Exponential Battery Drain?

[lonelocust]

I've only had my RadCity 5 for a few weeks, and, on just a couple of occasions ridden it almost dry.  Those rides have been about 50-55km, on mostly flat, city riding in Phoenix, AZ, with little or no wind.)

I've noticed a couple odd things and want to see if this is anyone else's experience.

• (It's been mentioned elsewhere) The battery gauge on the battery and the one on the handlebars often don't match, with the handlebar gauge being 1-2 bars lower than the indicator on the battery - until it gets to about 50%, then they seem to sync up all the time.
• The battery drain seems to be exponential.  For example, the first 15-20km only uses 10% (90% full).  By the time I'm at 30km the battery reads about 70% full.  Based on that, I would expect to be getting closer to 90km range, nonetheless, by the time I hit 50km, the battery is on it's last bar. (Note: I'm using the readings from the battery rather than from the handlebars, since... I don't know... that one should be more accurate?)

Any idea if this is typical or is this a service issue I need to take up with Rad?

[Monty]

Looks like normal battery drain. The same thing happens to phone, laptops, etc. I don't know why though but as long as you have normal range, I wouldn't worry too much about the battery.

[Eric7]

A lot of these battery gauges are psychological tricks to make people feel good. 

For example, you can sometimes use your phone for 15 minutes and the battery gauge still shows 100%.  This is impossible because if you do the math it means that you can use the phone for 25 hours straight.  I think customers feel really bad if the gauges were accurate.  This is true for all electric devices like Monty says.  Phones, computers, cars, etc.

The same thing is happening with ebicycles.  If 20km only uses 10%, then the bike should go 200 km.   

I think the first bar represents 35% of the battery capacity based on another post (not by me) in this forum.  It means that when the first bar goes, you have only 65% capacity remaining. 2nd bar is 20%.  3rd bar is 20%.  4th bar is 10%.  The last bar is 15% but at reduced power.  So, I guess around the middle, the guage is "linear".  But be careful, near the end the battery will die rapidly.

I would not trust any of the gauges.  They are probably all "calibrated" using different plans to make customers happy.

[DickB]

This chart was made from testing of my 2021 Rover. The battery gauge readings were determined by substituting a variable power supply for the battery, to determine at exactly what voltage each bar is lit. The display gauge is not linear.

[Eric7]

This chart was made from testing of my 2021 Rover. The battery gauge readings were determined by substituting a variable power supply for the battery, to determine at exactly what voltage each bar is lit. The display gauge is not linear.

Yes.  This is the chart I am referring to.  Thank you for your wisdom DickB.

[NaturallyRC]

Relevant to your observations, I ask Rad about this recently and this is what they said:

Thank you for reaching out to us at Rad Power Bikes Product Support. I understand from your message that your battery is only giving you limited power when on the last bars of capacity. This is a built-in feature of the system to attempt to give you power for as long as possible. The system will limit the output watts to try and keep you from having a total loss of power when nearing the end of the battery's power. Here is some information to hopefully explain what is happening.

Rad Power Bikes and other li-ion powered electronics get what's called Voltage Sag when the battery is low. As your battery dips into the lower charge levels so does your motor output power down to the road. The lower end of the battery charge will also drain a touch faster as well. Unlike a gasoline vehicle where it has 100% power to the wheels until the fuel is Empty, e-bikes and Rad Power Bikes decrease their output as the charge level of the battery goes down. That's a normal occurrence. Just as a smartphone will lose charge faster if at 50% rather than 100%. If at 50% charge, it's not just it's half full and will power off sooner it also loses charge faster. That's the normal technology of li-ion battery packs in this day in age.

Fewer volts held in the pack result in fewer volts outputting. You'll see torque, hill-climbing, and off-the-line acceleration lessons. It might feel like something is wrong but it's acting normally. This happens with laptops or smartphones as they decrease in battery power they decrease in processing speed. Especially in smartphones, you'll see a dip in cellular serviceability because the phone doesn't have enough power to grab a signal. You might be familiar or experienced with this but on an e-bike pushing the weight with a rider can lessen the power-to-weight ratio illuminating this occurrence. So in the closing of that thought, when you have 1 or 2 bars left in the battery the bike won't be as powerful as if it has 4 or 5 bars of battery charge. Even if it's outputting 740+ watts it's not outputting as much torque down to the road. The Rover Bafang motor outputs 60 lb-ft of torque so it's a lot more evident when the power dips down as opposed to the RadCity outputting 30 lb-ft of torque.

