What makes an efficient on road e-bike?

[coops]

Thanks flecc, well you can guess what's coming next ...

I'm still concerned about trying to make a battery pack, and fix it to a standard frame which, even then, is probably not designed to take the weight behind seat-tube for instance (panniers would be ok, but I'd rather avoid that if possible) - especially after what prState said about frame strengthening & weight saving on normal vs electric bike frames.

ADDED: Especially if NiMH is the only realistic option, since its heavier than Li. Do you think the ezee cases could be extended to take a higher capacity of NiMH batteries, e.g. 36V 12-14AH ?? 50% extra weight, but 50% extra capacity! ~ 432-496Wh and would the frame take the weight?
EDIT: Or higher voltage at current Ah e.g. 48V+ 9Ah? thats ~1/3 extra weight & range...

What about a hub motor geared for 26" wheel & fitted into the Torq frame as rear wheel, keeping the front 28"? would spoil the looks a bit, I admit, but practicality before fashion! no battery issues, but would ground clearance be ok you reckon? I tried to judge it, and seems not much lower with only rear wheel reduced to 26"?

I suppose the Torq motor could even be used, except that the motor gearing improvement would be small, and would be difficult to engineer as flecc has said. Rear 24" might be stretching it (!) ground clearance would be as low as 2x26" and though the gearing payoff would be better, the technical trickery involved remains the same .

Stuart.

 

[flecc]

The only way to get that sort of capacity is by using F cells, and there's no chance of getting those into the eZee case. Even increasing it for a couple of extra cells as I've done gets it very close to the saddle. The same goes for voltage increase, no chance at all. The voltage could be got by using C cells with 5 Ah, but that's severely limiting. The rear frame probably isn't strong enough for a motor and F cells, even if they could be used.

F cells are 50% longer (91mm) than D cells (61mm) and heavier, 246 gms against 155 gms typically for NiMh. Mounted in the eZee vertical form that the case is designed for means the battery would be half again the height, impossible to accomodate.

I don't think a 26" rear wheel would spoil the Torq's looks, the carrier disguising the wheel size to a fair extent. The same goes for a 24", and I'd previously checked that ground clearance is viable for either, though with shorter cranks with the 24". However, dropping only the rear alters the steering head angle, so handling would change slightly, basically the steering would become a little slower, but not problematic.

Using the Torq motor in the rear is out for anyone without the necessary tools and equipment and the experience and knowledge of what was wanted at each stage and how to achieve it. That simply can't be relayed via a forum.
.

 

[coops]

Thanks flecc

My saddle height is rather high, and there has always seemed a large free space above the top of the battery - about 16cm without saddlebag - less than half the height again of the case, but more than 1/3, so any chance I could add D cells for extra voltage, say up to ~ 48V, for use with a 48V controller & motor, & do you think 1/3 extra weight be excessive?

If that weight/size is an issue, then it seems the only plausible way to increase the capacity within the case alone is with Lithium... I'm looking into that more, does look complex - the ezee batteries look very intricate! But they are all in series aren't they? I'll have to look into it more - have read one opinion that Li-manganese is more suited for evehicles, Li-nanophosphate for RC model use (though not why, as yet..!) - but it looks a possibility to fit over 37V of Li-manganese batteries in the case, if extended, & if I can sort the electronic circuitry & charging... we'll see .

On the frame ground clearance, I guess the sort of terrain it'll be used on won't be too uneven, so that should be ok & shorter cranks is an option as you say: I'd be happy with 26", especially given the change of steering geometry; would 24" rear be the same overall drop in ground clearance as 26" front & rear? Though that would need 26" front forks, plus a new front wheel rim, I guess, so not worth it?

Do you think that a 26" rear motor is a better idea than 28"? Its still just an idea, but given I could use the extra torque & also am unable to rear mount the torq motor as in the T-bike, is it on the right lines would you say?

Stuart.

 

[flecc]

You might just scrape in the extra D cells for 48 volts, it adds one third to the height of course.

You won't sort the eZee battery Li electronics, or those on any other e-bike into a viable larger size, but you could adopt the kraeuterbutter parts solution to build a system using all those separate charging and control modules. I wouldn't risk the money on that though, and it would be a mess.

