding, we will solder the current leads to the new cans. To do this, first align the current leads with pliers. The soldering algorithm is as follows: we drop a little flux (soldering/orthophosphoric acid) onto the contacts and tin (cover with tin/sold), the main thing is not to keep the soldering iron tip on the contacts for a long time. If something doesn't solder, don't try to solder it the first time. Cool the jar contacts and try again. If all contacts (plus and minus) are tinned, then solder the minus first, then the plus.
Straightening current leads and tinning them on both sides:


We will solder these cans (1, 2 and 4):


We drip flux, pick up solder with a soldering iron tip and press it to the contacts ( don't overheat!)


The current leads are soldered tightly:


You can insulate the terminals with electrical tape:


Lastly we solder the “plus”:


To reduce losses, I recommend bridging the negative current lead with an MGTF wire:


We clean all contacts from acid with alcohol/acetone. If nothing works, briefly turn on the power bank to charge, because The controller is blocked, everything should work:


Next, insert the board into the frame and tighten it with two screws. We secure the negative current lead with the supplied tape, and also fix the temperature sensor:


We insert all this equipment into aluminum frame and tighten with screws. Take this double-sided tape or similar:


We cut the strips to the size of the ends of the PB and glue them on:


We tear off a little protective film tape and use a sharp blade to cut holes for the ports. You can completely tear off and cut out the tape itself, but the edge of the knife will stick and the result will be worse:


We glue the strips, it turns out like this:


It’s quite easy to glue the other strip; there are no ports there.

Measurement of the “useful capacity” of the modified Xiaomi PB:
The discharge current is 0.95A, the output voltage is also 5.04V (“useful” capacity 4190mah):


The discharge current is 1.7A, the output voltage is also 4.98V (“useful” capacity 4085mah):


As you can see, it’s a little smaller, but the difference in capacity between banks is significant (2000mah vs 3350mah). Well, the test of the “flooded” container:


Let’s summarize the performance characteristics of “skeletal” VS “modified”:


That's all, it's up to you to decide whether it's worth redoing. Personally, one single-can cylindrical PB was enough for me, so I didn’t notice any difference from reducing the “useful” capacity. If you live, for example, not in the city, then perhaps this alteration will be unnecessary. At least, my main task when purchasing this PB was completed, namely “mining” high-capacity cans...