Hummina Shadeeba
10 MW
at what speed is it most efficient to travel? Does the motor care?
Most efficient speed?
- Thread starter Hummina Shadeeba
- Start date
dustNbone
10 kW
In my experience, on an upright bike, wind drag increases very quickly over 16MPH or so. Even slower would probably be even more efficient, but you're really looking for a compromise somewhere between efficiency and speed. Another consideration is motor speed, if your motor is running way below it's efficient range, even though you're not suffering much wind drag you might be wasting alot of energy to motor/controller heat.
Jon NCal
1 kW
The lower the speed, the less wind resistance, but at very low speeds the motor will lose efficiency. I looked at the motor sim tool on grin tech site for a mountain bike and I think around 4 to 5mph is lowest wh/mi. on flat, no wind.
motomech
10 MW
For me, the low to mid 20's mph is the best compromise between going fast enough to be fun and maximising battery range.
Added benefits of staying in this range are;
1)it's fairly easy to get the gearing that allows pedal input to be meaningful.
2)Traffic isn't as likely to mis-judge your speed and cause you harm.
Added benefits of staying in this range are;
1)it's fairly easy to get the gearing that allows pedal input to be meaningful.
2)Traffic isn't as likely to mis-judge your speed and cause you harm.
Hummina Shadeeba
10 MW
Looking at the grin tech simulator a motor will be less efficient at slower speeds. Why ?
Hillhater
100 TW
Because its operating outside its ideal efficiency speed range.Hummina Shadeeba said:
FYI at 10 kmh a typical bike rider needs only 20W to maintain speed.
For every 5kmh faster you have to double the power needed....such that at 30kmh (20mph) you need 240W .
And at 50kmh (30mph) a 1000W would be required
The majority of this extra power is all needed to overcome wind and rolling resistance at the higher speeds.
Hummina Shadeeba
10 MW
That's it.
dogman dan
1 PW
Re the simulator, it is showing motor efficiency, not the watt hours per mile needed to go a certain distance. They are apples and oranges, One is motor efficiency, the other is system efficiency. So the wh/mi way of looking at efficiency, slower is nearly always less wh to go one mile.
Except when the low speed is caused by overloading the motor.
This is why the most efficient way to climb a big hill can be to go as fast as the motor setup can. Or not,, depending on how overloaded the motor is.
On the flat,, with no overload, slower always takes less power, because wind drag is less. And for bikes, because you can pedal up most of what is needed.
So the way to get there on less battery, is nearly always to slow down, even if motor efficiency drops by 20%. Lose 20% on the motor, but gain 75% on drag. Net gain 55%. Mid drive bikes have an advantage, they can gear down, and go slow with a higher rpm. this is also true for geared hub motors. So either one can be set up to be very motor efficient at low speed,, or not depending on the gear ratio used. DD hubmotors are set in one gear, so there are advantages to using a smaller, lower gear, wheel.
Except when the low speed is caused by overloading the motor.
This is why the most efficient way to climb a big hill can be to go as fast as the motor setup can. Or not,, depending on how overloaded the motor is.
On the flat,, with no overload, slower always takes less power, because wind drag is less. And for bikes, because you can pedal up most of what is needed.
So the way to get there on less battery, is nearly always to slow down, even if motor efficiency drops by 20%. Lose 20% on the motor, but gain 75% on drag. Net gain 55%. Mid drive bikes have an advantage, they can gear down, and go slow with a higher rpm. this is also true for geared hub motors. So either one can be set up to be very motor efficient at low speed,, or not depending on the gear ratio used. DD hubmotors are set in one gear, so there are advantages to using a smaller, lower gear, wheel.
Jon NCal
1 kW
re: the motor sim
Look in the box at the lower right of the screen. You will see watt hours per mile, range and time to overheat.
Look in the box at the lower right of the screen. You will see watt hours per mile, range and time to overheat.
Hummina Shadeeba
10 MW
thanks for the real world situation and there's also the motor simulator. but...looking at the simulator there isn't an overload scenario showing up. I can add to the load with greater steepness or weight but the efficiency is still tied solely to the rpm and unaffected. It makes sense that a small motor will become more inefficient with a big load, being overloaded, but Id like to know why. a large magnetic field needs to be formed with a large load and a smaller motor will have more resistance in the windings and losses there I imagine but anything else I could point to as to what is happening in an overload situation?
I gather theres no way around the low efficiency at slow motor rpm and its a mechanical limitation of a hub motor as it has less speed to utilize for power at low rpm
I gather theres no way around the low efficiency at slow motor rpm and its a mechanical limitation of a hub motor as it has less speed to utilize for power at low rpm
spinningmagnets
100 TW
Below 20-MPH, light weight is a major benefit, above 30-MPH aero becomes very relevant (though rarely used). If you limit yourself to speeds at 20-MPH or power, the way to increase motor efficiency is to speed up the magnets, and the most common methods are to A) go to a geared motor (typically a 5:1 gearing to speed up the motor), or B) with a direct drive hubmotor, spec a higher Kv motor and swap to smaller diameter wheel.
