Used and new battery performance

[Wisper Bikes]

I have just received this email from a BEBA member, I would be grateful for any thoughts.

"I wonder if you could give me your opinion on this. Across the industry there are varying claims for the Ah capacity of batteries. I think we need (if there isn't already) a standardised average discharge rate in Amps upon which the capacity can be stated. As a 10Ah rated battery might produce a lot less than 10Ah until the BMS cuts out if it were on a bike constantly drawing, say, 12A.

My own thought is that a constant drain 4A is about right, given the power consumption averaged out over a typical ride.

Obviously there is no 'standard' rider, bike or journey, but some sort of realistic benchmark is desirable so that real effective battery Ah ratings can be achieved.

One part of my aim here is to consolidate methods and reach agreement about how a well used battery's performance can be quantified."

Many thanks

David

 

[banbury frank]

Hi

Yes we need a standard to test to at a fixed current load over 0ne hour Plus a temperature reading

Maybe 2 readings at high and low current draw and temperature

so when winter hits the capacity we will no the drop in range


so 36 volt 14 AH at 4 A constant Load at 20 degree C should last 3.5 hours

36 volt 14 AH at 4 A constant Load at 0 degree C Approx 25 % less 2.6 Hours

Will need putting in a freezer to run test




well Done David

More ideas Please from the forum Please


Frank

 

[tillson]

Because battery voltage is different depending on which brand of bike a person buys, I think that an Amp value based on a nominal power would be a good starting point. 4 Amps from a 37 volt battery is a lot more useful than 4 Amps from a 24 volt battery for example.

I don't know what nominal power value would be fair, but as a starting point, would it be reasonable to say that an ebike with an average rider draws about 100 Watts whilst going along gently undulating terrain? If so, the Amp value for a 24 Volt battery using this Nominal Power Value, would be 4 Amps and the 37 Volt battery 2.5 Amps.

I'm not saying that 100 Watts should be the Nominal Power Value, but the battery capacity should be based on a typical power value that it is expected to deliver in service. If a standard power value was used, it would give a good comparison of relative battery capacities.

 

[Wisper Bikes]

I have spoken to Anthony (Wisper NZ) and his idea is as follows;

Instead of setting a standard current of 4A would it would be better to set a standard power of say 150W? This would put batteries of varying voltages (24, 36, 37, etc) on a level playing field, because power = voltage x amps. A 24v batt will be tested at 6.25amps (24 X 6.25 = 150w). A 36v batt will be tested at 4.16amps (36 X 4.16 = 150w).

Anthony also says he would rather see the tests done at 200W as larger batteries would be more efficient under a higher current!

 

[NRG]

A can of worms this one. It would be ideal if there was clear labelling by manufacturers at what discharge rate the capacity quoted was based on. In some cases the C/10 rate is quoted so 10Ah capacity at 1amp discharge.

Its very unlikely on our E-Bikes the average discharge is anywhere close to this; the average is in reality higher and a C1 rate would be better but it still wont take into account the varied nature of one individuals usage to the next who's average and peak draw will be different.

However, there are many graphs on-line if you Google them that show cells capable of or rated for < C1 discharge rates having a stated capacity that's more or less correct. Some I've seen show just a 5% drop in capacity if used at the C5 rate. As our E-Bikes are current limited by the controller and will probable never draw more than C1.5...and given a reputable manufacturer IE: who uses quality cells, I would say the stated capacities used are more or less correct given a few percentage points either way...

...but this is probably all meaningless as what everybody wants to know is how far they can go and in this Wh's are a far more useful metric as the figure represents the available energy to do work and not just capacity.

 

[NRG]

I have spoken to Anthony (Wisper NZ) and his idea is as follows;

Instead of setting a standard current of 4A would it would be better to set a standard power of say 150W? This would put batteries of varying voltages (24, 36, 37, etc) on a level playing field, because power = voltage x amps. A 24v batt will be tested at 6.25amps (24 X 6.25 = 150w). A 36v batt will be tested at 4.16amps (36 X 4.16 = 150w).

Anthony also says he would rather see the tests done at 200W as larger batteries would be more efficient under a higher current!

Maybe I need to dwell on this a bit more, however, David, I assume you know what cells go into your batteries and so would also know how the capacity was derived from the data given by your supplier....

 

[flecc]

With such wide variations in motor power/consumption rates that we have, I'm not convinced that a fixed figure of this kind is useful.

Current legal hub motor system gross powers range from 300 watts to more than two and a half times that. One legal bike flattens a 37 volt 10 Ah in 15 miles or less, regardless of who's riding it and easily covering that within the hour. A 4 amp figure is worthless for that bike.
.

 

[Wisper Bikes]

Naturally NRG we have all the specs and know exactly what goes into our batteries.

The BEBA member in question is just as interested in older batteries as in new.

Thanks for your input, as always it's much appreciated!

 

[Wisper Bikes]

With such wide variations in motor power/consumption rates that we have, I'm not convinced that a fixed figure of this kind is useful.

Current legal hub motor system gross powers range from 300 watts to more than two and a half times that. One legal bike flattens a 37 volt 10 Ah in 15 miles or less, regardless of who's riding it and easily covering that within the hour. A 4 amp figure is worthless for that bike.
.

My position exactly Tony, I wanted to air the discussion again to see if anyone has any new ideas.

 

[Mussels]

I like the idea of standard measures but will it be good for the consumer?
When metrics were introduced for cars (CO2, km/l etc) manufacturers designed cars to look good in the tests and the results were quite different to real life driving. If metrics are introduced for batteries then will we just see manufacturers adopting the same cell types across the board because they produce good figures?

