How much power do I need?

[mxh]

I've been riding a E8000 based emtb for the last few years and am thinking of getting a new bike. Ideally I'd like one of the new lighter bikes, but don't really want to have less power than I have at the moment (probably a common request!)

The graph below I got from this article - e-MTB and STEPS Riding Characteristic assist levels explained | BetterShifting.com

I've got both my Boost and Trail modes set to Medium, so I'm getting 200% and 90% Assist Ratio respectively. But what does this actually mean in the real world? 200%, 90% of what exactly?

And how would the current crop of lighter bikes (eg the 60NM EP8 RS in the Orbea Rise with Boost set to Max) compare in feel on the trail to my current setup - considering that I'm not using the 'High' setting in Boost, so theoretically not using the maximum assist that the motor could give.

 

[Rubinstein]

Only way your gonna find out for sure is get a test ride. I was in the same boat this year and very nearly bought a rise but instead rented one for the day and although the bike was very capable it didn’t have the fun factor after coming off a 80nm bike. I think I could drop to 70-75, but 60 is to low for me. Stats are great for comparison but nothing beats being in the saddle for a few hours to make your mind up.

 

[mxh]

Only way your gonna find out for sure is get a test ride.

Yes, I agree.

But putting aside test rides, I'd still like to understand what that graph is trying to tell me. I could read it as I'm only using 2/3 of the assistance available to me (200% rather than 300%), but without knowing what Assist Ration actually means in the real world, I'm not sure what to make of it.

 

[steve_sordy]

@mxh emtb motors are rated at 250Watts of power. In fact, "rated" is the wrong word, "restricted" would be more accurate! That restriction is a requirement in order to get the ebike to be still considered a pedal bike and not require tax, registration and insurance like mopeds, scooters and motorbikes do. But there is a loophole! I have no idea whether the loophole was engineered in there by some clever bureaucrat or just an oversight. It's all to do with how the 250W is measured. I am happy to be corrected but I'm fairly sure It is the average rating over two hours. That allows the motor to deliver multiple times that rating, but for a proportionately shorter time. So, a motor with a 500W-hr battery will be able to deliver a steady 250W for 2 hours, but 500w for only one hour.

Therefore, in answer to your question of "200%, 90% of what exactly?" It will be 500W and 225W.

The motor's clever software will use that power to deliver useful torque to you, which is affected by your cadence and what effort you are putting in. Different brand and model motors have different software and each in turn can be modified by the user to different extents (before hacking). That is why it is worth checking out bikes that have different motors before you make a buying decision.

I can offer no firm guidance on what the lower torque bikes will feel like, but I did hire a Kellys bike earlier this year that was offering 60Nm (an e7000 I believe). Compared to my current 85Nm EP8 equipped bike, I was expecting poor performance. But what I got was an amazing bike! I never once felt a lack of urge from the bike. Maybe I was working a bit harder as it was a hot day (South of France), but I never felt that I couldn't get up any climb because the bike lacked anything. The usual advice applies: You must test ride them if at all possible.

LATER EDIT: I withdraw some of this post as it is most probably wrong! See later Post #10

 

[mxh]

Therefore, in answer to your question of "200%, 90% of what exactly?" It will be 500W and 225W.

OK, thanks - that makes sense.

Different brand and model motors have different software

So does that mean that, for example, my 200%, 70NM setting (which you've clarified means 500W) can't be compared to similar figures from a different motor to make any meaningful comparison?

The usual advice applies: You must test ride them if at all possible.

Well yes, in an ideal world. Here in Aus, there's virtually no stock to buy, let alone have the luxury of a test ride

I did hire a Kellys bike earlier this year that was offering 60Nm (an e7000 I believe)

The article I linked to had a graph for the E7000 as well - it shows that max boost would give 250% (so 625W?) and 60NM. Considering it's also a Shimano motor (so should use similar, if not identical software, to the E8000) how would that compare to my 500W, 70NM setting. What's more important, power or torque?

 

[S13]

Well, that is not how it works.
The percentage rating means the amplification of your own input. So if it is set for example at 200%, and you apply a force of lets say 30Nm on the cranks, the motor will output 60Nm. If it is set at 90% and you apply 30Nm, the motor will output 27Nm.
But then there is also the limit of the motor. For example if it is set at 30Nm (like in ECO) then the motor will never output more than 30Nm, no matter how hard you pedal.
Likewise, the motor can never output more than its maximum rating (70Nm for the E8000 in this case).

