Power Lost to Electric Drive Train Inefficiency

Hi,

I am new to the forum. I have two questions about electric drivetrain efficiency loss:

(1) FIRST SCENARIO:


Imagine a Wrightspeed drivetrain. A 300 KW battery pack sends electricity through wires to an inverter, which powers the 250 HP hub electric motor; but the motor also has a “Geared Traction Drive” -- see http://www.wrightspeed.com/technology/. Could we say that the electric motor operates at only .98 % efficiency, the inverter deducts another 2 %, and the hub-gearbox another 3 %?

In other words, what would be the “Battery to Wheel” energy loss? 5 %? 7 %? 10%?


*****************************************************************

Now imagine a much more “theoretical” scenario.

(2) SECOND SCENARIO:


A 300 KW battery pack sends electricity through wires to an inverter, which powers a 250 HP hub electric motor, which also has hub reduction gearing. But the year is 2020, and everything has been perfected even further: motors inverters, hub-motor gearing, etc. In such an ideal scenario, how much energy would be lost “Battery to Wheel”?


*****************************************************************


Engine-to-wheel energy loss in the world of ICE is a common topic, where lots of energy gets absorbed by transmissions. But it’s also a controversial topic. There is no “typical” figure for the percentage lost to a transmission and other components in a drivetrain, although some will cite 15 % for cars, and 30 % for trucks. Apparently in racing there is a "15 % inefficiency" rule, but a very detailed, well-written, well-informed, and insightful article suggests that the "15 % rule" is pure rubbish -- see http://www.superstreetonline.com/how-to/engine/modp-1005-drivetrain-power-loss/. Another webpage that is probably not to be trusted pegs the drivetrain loss as just 5.6 % -- see http://www.consumerenergycenter.org/transportation/consumer_tips/vehicle_energy_losses.html \. A more empirical article that tested two cars, came up with figures for driveline inefficiency that varied from 17 % to almost 40 % -- see http://www.hotrod.com/how-to/transmission-drivetrain/ccrp-0311-drivetrain-power-loss/ . There's a really terrific thread in which some seemingly well-informed posters claim that modern car transmissions can lose as little as 10 %, while others who sound just as informed will say that 30 % is more nearly right -- see http://cr4.globalspec.com/thread/36270/Overall-efficiency-of-gasoline-powered-cars . And that's just cars. I wonder where the guesstimate of 30 % lost to the transmission for trucks came from?

In any case, what I am trying to determine is the same for an electric drivetrain. Between the surge of current provided by the the battery pack, and the final power output via the electric motors "at the wheel", how much energy do you think will be lost, due to various kinds of “inefficiencies”? Or what would be a reasonable rough-ball-park figure to work with? So far, the only thing that a friend has been able to find on the web is a Master's thesis written in 2011 -- see http://scholar.lib.vt.edu/theses/available/etd-05162011-220140/unrestricted/Gantt_LR_T_2011.pdf. So the question is a rather "new" one, if only because electric drivetrains are still rather new.

My motivation is the following. At present I am using a reasonably good web-calculator to try to figure on the different power requirements for a serial-hybrid electric bus-type vehicle driven at different speeds, altitudes, and ascending different kinds of slopes. There is a space where one needs to insert a percentage figure for “mechanical transmission inefficiency" -- see http://buggies.builtforfun.co.uk/Calculator/index.html :

Mechanical Transmission Efficiency

Effy ???? %

The calculated power is the ideal mechanical power required at the wheels to drive the vehicle under the conditions you describe. The motor output power will need to be more than this - because some is lost in the mechanical transmission. If you specify a transmission efficiency (in %) a figure for motor output power will also be calculated.

Click to expand...

So I would greatly appreciate it if those on this forum who are knowledgeable about “electric drive train efficiency loss”, might suggest a good percentage figure into insert in this part of the web-calculator. Even an informed guess would be better than my guess, which would be utterly arbitrary, because I am not an electrical or automotive engineer.

