Porformance drop off with lower battery %age.?

[Zlatan]

A couple of posters have reported poorer performance when their batteries are hitting around 10%. Personally I have not experienced this until well below tho figure.( Sub 3 or 4% I have noticed a slight drop off)
I was under impression with brushless motor and controller the working voltage ( supplied to motor) is below that supplied by battery, consequently as battery looses power the controller can (within reason) compensate for the dropping capacity of battery..

Wondering if one of the experts ( d8?) could explain if this is case or not..
If its not the case surely once battery begins to lose any significant power we would see immediate drop in power available...??? ie...the only time we,d get full power would be on 100% ???

 

[Deleted member 4366]

As you use up your battery, the voltage goes down until it's 75% of what it was when fully charged. For most bikes, that has a direct effect on power. they're not clever enough to compensate the current for the reduced voltage.

Without testing all bikes with a watt-meter, I couldn't say whether any use any sophisticated algorithms to compensate. I'd be surprised if any do, though it's conceivable that some could do it.

25% difference isn't easy to detect, especially as it happens over a few hours, except that I've ridden a couple of bikes where there was a very noticeable reduction when the battery was nearly empty. I assumed at the time that the low-voltage cut-off wasn't working properly and the battery was at the level when it couldn't deliver power anymore.

 

[Zlatan]

So from 100% to 75% will give best power ?

And what about from 75% to lets say 10% ?

So in effect when bike asks for full power controller gives battery voltage to motor ???

Recently bought a brushless drill and it seems to give really good power , and just stops...rather than gradually loosing torque is brushed alternative...?

D8, don't assume I,m questioning the validity of your answers..just questioning what's happening...
Thanks .

 

[Deleted member 4366]

The battery is 42v when fully-charged. You can use it until 31v, at which point you have 75% of the power you had at the beginning. you mustn't use a battery below 31v, so both the controller and the BMS are supposed to shut it off at around that point. That means 100% of the battery's usable charge is between 42v and 31v.

 

[cosybike]

Ryobi drills do that. They maintain full power then cut out. My bike feels similar. There is a tiny drop off in power as it selects eco mode on your behalf then it stops dead eventually. With 3 Kms left on the display I know I have about 1km climb ability remaining.

Sent from my Lenovo YT3-850F using Tapatalk

 

[Zlatan]

The battery is 42v when fully-charged. You can use it until 31v, at which point you have 75% of the power you had at the beginning. you mustn't use a battery below 31v, so both the controller and the BMS are supposed to shut it off at around that point. That means 100% of the battery's usable charge is between 42v and 31v.

And when battery gives 42v will that be the voltage applied to the motor ? ( assuming max load/max bars)

 

[Zlatan]

Ryobi drills do that. They maintain full power then cut out. My bike feels similar. There is a tiny drop off in power as it selects eco mode on your behalf then it stops dead eventually. With 3 Kms left on the display I know I have about 1km climb ability remaining.

Sent from my Lenovo YT3-850F using Tapatalk

That's what I thought my Haibike does, but a couple of posters have different experience..

 

[Deleted member 4366]

And when battery gives 42v will that be the voltage applied to the motor ? ( assuming max load/max bars)

The battery voltage is always applied to the motor. The controller has a series of gates that open and close very quickly in short pulses of battery voltage. All it can do is control the time that the gates are open. That's why the power diminishes as the battery voltage runs down. the energy in each pulse is basically the voltage times the time of the pulse.

 

[chris_n]

A couple of posters have reported poorer performance when their batteries are hitting around 10%. Personally I have not experienced this until well below tho figure.( Sub 3 or 4% I have noticed a slight drop off)
I was under impression with brushless motor and controller the working voltage ( supplied to motor) is below that supplied by battery, consequently as battery looses power the controller can (within reason) compensate for the dropping capacity of battery..

Wondering if one of the experts ( d8?) could explain if this is case or not..
If its not the case surely once battery begins to lose any significant power we would see immediate drop in power available...??? ie...the only time we,d get full power would be on 100% ???