I wish the bike did operate like a gasoline engine holding an RPM for pushing power out to the wheels at top speed but the programming doesn't work that way. It's a smart system that can systematically drip that exact amount of fuel for the speed. 30 watts is plenty to push 20 MPH without load. 300 watts is plenty to get a person around 150 lbs or even right below 200 lbs up to speed but ours has a peak power of 750 watts. The smart powertrain system will fluctuate its watts constantly to provide the most efficient delivery of power possible. If it senses a lot of weight under throttle it might need 750 watts to launch you from a stop but most vehicles don't need a lot of power to stay at a constant speed on flat ground so it backs off. If you're at 20 MPH it will sense it's reached the top speed and back off and not hold the motor pumping out fuel. It's designed to be the most fuel-efficient.

I hope this information is helpful to you. If you have any other questions or concerns please feel free to reply to this message and I will get back to you as soon as possible.

[Eric7]

Naturally RC, thanks for the email from Rad. The following is not about you and I am just referring to Rad's letter. 

I understand what Rad is saying but I think it is wrong and it is just some fancy way of saying the gauge cannot be relied upon for range.  If it is not range, what is the gauge measuring?  How can it be useful for the average ebiker? 

Any battery gauge with 5 bars will have to make some interpretation as to what constitutes a bar.  I think the bars should be proportional to the amount of work the battery can still do. To put it another way, the bars should represent the remaining distance or range.  This is what the public is used to in a gas gauge.  The public, used to gas gauges, will understand this.  I know about voltage sag.  If there is voltage sag in the system, the gauge should show an even lower percentage so the user will not be stuck.  The power meter is not a voltage meter - otherwise, just call it a voltage meter and make no reference to "battery gauge."

If we use any other type of standard, my question is, what does 50% mean?  50% of what? Alternative question: 4/5 bars of what? 3/5 bars of what? 2/5 bars of what? 1/5 bars of what?

The simplest way to visualize this is that a user should be able to drive away from home on flat ground until the battery is 50%.  Then the user should be able to turn around and drive the ebike back home, using roughly the same battery assistance and under roughly the same conditions (flat ground in this hypothetical), and be able to reach home.  Any gauge that cannot do this is inaccurate and useless in my mind. 

In fact, in a car, there is reserved capacity.  In any modern car, I can drive away from home until there is 50%, drive home, and still have some in reserve for another 10% of driving range.

I think most (probably all) modern ebikes have inaccurate meters that exaggerates range for sales purposes.  I do not single out Rad.  Everybody does it.

[NaturallyRC]

No disagreement from me, Eric. Considering how crucial it is, the readout is a surprisingly weak area. The lack of complaints (that I can find) leads me to believe that most folks never drop lower than about half anyway, and are unaffected by what happens when the tank gets real dry. Or, they intuitively learn to (mis)interpret those uneven bars, and thereby avoid pedaling home with no assist, which was how I leaned my lesson. Cheers.

Naturally RC, thanks for the email from Rad. The following is not about you and I am just referring to Rad's letter. 

[lonelocust]

I understand what Rad is saying but I think it is wrong and it is just some fancy way of saying the gauge cannot be relied upon for range.  If it is not range, what is the gauge measuring?  How can it be useful for the average ebiker? 

Any battery gauge with 5 bars will have to make some interpretation as to what constitutes a bar.  I think the bars should be proportional to the amount of work the battery can still do. To put it another way, the bars should represent the remaining distance or range.  This is what the public is used to in a gas gauge.  The public, used to gas gauges, will understand this.  I know about voltage sag.  If there is voltage sag in the system, the gauge should show an even lower percentage so the user will not be stuck.  The power meter is not a voltage meter - otherwise, just call it a voltage meter and make no reference to "battery gauge."

