Basic electrical theory. Motor and controller

Get all your technical information about electric bikes here.
Post Reply
Builditgood   10 mW

10 mW
Posts: 26
Joined: Aug 29 2015 4:36pm
Location: Southampton, UK

Basic electrical theory. Motor and controller

Post by Builditgood » Apr 08 2018 2:12pm

Hi. A friend and me we're discussing which components govern which factors and arrived at the following. What do you folks think?

The motor and it's windings have a certain resistance when cold and so draw a max current for a fixed voltage (e.g. plug a 1000W hub into a 50V battery and it will ask for 20A). We surmised that if you hook a motor up to a controller that cannot supply the current required then you risk damaging the controller and conversely if a controller is rated to more than what the motor requires then the motor draws as much as it needs and no more. The wattage of a system is(should be) therefore determined by the motor and not the controller.

The current draw will increase as the motor heats up because the resistance in the windings gets greater with temperature. This can lead to bogging down, the theory behind which we haven't figured out yet.

The controller will pull the current it requires from the battery and the battery must have enough capacity to supply what is required. The ability of a pack to supply current is calculated by either multiplying the Ah by discharge rates (C) and f.o.s. or multiplying the cell current capicty by the number of parallel strings.

That's as far as we've got for now☺
Full sus specialized, tangent 3000w, 16ah 12s lipos.
Full sus yeti, cyclone 3000w, 12ah 72v lifepo4.
Folding shopper, 1000w hub in 20" rim.
DMR trailstar, 2000w hub in 17" moto rim, 21ah 50v lion.
DMR sidekick, gng 2015, 16ah 12s lipo.
Full sus giant, 2000w hub in 26" rim, 12ah 50v lifepo4.
Spare Divinci 6000w midrive waiting to be built into something!!

User avatar
liveforphysics   100 GW

100 GW
Posts: 14143
Joined: Oct 29 2008 1:48am
Location: Long Beach, CA, USA

Re: Basic electrical theory. Motor and controller

Post by liveforphysics » Apr 08 2018 3:39pm

The motor has winding resistance, but it's generally sub 0.1Ohm, and can be much lower depending on motor size and Kv (motor RPM per volt).

This means if you shorted 50vdc of pack across it, it's going to draw 500amps and become a 25kW copper thermal gain rate (assuming the 50vdc source is capable of 500amps).

Fortunately, the as the motor features loops of conductor around a magnetic core, it has an inherent inductance which limits the rate at which that current attempts to reach 500amps, giving it a sloped current rise time.

The motor controller functions by observing current as it's rising up that slope, and when it achieves it's set-point, it disconnects shorting the battery voltage into the motor winding, and the magnetic field stored in the inductance of the motor collapses, driving current back against the controller MOSFET's body diodes to maintain current in the coil. When the controller observes that current has fallen back below the set-point, it connects the pack back across the motor winding.

The voltage waveform fed into the motor just looks like patterns of spikes of full pack positive or negative, but if those spikes are timed correctly, the resulting kicks of voltage absorbed into the inductance of the winding/core assembly may have >99% of being a perfect AC sinewave. As the motor's torque is only a product of this resulting current, as pack voltage goes up or down, only the "ON" time width of the spikes change to achieve the motor winding current.

If the motor is also a brushless type, then along with this current control mechanism, it also flips which side of the pack it's connecting the motor winding to, depending on what the polarity of the magnet passing by on the rotor happens to be, so it's always making the correct magnetic field to be pulling/pushing the rotor magnet aligned with core at that point of rotation.

If you feed the same motor 1,000Watts from a 500vdc pack or a 50vdc pack, it ends up making exactly the same magnetic field in the core (the aspect that makes the torque), the difference is the 500vdc pack is feeding spikes aproximately 1/100th the "ON" time of the 50vdc pack, and hence the 500vdc pack averages 2amps battery current while the 50vdc pack averages 20amps battery current. Either way the energy consumption rate from the battery is the same, and motor heating and efficiency is the same (realistically the 500vdc system would be more lossy from using IGBT's vs MOSFETs in it's controller.)

No matter how a given motor core is wound, it's efficiency and performance is only a product of the % of the slot area filled with copper.

Voltage does not create any of the traction-useful forces inside the motor, just the current in the winding. As an example with deathbike's motor, it draws around 7amps off the battery to sustain 660amps RMS in the motor winding while the motor isn't rotating. As the motor rotates, it builds it's own voltage across it's phase leads(BEMF), which means the controller must no supply this voltage as well as create the resistive voltage drop across the winding, so the "ON" time must increase to maintain the same winding current. This means the power consumed from the pack increases, along with the power output of the motor.

-Luke
Each carcinogen vapor exposure includes a dice roll for cancer.

Each mutagen vapor exposure includes a dice roll for reproductive genetic defects in your children.

Each engine start sprays them into a shared atmosphere which includes beings not offered an opportunity to consent accepting these cancer experiences and defective genetics life experiences.

Every post is a free gift to the collective of minds composing the living bleeding edge of LEV development on our spaceship.

Builditgood   10 mW

10 mW
Posts: 26
Joined: Aug 29 2015 4:36pm
Location: Southampton, UK

Re: Basic electrical theory. Motor and controller

Post by Builditgood » Apr 08 2018 5:35pm

Brilliant explanation thanks Luke. So if I've understood it correctly the resistance of the windings is low so it is not a matter of the motor taking as much current as it needs but the controller cutting the supply when it reaches a preset limit. Based on the fact that the power of a system is determined by the controller and the current limit it has in place, it seems to be the case that one should choose to supply a motor only with as many Watts as it can cope with. I guess best practice is to limit the Wattage supplied according to manufactures recommendations with the option of overclocking it if conditions allow.
Full sus specialized, tangent 3000w, 16ah 12s lipos.
Full sus yeti, cyclone 3000w, 12ah 72v lifepo4.
Folding shopper, 1000w hub in 20" rim.
DMR trailstar, 2000w hub in 17" moto rim, 21ah 50v lion.
DMR sidekick, gng 2015, 16ah 12s lipo.
Full sus giant, 2000w hub in 26" rim, 12ah 50v lifepo4.
Spare Divinci 6000w midrive waiting to be built into something!!