The ground clearance difference between 28" / 24" and a pair of 26" is small, a touch less at the cranks with the former.

As already said, 26" is preferable for a robust motor wheel build.

To summarise, I don't think this high voltage approach makes sense. It's the sort of thing constantly done by the performance chasers like Mark at Team Hybrid and many in the USA who are obsessed with getting a bike to go very fast for no apparent reason. They don't ever use these bikes, just film one demo and move onto the next harebrained scheme. Basically toys for boys, and it's not an area I've ever entertained. I think the only reason for a project is to create a practical working bike, fit for purpose and to be used over time, and high voltage isn't sensible for that, given the bulk and weight constraints. An electric bike should remain a bike, pleasant to ride, and heavy bulky solutions are never that.
.

 

[coops]

Thanks flecc

I think what I said on Li in my last post looks very unintentionally misleading, especially the reference to ezee Li's & putting over 37V in one case: I have never considered making a larger capacity Li by adding to the ezee Li . I think I'll forget the "self-build" route too - too risky, messy & pricey, as you say. If I were to try a Li/other (in time) it would most likely be a purpose built ebike one; they are rather pricey and still a risk involved, but look less messy .

It doesn't need to be 48V, just some extra batteries for a V boost, and I wouldn't want to sacrifice too much of the lighter "bike" feel for a "heavyweight" performance gain, if possible: I wanted to see if the Torq could be used as a basis for an efficient, performance bike with a motor geared with more torque for hills for the weightier rider, which could be fitted by anyone, and extra voltage for a little bit of speed on top of 15-16mph and/or some extra range. The motor & batteries would add some weight (though Lithium, if used, would weigh ~2/3 of NiMH & offset some of the motor weight) but hopefully not an excessive amount, say 2-3kg between them?

Adding to the battery seems not strictly necessary: one could keep it at 36V, get a motor geared for the top speed you want, say 20mph, and then rely on power alone - a higher current limit - for low speed torque, a practice with bikes in the U.S. and elsewhere too, it seems, but that would tend to eat up the range in hilly terrain and be unsatisfactory, so I'm trying to aim for a compromise between added weight and low speed torque at minimal expense to range or speed.

Ironically, a 26" low profile tyre on the rear is barely 25", if that, and might work well with a rear-fitted Torq motor if one had the skills to do it?! (Did you keep with 28" for convenience, and because narrower tyres gives better motor gearing so the necessity was not there: all the small changes add up?!)

I appreciate your advice & comments, and above all I definitely want a "commuting" type vehicle, not a toy! This is still brainstorming & an evaluation of how one can best or most easily achieve that, either by modding an existing ebike or converting a normal one: I'm still very inclined to agree that the former is a better start point; just need to find the right candidate & parts for a (minor) job which can be done by a less skilled person .

John, how's your thoughts coming along? Any progress?

Stuart.

 

[flecc]

It's already on the market Stuart, and at under £1400 it's cheaper than you could build yourself.

It's the eZee Forte (or Forza)

The Velovision test in very hilly Sheffield returned 15 miles range, same as the Torq in hilly territory, and from the same 36 volt battery. But it speeds to 20 plus derestricted like the Torq and it had no trouble zooming up a 1 in 5 (20%) in Sheffield, pedalled by a vicar, so exactly what you are asking for, a Torq that climbs hills easily.

Problem solved!

The eZee approach with the F series is a US style motor restricted down to UK levels of course.
.

 

[flecc]

Further to the above, I kept to 28" on the T bike simply because I was making the bike much lighter and at the same time drastically reducing the rolling resistance. Those two would easily give the hill climb increase wanted, so any gearing reduction wasn't necessary. The extra cells were just an added insurance against voltage drop, but not strictly necessary.
.

 

[coops]

Thanks for the information flecc,

the eZee Forte ...returned 15 miles range, same as the Torq in hilly territory, and from the same 36 volt battery.

Sounds useful, though I'd prefer better range and hillclimbing than the Torq in hilly areas, which seems very possible with motor gearing matched to the wheelsize, and to keep some of the speed (20mph is ok), which would then require boosting the voltage in proportion to wanted speed increase: that alone would incur a small weight penalty with a "safe" NiMH option, or could even be lighter than existing 36V 9Ah NiMH if lithium is used, but at a cost & risk .

