The "best" battery gauge is an amp-hour meter.
Li-ion batteries like we use tend to have extremely high coulomb efficiency meaning the number of electrons put in comes very, very close to how many electrons come out. Over time the number will decrease but from day-to-day however in simplistic terms it's the number of electrons left in the battery from the charge cycle that gives you the best estimate of state-of-charge (SOC)
If Li-ion are so extremely coulomb efficient why aren't they just as energy efficient? It takes more voltage to push the electrons in than the voltage they come out at. Since power = volts * amps and energy = power * time if the charge/discharge are at constant currents then the time will be nearly the same since the coulomb efficiency is so high so the difference will be less energy out than in because the voltage during discharge is lower than charge. There are other ways of looking at it but that's an easy way and it explains why measuring energy used while a very useful measurement isn't as good as amp-hours when it comes to measuring SOC.
Not everyone has a amp-hour totalizer on their bike. You need to wire in a shunt in order to measure current and while not expensive they aren't free and usually don't come standard
So if we don't have a good way to measure current and totalize it (integrate over time to get amp-hours) then we're left with only one thing left to use - voltage
Now that curve in the prior post is BS. It's not so flat in the middle. A li-ion battery like we use has a discharge curve that looks more like this:
And we don't usually use the last 10-20% at all (almost no point to it since the power is dropping so fast and we get more voltage sag) so that the part of the curve we're really interested in has a fair amount of slope that's constant enough once we get past the first 10% or so.
So if all we have is voltage we can get a reasonable idea where we are - especially once we get some time with a given battery and learn what the voltage display (whether bar graph, % or decimal voltage) means for our setup with regards to SOC.... then if we are relegated to using voltage should we choose bar graph, % or voltage? I say whatever gives us the most resolution as long as it covers the entire range of the battery. Typically we get 0.1V resolution with a voltmeter although some have 0.01V. On a percent gauge if we are going with a 52V battery and calling 58.8V 100% and have an LVCO of 43V at 0% then each percent = 0.158V and the voltmeter wins the resolution contest even if only one decimal. Bar graphs aren't even in the game.
In summary, voltage is a rough method that when a user "learns" their machine can have a repeatable and pretty good idea of where they are in the discharge cycle. A voltmeter gives the best resolution and is more likely to accommodate the entire range of the battery. However an amp-hour meter is the best way if we want accuracy and a linear representation ...
Li-ion batteries like we use tend to have extremely high coulomb efficiency meaning the number of electrons put in comes very, very close to how many electrons come out. Over time the number will decrease but from day-to-day however in simplistic terms it's the number of electrons left in the battery from the charge cycle that gives you the best estimate of state-of-charge (SOC)
If Li-ion are so extremely coulomb efficient why aren't they just as energy efficient? It takes more voltage to push the electrons in than the voltage they come out at. Since power = volts * amps and energy = power * time if the charge/discharge are at constant currents then the time will be nearly the same since the coulomb efficiency is so high so the difference will be less energy out than in because the voltage during discharge is lower than charge. There are other ways of looking at it but that's an easy way and it explains why measuring energy used while a very useful measurement isn't as good as amp-hours when it comes to measuring SOC.
Not everyone has a amp-hour totalizer on their bike. You need to wire in a shunt in order to measure current and while not expensive they aren't free and usually don't come standard
So if we don't have a good way to measure current and totalize it (integrate over time to get amp-hours) then we're left with only one thing left to use - voltage
Now that curve in the prior post is BS. It's not so flat in the middle. A li-ion battery like we use has a discharge curve that looks more like this:
And we don't usually use the last 10-20% at all (almost no point to it since the power is dropping so fast and we get more voltage sag) so that the part of the curve we're really interested in has a fair amount of slope that's constant enough once we get past the first 10% or so.
So if all we have is voltage we can get a reasonable idea where we are - especially once we get some time with a given battery and learn what the voltage display (whether bar graph, % or decimal voltage) means for our setup with regards to SOC.... then if we are relegated to using voltage should we choose bar graph, % or voltage? I say whatever gives us the most resolution as long as it covers the entire range of the battery. Typically we get 0.1V resolution with a voltmeter although some have 0.01V. On a percent gauge if we are going with a 52V battery and calling 58.8V 100% and have an LVCO of 43V at 0% then each percent = 0.158V and the voltmeter wins the resolution contest even if only one decimal. Bar graphs aren't even in the game.
In summary, voltage is a rough method that when a user "learns" their machine can have a repeatable and pretty good idea of where they are in the discharge cycle. A voltmeter gives the best resolution and is more likely to accommodate the entire range of the battery. However an amp-hour meter is the best way if we want accuracy and a linear representation ...
Attached Files
Comment