Determining electric bike range and hill climbing ability

[Jed]

flecc, how many electric bike suppliers actually make the motors themselves? it seems to me that many parts we see on bikes over here that are made in china are used by lots of different companies. Are there companies there that just manufacture the motors and then the bike companies obviously buy them and fit to their bikes? Could we not go straight to them for the specs

thanks
Jed

 

[flecc]

Yes, the parts bin bikes are commonplace Jed, but Chinese companies often aren't very easily contactable by individuals. There are intermediary companies like Abracadabra who act as the web forwarding agent.

Enter a Chinese company name in Google or similar and you'll end up with the intermediary who forward messages. If you're not a company with a serious quantity inquiry, you don't get an answer. I've tried many times. The companies that are more easily contactable are all too often the junk manufacturers.

Just finding out who does what for whom is often impossible.
.

 

[Jed]

thanks Flecc, so for the most part it looks like we are back at square one then back to us to determine the range etc as best we can.

thanks
Jed

 

[chazpope]

Guys

I still think you can get consistent results that would be indicative of the bikes' performance under different conditions by just doing simple tests. The conditions of the tests can be consistent enough I think - certainly car tests don't happen in vacuum and some spread is assumed there too.

Most towns (sorry flecc) have an 'industrial estate' with straight asphalt roads that is unused on weekends. I appreciate that bikes can have a varying top speed therefore we could have a 0-to-distance test with the result in seconds. I have been doing a similar thing on my bike every day on the same route and get fairly consistent results.

So how about this:

Standard Test 1 - 0 to 100 meters. From a full stop simply turn the throttle to max and measure the time when you reach the 100m mark.

this will give you a pretty good idea of the bike`s power delivery from start and at low revs, torque curve and this sort of thing which is important in everyday use and pulling out of junctions.

Standard Test 2 - top speed sustained over 100 m. Get to the throttle-only top speed before the first mark and measure the time to the second mark - then work out the speed in mph.

How to achieve consistent test conditions:

Cyclist weight: 75 kg - simply weigh yourself wearing a rucksack full of bricks/water bottles (that you can get rid of on the way back)- untill you get to 75 kg. If you are heavier get your kids/wife/gran to do it.

Battery condition: Charge the battery to full and cycle to the 'test range' on muscle power only

Wind / Not-quite flat effect: do the test in both directions on a non-so windy day.

Tyres: should be pumped up to bike manuf specs

Time measuring: use a simple stop watch

Data recording: do several tests and average - excluding any single wildly inconsistent results

Distance measuring: preferably do not use bike computer for this unless you measured and entered the actual distance covered per 1 rev while you are sitting on the bike as circumference

Now on the 'Range' test - I see it can get fuzzy - you will need to ride the bike around the block until it stops - obviously speed here will have a huge effect on results i.e. there would be speeds that get you further than others.

However I think the above 2 tests will also indicate the range performance - because all bikes are rated at constant 200w the only other things that would affect the range are battery size and 'mechanical efficiency' - which is tested in tests 1 and 2 especiallly.

obviously I intend to lead by example and have a go at this myslelf

 

[flecc]

As I posted before, the 200 watt rating is meaningless in performance terms, so comparison there is invalid. There are huge differences in the bikes on the market, some power curves being over double others, a more than 2 to 1 difference. Had you realised that peak motor powers on the market go from 272 watts at the bottom to 700 watts at the top? Any attempt at range assessment using 200 watts as a guide is worthless.

While I agree that a precision acceleration test with a fixed weight gives a roughly valid indication of hill climb ability, it's worthless to potential buyers. How do they relate the result to their own hill climb circumstance? In short, they can't.

The three tests I've just read do make sense for them. On the same 1 in 5 hill, one bike barely had the power for it's own weight, leaving the rider working as hard as if they'd ridden an ordinary bike. The second bike was better but still demanded quite a lot of work from the rider, while the third bike climbed it easily with only moderate assistance. That customers understand, your results they wouldn't.

The other problem is that people are notoriously bad at judging a flat road. Probably no-one can judge 2% from flat, and that makes a huge difference to performance. On the Torq theres a 4 mph differential between the up and down directions on 2%, on the Quando or Twist it's 3 mph differential. The results would again be worthless whenever someone inadvertently included a gentle gradient not perceivable by eye. The acceleration time difference between one on 2% up and another on 2% down would be very large.