Power is a different number and also takes in consideration at what cadence you are pedaling (its a function of RPM and torque). Its not really something we are too concerned with generally (though maybe we should). That 250W rating is just a lie to conform to regulations. Somehow that is supposed to be an average maximum that the motor could output. But in reality all motors will output way more than that, and will happily do so until the entire battery is drained. Shimano and Bosch are at ~500 to ~600W max, Dyname goes up to 750W, and Bafang goes even higher.

Now, in the real world the numbers arent that precise. Motors dont really have flat torque responses, they are usually curved. So that 70Nm might be achievable at a certain RPM, but not at others. Also, the amplification % isnt exactly linear either. At all depends on how the torque is sensed at the pedals, what support curve the designer has implemented etc... Every motor is different in this regard.

We use those simple torque and % numbers to make it easier to understand and compare. But just know that the reality is a little more complex than that.

 

[Mteam]

@mxh

Therefore, in answer to your question of "200%, 90% of what exactly?" It will be 500W and 225W.

The assist percentages are not percentages of the nominal 250w, they are percentages of what you are putting into the pedals, as per post from s13 above.

ie if you have the assist set to 200%, and you are putting in 100watts, then the motor will output 200watts, so you and the motor combined will be propelling the bike with 300watts in total. This is subject to a power ceiling which is either specific to the mode you are in, or the EU nominal 250w limit (which doesnt mean the limit is 250w, but I have no idea how this nominal power is actually arrived at,I think its an average over 30 minutes of something like that) , or other limits within the motor - maybe lower limits will apply at certain motor temperatures for example, then there is the speed based limit where motor power is capped at 0watts regardless of what you put in etc etc


OP you need to ride the bikes to decide, no amount of looking at specs will really help decide if you would be happy with a lower powered lighter bike.

 

[pagheca]

So, a motor with a 500W-hr battery will be able to deliver a steady 250W for 2 hours, but 500w for only one hour.

I was looking for an explanation like that (thanks very much!), but do you have any reference for that? It looks to me very strange that the bike record what was done in the last hour and can suddenly cut-off all power if you "drunk" too much power in the past hour.

 

[steve_sordy]

Thanks to all above. Ever since I got an emtb (Jan'19) I have always believed that the percentage was aimed at the 250W nominal. I can't now think where I got that belief, so I must have read it somewhere - it's not the sort of thing I would make up. It all made perfect sense, so it was easy to believe. Why are the documentation and publications on this topic so vague. @mxh is asking a perfectly good question "%age of what?" I had thought it was easy to answer, seems I was wrong!
My apologies for the confusion caused and thanks to those that corrected me without flaming me!

It seems equally believable that the percentage is expressed as being applied to the torque that the rider puts in. I would very much appreciate it if anyone could paste a link to a reference to where they got their belief from. I need to see it written down to get it firmly into my head!

 

[Mteam]

It seems equally believable that the percentage is expressed as being applied to the torque that the rider puts in. I would very much appreciate it if anyone could paste a link to a reference to where they got their belief from. I need to see it written down to get it firmly into my head!

I think the primary sensors the motor has are a torque sensor and cadence sensor, so I think technically it does base its output calculation off the torque you are applying to the pedals, but I suspect it then combines the torque value with the cadence its is also measuring/sensing to derive your pedalling power in watts, to then determine how much the motor needs to output based on the assist percentage - as far as I know none of this info is in the public domain, so I'm just speculating.

 

[steve_sordy]

OK, thanks - that makes sense.

So does that mean that, for example, my 200%, 70NM setting (which you've clarified means 500W) can't be compared to similar figures from a different motor to make any meaningful comparison?

Well yes, in an ideal world. Here in Aus, there's virtually no stock to buy, let alone have the luxury of a test ride

The article I linked to had a graph for the E7000 as well - it shows that max boost would give 250% (so 625W?) and 60NM. Considering it's also a Shimano motor (so should use similar, if not identical software, to the E8000) how would that compare to my 500W, 70NM setting. What's more important, power or torque?

