What makes an efficient on road e-bike?

[flecc]

If it's got disc brake facilities, no problem, and the chainline isn't a problem since the chainring can be shuffled about the spider, mounted outboard instead of inboard as at present, or with spacers to bring it still further out.

The 28" versus 26" wheel isn't much of an issue in the rear wheel, which is why I happily kept to that. It's in the front that things are more critical.

However, that torque problem with that high geared motor could be a problem, and I don't think it's worth discussing anything else until you definitely resolve the motor you want to use. Otherwise we'll just be going round in circles with "what if this, what if that" for many more postings, and I think we've done it to death already!
.

 

[coops]

Thanks flecc

Yes, agreed but it has been very useful & informative process - especially from the design point of view, though I suspect if it ever comes to fruition their may be many potential hitches or difficulties to resolve in the practical details... but thanks for all your help & advice, really! I got a bit too enthusiatic about that motor before realising its possible limitations: as we've said elsewhere, specifics are needed to judge performance properly, and I don't have any right now...

I'm still hopeful though of finding something suitable though.

What do you reckon, John? Do you have any more ideas? Coverting a normal road bike is still an option, but so much more difficult I think with battery & controller mounting problems, worries of frame stress due to weight etc. - not to be approached lightly I think, given the potential for cost & disaster aswell as hazards of the electrics . Still, I'd be interested to hear your opinion?

Stuart.

 

[JohnInStockie]

For me I am still undecided. I cant see any ebike out there that is efficient enough. For me, the best of the technology at the moment is the Wisper and Ezee offerings, which may be better for adapting than any other at the mo, if you were to bother at all.

Aside from that, I am still interested in the idea of adapting a normal touring bike, if one can be found cheap enough. Im sure they can take the stresses and the weight, and for me, the battery mounting issue, simply isnt an issue as I dont want the battery integral to the frame (just suspend it between all the bars in the centre of the frame, 1 or 2 batteries thick in an adapted bag, or put in panniers, or around the central post).

Then there are 2 remaining issues, 1 range, and 2 ridability, and these are in conflict.

If I want range, lets say I wanted 60 miles of assisted power. Then on my Twist (in pedelec mode always) I am going to need roughly 4 times the capacity that I have now, which simply rules out NiMH through sheer weight (circa 20 Kg). If I use LiPo, then that would be 2/3rds, or around 13Kg, more realistic. The frame for this would have to be substantial and strong. I think that a bike that could do this would be very useful, especially on long haul days out riding through the countryside.

If I want ridability, then I want the bike to act as a bike, and to be lithe meaning as little additional weight as possible, and efficient. This is where a touring style bike comes in, lighter motor perhaps and batteries with maybe only a range of say 12 miles (5Ah?) but the batteries need to be cheap as they will be replaced regularly, or go with the 80% LiPo's?

So it looks like its 2 different bikes with different characteristics perhaps.

Im afraid I have read and re-read the posts here and have come to the conclusion that we may need an electric motors 101, or rather 102. I just dont get it yet. Was the conclusion that ideals are a motor that can be adjusted, or a preconfigured motor?

Realistically, it looks to me that I should buy another bike (Wisper SE, Forte or Forza), and give up on the idea of building or converting anything to have an efficient e-bike, for now...

I think I will buy a non elec tourer though, but first I am going to need a bigger shed!

John

 

[flecc]

No, the motors can't be adjusted John. Some kit motor manufacturers offer versions geared for different wheel sizes so that each size runs to legal speed, but there's no after adjustment available. The controllers can sometimes be changed to deliver different performances within certain limits.

Your conclusion that two different bikes are needed to meet your requirements is right. Range and power mean weight, rideability as a bike means lightness.

Only lithium batteries and a moderate power motor stand a chance of getting near to your range requirement. Giant have tried for it with the new Twist 1 with twin Li-polymer batteries which they claim can do up to 100 miles on the flat in eco mode in ideal conditions, but on test in an area with only moderate hills it turned in about 32 miles, with an assessment that it might do 40 miles with care. If that's the best that very big company can manage with two large pannier batteries when range was their objective, you can see you have your work cut out to improve on it by much.
.

 

[JohnInStockie]

Flecc Thanks for your comments, unfortunately your summising of the situation is bang on. Frustrating isnt it

 

[flecc]

It certainly is John, reflecting the many years lack of real progress on battery technology, despite all the hype.
.

 

[flecc]

Further reply to John. If you took a Powacycle Salisbury in it's Li-poly guise, and bought two spare batteries, the all up weight with the spares in panniers would be 28 kilos only, 1 kilo less than an eZee Sprint for example, so very reasonable. Put M/Plus tyres on it and it would perform well as a bike.

And the total cost? Just under £1000 the lot, again matching the Sprint, so again very reasonable.