I realise that you are trying to set up a system where we can publish results from our own bikes which can be compared, but the chance of actually achieving people sending in truly accurate timings on absolutely flat land is close to zero I think, especially given the weight adjustment procedure you propose.

In my case, I'd have to ride with the sack of bricks, bottles of water etc for six miles to get to the suggested factory estate, and even there the flat runs aren't precisely so, and are very short as much of it is on sloping ground.
.

 

[coops]

In my case I'd have to wear such a huge helium-filled balloon that increased wind resistance at any speed would invalidate any results .

Stuart.

 

[flecc]

Blame the chippie Stuart!
.

 

[coops]

Now if only I could get my bike sorted & get some decent exercise...

 

[Ian]

While I agree that a comparison of different bikes would be a great help to potential buyers I do have to agree with Flecc in that there are too many uncontrollable variables to make meaningful testing by users viable.

As Flecc has pointed out it would be very difficult to ensure the testing ground is perfectly flat, but in addition to that the biggest uncontrollable factors influencing results would be rolling resistance and air resistance, the former being influenced by road surface type and condition and the later by rider position and clothing. At a constant speed on a level road almost all of the power delivered to the drive wheel is expended overcoming the above and therefore any variation will heavily influence the results. In addition there is the effect of ambient air temperature on battery performance and, very significantly, the age of the battery to consider.

Even if it were possible to test and compare bikes under closely controlled conditions I'm not sure how relevant the results would be to the very variable conditions that exist in the real world. The tests may may reveal those bikes that are either very good or very bad, but I fear would be meaningless for the vast majority.

 

[chazpope]

yes I agree that judging the range would be difficult - however I disagree with your interpretation of 'power'. During an imaginary range test we can assume the motor will be running for an extended period of time at constant revs and the power that would be employed over the whole test would not be 'peak power' however high or low that may be. It would be the constant power that the motor is capable of - which for all purposes should be the nominal rated power of 200W. That quoted number must mean something after all; now whether it is quoted truly by manufacturers is another matter.

flecc I did not read the tests you are refering to - but I think if there is any pedalling involved it would be highly subjective and therefore worthless.

About 'judging the flat road' - I did propose a solution - do it in both directions and this would cancel the effect of any inperceptible incline and to a certain extent wind, don't you think ?

On the question of road surface unless it is a totally smooth brand new road I don't think you would get too much difference, on my previous bike I did get a more or less consistent 13mph computer reading of top speed on most city roads. The point is - on other bikes this speed would be higher or lower and this would be a very important consideration when buying a bike - especially when no such data is available at the moment.

FYI I live in Cambridge fenland where most roads are flat as is most of this country I find

 

[slimtim]

science teachers searching for "fair testing" ideas could start here!!

how about I get to ride all your bikes at the New Milton rally

while you all have a nice cup of tea. I will take the winning bike home as a prize...

teaching year 10 kids the 'way to do science' is supposed to get more real life contexts. This kind of discussion is exactly what we should be doing with them but it is difficult

 

[coops]

Hi chazpope

What Ian says is true: the nominal motor rating e.g. 200W/250W is the power the motor uses when running at maximum efficiency which occurs at the maximum loaded rpm of the motor i.e. top speed on a bike. Any extra load (work) put on the motor e.g. friction from knobbly/profiled tyres, air resistance from loose clothing, upright ride position, large frontal cross-sectional area etc. will increase the power output required of the motor and reduce efficiency too, so the exact power being used by the motor is unclear.

There are meters (watts-up, drainbrain) that gauge the power output, current, voltage etc. if you want to go to such lengths.

As flecc said earlier this thread...

What we really need is for the manufacturers/suppliers to give us three things the correct net power output in watts, the point at which peak torque occurs in mph, and the precise weight of each bike as supplied. Then a very accurate comparison chart would be easy to produce.

...if bike suppliers gave these very basic specifics, the overall bike performance would be very easy to calculate fairly accurately - probably much moreso than tests would reveal given the huge number of variables that need to be controlled for any results to be accurate or useful, and thats once you've decided what to test and not accounting for possible errors in measurements etc.: it really is a minefield!