@mxh As you will know by now, what I said so confidently appears to be wrong. Sorry about that, I really did believe in what I said. Every day is a learning day!

Ref your final question "what is more important, power or torque?" I would say torque, because that is what you feel most clearly. emtb motors are advertised on what torque (Nm) they have and what battery size (W-hrs) and not on what power they have.
If you use the correct units, Power = Torque x Revs. If you use the wrong units Power is directly proportional to Torque x Revs. The motor software is a very important factor in the ride feel and each motor has no doubt had zillions of man-hours of testing, thought and programming time built in to the result. They are all good, but as I said earlier, they each have their own features, profile, response....whatever you want to call it.

If there is little opportunity to test ride new bikes, try to ride the emtbs of your riding buddies. Failing that engage fellow riders in conversation about their emtb and after showing admiring interest, a bit of flattery (whatever you think will work) ask if you can have a quick go. Believe it or not I have done this and most of the time I got a yes. Sometimes I didn't ask because I just knew that the answer would be no. But nothing ventured, nothing gained!

 

[steve_sordy]

I was looking for an explanation like that (thanks very much!), but do you have any reference for that? It looks to me very strange that the bike record what was done in the last hour and can suddenly cut-off all power if you "drunk" too much power in the past hour.

I got some bits wrong (see my two previous posts).

You are correct, it would be strange. I very much doubt that is what happens. The motor software keeps you legal unless it has been hacked (loads of products to do this by the way). The only time my motor has ever cut me off was when I ran out of battery, but it was not without warning. I could see the power levels falling on the display, then I got the red bar, then I got limited to Eco, only then did it go dead.

 

[pagheca]

there is a strict relation between power (P in W), torque (T in Nm) and cadence (C in rpm). In an analog bike: P =T*π*C/30. Example: if C=80 rpm and P=150 W, one obtains: T = 30*P/(*C*π) = 30*150/(80*π)=18 Nm.

Now, let's assume that P=150 W is the power read on the e-bike display, that this figure represents the contribution by the motor only and that in the selected mode the motor applies 200% of the power excerpted by the human. If this is the case, the total torque will be: T = 30*150*(1 + 1/200%)/(80*π) = 30*150*1.5/(80*π) = 27 Nm and the total power at the wheel is 225 W, of which 150 applied by the motor and 75 by the human.

If this is correct, by applying the maximum rated power (250 W) we obtain the maximum torque available for each value of the cadence, i.e., if again C = 80 rpm: T=30*250*(1+1/200%)/80*π) = 44.8 Nm, that is still well below some motors rating. To reach it we need to largely increase the cadence.

Am I wrong??

 

[Mteam]

@mxh I would say torque, because that is what you feel most clearly. emtb motors are advertised on what torque (Nm) they have and what battery size (W-hrs) and not on what power they have.

as a complete aside I would say the reason that all the ebike marketing is based on torque is because the EU (and other jurisdictions I guess) set a power limit of 250w, so it would be a very difficult/confusing message to say your motor outputs 750w yet is still legal in the EU.

By the way - I'm sure I remember seeing an early (very early) specialized turbo levo ad in a UK magazine that talked about their motor having 750w, and at the time I couldnt work out how it was legal when the limit was supposed to be 250w, I wondered whether it was an ad designed for a different market and that uk/eu bikes would be limited to 250w but the ad obviously didnt state it, it was v confusing

 

[p3eps]

I'd go with the above that the percentage of assistance is based on the rider.
I'm purely going by all the original Levo SL literature and marketing campaign as my source - as I went through this quite thoroughly almost 3 years ago when purchasing.

The SL was marketed as you times 2. It all pointed to giving you back 100% ontop of what you put in. If you had the SL set to Turbo (100/100), then it would double your input. You put in 100 watts, you'd get 200 watts out.
I was led to believe this was the same with all eBikes, however some FF bikes could offer you an additional 300%+ ontop of what you put in.
You put in 100 watts, and get 400 watts back.

 

[S13]

If this is correct, by applying the maximum rated power (250 W) we obtain the maximum torque available for each value of the cadence, i.e., if again C = 80 rpm: T=30*250*(1+1/200%)/80*π) = 44.8 Nm, that is still well below some motors rating. To reach it we need to largely increase the cadence.

Am I wrong??