 

[Straylight]

One question I'm repeatedly asked by people interested in my bike is not "how much power is in the battery?" but "how far can you go on a charge?". On this basis, I think a simple chart detailing different styles of usage vs distance travelled, would be a lot more useful to the customer. They could then estimate their own expectations a little more realistically, than by using the current marketeering of say "60 miles range!"... "based on a naked five stone rider using minimal power assist in a pancake flat car-park, ambient temperature 30 degrees."

Or, to put it more in the context of this discussion, the power required to sustain the different styles of riding, coupled with a chart of how long the battery will last at said rates of use, then the outcome of how far this equates to in distance travelled.

 

[tillson]

With such wide variations in motor power/consumption rates that we have, I'm not convinced that a fixed figure of this kind is useful.

Does this matter? As I understand it, we are trying to establish a method of drawing a comparison between batteries from different manufactures, of different voltages and of different claimed capacities. An ability to deliver a nominal amount of power, which is a standard value across the range, would give an indication of the merits of one battery compared to another.

I know that there are peak current values and peak power demands outside of the legal maximums, so the above method isn't perfect, but it is better than what we have at the moment.

 

[flecc]

Does this matter? As I understand it, we are trying to establish a method of drawing a comparison between batteries from different manufactures, of different voltages and of different claimed capacities. An ability to deliver a nominal amount of power, which is a standard value across the range, would give an indication of the merits of one battery compared to another.

I know that there are peak current values and peak power demands outside of the legal maximums, so the above method isn't perfect, but it is better than what we have at the moment.

If there was battery choice my objection wouldn't matter, but for most people there is no choice, just the e-bike manufacturer's supplied battery. That makes such measurements at one fixed level for all a bit pointless.

The bike I mentioned has a battery that will easily show 10 Ah at a 4 amp rate, but on the bike in typical use the result is more like 7 Ah due to it's high consumption rate.
.

 

[Mussels]

One question I'm repeatedly asked by people interested in my bike is not "how much power is in the battery?" but "how far can you go on a charge?". On this basis, I think a simple chart detailing different styles of usage vs distance travelled, would be a lot more useful to the customer. They could then estimate their own expectations a little more realistically, than by using the current marketeering of say "60 miles range!"... "based on a naked five stone rider using minimal power assist in a pancake flat car-park, ambient temperature 30 degrees."

Or, to put it more in the context of this discussion, the power required to sustain the different styles of riding, coupled with a chart of how long the battery will last at said rates of use, then the outcome of how far this equates to in distance travelled.

If everyone tests with a naked 5 stone hairless rider on a flat smooth surface then it will be a good comparison.
What nobody but the buyer can answer is 'how far will it take me?'.

 

[Straylight]

LOL

Well, yes, you get it. The current level of detail provided means next to nothing in real terms. It may serve as a basis for product comparison, but only that in isolation, if, as you say, everybody used the same testing parameters. What's missing is how advertised figures relate in any way to real world use.

What I was suggesting was a means of roughly bridging this gap in the data, by publishing distances of actual test rides, in a variety of terrain, and riding style, from derestricted - feet up on the handle bars, to slogging it for all you're worth.

Also the way to gather such data cheaply would be to ask the existing user base, none of whom are 5st, and given to pootling about in the buff (or are they???).

Surely then some geek could come up with an algorithm based on rider weight, level of effort, and terrain, all roughly quantified of course to at least give the prospective buyer some kind of approximation of how useful their purchase might be? I get that no level of accuracy could be guaranteed, but still think it would be a bit more useful than what we have currently.

 

[Mussels]

Also the way to gather such data cheaply would be to ask the existing user base, none of whom are 5st, and given to pootling about in the buff (or are they???).

I wouldn't mind helping (clothed unless it's really warm) but I'm not going to intentionally run my battery flat when I don't have a spare.
The other problem is the controller cutoff, if headline figures are what matters then controllers will be set to run to a lower voltage which will shorten overall battery life.

 

[Wisper Bikes]

They could then estimate their own expectations a little more realistically, than by using the current marketeering of say "60 miles range!"... "based on a naked five stone rider using minimal power assist in a pancake flat car-park, ambient temperature 30 degrees."

Absolute rubbish Strayight, we rub the rider down with fish oil too!

David

 

[Wisper Bikes]

Does any one know how the motor industry arrive at their figures?

As you can imagine I am particularly keen to find a fair test method, that we can all can subscribe to.

Short of having a robot and a ride simulator that could include hills, I can't see any other way of working out the range.

I think that is why I have been asked to concentrate on battery output.

Regards

David

 

[flecc]

Life is so easy for the motor industry, they only have to consider the motor driven performance.

It's the hybrid human + motor nature of the e-bike that makes them impossible to generically quantify in performance terms. It really is "How long is a piece of string?"
.

 

[indalo]

Does any one know how the motor industry arrive at their figures?

Regards

David

I shouldn't bother Mr Miall. The cars are tested in a simulated set of conditions in what is effectively a giant laboratory. Golf clubs and indeed golf balls are tested under laboratory conditions also.

It hasn't gone unnoticed that the vast majority of drivers never achieve the mpg figures advertised by the manufacturer. Neither does the use of a leading golf ball (or clubs) make one a better golfer.

You could have all the robots, dynamometers and other measuring devices at your disposal but it all comes back to the "How long is a piece of string?" syndrome. How far a bike may travel on a charge is an indeterminate figure which may vary even for the same rider on the same course and even on the same day.

The current arrangement seems like a pretty fair indication to me of how far a new bike and battery might convey one. If yours, or other companies, made claims about potential distance that were way off the mark, I think you'd meet with considerable hostility and you'd probably also meet a trading standards officer at some point too!

Regards,
Indalo