Battery % Readings - Is it really that hard?

[6zfshdb]

I've been wrestling with this non linear voltage drop issue ever since I bought my first e-bike. The displays on some e-bikes are better than others at estimating range but most still rely on voltage to compute the results. My bike has only a simple 5 bar gauge which is practically useless to determine range

The only real way to determine SOC is to measure ampere hours consumed. I eventually found this gauge which does just that. It's mainly used in marine or R/V applications.

New 80V 50A 100A 350A Battery Capacity Tester Coulomb Meter Battery Indicator | eBay

We will send you a new one after receiving the defective item.

www.ebay.com

You set the ampere hour capacity of any battery up to 80V and the gauge measures the AH consumed giving you an accurate percentage of power left. It has a 10 bar visual indicator as well as the actual % remaining. It also reads voltage and real time current drain. I like this feature since you can easily see how your pedal effort affects current consumption.

You can enter any value for AH capacity which means you can make allowance for battery cutoff voltage and loss of capacity due to age. For example, I enter 13.78 AH for my 52V 15AH battery which allows for the cutoff voltage set by the controller. For those who practice the 80 / 20 rule, It can also be set to reach zero when 20% charge remains.

Unfortunately, some DIY is required to mount it on a bike. The face of the gauge is weatherproof but the housing is not. It must be mounted in a weatherproof enclosure. It also requires installing a shunt in bike's the motor circuit.

Anyone interested can view this post on how I mounted the gauge:

Accurate Battery Gauge Project

I've searched in vain for some time now for an accurate, programmable, handlebar mounted battery gauge. Some high end bikes come equipped with this type of gauge and the displays on many others are up-gradable. Unfortunately, the controller on my bike isn't one of them. The battery gauges on...

electricbikereview.com

 

[AHicks]

Regarding the idea of riding down to 40v or so, especially for those running bigger motors. When riding the battery down that low, I'm always concerned about the potential for voltage sag, and what might happen if I needed to cross a busy road for instance, and my mind doesn't consider the low state of charge prior to doing that. If my battery is down, to say 42v, and I punch my throttle to wide open throttle, I can just about guarantee something (the BMS or contrtoller) is going to see the voltage sag, and shut the battery power off. The idea of this happening 10' into my run across a busy road is a little disconcerting....

For this reason, and the fact I have no good reason to ride the battery down so low to get extended mileage on a charge, I charge my batteries when they get down to 46v or so.

My point is, you really need to keep the potential for sag, and unintended power shut offs in mind, in addition to actual battery voltage readings, when it comes to developing a practical set of rules that you can live with - safely.

 

[Jon A]

Yes, that is one reason I lowered the "Low Battery Protect" in the controller to 39V. I should be able to run it down to 42V and still have a decent cushion.

 

[AHicks]

Yes, that is one reason I lowered the "Low Battery Protect" in the controller to 39V. I should be able to run it down to 42V and still have a decent cushion.

Depending on a few things, including the size of your motor, you may find 42v isn't quite enough "cushion". If you feel you must run the battery down that far, try it and see what happens. Run the battery down to 42v and then give the throttle a shot of WOT for 15-20 seconds (the length of time to clear a busy road for instance). See if it shuts down on you.

 

[Jon A]

Yes, I need to do some experimenting (mainly to verify I actually can run the battery that low). But the point would be if I mentally use 42V as 0%, if I ever get caught out somewhere and run it down to 42V it would be time to put it in PAS 1 and limp home. PAS 1 limits my power to 225 peak watts so voltage drop should be minimal. I've ridden the bike a fair amount and even climbed some steep hills with the motor off, so I figure I could go a long way on PAS 1 while sipping on the battery--it would just be much slower than I usually like to go.

I'm also thinking (hoping?) if/when I get my dual battery system up and running, that should reduce voltage drop at all power levels. I'll have to see if that works....

 

[AHicks]

Jon, I don't know your situation or what you are trying to accomplish, but would like to share MANY new riders (myself included) suffer some paranoia regarding their battery range (range anxiety). Most, if not all, find it's a total non event once they get more familiar with their bike - it's capabilities.

Maybe you already know this, but while limping a bike home on a very low battery is entirely possible, running the battery down to the level where you are required to do that is hard on the battery, especially if it's required to do that frequently. Consider the idea that if just 1 cell is discharged a little too far - you'll have a really serious (expensive) battery problem on your hands. Depending on the controller and/or the BMS to avoid that situation regularly is like playing with fire..... most would consider that a bad plan. FWIW, -Al

 

[Jon A]

Thanks, but that's not my plan at all. My plan is to make the bike as capable as possible and give myself accurate information with which to use it. That's why I'm trying to get a handle on the display readings--freaking out every time the battery dips below 50% actual capacity because the display turns red and reads a scary low % is not productive. Adding the dual battery will cut down on the charge cycles of the main battery, and at an 85% charge I'll still have about a 45% increase in range vs. the main battery charged to 100%--which will mean I don't run the batteries nearly as low for any given ride and feel the need to charge to 100% much less often.

Just because the bike will be capable of it and I'll have accurate information about it doesn't mean I plan to abuse that capability.

 

[alphacarina]

We're in the third decade of the 21th century, folks. I can't believe we still have to mentally map a reading of, say, 50.7 volts on our 48 volt battery (that alone is crazy) to a 75% battery charge - and that the 75% battery charge itself doesn't actually relate to usable battery percentage!

Now, there's the issue that voltage drop against usage (miles ridden is the better proxy) isn't linear. That is, a 0.5 volt drop when the battery is almost full is the result of more miles being ridden than a 0.5 volt drop when the battery is half or less full.

You also have a voltage drop when a load is applied to the battery - The bigger the amp draw, the bigger the drop. Getting an accurate percentage reading just by looking at battery voltage alone is . . . . impossible in the real world. The only way to do it is to put a shunt in the circuit and measure how many amps have been used since the battery was last fully charged. Why don't we use this method on eBikes? It would cost a good bit more

Don