Is that correct?

Also, in terms of efficiency, the F-series are front motor & suspension, and wider tyres; though I think the sus works quite well, I'm not sure if narrower tyres would be an option, but that would also (I think) help bring the motor gearing into line with wheelsize aswell as lower rolling-resistance to boost efficiency. The comfy sus seat would have to go too.

Its no compromise now, this is getting serious .

at under £1400 it's cheaper than you could build yourself.

I haven't even begun to fully explore the feasibility nor the total cost of a Torq mod, but you might be able to build a second, very light ebike for the added cost of a battery, and a suitable frame, if you don't have one already .

Stuart.

 

[flecc]

I'm not you're understanding the gearing Stuart. You keep coming back to matching the gearing to wheelsize on the Torq, but if you did that you wouldn't have 20 mph, you'd be back to the legal limit. In other words, the gearing is matched if you want 20 mph.

To repeat, it's motor is normally geared in a 20" wheel for 15 mph and good hill climbing, or geared up 40% or so for 20/22 mph and poor hill climbing, and that's it, you can't have both.

If you're thinking extra volts on the lower geared version to get back to 20 mph that way, there's a number of issues. For example, the need for a replacement higher voltage controller, and the motor excess temperature, it already runs hand burningly hot in hot summer weather and might not stand more.

The only sensible approach for a fully equipped bike with a fairly heavy rider for 20 mph and better hill climbing is a more powerful motor, as in the F series. As for the range, that reflects the hills and that power, they can't be climbed for free, and you'd have the same range if you provided the same adequate power on a home build. Short range is just a powerful e-bike fact of life, and the answer to that is to carry a second battery when necessary, or permanently reduce the performance by having a very large and heavy battery all the time.
.

 

[coops]

I think I'm clear on the gearing issue, and I understand what you've said about the relation of wheelsize, top speed & torque with the T/Q motor, and the limitations on either top speed or "low end" torque depending on wheelsize, for that motor with its current power source (36V 15A) and given internal geared reduction of 7:1 or 260-280rpm or thereabouts.

What I'm seeking is a way to be more efficient on hills by applying power using the best balance of voltage & current, to an appropriately internally geared hub motor for, say, max speed of 15-16mph in a 26" wheel at 36V, then...

If you're thinking extra volts on the lower geared version to get back to 20 mph that way...

Yes, I was . I asked about this in the hub motor thread , and your reply a few posts on.

...there's a number of issues. For example, the need for a replacement higher voltage controller, and the motor excess temperature, it already runs hand burningly hot in hot summer weather and might not stand more.

Fitting a new motor to the Torq rear (I can't modify the existing Torq motor for rear wheel use) would require a new controller anyway (couldn't use the existing one). Hopefully the new motor should be cooler running and allow a moderate voltage increase, and I'd hope current could be limited to 15-20A maximum.

The only sensible approach for a fully equipped bike with a fairly heavy rider for 20 mph and better hill climbing is a more powerful motor, as in the F series. As for the range, that reflects the hills and that power, they can't be climbed for free, and you'd have the same range if you provided the same adequate power on a home build.

A more powerful motor is needed, yes certainly, but would a lower internally geared motor be more efficient than a higher internally geared one, on hills, for the same total power output (but at higher voltage and lower current) and at the same speed, so would give a better range? As far as I can see, the f-series motors are similarly internally geared to the Torq at around 260-280+ rpm, aswell as similar wheelsize, for similar speed i.e. ~20-22mph+ at max revs, but have a higher current limit (20A vs 15A) for torque which makes them more powerful as you say, but geared very similarly from what I can see?

What if a motor was internally geared for, say, 15-16mph at max revs, in 26" wheels, and at the same 36V, but had a 15A limit (c.f. the Sprint's internal motor gearing ~ 200rpm ~15mph @ 36V 15A limit for good torque) for efficient low speed torque & hill-climbing & range, then the rpm scaled up towards torq/F revs of ~260-280 rpm by voltage increase, not a gearing increase!? Would you get better range in mixed hilly/flat terrain? My apologies if I'm not putting this very clearly, its hard to put into words! I think you know exactly what I'm saying, but as to whether it gives better range I'm unsure... maybe some tradeoff with power consumption on the flat too?