No you are not wrong, but its not the cadence that has to go up. Its the electrical power. That is why i mentioned before that most motors actually consume anywhere from 500W to 750W at peak, resulting in ~70 to ~110Nm at the cranks.

 

[pagheca]

but, isn't the power limited to 250 W (at least in EU)?

 

[Mteam]

but, isn't the power limited to 250 W (at least in EU)?

yes, but its not a hard limit that a motor must never output more than 250w, its referred to as max continuous rated power must not exceed 250w. so motors can exceed 250w, as long as the max continuous rated power doesnt go over 250w. I dont know what defines max continuous rated power.

edit : heres some background reading for you The Myth of Ebike Wattage

 

[S13]

but, isn't the power limited to 250 W (at least in EU)?

Theoretically yes. In practice, no.

Im not sure how emtb manufacturers are getting away with this. But hey, you dont hear me complaining...

 

[pagheca]

This explain a lot of things. I found a lot of peaks of power in the record of my Bosch CX motor well above 250 W. Thanks! I wasn't aware about that.



However, it is still not clear to me how the motor complain with the 250 W rule...

 

[steve_sordy]

Theoretically yes. In practice, no.

Im not sure how emtb manufacturers are getting away with this. But hey, you dont hear me complaining...

I agree that we should not create a fuss about this. I am still uncertain as to whether the loophole was by accident or design. I would like to think that some clever EU bike riding bureaucrat pulled the wool over the eyes of his fellow legislators. Maybe he had been briefed by one of the bike lobbyists. Stealthy steps, grandma is sleeping!

 

[billwarwick]

My brain hurts

 

[steve_sordy]

My brain hurts

I suspect that is why the bike companies don't spell it out. They don't want to confuse their customers with technical facts, but more importantly, they don't want to alert the legislators that they may have missed something.

 

[pagheca]

but did someone ever found a continuous power record above 250 W? Or just peaks (like in my previous screenshot, but I am not very fit...).

If this is not the case, it could just be a limitation of the motor management system that try to instantly respond to a sudden high demand of torque and overshoot for a second.

 

[S13]

but did someone ever found a continuous power record above 250 W? Or just peaks (like in my previous screenshot, but I am not very fit...).

If this is not the case, it could just be a limitation of the motor management system that try to instantly respond to a sudden high demand of torque and overshoot for a second.

oh yes. The dyname motor actually shows the output power, and it actually does consume 750W almost continuously on the highest power setting:

Source:

And it seems to make sense. With my previous E8000 i was able to ride a 500Wh battery flat in 1 hour, meaning i was consuming 500W on average.

 

[pagheca]

oh yes. The dyname motor actually shows the output power, and it actually does consume 750W almost continuously on the highest power setting:

but this is sold on the US market, not EU, isn't it? Sorry to look picky but I really want to understand...

 

[S13]

but this is sold on the US market, not EU, isn't it? Sorry to look picky but I really want to understand...

No this is sold on the EU market as well. In fact i bought one, and there is a whole network of distributors and shops that sell it in the EU.

 

[Zimmerframe]

but this is sold on the US market, not EU, isn't it? Sorry to look picky but I really want to understand...

Another example would be the brose. It pulls 20amps from the battery. So at 36v (will be slightly higher when fully charged) then 20*36 is 720w - that's power drawn, not power produced, which is about 80% of that.

There is a good thread somewhere which explains in detail how the 250w is tested and complied to - they do comply. It isn't a max. I'll try and find it later.

 

[pagheca]

Thanks, I am really confused now and would be very interested in reading it

The regulation is this: EUR-Lex - 32013R0168 - EN - EUR-Lex and it appears very clear to me:

(h) pedal cycles with pedal assistance which are equipped with an auxiliary electric motor having a maximum continuous rated power of less than or equal to 250 W [underlined mine], where the output of the motor is cut off when the cyclist stops pedalling and is otherwise progressively reduced and finally cut off before the vehicle speed reaches 25 km/h;

I'm new in the field so pardon me if I write something trivial for most of you.

 

[Mteam]

max continuous rated power = the max power that the motor can output for an infinite (sort of) amount of time (assuming an infinite battery) without overheating or failing .

That doesnt mean that the motor cant put out more than that for shorter periods - read the link I posted up there it goes into more detail

max continuous rated power != max power