And the range? Well, combining reports from two sources, you could expect about 69/75 miles from the three batteries at least, and as much as 81 miles.

So all is not lost, there is a way without the bank manager dying of shock or you exhausting yourself in making a special.
.

 

[coops]

Hi John

If I want range, lets say I wanted 60 miles of assisted power.

Now then... how much assistance and in what sort of terrain, and at what sort of "average" speed...?!...

In principle two bikes with very different power output levels could give you the same range for the same energy use & more or less the same weight (motors may be slightly different weights), depending how you ride them, but the weaker one will always require you to supplement its power when its insufficient whereas with the stronger one you have a choice of how much assistance to give, so its more versatile .

Are you sure of those weight figures for the Twist: 20kg of NiMH for 60 miles range seems very steep? I think I could do 60 miles at 20mph+ on the Torq for that!!! I think you need to recalculate a bit, and so you can adjust the frame strength required accordingly too.

Also, I think positioning of battery weight is crucial to good bike feel & handling: if done well, the weight itself is hardly noticeable when riding I find, so you could almost forget its there.

ADDED: Like flecc said though, I don't think you'd get close to that sort of range without dabbling in lithium... reminds me of protanium now!

So it looks like its 2 different bikes with different characteristics perhaps.

I see no reason why one bike can't fulfil both tasks though, or am I missing something? One bike with 2 or more different battery setups! Like flecc said.

Stuart.

 

[flecc]

I see no reason why one bike can't fulfil both tasks though, or am I missing something? One bike with 2 or more different battery setups! Like flecc said.

Stuart.

Afraid so Stuart. Although I've given the Powacycle Salisbury solution for range at reasonable weight and cost, there is one thing missing for John.

At just over 290 watts peak power, it's a low powered bike.

To have a more powerful bike with a long range would drastically up both weight and cost, and incur the problem of lithium cutting out under load with a powerful motor. For example, a Torq with three lithium batteries would weigh well over 33 kilos and not be at all pleasant to cycle. What's more, it's hill climbing would be very compromised with all that weight, and it's range could still be as little as 45 miles in hilly country.

But two bikes would answer this, the Torq or similar for power riding, and the Salisbury for range for example, but there are other possible pairings.

The nearest to a universal bike yet is the Q bike, but the range is still at best 35 miles. However, with a second lithium battery it would weigh just under 30 kilos and could cover 60 miles while the batteries were new, though that would be almost down to 50 miles after about six to eight months and 45 miles at one year, so even that successful special doesn't measure up to what John needs.

The battery technology needs to move on quite a bit yet.
.

 

[coops]

Ok, lets brighten this up, its getting me pessimistic...

Yes battery technology needs to move on, if we're not to fork out a fortune every year or so for light batteries which might cut-out anyway... but given one goes "lithium" then...

ADDED:

I've given the Powacycle Salisbury solution for range at reasonable weight and cost

Except that wasn't the "one bike" I was thinking of (I was aware of the power limitations), which is why I said

two bikes with very different power output levels could give you the same range for the same energy use & more or less the same weight (motors may be slightly different weights), depending how you ride them

What about a bike with... the frame weight of Torq or Wisper; rear-mounted motor geared for 26/28" wheel with same efficiency as Sprint's, lithium battery capacity of Wisper 905se (13.8Ah 37V) in one battery - should be good for close to 40miles at 15mph with light pedalling, but not too heavy for "rideability"

And then either a second battery (at those capacities, both together would do at least 60 miles for over a year) or just a bit of exercise, to make up the range?!

Personally I think 60 miles, unless you're pedalling a good amount, is an unrealistic target to aim for unless you're flash for cash. I'd settle for one battery, 40 odd miles, which I could probably manage with a modified NiMH battery, and it'd last longer & be cheaper, though a bit heavier. Hopefully by the time its deteriorated, Li or something else will be more established & cheaper & more durable.

For now just carry your charger & top-up as required .

Stuart.

Stuart.

 

[flecc]

That Wisper with two of those batteries will be at least 31 kilos, so I'd question that pleasant to ride assumption Stuart. Most e-bikes aren't too good to ride anyway, judged by good normal bike standards.

By the way, we had a discussion on our rider powers recently. Today after a range check I rode the T bike with flat battery for three miles home, ending with the 14 % final hill. Ridden up that at 6 mph in the 35" low gear standing on the pedals but without undue effort, I calculated that climb accurately afterwards as a continuous 350 watts from me, allowing 30 watts for rolling resistance and friction etc. So I'm a touch fitter than I thought, and not too bad for 71.

NASA rating for a fit young adult for 5 minutes is 300 watts.
.

 

[coops]

That Wisper with two of those batteries will be at least 31 kilos, so I'd question that pleasant to ride assumption Stuart.