One caveat: I realize that on this forum there exists a long thread whose subtitle might be "To gearbox or not to gearbox?". I got the impression reading that thread that if a hub-motor's gearbox is well-designed, it can be very efficient, and the power loss will be minimal. See https://endless-sphere.com/forums/viewtopic.php?f=30&t=47930&start=30 . But I didn’t understand the half of it. My overall impression is that using a direct drive motor without a gearbox at low speeds kind of abuses the electric motor, if the electric motor is not large enough. So the choice seems to be between a smaller electric motor + gearbox, or a larger electric motor that has enough power and torque at slow speeds such that it won't get abused when used to direct drive. Tesla settled for the second alternative, the direct drive approach. I am guessing that Wrightspeed developed its two-speed "Geared Traction Drive" primarily because garbage trucks do so much starting and stopping, and are working primarily at such slow speeds. Apparently a gearbox is also important for some reason when there's a headwind, and when climbing hills. A direct-drive electric motor is not optimal when climbing hills, for reasons that I don't understand.

So for the present purposes let’s assume that a hub motor will have a gearbox, and that direct-dirve is not an option? In the two scenarios described above, what would the efficiency loss? What percentage for drivetrain inefficiency should I plug into the web-calculator at http://buggies.builtforfun.co.uk/Calculator/index.html, and why?


All best wishes,



Biotect

Brief addition,

Again, I really know nothing about electric drivetrains, although I am a transportation designer. I am on the "aesthetic" ends of things, and professionally I focus on interiors. So all that I am really looking for, is for someone to make an "informed guess". It can be very rough-ballpark. If most electric motors seem to be somewhere between 95 - 98 % efficient, with the mean something like 96.5 % efficient, that's all I would need to know. And the same with wiring, the inverter, hub gearing, etc. I just need a reasonably informed guess; a very rough ball-park sort of figure.

Or, actually, two of them: one for the first scenario above, and another for second 2020 scenario.

All best wishes,


Biotect

biotect said:

Could we say that the electric motor operates at only .98 % efficiency, the inverter deducts another 2 %, and the hub-gearbox another 3 %?

In other words, what would be the “Battery to Wheel” power loss? 5 %? 7 %? 10%?

Click to expand...

Losses multiply.

Actual efficiency loss you'd ahve to measure. It is unlikely that any motor will be anywhere as low as less than 1% efficient as it would melt from any substantial use, so I am assuming you mean 98% (rather than .98%) efficient. That is also an unlikely figure; even at peak efficiency most are much less than that. You'd either have to get *real* dyno'd test data on the motor in question, or do those tests yourself, to get a true efficiency figure for it, at the motor speeds and loads you will be using. (they are not the same efficiency across the entire speed range).


You can get some idea of hub efficiencys for ebikes, to give you a little idea of how this can work for *any* motor, over at http://ebikes.ca/simulator.


Drivetrain efficiencies post-motor you'd also have to get the actual tested data for your actual part speeds and loads, as again they may vary over a range.

Hi amberwolf,

Interesting simulator. But in a sense my question is much simpler than that.

I am just looking for a single percentage number for "drivetrain inefficiency" that I can plug into a web-calculator used to calculate a vehicle's Power Requirement -- see http://buggies.builtforfun.co.uk/Calculator/index.html . As designers, we can keep things very general and vague during the initial stages; only engineers need things to be more precise. For previous calculations I just figured that the electric drive-train on a serial-hybrid bus would "probably" be better than an ICE drivetrain, and the oft-quoted figure for ICE drive-trains is somewhere between 15 - 30 %, lost to the transmission, etc. So I just guessed (on the basis of nothing whatsoever!) that the electric drive-train in a serial hybrid bus would lose 10 %. And so I have been entering 90 % as a number into the web-calculator. Does 90 % efficiency seem roughly correct, for an electric bus drivetrain? Does it seem like a serial-hybrid electric bus would lose 10 % in total: the electric motor inefficiency x inverter inefficiency x hub-gear inefficiency x wiring inefficiency?

My question is a very simple one, and all that I am looking for is a reasonably informed guess. A guess that would be better than my guess!