Stuart.

 

[coops]

I think I can put it more succinctly: the Q-bike and the Sprint both have motors that are internally geared such that, in their respective wheelsizes (20 & 26"), they both give good range in flat and are efficient in mixed terrain.

If a motor identically internally geared & efficient as the Sprint & also in a 26" wheel, but run with a higher voltage battery & controller (same current limit of 15A though), wouldn't it have more or less the same performance - efficiency & range - as the Sprint in the same terrain at the same speeds, but also have a speed "boost" option if the throttle was opened further (beyond the Sprint's max speed) & the extra voltage utilised? The higher speed requires more power and range would then decrease, as you'd expect, but I think there is only a possible slight efficiency decrease at that higher speed & voltage compared to a likely bigger loss of efficiency due to running a high internally geared motor at lower speed & using high current for torque rather less efficiently than gearing would do the job... [ADDED:] so overall wouldn't the lower internally geared motor give better range, because its more efficient at the points when the motor uses most power and is normally least efficient (so for any added efficiency losses there, the energy wasted is proportionally much higher e.g. 75% efficient at 600W gross peak output is 450W net "useful" power, while for a 60% efficient motor, the same net power requires a 750W gross power output from the motor, thats 150W or 25% more input for the same result and so range would be reduced proportionally at that level of power usage.

Stuart.

 

[flecc]

In short, yes.

But I doubt if the difference would be worth very much. A bike overgeared like the Torq has the effect of encouraging the rider to contribute. As you know, when it accelerates, it's sounds and feels as though it wants the rider working with it. Indeed, it virtually demands it.

Conversely, a bike like that you propose would not have the same factor and would have more of a moped character, and indeed, the F series are reportedly like that, in contrast to the Torq. It's less likely the rider contribution would be as effective and so the combined performance outcome could be very similar.

This problem of assessing these pedal assist bikes crops up all the time of course, the unknown of the rider making them impossible to classify with any consistency.

By the way, I'm back onto two extra cells in the radical battery now. Not much speed gain due to the controller limiting, but decidely stronger when hill climbing, not so much seen as speed as felt in it being obviously more effortless.
.

 

[coops]

In short, yes.

Now thats concise! Why can't I do that?!

This is still on topis, isn't it John? We're talking efficiency of motor + bike+ rider combination: they're inter-related, so should always be spoken of together...

I think the degree to which rider input helps depends a lot on the terrain: the common figures of e.g. 20 miles motor only vs 30 miles with pedalling apply mostly to the flat, and the 3:2 ratio would markedly decrease (aswell as the range) in hills, where the rider input is significantly less would you say?

To quantify a little to illustrate, if total power needed is say 160W on the flat at 15mph, at 80% efficiency the motor uses 125% of that, so 200W, to produce that net output, but if the rider pedals only 40W worth with light pedalling, the motor only need output 120W & so the total motor power used drops to 125% of that, so 150W: so with pedalling uses only 75% of the power used without pedalling, and would give 4/3 or 1.33 times the range.

On hills the motor power output can be much higher, up to ~500W or more, and the motor's efficiency is lowest at peak torque point and below, so rather less than 80% efficiency would be expected and near to 400W output from a 500W peak output motor would be good going; clearly from a numbers point of view the same 40W output (in lower gear) will make a much smaller saving on the power use of the motor and so increase range a proportionately small amount.

Conversely, a bike like that you propose would not have the same factor and would have more of a moped character, and indeed, the F series are reportedly like that, in contrast to the Torq. It's less likely the rider contribution would be as effective and so the combined performance outcome could be very similar.

Since the bike I'm proposing would be geared like a Sprint, or maybe if the internal motor gearing is a bit higher, between a Sprint and a Torq, it should behave somewhat like that. If the current limit was increased to give a higher torque like the f-series, it may feel more moped-like as you say, but that would not be necessary for hill-climbing for my proposal as it appears to be for the F's gearing, and so it need not have the same higher power consumption and reduced range.

I've not ridden a Sprint though, or a bike with similar internal motor gearing, so I don't know how much rider input they "demand", but I'm confident they can be ridden with useful rider input for power economy, especially with experience .