I said it should be not too heavy to ride with one of those batteries. I wouldn't like to try to ride it with two, and point taken on the weight: maybe one large and one small Lithium would be better (and more versatile for short trips) i.e. same total weight & capacity as your Q with 2 ezee li's, but same problem with deterioration after a year or so like you said... but then nor would I try to get 60 miles range! Unless I only needed 40miles or so and therefore would manage for longer on that rate of deterioration.

So, realistically, 60mile ranges between charges is out, but I think one bike can be enough to remain rideable & with up to 40miles or so range.

Nice going on the T-bike climb flecc! So the "get me home" gear now tried & tested & it works then I could have used one of those the other day: got only 18 miles range with hills & some slight offroad (honest!) speeding , not too bad, except I was 12miles from home at the time... still, my (relatively) "low" 44.8" helped a bit could've used a few hundred watts too though ...

Stuart.

 

[coops]

Another unresolved issue with the particular hub motor I had my eye on is that it seems to have a very fast rpm given it has 6:1 geared reduction already: seems to run at ~25mph in a 26" wheel at 37V, so I'm questioning how much low speed torque it will give even if one could be made internally regeared for 15-20mph ish. I haven't seen a torque graph so I just don't know.

I think finding a good & correctly geared motor may not be easy...

I think I miscalculated & the puma is still on the shortlist the speed was high because the power curve was so steep due to a high current limit at that speed of ~35A. A more usual & moderate limit of 15A would give likely ~22mph top speed, and further geared reduction from 6:1 to 7:1, same as the Torq, would also bring it into line with the Torq/f-series for speed, so the puma torque seems similar to the Torq's, at the same voltage & controller current limit .

It would still be better to get one with lower internal motor gearing though, especially if its to go in a 28" wheel, but the reduction necessary isn't as big as I thought, and the motor should have lots of torque even at 15A.

Stuart.

 

[coops]

Ok, I have a question on a big issue of efficiency/power economy: it has come up before I think, but not exactly in this context I think?

It seems there must be a flaw in the logic/maths here, but is it more efficient to go up hills faster, up to a sort of optimum speed for a given bike, if your bike can do that?

Its been said that climbing a hill at twice the speed takes about twice the power, but half the time, so the same energy is output overall. Rider energy makes only a small contribution in each case, and though it will be a larger proportion of the power needed at lower speed, and at higher speed more power is needed to overcome air & rolling resistance in addition to the climb, the higher speed climb runs much closer to the maximum efficiency revs of the motor, so much less energy is wasted - I'd estimate an efficiency saving of at least 10-15% (of total power) over climbing using less efficient lower speed torque; for some hills the rate of energy loss at both speeds would be comparable e.g. 700W output at 70% efficiency is 1000W input so 300W loss, 400W output @ 60% efficiency is 666W input so 266W loss, but the faster climb wastes only about half as much energy (because it takes only half the time), and that could more than compensate for the disproportionately higher power output required to overcome the extra resistance at speed mentioned above .

But that can't be right, can it? And you'd need a very high torque motor to climb some hills at those speeds, but surely it can't save power to use power?!!?

Stuart.

 

[flecc]

There you go again Stuart, tying yourself in knots! Try it more simply!

Yes, we've covered it before. The most efficient climb is done at the speed at which maximum torque/power occurs, which on a bike designed and therefore geared for 15.5 mph using a Hall effect motor is about 8 mph.

The slower you climb below that, the greater the proportion of wasted power, since the power used at full throttle is constant at it's maximum level from 0 mph to that maximum torque point mph.

The faster you climb above that ideal point, the more you must contribute, since both the motor output drops with increasing speed, and the higher speed demands more climb power, a double whammee.
.

 

[coops]

Yes, simply is probably best .

The most efficient climb is done at the speed at which maximum torque/power occurs

That's what I'm questioning, because that is the least efficient part of the power curve, except for below that speed as you say.

The faster you climb above that ideal point, the more you must contribute, since both the motor output drops with increasing speed, and the higher speed demands more climb power, a double whammee.

Not if the torque was high enough to climb faster, with the same or less help from you than at lower speed: that would require a rather high torque motor as I said, but the higher efficiency at the higher speed means lower energy losses for a shorter time and so overall a faster climb for the same or less energy output.

If I'm knotted or its flawed please point it out, but I can't see it!

Stuart.

 

[flecc]

That's what I'm questioning, because that is the least efficient part of the power curve, except for below that speed as you say.

No it's not really the least efficient. The difference in efficiency between that maximum torque point and thereafter to the point of maximum efficiency is very small. We are talking about climbing, so increasing the climb speed would demand more power with each 1 mph, which would more than offset the tiny difference in efficiency gain from the preceding 1 mph step. Therefore in work required terms, the most efficient is the point of maximum torque.