I'm really of the view now that dealers and manufacturers simply must give this information: if not, how can anyone judge a bike's capabilities? Relying on impressions or reviews simply isn't good enough. I have to consider lack of information as either an indication of weakness of product for purpose or lack of expertise/professionalism, and view it exactly as I would lack of product support & development when problems are identified.

For now, its probably best to have a good idea of a bike's performance if possible e.g. from asking/reading here, then try it out, preferably on terrain similar to your proposed use.

Stuart.

 

[coops]

teaching year 10 kids the 'way to do science' is supposed to get more real life contexts. This kind of discussion is exactly what we should be doing with them but it is difficult

Just to add to my last post, the upside of the dificculty of getting useful results from "tests" is that, so long as the bikes behave the way they "should", it is really quite easy to estimate their performance from those few specifics.

It is far simpler to start with bike & motor performance specs & then guestimate, with good enough accuracy, how a given rider will affect performance, than to try do it the other way round i.e. estimate a bike & motor performance from tests which by nature will have many (unnecessary) variables reducing accuracy, making measuring much more difficult & harder work than it need be. Much better to isolate the unneeded variables e.g. road surface, gradient, tyre type, rider weight, frontal area, clothing, wind speed, temperature, battery age & charge state, bike condition, bearing friction, sticky brakes etc. etc. for clear, accurate results & then extrapolate to the other variables as needed.

Which is why we need those specifics.

ADDED: a motor torque graph plus bike weight is enough.

Stuart.

 

[Ian]

flecc I did not read the tests you are referring to - but I think if there is any pedalling involved it would be highly subjective and therefore worthless.

Some bikes (all, if the EU bureaucrats had their way), including the new Ezee Forza are pedelec only, ie they must be pedalled, so comparative testing would be difficult.

FYI I live in Cambridge fenland where most roads are flat as is most of this country I find

I agree the fens are mostly flat, plenty would disagree about that applying to most to the country though.

What Ian says is true: the nominal motor rating e.g. 200W/250W is the power the motor uses when running at maximum efficiency

Flecc said it actually Stuart, but it is certainly true so I don't mind

The problem is that by definition the motor is there to assist the rider, these bikes are not motorcycles and are meant to be pedaled(One of the fastest bikes available can hardly start without the rider pedaling), therefore testing without pedaling will not show the performance of the machine in the context which it is designed to be used and the results could be misleading. likewise, as you say, pedaling does introduce another imeasurable variable making the results meaningless.

If all bikes were designed to be the same then there may be some value in comparative testing but that is not the case, bikes are as different as the people who ride them and the terrains over which they are ridden.

 

[flecc]

And of course, some have to be pedalled with some effort put in for them to work at all, the Giant Twist series for example, based on Japanese law requiring rider and motor power are at 50/50. For those bikes the test would be meaningless.

 

[chazpope]

What Ian says is true: the nominal motor rating e.g. 200W/250W is the power the motor uses when running at maximum efficiency which occurs at the maximum loaded rpm of the motor i.e. top speed on a bike. Any extra load (work) put on the motor e.g. friction from knobbly/profiled tyres, air resistance from loose clothing, upright ride position, large frontal cross-sectional area etc. will increase the power output required of the motor and reduce efficiency too, so the exact power being used by the motor is unclear.

Yes Coops but this is the whole point - when the bike is constantly running at its max speed - then it is using its full power - and this is what its nominal power is - because through gearing, wheel size etc the bike is designed to do just that. When you turn the throttle from stop - the bike accelerates until the increasing load consisting of friction, gears wind etc becomes equal to the power that the motor can supply in a continuous (not momentary, peak etc) mode - then you have a ballance of power whereby the power supplied at the motor rotor is equal to the torque * rpm required to push the bike ahead at that particular speed.

so my point is - when the bike is running at its full/throttle/speed - then it is using it full and nominal power, give or take a few design faults. This assumes of course that the speed limit cut off point is not reached.

in other words if two 200W bikes of different make attain and maintain their respective top speed you will have a case where both motors are supplying the same 200W of power. The more efficient bike will be going at a higher speed but the power used will be the same.

so therefore - the bike that can attain a higher speed on the flat will also have a longer range - because (given the 2 batteries have the same capacity) the range-running time for both bikes will be the same, proportional to an identical capacity/power ratio.

hope that clears this academic point

Re whether or not we need these tests - it is the same as with a car - it has a nominal bhp rating but also a 0-60 and top speed rating that tells you about its mech and aero efficiency and overall performance - and those fugures do matter when choosing a car.

