AiLi battery monitor percentage incorrect

[pjm511]

Long time listener, first time caller. I have a AiLi battery monitor I installed recently. I have noticed some odd behavior I was hoping I could get some help diagnosing.

When I first installed it, I did a test by fully charging my battery (100 ah battleborn battery) and then turning on lights and my fridge and watch the behaviour of the AiLi battery monitor. It performed as expected: the percentage on the battery monitor tracked downwards with the expected voltage. All is good..

I then charged my battery back up and turned off everything (lights, fridge, etc.). There is a small draw from some passive electronics in my fridge and other misc. draws in the range of 0.09-0.15 amps so, now, around a month since charging the device my voltage is down below 12.0V which is approaching 0% for my battery. This is expected due to the small draw on the system X the few weeks of time. The odd thing, though, is the AiLi battery monitor is still reading ~96% full despite seeing the ~0.09-0.15 amps of current on the battery monitor. Does anyone have any thoughts as to why the battery monitor isn't tracking the correct percentage despite seeing the draw on the monitor (albeit, it being small)?

Note: Wiring is correct and the preset voltage setting was left at the 0v so I don't think that is contributing to the issue.

Battery monitor being discussed: https://www.amazon.com/AiLi-Battery-Monitor-Voltmeter-Motorhome/dp/B07CTKYFTG?th=1

 

[MrThisIsME]

Looks like a nice unit for the price.
Have to ask the obvious as I did not see you mention it. (I have no experience with this particular unit). With most such devices there is a configuration and a synchronization requirement. Was this completed?
For example on the Victron unit you set the Ah, and can calibrate SOC.

 

[pjm511]

Yes, there is a configuration that was done. Essentially, you set the rated capacity of the battery (in my case, 100 ah) and then set the percentage to 100% (if the battery is known to be fully charged which, in my case, it was).

 

[perryg114]

Here is the deal. The meter has a resolution limit as to what the smallest current it can read. Let's say it has a 50mV output on that shunt at 100A. That means that one Amp is .5mV which is getting pretty small. Just about every piece of instrumentation is measuring voltage that voltage only be read down to a certain point. There is a think called and A to D converter that converts analog voltage to a digital number. Some are 12 bits and some are 16 bits. 12 bits means it can read 1 bit in 12. So that is 4096 steps of voltage. So lets say the A to D converter range is 0-10V. So that is 10/4096 = .0024 V or 2.4mV. If it is 16bit then it the number is 65536. So 10/65535=.00015 or .15mV resolution. But there is usually noise that might add up to 10 times that so maybe you get 1.5mV out of it. So you see that in order to measure .1A that is like .05mV which is probably in the noise. So you are basing the state of remaining charge on a number that might not mean anything. Yes with the right equipment you can measure microvolts but I doubt that is what you have.

 

[pjm511]

Thanks for the response - I certainly understand what you are saying. Actually, I at first thought it was a resolution issue as well but the meter is reading the current reasonably correctly (you can monitor the instantaneous current via the screen on the meter and that is where I am seeing the 0.09-0.15 amps I mentioned in the original post). So the meter is picking up on that low current (whether that is accurate or not with the amount of noise that could be present) but it seems that it is not using that draw correctly to calculate the percentage of battery left (i would think it would purely be a calculation of that instantaneous current multiplied by time to give ah and then that number is divided by the battery capacity [100ah in my case]).

For example, I am reading 0.15 amps on the meter so 0.15 amps x 21 days x 24 hours = 75.6 ah so the monitor should be reading 100-75.6 = 24.4% battery capacity left but it isn't. It is still reading >95% battery capacity left. So I guess it could be a resolution issue but it seems odd that it has the instantaneous current but isn't using that value correctly to calculate overall consumption until it sees a higher draw (for example, when lights + fridge are running).

 

[perryg114]

Another way of totalizing is a coulomb meter. Most battery manufactures have a voltage chart showing the state of charge as a function of voltage so that is a rough guide. You can discharge your batteries down to their limit and measure what is left. If you measured capacity when you put the system together then you will know what was left in the batteries.

 

[pjm511]

Right. That is why I was hoping someone with this battery monitor has had a similar issue and found a solution. To reiterate what is going on in my situation: the voltage suggests <10% battery life left (per the voltage chart), the instantaneous current from the battery monitor X the amount of time since the last charge suggests <10% battery life, but the battery monitor says it is at >95% charge.

 

[TacomaJoe]

I've seen the same thing with mine. I think the monitor sees the low current but the value it below the threshold of it's counting algorithm. Not an issue running in my trailer because I'm always getting a fair cycle current, but I did see it when my battery was just sitting in the garage.

 

[pjm511]

Thanks @TacomaJoe that’s what I figured might be happening. It certainly is annoying that it does that. If it goes back to 100% correctly though with a charge, then it isn’t the end of the world