Going to a higher voltage system can sometimes reduce the max amp-draws from the battery. Lowering the amps that are pulled from a battery can sometimes provide more range from the same pack.
Going to a higher voltage system can sometimes reduce the max amp-draws from the battery. Lowering the amps that are pulled from a battery can sometimes provide more range from the same pack.
Hummina Shadeeba
10 MW
why heat and not flux? Is it largely copper losses because a small motor has more electrical resistance in the windings given the same kv as a larger motor? And why would the windings be more inefficient at producing flux at slower speeds? If its all copper losses cant you over-compensate with huge motor that will have less resistance and lower copper losses at least? whats the biggest hub motor on the grin simulator? I read a motor is ideally sized if the copper losses and iron losses balance and I'm assuming with a hub motor it's largely copper losses being a slow turner. simply need a bigger motor?
for context I've been selling/riding these little skate hub motors with 4725 stators. steelhubs.com Riding a pair they get hotter than I'd like and moving on to being able to run a single 4770 stator. Long wheel 80x101mm. 12x14. 90kv for the 30mph speed on 12s and don't want to go slower! Its a done deal now and I'm impatiently waiting for them to arrive but in the meantime trying to figure out whats going on and dorking out on the computer waiting for turkey.
looking to figure iron losses too with this app and curious if any of you have.
http://www.infolytica.com/en/applications/ex0166/
for context I've been selling/riding these little skate hub motors with 4725 stators. steelhubs.com Riding a pair they get hotter than I'd like and moving on to being able to run a single 4770 stator. Long wheel 80x101mm. 12x14. 90kv for the 30mph speed on 12s and don't want to go slower! Its a done deal now and I'm impatiently waiting for them to arrive but in the meantime trying to figure out whats going on and dorking out on the computer waiting for turkey.
looking to figure iron losses too with this app and curious if any of you have.
http://www.infolytica.com/en/applications/ex0166/
dogman dan
1 PW
I'm such a dumbass, I'll never understand the motor theory. I can only evaluate what I have in hand on the road. Is it working, is it failing?
So, your motor gets hotter than you'd like, that's clearly not ideal efficiency. Not if it does that on flat ground anyway. What is too hot? Does it sizzle your finger if you touch it? That's too hot. If its just very very warm, Id say you hit it on the button. Since you are stuck in one gear, you have two choices.
One is to lower the ideal rpm with the winding, or increase the rpm with wheel size or gearing, and be stuck with a much slower result.
The other is to live with a moderately inefficient setup, but one that does not melt down if you climb some hills. Ideal efficiency is only going to happen if you have good strong power, but the motor winds out at a low speed,, or, you go faster to turn the motor as fast as the load will allow. Both also require that its not an overload.
The fun choice, the one most will prefer, is to go for the faster motor, and live with some inefficiency when they happen to travel slower. They want to cruise at a comfy speed, whatever that may be for them, yet still have 25% more power there if they want or need it, like for a hill. So you cruise in a less efficient rpm, but with the load light enough so that is not a big problem. Your best choice if the motor is just getting too hot, is perhaps to get more copper and more magnet. So the ratio between motor size and load is not so overloaded. To stick with the same size and same winding motor, then you must gear down and go slower, which also lowers the load.
So, your motor gets hotter than you'd like, that's clearly not ideal efficiency. Not if it does that on flat ground anyway. What is too hot? Does it sizzle your finger if you touch it? That's too hot. If its just very very warm, Id say you hit it on the button. Since you are stuck in one gear, you have two choices.
One is to lower the ideal rpm with the winding, or increase the rpm with wheel size or gearing, and be stuck with a much slower result.
The other is to live with a moderately inefficient setup, but one that does not melt down if you climb some hills. Ideal efficiency is only going to happen if you have good strong power, but the motor winds out at a low speed,, or, you go faster to turn the motor as fast as the load will allow. Both also require that its not an overload.
The fun choice, the one most will prefer, is to go for the faster motor, and live with some inefficiency when they happen to travel slower. They want to cruise at a comfy speed, whatever that may be for them, yet still have 25% more power there if they want or need it, like for a hill. So you cruise in a less efficient rpm, but with the load light enough so that is not a big problem. Your best choice if the motor is just getting too hot, is perhaps to get more copper and more magnet. So the ratio between motor size and load is not so overloaded. To stick with the same size and same winding motor, then you must gear down and go slower, which also lowers the load.
Hummina Shadeeba said:
I'll have a stab at this...
Yes, small motor = less copper = greater resistance = more copper loss = more heat
The flux produced by the winding is only a factor of the current flowing through it - rotational speed of the motor is irrelevant
Yes, bigger motor = more copper = less copper loss
Yes, IIRC peak efficiency is always at the point when copper losses equal core losses (apparently this is usually 80-90% of no load speed)
Yes, most hubmotors are rarely spun fast enough to achieve their power and efficiency potential (hence why many advocate using as small a wheel as possible). They do not have to be this way: many common hubmotors and controllers are cheap and poorly designed.