 

[Wisper Bikes]

Tests

In my humble opinion all tests under throttle only and otherwise are very important. These could easily be achieved if all bike retailers, agents and manufacturers were to get together and trial their bikes under strict supervision. We could all meet at a central point with a flat low friction surface - say Silverstone, and run all the bikes through their paces. Maybe we could ask a local university to run the trials for us. We could also get one rider to ride the bikes flat out over half a mile to get some results for pedelec mode.

Wisper would be delighted to take place in such trials but I dare say we would have butterflies all day!

I would imagine however that it would be tough to get every one to enter their bikes as there has to be a winner, and a looser at the end of the trial! I would imagine that some myths would be exposed.

There would of course be an expense but I would think that any one representing a particular bike would be happy to stump up £100 or so for the day.

I believe that tests like this should be carried out annually and maybe Pedelec would like to host them and give a stamp of approval to all bikes tested.

We for one would love to have fair comparisons done. If we weren't the best it would spur us on to improve for the next trials, I imagine everyone involved in the industry would feel the same. If anyone was particularly nervous then perhaps the first year's results could be kept private? Wouldn't that be a great way to ensure that we are all doing our utmost to ensure our bikes are as good as they can be?

If anyone else is interested please drop me a line and I will get it organised.

I imagine you have all heard that the Chinese government have just introduced a 4% export tax on electric bikes. This will be in place from the 1st July so expect price increases.

Best regards David

 

[Ian]

so my point is - when the bike is running at its full/throttle/speed - then it is using it full and nominal power, give or take a few design faults. This assumes of course that the speed limit cut off point is not reached.

in other words if two 200W bikes of different make attain and maintain their respective top speed you will have a case where both motors are supplying the same 200W of power. The more efficient bike will be going at a higher speed but the power used will be the same.

almost all bikes will easily reach the UK limit without reaching maximum motor power simply because it requires somewhat less than 200W to overcome friction, drag and rolling resistance at 15.5 mph. The limiting factor is usually the maximum rpm the motor can attain at the applied voltage. In cases where the motor can achieve higher rpm an electronic limiter is fitted. If one bike can attain a higher speed than another on the flat then that is not an indicator of better efficiency, just a difference in design or materials. An example would be if a motor had weaker permanent magnets then it's maximum rpm at a given voltage would be faster than the the same motor with stronger magnets, but it would be less efficient under most conditions.

The 200/250W rated power of e-bike motors has absolutely no meaning other than to satisfy the legislation and is derived at simply from the fact that an average of 200W is all that a motor needs to supply to propel a 15.5 mph limited bike on a typical journey. Most motors are capable of supplying 2 to 4 times the nominal 200W and will do so continuously in sustained hill climbing, but as stated above will only be delivering something like 170W at the legal limit on the flat.

 

[flecc]

in other words if two 200W bikes of different make attain and maintain their respective top speed you will have a case where both motors are supplying the same 200W of power. The more efficient bike will be going at a higher speed but the power used will be the same.

hope that clears this academic point

Clearing an academic point? You have to be joking.

A motor's nominal power of 200 watts is NOT it's peak power, and peak power is NOT a momentary peak.

For example, if the nominal 200 watt Powacycle Salibury and nominal 200 watt eZee Sprint accelerate to their respective top speeds, they aren't using 200 watts. The peak power of the Salisbury is 272 watts and the Sprint 500 watts, but both delivered at lower speeds.

I've pointed this out to you before and observed that the 200 watt legal figure is purely notional, not a scientific measure of motor power. You've chosen to ignore that which is fine is you wish to believe in a fantasy, but it's wrong to potentially mislead others in this way just to promote this performance measurement idea.

If your performance measurement idea has merit, it should stand on it's own, and not be supported by pseudo science.

P.S. I've just seen that Ian has got in before me with the same correction. I do hope you take note of it this time.
.

 

[flecc]

I imagine you have all heard that the Chinese government have just introduced a 4% export tax on electric bikes. This will be in place from the 1st July so expect price increases.

Best regards David

And there was us thinking, expect margin decreases.
.