All best wishes,



Biotect

It's not really completely clear what kind of question you're really asking at this point (because no one can give you an efficiency number unless they know what parts you're using and what the numbers for those parts end up being) so:

If you are looking for whether electric is more efficient than ICE, then yes, by a lot--but if you're looking for whether the *transmission* is more efficient, well, if you use the same transmission in the same way, then the answer is they are the same, for that very reason.

The electric motor/controller itself *would* be much more efficient than the ICE in conversion of it's "fuel" to motion, because the ICE itself is probably not even 10% efficient in that conversion, *and* the powerband for the electric is probably much wider than the ICE. (depending on exactly which ones you are comparing and how you use them; values vary widely)

But again, the transmission itself is the same losses, if it's the same transmission and the same way of using it. Unless you know what you're using in each one, you can't compare them directly.


For guesses, as they say "your guess is as good as mine".

If I was just guesstimating things, I would always err on the side of worst-case, and then if I got better real test data numbers once built, then "yay!".

But most people don't want to hear that, so my one example below is not all that pessimistic; you could get a lot worse in reality, depending on what parts you actually use and how you use them. My guess is it probably wouldnt' be much better than that, though.


So I would go with 75% for motor efficiency, because it won't be used at it's efficient zone all the time, and if it's used in typical stop-and-go traffic it'll be worse than usual, as even with a multi-gear transmission it'll still have to restart from zero or near zero speed a lot, and that's really inefficient. Again, this depends on your specific motor and how you use it, so it could be better or worse in real life, by an unknown amount.

Controller efficiency is probably in the high-90s, let's call it 95%, but depends on how it works and what kind of motor, etc., so this could be a lot different, yada yada.

So here you have motor of 0.75 * controller of 0.95 gives you 0.7125, or 71.25% efficient at that point.


Then the transmission-to-wheels-on-ground sections of the powetrains themselves, if they are identical and used identically to an ICE, will be exactly the same as with an electric system. So for that, you'd need to know what the efficiency of your actual transmission is.

A chaindrive system, like many motorcycles and bicycles use, and some other vehicles, is around 98% efficient for each stage, so a single reduction stage would be 98%, and if you have two stages for whatever reason, then it's 0.98 * 0.98 = 0.9604, or 96.04% efficient.

Then that 0.98 single-stage chaindrive * 0.7125 (motor+controller) would be 0.69825, or 69.825% efficient.

If you use a two-stage chaindrive, you get 0.69825 * another 0.98, and that's down to 68.4285%.


I don't know what the losses are in say an automatic transmission and differential from a car, but let's say some of those people are right, and it's 70%--then you have the motor-controller losses of 71.25% * 70% = 49.875% efficient for the whole thing from battery to wheels.

But if you had an ICE instead of the electric motor for the exact same transmission and usage of it, you'd end up with 10% * 70% = only 7% efficient from "battery" (gas tank) to wheels.

Same goes for the chaindrive; you can run those numbers yourself from the above.

So the electric vehicle would be 7.125x as efficient as the ICE...which you can see without even involving a transmission, just by comparing the original numbers of efficiency of the motors themselves.


Again, if you really wanna know how efficient it is, you have to know the efficiency of hte actual parts you're using, in the ranges you'll be using them in.



As one example, the NuVinci 171b CVT hub, when used at low torque/power levels, has been noted in various builds to be 70-80% efficient. But at higher torque/power levels, efficiencies as poor as only 40-50% or worse have been seen, for the exact same hardware.

So your usage of your transmission could make a huge difference in the total efficiency of your system.

OK, that's clear. Perhaps all that I really need to do is contact Wrightspeed, and ask them to give me a number for the efficiency of their "GTR" -- their geared traction drive. See http://www.wrightspeed.com/technology/ .

So that I might sound "technical" and like an inquiring engineer, any thoughts about how I should write the email? I don't want to sound like some flaky transportation design student who knows more about color and clay than he does a about mechanics and electricity.... even if I am one!

All best wishes,


Biotect

Be honest in all your communications and you're more likely to get an honest answer.