I guess I sparked this debate off.
Which battery do you have? My Yamaha 400Wh has always been terrible for me up my home hill with anything less than 15%, it is particularly bad at any lower than 10%. I always aim to get back into the village at 20% to give me plenty of leeway.
I hired a 500Wh battery last week and it was fine down to 5% though the hill I was on at that level was not quite as steep as my home hill.
I must qualify this by saying I still find the Yamaha system very good and I am giving the battery a really hard time, most of my rides average 10%+ climbing for 50% of the ride over 40 to 70km (there isn't a lot of flat around here).
I'm just waiting for d8veh to give more details of his battery build before I decide what to do for an extender battery. Do I use a power tool battery or an ebay 10s4p package from Germany or a build like his with 10s2p Lg 3000mah cells.

 

[Zlatan]

The battery voltage is always applied to the motor. The controller has a series of gates that open and close very quickly in short pulses of battery voltage. All it can do is control the time that the gates are open. That's why the power diminishes as the battery voltage runs down. the energy in each pulse is basically the voltage times the time of the pulse.

And does the controller have the ability to provide longer pulses with diminishing voltage ?

I,m struggling getting my head around fact that when fully charged motor is getting 42 v and yet by 25% its only getting 31v...I do realise brushless don't exactly follow normal rules but I,d expect drop off in power to correlate with drop in voltage...like you said earlier , its already down 25% ? Unless controller pulses longer or perhaps we just don't notice drop in power ??
Ps Lecturers charge for this type of instruction d8 !

 

[Deleted member 4366]

And does the controller have the ability to provide longer pulses with diminishing voltage ?

I,m struggling getting my head around fact that when fully charged motor is getting 42 v and yet by 25% its only getting 31v...I do realise brushless don't exactly follow normal rules but I,d expect drop off in power to correlate with drop in voltage...like you said earlier , its already down 25% ? Unless controller pulses longer or perhaps we just don't notice drop in power ??
Ps Lecturers charge for this type of instruction d8 !

First question is easy. the answer is no. Voltage is what pushes electricity down a wire. The more volts, the harder it pushes. the gates in the controller can only let the electricity out. they can't change the push.

The second part you have the wrong way round. Let's change the explanation using the above. The push of the electricity down the wires is 25% less when the battery is empty compared with when it's full. Does that make more sense?

If you can imagine the energy of each pulse as being a rectangle having a width dependent on the time the gate is open and a height proportional to the voltage, the energy is the area of the rectangle. whether you have a short fat rectangle or tall thin one makes little difference, so the controller has the capability to compensate for the reducing voltage by making the pulses longer. The normal Chinese controllers don't do that, but it's quite feasible that Bosch would do that.

 

[Zlatan]

First question is easy. the answer is no. Voltage is what pushes electricity down a wire. The more volts, the harder it pushes. the gates in the controller can only let the electricity out. they can't change the push.

The second part you have the wrong way round. Let's change the explanation using the above. The push of the electricity down the wires is 25% less when the battery is empty compared with when it's full. Does that make more sense?

If you can imagine the energy of each pulse as being a rectangle having a width dependent on the time the gate is open and a height proportional to the voltage, the energy is the area of the rectangle. whether you have a short fat rectangle or tall thin one makes little difference, so the controller has the capability to compensate for the reducing voltage by making the pulses longer. The normal Chinese controllers don't do that, but it's quite feasible that Bosch would do that.

Good explanation. Thanks.

 

[anotherkiwi]

I guess I sparked this debate off.
Which battery do you have? My Yamaha 400Wh has always been terrible for me up my home hill with anything less than 15%, it is particularly bad at any lower than 10%. I always aim to get back into the village at 20% to give me plenty of leeway.
I hired a 500Wh battery last week and it was fine down to 5% though the hill I was on at that level was not quite as steep as my home hill.
I must qualify this by saying I still find the Yamaha system very good and I am giving the battery a really hard time, most of my rides average 10%+ climbing for 50% of the ride over 40 to 70km (there isn't a lot of flat around here).
I'm just waiting for d8veh to give more details of his battery build before I decide what to do for an extender battery. Do I use a power tool battery or an ebay 10s4p package from Germany or a build like his with 10s2p Lg 3000mah cells.

