Hey all,
I have a 3.7v lipo battery connected to a custom PCB and a bluetooth module and I want to be able to track the battery charge and transmit the info to the user's phone. Is there a way to easily do this? The hardest part is the space confines, as the whole project has to fit into a cylinder of diameter 22mm. If anyone could help, I would really appreciate it. I've attached a photo of my current schematic for reference.
Thank you everyone,
Max

 

You need a coulomb counter.
In other words, you measure the battery current and integrate over time. That gives the charge that gone into or out of the battery. If you know the capacity and start counting at a known level (usually fully charged) then you know the state of charge of the battery.
There are ICs that will do it for you.
https://www.mouser.co.uk/c/semicond...y-management/?q=gas gauge&product=Fuel Gauges

 

Analog devices has over 30 ICs to chose from with four different interfaces. Prices start at $1.18 U.S.D.

https://www.analog.com/en/parametricsearch/12979#/p5380=1 Cell Li-Ion|1-2 Cell Li-Ion

 

To expand (very slightly) on @Ian0's comment...

It is notoriously difficult to know the State of Charge (SoC) of a lithium chemistry cell/battery by measuring anything about its current state. The cell voltage is a completely unreliable way to determine the actual SoC—open terminal or under load. For this reason the Coulomb Counter was developed.

A Coulomb Counter does literally that: it watches the number of electrons that pass through it and either adds to the SoC or subtracts, depending on the direction to the current. This means it has to characterize the cell/battery by first seeing a complete discharge from a full charge state.

The currently available CC ICs are very mature and sophisticated—they work very well. They can be quite small but they are the only game in town so you'll have to make one fit if you need the functionality.

 

My experience is that in real life they work well enough but not quite as well as people think they do.
Because they give an output percentage accurate to three significant figures, people believe the figure they are given, regardless of how accurate it might or might not be, and there is very little you can do to prove whether it is or isn't correct.

 

My experience is that in real life they work well enough but not quite as well as people think they do.
Because they give an output percentage accurate to three significant figures, people believe the figure they are given, regardless of how accurate it might or might not be, and there is very little you can do to prove whether it is or isn't correct.

The datasheet should include information about actual reliability of the CC's SoC report. They do depend on some vagaries to operate and the precision could be misleading concerning the accuracy. But then it's the only game in town and at least it is wrong to three digits, that's very precisely wrong!

 

The datasheet should include information about actual reliability of the CC's SoC report. They do depend on some vagaries to operate and the precision could be misleading concerning the accuracy. But then it's the only game in town and at least it is wrong to three digits, that's very precisely wrong!

Battery capacity (and therefore state of charge vs. Actual measured Charge) varies with number of discharge cycles, temperature, length of time in a discharged state etc. and perhaps there are devices that can compensate for this over a 10 year battery life, but most of the devices available haven’t been available for 10 years, so who knows?
And, for most devices, it is an advantage to the manufacturer if the battery needs replacing after a ”reasonable” lifespan, even if it’s not ready to die just yet.
It becomes a different ballgame if the cost of the battery is the majority of the cost of the product, SOC measurements then will have to get a bit more accurate.

 

Here's the evil-twist to the story ........

People don't care about the actual "State-of-Charge" as a percentage number,
they want to know exactly how many minutes until I'll have to plug this thing in to a Wall-Outlet.

Which will vary wildly, all over the place, because of fluctuating Loads.

So, it's a "guesstimate" in the very best-case-scenario.
.
.
.

 

Here's the evil-twist to the story ........

People don't care about the actual "State-of-Charge" as a percentage number,
they want to know exactly how many minutes until I'll have to plug this thing in to a Wall-Outlet.

Which will vary wildly, all over the place, because of fluctuating Loads.

So, it's a "guesstimate" in the very best-case-scenario.
.
.
.

Or will it run out of charge before you get to the next charging point, and leave you stranded by the side of the motorway?