Field Weakening VS Gear Box VS Higher voltage battery

onloop

onloop

10 kW
Joined
Jun 24, 2013
Messages
600
Hi,

I thought i would start a new thread on this topic....

I started reading the thread Conclusive proof gearboxes are awesome and I have been thinking about this topic for my skateboards.

Unlike bikes / cars or any larger EV... skateboards don't have much space to mount mechanical systems or extra battery cells to improve the DC motors range of performance.

So if electrickery can be used to improve motor speed & overall vehicle performance that would be fantastic... But i can't seem to get much info about.

So why don't more people talk about this Field Weakening topic?
1. is it just because it is not a reliable or efficient way to run an EV?
2. its cheaper / easier to use more cells in your batteries?
3. its cheaper / easier / more reliable to use mechanical systems?


I suppose the other idea i had was stepping up the voltage of your battery using a transimpedance amplifier... (convert capacity for higher volts?) to make motors spin faster when not under big loads... such as perfect flat (or slightly sloping downwards) ground.
 
Under a given speed with equal gearing, a field weakened motor will have higher copper losses and typically lower efficiency than a naturally faster motor. The best practice is normally higher voltage or higher kv combined with more gear reduction, up until gearbox losses or stator losses dominate or if the motor inductance is too low for the controller.

Directly to your application, a typical 14 magnet outrunner begins to be dominated by steel losses around 20,000 rpm. Gear your skateboard with a single stage with this in mind if you are wanting to get high performance face plants.
 
Field weakening/timing advance is a key to the performance of perhaps all current production EV's. You sacrifice some efficiency in exchange for sustaining high torque levels to higher than otherwise possible RPMs. It does what it does, I personally find it to be extremely handy and a pleasure to use provided your motor can handle the bit of additional heat.
 
so motor timing degrees is how you control the "Field Weakening" effect?
 
onloop said:
so motor timing degrees is how you control the "Field Weakening" effect?
Click to expand...

You can view the effect from many perspectives, but yes ultimately you are shifting commutation timing ahead of BEMF.
 
So has anyone ever tested / compared the results of:

running a buck booster device to increase voltage/motor speed

> VS <

maxing out the field weakening effect to increase motor speed.

I'm not a scientist but logic tells me Boosting voltage would give you more speed & same torque. So I suppose it would be a better option.
 
Field weakening, in my point of view, basically uses the voltage drop across the motor inductance to increase the voltage room for the back-EMF voltage source. It is a little bit like having a boost converter that uses the motor inductances instead of an external one.

As far as I am concerned, field weakening is bad for efficiency so I wouldn't do it, better to have a higher battery voltage or (not so good), use a gearbox (but field weakening is the worst).
 
say for example, using a higher voltage battery is not an option. Due to physical space restrictions and charging options are fewer and more expensive.

If one was trying to create a propulsion system that offers superior low & high range charactistics it would seem that having the ability to step up the voltage for a short period of time during situations of minimal load would allow you to permanently gear the drive train to be higher torque with a top speed that is marginally less then your desired top speed - Knowing that you can engage the voltage boost if higher top speed is required. Say when cruisING the flats. Or slight declines where load is minimal.

Can I have my cake and eat it too?
 
onloop said:
So has anyone ever tested / compared the results of:

running a buck booster device to increase voltage/motor speed

> VS <

maxing out the field weakening effect to increase motor speed.

I'm not a scientist but logic tells me Boosting voltage would give you more speed & same torque. So I suppose it would be a better option.
Click to expand...


If you care about efficiency you use neither option. Choose an appreciate motor wind for whatever you pack voltage may happen to be. You can make the power you want over the speed range you want it from any pack voltage.
 
Sounds like you want a step up converter, but only at light loads.

Technically possible, but you will take an efficiency hit.

Likely much better to devote the cost and extra space required to batteries to get the voltage directly.

But if range is no concern and you have the ability to create a step up converter that can handle the current you need for the motor, it's quite possible.

Just not best practice in my opinion.

Good luck with it!
 
Subscribed..

If this thread continue, it will get a lot attention I guess!

Doc
 
liveforphysics said:
If you care about efficiency you use neither option. Choose an appreciate motor wind for whatever you pack voltage may happen to be. You can make the power you want over the speed range you want it from any pack voltage.
Click to expand...

This is my "stock" answer to timing advance/ field weakening. But I also agree that it can be very useful in situations where the system can handle extra heat and more speed is needed. Maybe there isn't the exact wind of motor available you need, or you can't quite get the correct rollout, or maybe you just want to push things harder and have extra power on tap.
 
liveforphysics said:
onloop said:
So has anyone ever tested / compared the results of:

running a buck booster device to increase voltage/motor speed

> VS <

maxing out the field weakening effect to increase motor speed.

I'm not a scientist but logic tells me Boosting voltage would give you more speed & same torque. So I suppose it would be a better option.
Click to expand...


If you care about efficiency you use neither option. Choose an appreciate motor wind for whatever you pack voltage may happen to be. You can make the power you want over the speed range you want it from any pack voltage.
Click to expand...