I'd quite like a bike I can ride in the heat, even uphill, without the requirement to pedal if I don't want to, even at a small cost to range, since I struggle in the heat anyway, and it would be a shame to miss out on good, dry weather and automatic air-conditioning while mobile :-D. Of course, I'm probably just too impatient to lose weight, but I'd like to get the benefit of an ebike preferably within the next few years or so

That's good news on the T-bike flecc, thanks for the update, I meant to ask but I've been a bit preoccupied...!

Stuart.

 

[flecc]

You didn't have to expand to explain that, since it's very much what I live with. Yes, the rider input in a hilly area makes very much less difference often. For example, the Torq ridden by A to B in a flat area will only do 15 miles with minimum rider assistance, but doubles that to about 30 miles or more if the rider makes an effort.

In my hilly area, I also get around 15 miles with little assistance, but can only improve that to about 17/18 miles by putting in a huge effort, a very poor payback, so I don't bother.

That's the overgeared bike example of course. In this hilly area, the same motor in the normally geared Quando and Q bikes gave 15 miles unassisted in the Quando when the battery was new, and 31 miles with moderate assistance in the Q bike (35 miles if the battery was new). The Q bike is clearly more efficient than either the Torq or Quando, and tends to cancel the hilly area difference.
.

 

[coops]

You didn't have to expand to explain that, since it's very much what I live with.

Yes, I'm aware of that, but I just thought I'd post it for general information - and if your experience tallies with what I said, then I'm more confident of my grasp of the issues!

What do you think of the comparison I made of the performance, especially range, of the bike I'm proposing (I think I'll call it the Italian job!), with the Sprint, Torq and f-series?

Do you agree it would be more like the Sprint, in motor performance terms (with similar internal motor gearing & 15A current limit) up to 15mph at least, than an f-series bike?

Am I correct in thinking that the Sprint has a good range in mixed terrain (comparable to the Q? especially a Sprint on m+ ), because I can't recall that information from this forum?

Stuart.

 

[flecc]

Yes, correct and agreed on all points, The A to B test confirmed the Sprint's range is good at up to 27 miles at best, so it may get fairly close to the Q bike if on M+. I doubt it would match it though, hub gear efficiency lower and riding position not ideal.
.

 

[coops]

Yes, correct and agreed on all points

Thanks flecc, I needed confirmation that I'd sussed it all properly .

Is that 27mile range "at best" on fairly flat ground or any hills/slopes? If its flat, then I'm still unclear on the Sprint's range in mixed terrain.

One other question I need to find an answer for is whether throttling the motor to over 36V for speed above 15-16mph would be much, if any, less efficient than having a higher internally geared motor at those speeds, though I don't think it would be: from torque graphs of similar motors like the crystalytes it looks like the motors can run close to peak efficiency at higher voltages, even at the power output needed for 18-20mph and above (~300-400W+) if required .

Stuart.

 

[coops]

Efficient on-road ebike idea

It could be more sensible to buy a basically suitable electric bike second hand, then fit the things you want to that.

Flecc - Do you know of a production e-bike that would start out with the characteristics for a good basis as mentioned above, bearing in mind that most are unable to fabricate parts, and would have to be looking to replace inferior parts with more suitable ones?

My slightly radical idea (its only an idea) for an efficient, hopefully doable (by non-engineers only minor assembly/wiring) & on-road ebike then:

Efficiency & Performance points:

Capable of steep hills (even for us weightier commuters!).
Good power economy at moderate speed, even in mixed terrain; good range.
Taut, lightweight purpose-built Aluminium alloy frame;
Low overall weight & integrated battery.
Narrow, high pressure marathon plus tyres for low rolling resistance.
No front, rear or seat suspension .
No hub gears.

Rough idea: (needs some details fleshing out)

Start with a new or 2nd hand Torq & battery;
Get a good rear hub motor kit: motor geared for ~200-230rpm (can be more) spoked into 26" double-wall aluminium wheel rim, controller & throttle etc. and a suitable freewheel (up to 6-speed should be ok - see flecc's recommendations).
Disconnect & remove Torq front & rear wheel & motor.
Check rear hub motor & freewheel for fit to rear dropouts (should fit ok they're 135mm) & fit.