If we use any other type of standard, my question is, what does 50% mean?  50% of what? Alternative question: 4/5 bars of what? 3/5 bars of what? 2/5 bars of what? 1/5 bars of what?

The simplest way to visualize this is that a user should be able to drive away from home on flat ground until the battery is 50%.  Then the user should be able to turn around and drive the ebike back home, using roughly the same battery assistance and under roughly the same conditions (flat ground in this hypothetical), and be able to reach home.  Any gauge that cannot do this is inaccurate and useless in my mind. 

In fact, in a car, there is reserved capacity.  In any modern car, I can drive away from home until there is 50%, drive home, and still have some in reserve for another 10% of driving range.

I think most (probably all) modern ebikes have inaccurate meters that exaggerates range for sales purposes.  I do not single out Rad.  Everybody does it.

Thank you for that!

That is EXACTLY right, a gauge that tells you nothing useful may actually be harmful.

The gauge may be technically correct, but it is not communicating information that is particularly clear or useful to the rider.

Obviously, there are a LOT of other factors, like grade, PAS level, wind speed and direction that make it difficult to judge range, but that's actually true with a car, too.

The rider should not need to memorize the hockey-stick shaped output curve of a lithium ion battery and then apply that to the information presented to have a reasonable estimate of remaining ride time/distance available.

The simple fact is, if I ride out in ideal conditions, on flat ground, with no wind until the battery gauge reads 50%, I should have a reasonable expectation (NOT A GUARANTEE) of being able to turn around a get home.

As the gauge stands, this is not even close to true.  Maybe a charge indicator isn't even what we need?

They're monitoring motor output, battery output, distance, time, etc.  perhaps something a little more interpretive is what's needed?

My Toyota estimates the range I have remaining to the nearest gas station, it's not perfect, but it's close, and it's helpful, and it updates based on current driving conditions.

This is not an insurmountable programming problem.

This is especially true since the two gauges don't even match!  Have the true battery gauge on the battery, and something more helpful on the head unit.

(Then, all they'd need to do is make the left-hand readout ACTUALLY readable on clear days in sunlight.)


Sent from my iPhone using Tapatalk

[NaturallyRC]

Good observation, now that you mention it.

My Toyota estimates the range I have remaining to the nearest gas station, it's not perfect, but it's close....This is not an insurmountable programming problem.

[DickB]

The statement from Rad, "Even if it's outputting 740+ watts it's not outputting as much torque down to the road.", is nonsense.

The motor operates on current. If the motor is getting 740+ watts, it will perform the same whether or not the battery is fully charged or partially charged. The motor can't tell the difference.

The battery is DC. The motor is AC.  The controller uses Pulse Width Modulation (PWM) to rapidly turn the full battery power on and off to simulate a sine wave (AC) to the motor. As battery voltage drops, the same AC sine wave current can be delivered by simply extending the "on" times of the pulses.

The battery meter is a simple voltmeter. Rad even states this in their online FAQs, and I have verified it with a variable power supply.

[Eric7]

This is how I use the battery meter.  Just a mental trick to keep things simple when I am having fun so I don't have to remember too much.  This is just my way of dealing with it and, of course, it is not the most accurate but close enough for me.

1.  Just pretend the last bar (the lowest one) does not exist.  So now you only have a 4 bar meter.
2.  When the first 2 bars go, or when it starts to flicker between the 2nd and 3rd bar, it is 50% used up.

Or use the following guide:

***** Full Charge
_**** 75%
__*** 50%
____** 25%
_____* 12%

[NaturallyRC]

Reminds me of my spouse behind the wheel -- when the gas gauge hits half, she beelines it to the Petrocan. Anything below needle-half...there be dragons. I'm forever driving around on fumes. 

This is how I use the battery meter.  Just a mental trick to keep things simple when I am having fun so I don't have to remember too much.  This is just my way of dealing with it and, of course, it is not the most accurate but close enough for me.

1.  Just pretend the last bar (the lowest one) does not exist.  So now you only have a 4 bar meter.
2.  When the first 2 bars go, or when it starts to flicker between the 2nd and 3rd bar, it is 50% used up.

Or use the following guide:

***** Full Charge
_**** 75%
__*** 50%
____** 25%
_____* 12%