Don't leave out using a more powerful motor, which I view as the best option since more speed requires more power. It's often the cheapest option too. If voltage is fixed and you're not already dancing with the rpm limits, then a more powerful motor is simply a higher Kv wind pushed to higher current. Overshoot the rpm so you can gear it down too, and then you get the similar result suggested early of volting up and gearing down. Both result in all around higher performance...ie more power, more speed, and more acceleration. :mrgreen:

If you maxed out battery voltage and capacity for your space, the next stage is maximizing the motor size and weight you can live with. Then maximize the controller. At that point you'll know your current and voltage limits, and can start looking for the highest Kv of that motor without pushing the rpm limits of the single stage reduction JRH mentioned. Last would be to gear that as low as you can live with in terms of speed, allowing for appropriate headroom if you require acceleration and speed for passing.
 
Field Weakening is somewhat like Afterburners on jet aircraft. Nice for an occasional burst of speed when you can afford the lower efficiency. The rest of the system can be sized for the normal speed needs, so you're not carrying a heavier motor, higher voltage batteries, and facing additional higher voltage FET losses ALL THE TIME. FW adds no weight, it is just a software change.

50 amps of FW at 70Von a Cromotor moves the unloaded speed from 37 to 49 mph.
 
Since it seems that we have agreement that FOC is analogous to timing control, doesn't this also bring into question the accuracy of recognizing the timing from the BEMF using Hall sensors or what is being done with sensorless control? Doesn't the BEMF also get distorted as a function of RPM, more distortion as the RPM increases due to coil inductance? If your drive is sine or SVM, doesn't this mean that the drive waveform has to be distorted as well to compensate for the inductance change with RPM? If a lookup table is being used to create the drive waveform, then the lookup table coefficients have to compensate for the distortion as the RPM changes. The only paper that I have ever seen published (Pusan National University) indicates that this distortion is a significant factor in torque maximization. Motor and controller design are closely coupled issues and trying to deal with one separately seems too simplistic.

This discussion is long overdue in the motor and controller design community.
kenkad
 
in terms of getting from POINT A - TO - POINT B sometimes higher speed might be advantageous over high efficiency

I got there faster & by using more power, So in terms of travelling that is more efficient!.... Time can not be recharged, batteries can be.

So in other words maybe using power inefficiently is ok only if the result is more efficient travel. Thoughts!
 
Alan B said:
Field Weakening is somewhat like Afterburners on jet aircraft. Nice for an occasional burst of speed when you can afford the lower efficiency. The rest of the system can be sized for the normal speed needs, so you're not carrying a heavier motor, higher voltage batteries, and facing additional higher voltage FET losses ALL THE TIME. FW adds no weight, it is just a software change.

50 amps of FW at 70Von a Cromotor moves the unloaded speed from 37 to 49 mph.
Click to expand...

Reminds me of tuning mopeds. It was usually possible to make adjustments that would allow them to 'spin out' to a higher RPM with a tuck, going downhill, or with a tailwind.
 
ctirad said:
On my setup field wekaning barely returned the ~5km/h of the top speed I lost by switching from trapezoidal to a sinus controller.
Click to expand...

Depending on the setup, if there is excess torque available then field weakening can add considerable speed. If torque is already used up then increasing speed and reducing torque isn't going to help.
 
Can field weakening work the opposite way? More torque less top speed?
 
is salient structure common? what exactly does that mean? something about the stator being separated from each other>?
 
ctirad said:
On my setup field wekaning barely returned the ~5km/h of the top speed I lost by switching from trapezoidal to a sinus controller.
Click to expand...

Can you share more information? Is this on one of your scooters? Do you know the pole and slot count of the motor, stator width and diameter, rpm/volt with the old controller? Old controller settings, eg did you use an above 100% speed setting, voltage, battery and phase current limits ? Brand and model of the old controller? The effect on motor sound? The real biggie, if you've ridden the new controller enough for a longish term average what's was the effect on efficiency...ie wh/mile or km with the old controller and the new one. If you had a big gain in efficiency and a more quiet motor, then that alone could justify the change. Hell, for me finally getting controllable regen and great throttle control is enough to justify the cost if I had to buy a new controller.
 
onloop said:
is salient structure common? what exactly does that mean?
Click to expand...
I think the important aspect, for field weakening or the ability to utilise d-axis current (positive or negative), is that there is enough rotor iron at the gap. So, a rotor with recessed magnets is better and a rotor with internal magnets is best (reluctance torque component is greater). Maybe someone with a better knowledge of FOC can explain this?
 
So with small BLDC motors of 50 & 63mm diameter there would not be much effect, due to low volume of rotor iron at the gap?
 
You must log in or register to reply here.

Similar threads

V
Value of DD Motor Phase Wire Upgrade: Winding Comparison, Field Weakening, and Motor Simulator Questions
Replies
7
Views
960
vanturion
V
R
    • Like
Really good article comparing Bafang vs. Tongsheng vs. CYC
Replies
12
Views
10,379
999zip999
999zip999
amberwolf
The HI-Lebowski: a Lebowski SMD brain running a zombified Honda IMA Inverter: *a HOW-TO guide*
2 3
Replies
64
Views
15,243
amberwolf
amberwolf
S
Lightweight Retro Electric Motorcycle Design
Replies
5
Views
1,648
amberwolf
amberwolf
S
I bought a used Chevy Volt - and you probably should too!
2
Replies
26
Views
8,892
neptronix
neptronix
Back
Top