Not if the torque was high enough to climb faster, with the same or less help from you than at lower speed: that would require a rather high torque motor as I said, but the higher efficiency at the higher speed means lower energy losses for a shorter time and so overall a faster climb for the same or less energy output.

This is where you get into a tangle. You say, "if the torque was high enough to climb faster". Since climbing faster demands more torque/power, if the torque was enough for that higher speed, the first point would not have been the point of maximum torque after all, would it? It can't be in two places at once!
.

 

[Ian]

I calculated that climb accurately afterwards as a continuous 350 watts from me

Nice going Flecc, I reckon my max is about 250 watts, but not for long, and I'm a mere youth of 52.
I'm off to Cornwall tomorrow, so I may learn how to produce a few watts on them there hills.

 

[coops]

I just want to say that, in case anyone thinks these are questions just for the sake of it, I'm asking in order to work out the design of a reasonably efficient, high performance (within limits!) hub motor ebike .

The difference in efficiency between that maximum torque point and thereafter to the point of maximum efficiency is very small

This is where we need that information on motors: the only figures I've seen for vaguely comparable motors like crystalyte shows some variation but are as low as 50-60% at the point of peak torque & power, gradually increasing to around 80-85% efficiency @ maximum revs/speed. My estimate (previous 2 posts) of power loss vs speed was very conservative, based on only a 10% efficiency difference e.g. 60% to 70%: that is small, and it could be much more - up to 20-30% from those figures - that would increase the efficiency difference & the make the effect more pronounced.

...increasing the climb speed would demand more power with each 1 mph, which would more than offset the tiny difference in efficiency gain from the preceding 1 mph step. Therefore in work required terms, the most efficient is the point of maximum torque.

But I showed the (very conservatively guestimated) figures & explained in my last two posts that it looks to be the other way around: the efficiency gain is more than the extra energy needed to do the climb faster i.e. more energy is needed, but even more energy is saved! Please check the figures and correct me if I'm wrong, flecc!

This is where you get into a tangle. You say, "if the torque was high enough to climb faster". Since climbing faster demands more torque/power, if the torque was enough for that higher speed, the first point would not have been the point of maximum torque after all, would it? It can't be in two places at once!

Yes, clearly not, unless the power curve has two "camel humps"! No, I didn't mean two peaks, rather I meant that if the motor's torque/power curve was steep and high enough, the motor would have sufficient torque/power to climb the hill faster and rather more energy efficiently, before the speed drops to the peak torque point i.e. at a point on the power curve between point of min power/max revs/max efficiency and the point of max torque/power/"intermediate" efficiency & speed.

It still seems like something for nothing, and shouldn't be, and while the efficiency difference may be smaller for good motors, at the power levels we're talking about (~500W+) and the duration difference to climb the slope, it makes a significant difference.

I hope I've expressed it more clearly now? I still don't see anything incorrect, but the figures could be a source of error: the efficiency for even a high torque motor will be below peak efficiency on a climb, so the efficiency difference between the climb speed & maximum torque speed might not be as much as I estimate; I'll check my sums before I say any more . If you can see any errors please point them out .

Stuart.

 

[flecc]

If the torque curve is steep and high enough as you say, then the fall off the other side of the hump is just as steep, the Powabyke motor beinga classic example of this. So you wouldn't get that "carry over" power.

I prefer to go by a generic two line consumption and power graph. As I've posted before, these motors are bound by the laws of physics and design is correspondingly confined. I'm a sceptic where manufacturer's data is concerned, since there's far more room for manouvre in the presentation of statistics than there is in the actual designs, something they freely exploit. The claims of 97% efficiency and the like amply demonstrate that.

Yes, theoretical efficiency does increase markedly if you view it in isolation, but it's only when there is a product, an output, that efficiency has any meaning.

Here we are speaking of hill climbing. I don't care whose motor it is or what any manufacturer says, there's a point at the maximum torque on a power curve at which a bike motor has a gradient climb limit or maximum. It will not climb faster on that gradient limit, and that's the maximum climb efficiency.

The gradient has to be reduced quite a lot for a faster climb speed to be attained by the bike, the motor having a natural tendency to settle for the maximum torque point, and when a faster climb speed is on a less steep hill, the lesser steepness must form part of the climb efficiency calculation. That will set back the result to a lower efficiency than that at the point of maximum torque.

If the lesser gradient isn't taken into account, the calculation isn't work/output related and is therefore meaningless theory.

Even if a higher efficiency could be shown on a climb at a speed above the maximum torque point, I still don't see where all this leads. If the bike can climb a given hill at a higher speed with the throttle open, it will.

The alternative is riding around closely watching a Brain Drain device and a gradient indicator to constantly monitor the current and climb relationships to save a few milliamps or gain half a mile an hour. I prefer to enjoy my riding, and use whatever watts are needed to do that.
.