Low Voltage UPS Board with Boost Converter and Charge Controller

Thread Starter


Joined Jan 27, 2019
This board is a cheap and cheerful way to get UPS functionality for small projects,
and you get a charger and a boost converter thrown in with the bargain.
It is not all roses, though, and a small problem might be a big one for some.
In the final analysis, I found the little extra work to make it operate as
advertised worth the trouble. If the functionality sounds good to you, read on...

Prompted by another thread here, I decided to investigate these little boards because they are cheap and have an interesting combination of functions. It seemed pretty useful to have that functionality in a small (23x33mm) package for low powered projects that need battery power (e.g.: MCU devices, particularly wirelessly connected things that need to keep running if the local power goes out) or to convert low power devices to lithium chemistry operation.

IC Complement
The board uses the TP4056 charger IC, which can deliver 1000mA charge current (and it is programmed to do it on this board), features thermal protection (not implemented on this board), and uses a constant voltage / constant current charge regime at 4.2V. I tested it with both LiPo and Li-Ion cells, and it worked well cutting off the charge at just under 4.2V battery voltage.

The output side uses the SDB628 boost converter IC. It operates at 1.2MHz allowing it to have the very small inductor you see on the board. It claims a capacity of 2A, though I didn't test it beyond about 700mA which it coped with admirably. However, the board got quite warm and I don't think I would try to run it at 2A for very long without some cooling. The underside of the board is an almost unbroken copper flood, used as the ground plane. I think this aids heat dissipation.

Testing and "Repair"
The board's construction is reasonably good. I would say above average for this class of module. The components appear to be of good quality, the soldering is clean, and the PCB seems to be FR4 or similar. Nothing really negative is present.

The SDB628's output is programmed with a voltage divider on pin 3 (FB). On this board, R2 of the divider is a variable resistor to allow user adjustment. This turned out to be the only problem I experienced with it. When I received the board, the first thing I did was to apply power and check the output voltage which was ~5.7V. I wanted 5V so I attempted to adjust it with (V)R2, but to my dismay I could go down only about a tenth of a volt, the rest was all up! The board is sold for use at 5V, so I thought maybe I had a bad one. I tried another and it was the same.

I tracked down the datasheet and found out about the voltage divider. In the image above you can see the position of the fixed R1 labeled \[ \text R_{PROG} \]. I measured the value and it was 100K. The formula in the datasheet indicated I needed a lower value to get a lower minimum voltage. Conveniently, there are a pair of through holes connected to the resistor's pads. This allowed me to put a ¼W 100KΩ resistor in parallel easily.

This worked. I was able to adjust to 5V. The VR is not the cleanest for travel, so it can be a bit tricky if you are picky, but I managed to set it to 5.0051V which is good enough for me. I also replaced the SMD resistor with a 50KΩ on another board, also successfully.

I tested the board with a 12Ω load drawing about 675mA from a LiPo battery until the battery stopped due to under voltage protection. Note that the board protection shut it down, and it would not restart until I applied about 6V. It did restart though, without problems. During the discharge I applied power at various voltages to the input terminals and the battery started charging, then returned to powering when I disconnected . I saw no fluctuation in the output voltage. For the record, at one point I measured a battery current of about 500mA (at 3.7V) for an output current of 350mA (at 5V).

Once the battery was flat, I left it to charge. It did, at a peak of the rated 1A. The charge curve looked right. Eventually, it hit 4.2V and the charging stopped. There are two LED indicators, red and green. For charging it is red while charging and green when done. They also have some diagnostic uses detailed in the datasheet.

I repeated the same test with an 18640 cell with similar results.

All in all, I think these boards are a good deal considering price and capabilities. The UPS function works very well, and the construction is good quality. The problem with the programming resistor could be a show stopper for some, but it wasn't such a big deal to me. If you don't want to play around with the tiny SMD resistor a parallel through hole does work. I think I will find a lot of uses for this power within the voltage and current range it offers.