Remember that a motor can only produce so much torque (before it magnetically saturates or the winding overheats). If you spin it slowly then power output will be low (Power = torque x speed). Use gearing (or different motor geometry for a hubmotor) and that same torque at higher speed give you much more power. Peak power is always at 50% of no load speed, and efficiency here is also 50%.
Hummina Shadeeba
10 MW
thanks. Does the stator saturate in an overloaded slow moving motor because recently I'd read that the point of saturation is way beyond a field strength we would be achieving (2T according to wiki is the saturation point of silicone steel). I dont know what we would be achieving, I imagine it's some formula including amps and turns and maybe airgap and magnet strength and steel mass.
I think I recall seeing some FEMM models that showed flux approaching the 2T level, so achieving that level of flux sounds entirely feasible.
Only amp-turns make flux, voltage and speed do not matter. If you take a given motor with a given winding and iron, it will saturate at Xamps whether it's rotating at 10rpm or 1000rpm.
Going slowly up a steep hill means asking for a lot of torque from a motor for a long period of time. If you go up the hill faster (steady speed) requires the same torque but for much less time.
Only amp-turns make flux, voltage and speed do not matter. If you take a given motor with a given winding and iron, it will saturate at Xamps whether it's rotating at 10rpm or 1000rpm.
Going slowly up a steep hill means asking for a lot of torque from a motor for a long period of time. If you go up the hill faster (steady speed) requires the same torque but for much less time.
Hummina Shadeeba
10 MW
Id have thought a motor would require less torque and therefore need to produce less of a magnetic field to get needed power if it has a greater speed. No? I thought that's why a hub motor was less efficient as at slow speeds it's forced to produce torque with less speed.
Yes and no. If output *power* is kept constant, then yes, you need less current/flux/torque as speed increases, but high speed at fuckall torque won't climb a hill, accelerate a load or overcome air resistance unless gearing is used (which simply reduces speed and increases torque - putting you back where you should have been in the first place...).
A motor has peak and continuous torque capabilities, if you exceed either then the motor will overheat. The torque required to propel a vehicle can be calculated without too much difficulty.
A motor has peak and continuous torque capabilities, if you exceed either then the motor will overheat. The torque required to propel a vehicle can be calculated without too much difficulty.
dogman dan
1 PW
Now that puts it in terms even I can understand. Exceed the continuous torque number, whatever that is, it's overloaded. Then it gets all hot.
Most typical bike hub motor stuff, can easily be overloaded. Fat guy, or just a steep enough hill.
And of course more copper just makes common sense. More copper, wider magnets, higher continuous torque number.
I keep thinking it would be cool, if they made a much larger geared type motor. 30 mm magnets, twice the copper, and a much larger diameter. 16 mm wide stators can only have so much copper, and end up more overheated that the typical smaller DD motors.
Likely would end up too wide for a regular bike, but a pedicab with a fat bike fork,, yep.
Most typical bike hub motor stuff, can easily be overloaded. Fat guy, or just a steep enough hill.
And of course more copper just makes common sense. More copper, wider magnets, higher continuous torque number.
I keep thinking it would be cool, if they made a much larger geared type motor. 30 mm magnets, twice the copper, and a much larger diameter. 16 mm wide stators can only have so much copper, and end up more overheated that the typical smaller DD motors.
Likely would end up too wide for a regular bike, but a pedicab with a fat bike fork,, yep.
It was reading Miles' motor design threads that made me aware of a motor's inherent torque capability. I'd see him using XNm/kg cont. to benchmark the power density of a motor design. Well-designed non-hubmotors are putting out ~5Nm per kg of motor mass continuous, whereas our typical Chinese hubmotors are capable of maybe 1-2Nm/kg. Combine that lowish torque with typical slow hubmotor speeds and the result isn't much power and you see the power/weight appeal of a big Astroflight motor with gearing working as a mid-drive.
As people like Luke have pointed out, there's no reason a hubmotor has to be like that. Sure, you can't spin it any faster in a given wheel size, but you can often increase the motor diameter and pole count to increase the velocity the coils pass the magnets. The downside there is higher manufacturing cost and higher switching speed required from the controller (better/more expensive controller).
It's an interesting field and one I can't claim to fully understand, but I try to pick up the main points from those who do
As people like Luke have pointed out, there's no reason a hubmotor has to be like that. Sure, you can't spin it any faster in a given wheel size, but you can often increase the motor diameter and pole count to increase the velocity the coils pass the magnets. The downside there is higher manufacturing cost and higher switching speed required from the controller (better/more expensive controller).
It's an interesting field and one I can't claim to fully understand, but I try to pick up the main points from those who do
Similar threads
