Who wins in real world use?

[flecc]

ExtraEnergy say they fit a pedal containing a strain guage to provide torque sensing for their test data.

I'd previously seen that, but of course that crude method isn't anything like the same thing as Panasonic's uniquely sensitive magnetic flux sensor.

If the Panasonic torque sensors are sealed, presumably one could in principle get at the rider torque signal where it reaches the controller, since in the control this signal is used to adjust the power to the motor. Indeed if the system is exactly as you describe and in the Kalkhoff literature where I have read the controller arranges for a fixed factor of the measured rider torque to be added by the motor, then measuring motor power (Volts and amps with meter) would give a handle on rider power.

The controller is within the unit and the motor and battery connections are integrated into the unit, so there really is no access to any relevant interconnections without opening up the unit and making alterations. Such is the sensitivity of the magnetic flux torque system and the tiny signal variations it produces, any external interference would be unlikely to give usable results. That of course is why it's sealed unit which is not intended to be serviced, repairs are by whole unit replacement.

I happily concede that it is possible to greatly improve on what we have in the way of e-bike testing and reviewing at present. However, I still maintain that the wide variations in subjective impressions due to the way the differing systems work makes personal testing the only sensible way to choose. For example, many if not most will try a Panasonic unit powered bike and like it, but some definitely do not and we have one member who having bought one, hated the way the system worked so much that he couldn't wait to get rid of it. Those personal preference variations simply cannot be tested and reviewed for, and it's where such as family cars for example are completely different since they do not have such widely differing ways of working. Indeed the common complaint is that they are too much the same!
.

 

[Deleted member 4366]

I'd previously seen that, but of course that crude method isn't anything like the same thing as Panasonic's uniquely sensitive magnetic flux sensor.

Are you sure that their 180,000 Euro measuring system that was specially developed for accuracy is crude compared with the unique Panasonic flux sensor?

 

[flecc]

Are you sure that their 180,000 Euro measuring system that was specially developed for accuracy is crude compared with the unique Panasonic flux sensor?

It's only as good as the pedal attached pressure sensors, and that sort of measurement is crude in comparison. No matter how much the following attached gubbins cost, GIGO can still have relevance.

When I start to see greater relevance in their findings I'll start to consider whether their hype has any value.

I trust you've seen Franks comments above. He is a highly experienced and technically very competent supplier of many different e-bike systems from both Germany and China, so I am far from alone in my criticisms.
.

 

[10mph]

Points above duly noted. What a useful place this is for tapping into such a range of knowledge and experience, not to mention perhaps a few engineers able to do things for pennies .

Frank's proposal for independent engineers to carry out test is on the ball, but maybe his proposal for robot riders needs adapting:

I think it is time that some relly independent group of engineer´s create a real test for e-bike motors. No meaning, only fact´s!!
No different riders, only robot´s as riders with different power and different terrain, so that every user find himself in the test.

You could design a rig to test a motor on its own. I suppose you could even make a rig that would drive bike pedals and send the bike output to some sort of rolling road test device only the lines of what they use in MOT brake tests. However we must be talking big money for such a rig.

I still like the ExtraEnergy approach to getting rider watts from an instrumented pedal. It has the advantage a test rider can go from bike to bike and if he is given a live display of his pedal power test bikes in roughly the same way over standardised test courses. The actual bike performance then being measured simple GPS and a recording of rider power and battery drain.

When I search for instrumented pedals I cant find anything cheap. Even it looks messy and costly to get the signal from the pedal to a display and a recorder fixed on the bike.

So let us turn the measurement the other way round. Attach a false sole to the rider's shoes using thin load cells. Feed the signal with wires up the rider's trouser leg to a plug perhaps at his wrist which can be connected to his handlebar recording/dislay unit. No expensive data transmission problems. You probably could send the force sensor's mvolts up screened cables to be processed in a little box on the handlebars.

Calibrate by pressing the instrumented shoe on scales.

A problem is that I cant find a really thin compression load cell - the going thickness seems to be 2 cm and more - a bit of a clumsy add on to a shoe sole!

So engineers, how can this be implemented?

 

[Caph]

:

1) Which bike is fastest over the course?
2) Which is quietest under power?
3) Ignoring price and other machines not listed here, which would you choose and why?
4) Is there one which excels in hill climbing?

1) The Aurora is fastest by a country mile. It's got a 350W motor so there's no comparison. You are not going to average 20mph on anything else. Bear in mind though that it is illegal to ride on UK roads.

2) I wouldn't get too hung up on noise. At 20+mph you're going to hear so much wind noise that it will be mostly drowned out. Maybe on the slower bikes it may be more of an issue.

3) is too subjective to be of any use.

4) The hills is tricky. On normal hills the Aurora is obviously going to leave everything else standing. On really steep hills/mountains there is only one option - crank drive. I recently went to the mountains in Austria and I kid you not when I say that the local bike hire shop had about 30 ebikes for hire all lined up outside. Of course every single one of them was a crank drive. I've been kicking myself ever since for not hiring one and hitting the steepest mountain I could find. Instead I took a normal bike up the mountain for two hours at an average speed of 2.5 mph! It took me 10 minutes to get back to the hotel though.

To be honest, I think you'd be mad to pay over a grand for a hub motor bike. However, if you're going to go for crank drive and have good reason then I guess the extra cash could be excused.

 

[flecc]

To be honest, I think you'd be mad to pay over a grand for a hub motor bike.

From a purely utilitarian point of view, paying over a grand for any e-bike is madness when a moped can be cheaper and so much more effective on all counts. However, when all the personal factors are taken into account, like exercise for example, things can change.

I'm not so easily convinced that hub motor bikes should be dramatically cheaper. A Panasonic crank drive unit for example has an alloy shell, an internal controller, a single nylon gear wheel, torque sensor and sprocket drive output and integrated battery platform. A typical hub motor also has the alloy shell plus a whole set of orbital nylon gears, a controller elsewhere with the added case and mounting costs, a separate battery mounting platform cost and very much more external wiring. In addition, most hub motor bikes need pedalshaft sensor discs, brake cutouts and many have twistgrip throttles, all with added cost and complexity. The mainly pedelec crank units don't have any of these, just the internal torque sensor. There are no dramatic part cost differences there, and it's arguable that the intrinsic labour costs of building the hub motor bike with it's complex external interwiring and insulation for motor, controller, battery, brake cutouts, pedalshaft disc, throttle etc are much greater. Maybe it's the hub-motor bike that should be dearer?
.

 

[Orraman]

10 mph,
A place to start?

Bicycle Pedal Force Measurement

BikeDynamics - Bike Fitting Specialists - Fit Guidelines - Hips Knees and Ankles

Dave

 

[10mph]

Orraman: That load cell equipped pedal would certainly do the job. The signals during each pedal rotation would need some clever analysis to convert it to power, so not simple. I assume it is something like this that ExtraEnergy do.

That link to BikeDynamics is very interesting indeed. I had no idea that there were businesses doing this sort of thing. I can understand why serious cyclists might pay £100 for a dynamic evaluation and rider postion adjustment/fitting on their own bike using one of these rolling road type of devices fitted with a power meter.

I realise now the possibilities of the turbo trainers which Streethawk was pointing me towards in post no 20 above. I did not realise that top end models come with a power meter measurement. May be comparisons of different bikes on such machines could help pinpoint some of the data that the OP Indalo was looking for. Yes a bit of engineering could be required to make different electric bikes fit.