Yesterday for the first time I paralleled 10 Ah LiPo with my very tired bottle battery. I was checking voltages on the batteries and as they were within 0.02v I decided to give it a try. Worked a treat, 0.5 V sag on the steepish hill on the way to the port and power to spare. Very heavy package, the bottle battery is much heavier than the LiPo thanks to the casing. But looks like the bottle battery gets a life extension for the time being.

 

[Danidl]

First question is easy. the answer is no. Voltage is what pushes electricity down a wire. The more volts, the harder it pushes. the gates in the controller can only let the electricity out. they can't change the push.

The second part you have the wrong way round. Let's change the explanation using the above. The push of the electricity down the wires is 25% less when the battery is empty compared with when it's full. Does that make more sense?

If you can imagine the energy of each pulse as being a rectangle having a width dependent on the time the gate is open and a height proportional to the voltage, the energy is the area of the rectangle. whether you have a short fat rectangle or tall thin one makes little difference, so the controller has the capability to compensate for the reducing voltage by making the pulses longer. The normal Chinese controllers don't do that, but it's quite feasible that Bosch would do that.

Very clearly explained.
The switches used in these controllers are remarkable devices, capable of switching from totally non conducting to fully conducting .. a fraction of an ohm in ten nanoseconds. That's one thousandth of a million of a second, and off again just as fast. They only absorb energy when actually switching. Theses times much much faster than any coil of wire can respond to, so by switching them on and off in specific sequences they can mimic a sine waveform into each of the coils in turn. The controller can also increase the duration of the pulses, so as to compensate for the reduction in driving voltage as the battery gets depleted.
While the switches are relatively expensive. Say a euro each and there would be 6 minimum, the microcontroller and firmware is cheap and once the programming is done, it's free. We can therefore expect better algorithms in future.

 

[Mal69]

I was out four hours today, knackered, anyways i took pictures and will try and post them in the relevant thread over the next day or two.

Just wanted to say my Oxygen battery cut out literally 400 metres from home, voltage was 32.3 when it happened.

 

[PH001]

I kind of get all this but surely the torque is defined by the current into the motor so it stands to reason that if the voltage is higher, the peak current is higher. Sure you can duty cycle the pwm longer to get the same AVERAGE energy into the motor but I'm not convinced the torque (i.e hill climbing ability) is as high when you have a higher mark space ratio at a lower peak value.

 

[Deleted member 4366]

No, the controller regulates the current, not the power, so the higher the voltage, the more you get. That affects torque and speed.

In theory the controller could adjust the current in relation to the voltage. I haven't noticed one that does, but I use and test mainly Chinese stuff. I'm not sure what Bosch or Yamaha do. Perhaps owners would like to comment whether they can detect a difference in power between a full battery and and empty one. If they did compensate, you'd see the battery going down a lot quicker as it drains down. I can't remember anybody commenting on that.

 

[PH001]

I would say it's more accurate to say the controller regulates the ENERGY into the motor in terms of amps x time.

Using an extreme example, if you considered two rectangular pulse waveforms. Waveform A has a very high peak current and a very narrow mark. Waveform B has a very low peak current and a very wide mark. The area of both (and hence energy delivered) is the same but which one gives you the most torque?

For the posh controllers to maintain everything perfectly as the battery volts drop they would probably need some kind of DC-DC converter in there which is fairly unlikely given the size / efficiency constraints.

 

[Zlatan]

The overall torque would be the same...in that higher voltage would provide higher torque but for very short time...lower voltage...lower torque but for longer...net result equal.power..

And I have noticed my first 25% of battery power gets me further than 50 down to 25%...which again makes sense...I,m extracting constant power from a diminishing power source...but I,m fairly convinced my Haibike climbs just as well under 10% remaining to the first 10%...but it would not climb as far...( if that makes sense???)
But some extra control comes in under 2% remaining...it definitely starts rationing what is available...I suppose in an effort to fend off no power...( or it might simply be at point where lower voltage can not be compensated for with longer pulses...ie pulses at max duration by 3% remaining ???

 

[PH001]

..but lower torque doesn't get you up that hill even if it lasts for much longer!