(I'm skipping the details so far, its only a schematic, and the details on the next bit are still very hazy to me!...)

Replace & rewire controller & throttle with existing battery...
If higher voltage battery wanted (for higher top speed) & space between battery & saddle allows, modify battery case (extend length) to either fit extra NiMH D-cells or a large Li battery (bought & soldered/wired earlier! )
Decide on, buy & fit 2nd set of brakes - options: rear disc, 2xfront V- etc.
Remove freehub from original 28" Torq rear wheel & fit to front forks.
Check ground clearance (rear is now 1-1.5" lower) and assess if cranks require shortening.

Anything else?

Choose motor rpm & drivetrain gearing according to your choice of rear tyre: don't assume 26" diameter as low profile tyres will be nearer 25", significantly altering gearing of motor & drivetrain!

Look at leftover parts (Torq hub motor, controller & throttle etc.) & consider a DIY job to make a front hub motor "baby" Torq in suitable frame... motor will need respoking into suitable size wheel rim & a battery will be required.

What do you reckon?

Stuart.

 

[flecc]

Comment on both posts above:

The A to B Sprint test was in the Castle Cary area and included their "mountain course". I wouldn't give too much credence to that, Castle Cary is on the Somerset Levels, and the Mendip Hills are only just about the edge of range on a return journey, so no really severe hills likely.

Your "rough idea" is fine, the only difficult point will be the rear brake in the Torq frame. Life will be much easier if you can get an appropriately geared rear motor for a 700c wheel. These hub motors don't generally accept a standard disc brake and there's not enough frame width for that addition. I've seen a large disc Heinzmann adaptation, but that was front wheel. If I'd adopted one I'd have cut out the disc centre and re-drilled and re-mounted it to the motor wall, but that's not an easy job and will transfer braking heat direct into the motor.

You won't be able to go too thin a section rear tyre with a rear motor for the sake of the structural health of the motor, spokes etc, I'm using 35 mm, so wheel diameter won't change too much, still around 25.5" on a nominal 26" wheel.

A very small lift in volts to overcome voltage drop under load might be ok, but I won't comment on 48 volt and other considerably higher voltage options, not an area of interest and I don't approve as you know.
.

 

[coops]

Thanks flecc

So we're not much the wiser re: Sprint range in mixed terrain then... I'll be surprised if the information isn't on this forum somewhere, but just don't recall it!

I thought rear brakes might be a little awkward: that hub motor I had in mind does have a disc brake ready "fixture point", though frame width is an issue as you say, but there may be other, more intractable problems: for one, the chainline on my Torq seems to be central on the rearwheel unlike reports I've seen. I haven't looked closely yet, so I can't tell if its just down to dishing of the rear wheel or if some more radical improvement has been made to it, but might that scupper a rear hub plan? Or could the rear wheel with hub motor be dished to allow for it, in which case would hub motor width be critical to not overweaken the wheel, especially if its a 700C/28"?

Life will be much easier if you can get an appropriately geared rear motor for a 700c wheel.

If it can be geared for 26", the gearing for 700c would be little different, so that shouldn't be a problem, but given that you said a 26" wheel build is preferable & more robust & "28" wheels using cycle components with a powerful and high torque motor is probably undesirable", do you mean 28" is ok if the torque is limited?

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...

As for the voltage, that depends on a number of factors: the battery weight, how high the current limit is set for low speed torque (at the expense of range), what the internal motor gearing is chosen to be & how much higher "offroad" speed is wanted if current limit & gearing are kept low. One could easily keep either or both of the latter two high and have enough power at 36/37V with lighter batteries, but at the expense of either range & efficiency due to higher current, lower capacity & unsuitable gearing for steep hills, or top speed.

I agree that "an electric bike should remain a bike, pleasant to ride, and heavy bulky solutions are never that" and I'm sure I'd regret making a bulky machine that was neither bike nor motorbike but lost between somewhere so I'll probably try to tone down the weight/speed/power as much as possible.

I'll just have to be patient, exercise more & improve performance in the best way: lose weight! oh, & keep away from the tempting waft of chip shops as I sail past... .

Stuart.