User avatar
spinningmagnets   100 GW

100 GW
Posts: 11853
Joined: Dec 21 2007 10:27pm
Location: Ft Riley, NE Kansas

Re: Basic electrical theory. Motor and controller

Post by spinningmagnets » Apr 08 2018 6:16pm

A long time ago, there simply was not any high-amp batteries available. these days there are many choices, so you can select a pack that can provide your desired volts and amps without overheating. Bear in mind, that...you can overheat a battery pack, so it's good to start from a position of knowing how many watts of power you want to use as a temporary peak during acceleration.

If the battery, controller, and motor are cool to the touch in all conditions (your commute and hills may be different than mine), they might be heavier, larger, and more expensive than necessary.

If those three main components are hot (or just one of them), then...they are undersized for the task you are giving them. So then, the question is...which component "should be" the limiting factor for you? When it comes to batteries, I recommend testing a similar ebike to the type of system you think will satisfy your needs. Once you determine that, you know the amount of watts you need, and you can spec a battery pack.

There is a natural tendency to spec an expensive battery pack that is as affordable and small (fittable) as you can find. I always recommend getting a battery that is larger than you need, because keeping the battery cool is one of the two factors that lead to having the absolute longest life (in years). The other is never charging to 4.2V per cell. Charge to 4.1V as a max, or...4.0V is even better.

Once you have chosen a bike frame and battery pack...we can move on to the controller and motor. A motor is a "dumb" device. By that I mean that...if you feed it too many watts, it will get so hot that it dies. It will take all the watts you feed it. Of course, it has a point where its heat-gain is less than it's heat-shed rate. To get the bast bang for your buck, be aware you can feed a motor more watts that it can handle continuously...for a few seconds. The light turns green, and you accelerate with high watts. Then...during the cruise-phase...it has a chance to cool off a little. Most sensorless motors have Hall-sensors to tell the controller the position of the rotor at any given moment, but...they are the weak link when it comes to heat (some exotic motors have optical sensors located away from the heat).

I would size a motor so it normally reaches 140F (60C) at a temporary max..for long life (140F on the interior, at the hottest part, with a digital temp sensor). I would never recommend getting a motor over 200F (93C). Many motors can survive that temp, but...you can cook food at 200F...if it is getting that hot, you are wasting battery watts by using your motor to convert electricity into heat. If you fry the Halls, you can slap-on a sensorless controller as a stop-gap measure, but doing that is a bad design.

The controller is where the magic happens, and it is crammed full of electronic components. Never get it hotter than 140F on it hottest part (The FETs?), if you want it to live a long time. The motors are three phase, and the the controllers typically use "Pulse width modulation " (PWM). A wider "on" pulse will push the motor to accelerate, and a shorter on-pulse will let it cruise or even slow down (depending on load). It uses a 1V-5V signal from the throttle to determine if you want it to slow down, cruise, or speed up.

The type of MOSFETs (metal oxide semiconductor, field effect transistor), and the number of FETs will determine how many max amps the controller can flow. They are electronic on-off switches. 6, 12, and 18-FET controllers are common. The lower the max voltage rating of the FETs, the more efficient they are. 3077's are good up to 75V (popular for 36V, 44V, 48V and 52V), and 4110's are good up to 100V (popular for 60V and 72V)

viewtopic.php?t=10655

You can't trust the manufacturers ratings. This is the biggest value of the forums, you can find out what the motors, controllers, and batteries can really do. Too cold, and they are probably too large, heavy, and expensive. Too hot, and they will die soon from being too small (and continue wasting watts if they live).

User avatar
liveforphysics   100 GW

100 GW
Posts: 14143
Joined: Oct 29 2008 1:48am
Location: Long Beach, CA, USA

Re: Basic electrical theory. Motor and controller

Post by liveforphysics » Apr 08 2018 9:45pm

Builditgood wrote:
Apr 08 2018 5:35pm
Brilliant explanation thanks Luke. So if I've understood it correctly the resistance of the windings is low so it is not a matter of the motor taking as much current as it needs but the controller cutting the supply when it reaches a preset limit. Based on the fact that the power of a system is determined by the controller and the current limit it has in place, it seems to be the case that one should choose to supply a motor only with as many Watts as it can cope with. I guess best practice is to limit the Wattage supplied according to manufactures recommendations with the option of overclocking it if conditions allow.


Any mfg motor power rating is largely an arbitrary number. In a cold wind it may be triple the continuous power of still warm air. Likewise, what it can output for bursts is determined by its magnetic core saturation point. If you're making a dragster you want to be holding that saturation pegged all the way down the track. For a commuter or road racing applications you typically want to run right near to core saturation point but not exceed it.
Each carcinogen vapor exposure includes a dice roll for cancer.

Each mutagen vapor exposure includes a dice roll for reproductive genetic defects in your children.

Each engine start sprays them into a shared atmosphere which includes beings not offered an opportunity to consent accepting these cancer experiences and defective genetics life experiences.

Every post is a free gift to the collective of minds composing the living bleeding edge of LEV development on our spaceship.


Post Reply