[samborambo]

For solar racing the BLDC motor would be perfect - predictable load.

Induction motors are typically more efficient across a wider speed / torque range compared to BLDC since they can be field weakened more easily. The 98% efficiency would probably be very hard to achieve on a BLDC in every day driving conditions.

So induction is probably a better choice for a practical EV. If you're intent is a wheel motor, have a look at axial induction motors.

Sam.

 

[samborambo]

I've been considering the alternator style motor, you drive it like a perm magnet but the rotor has a DC field winding instead of (or in addition to) the magnets. This lets you drop or use lightweight magnets to reduce that light load efficiency hit, but the field control seems great for a variable regen control and potentially much stronger field for DC-like torque. The power for this comes from 2 slip-rings which are much easier to deal with than normal series DC brushes, and could even be replaced with a wireless type setup if you are ok with putting rectifier diodes on the rotor.

The features/advantages are there but I'm not sure what the efficiency will look like compared to BLDC.

If you're considering a wound rotor synchronous machine (as you describe) you might as well go for a sepex (shunt wound DC). A sepex would have similar performance but with a much simpler / cheaper controller.

Then again, what benefit, other than power factor correction, does a wound rotor have over a pure induction machine?

Sam.

 

[samborambo]

What induction motor are you thinking of?
Pared with what controller?

I am trying to figure out ... personal curiosity ... what is the most efficient combination ... that is actually available to buy is ... so if there is another combination that can do better ... please let me know what it is.

The point about peak efficiency vs net operating efficiency is a good one ... but not a guarantee .... this combination has a pretty good operating efficiency range.

Definitely agreed the peak efficiencies will not likely to be sustainable in real world driving ... and real world efficiencies will always be lower than the peak best case.

But with the BLDC controller only dropping from full load ~99% down to ~98% at 25% Load ... the Wavesculptor Controller is not loosing much over the operational range.

The BLDC motor will loose more than the BLDC controller ... but what motor and controller out there is good enough to squeak out a net higher efficiency? ... I don't know of one.

If anybody does ... please let me know.

Let's clear something up. There is NO difference in power electronics between an induction motor controller and a BLDC controller. The main differences in FOC controllers is that a BLDC control algorithm must know the exact mechanical angle of the rotor WRT to the EM field and the induction control algorithm has an extra control step to compensate for slip.

Obviously a lot of R&D has gone into the Wavesculptor. I've seen one vehicle with two of these implemented as hub drive controllers and the performance, on paper at least, is excellent (that particular vehicle didn't push the limits of the controllers nor the motors as it probably would have damaged the chassis). That said, if you throw enough money at a motor controller, you can churn out some pretty high efficiencies. Engineering is the art of compromise - and usually cost vs performance. 98% efficiency may sound perfect but at about $7000 in quantity, is the extra money better off spent on batteries?

Remember that when you're assessing the efficiency of a motor, particularly BLDC motors, you need to consider efficiency vs load vs intended speed range. As I've said, as well as several others, BLDC motors are difficult to overspeed past their BEMF (base) speed. Even with phase advance in overspeed, a BLDC will incur some significant hit in efficiency due to reverse current flow. Induction motors will also have higher losses at higher speed due to eddy currents in the core but, comparing like for like, will not be as severe.

For most electric car conversions, if you decide you must run with BLDC, you'll be keeping your gearbox. To me the whole point of an EV is to do away with redundant ICE technology like the 5 speed shifter and create something much simpler.

Sam.

 

[samborambo]

Better power density and no pullout torque limit?

Nope.

Power density appears to be very misunderstood concept on this forum in electrical rotating machines since it relies on the operating speed range. Torque density is more appropriate. If maximum operating speed and voltage were no constraint, power (and therefore power density) in an induction machine will be limited mainly by winding insulation, core losses and bearing load. Mechanically commutated (DC machines) will always be limited by "zorching", ionising the air around the commutator and causing a rather eventful flashover which is a bit of a show-stopper.

Pull out torque for induction machines is usually only apparent when driven from V/Hz controllers or straight online utility frequency. FOC and DTC control schemes will decrease the frequency (for positive torque) in order to avoid pull out.

Sam.

 

[samborambo]

plus using it as a generator for regen, it blows away any other style motor. Like, it could make the brake pads unnecessary if you have a load to dump the juice into that the batts can't take.

Are you saying that induction motors aren't capable of regen?

Sam.

 

[samborambo]

they can be used as generators but are not great at it, unlike other motors this one acts like it's suddenly half it's size. A BLDC can acheive much closer to it's normal power rating as a generator.

The torque density advantage of DC or synchronous is that you are not limited by the magnetic field strength which can be induced, or what is present in magnets. A powered coil will achieve the strongest field, but since it's powered will likely not achieve the efficiency that a magnet would for most situations.

Can you cite any reference to induction motors only attaining half their rated performance when in negative torque? I've never come across such a statement before. I can't think of any technical reason why this would be the case.

Just did a quick comparison of peak torque between a netgain warp 9 and a 4 pole aluminium frame 25hp induction motor. 15% more peak torque in the induction motor and 25lbs less weight. These are standard industrial motors - not even optimised for EV use. Plus they're much cheaper than the DC motors I've seen.

Sam.

 

[samborambo]

There is a difference between individual specific controllers... whatever type of motor they are deigned for... not all controllers are equally as efficient.

Well if you want to nit-pick. Of course they use different components of cost vs performance. The topologies are identical.

Sense the question at hand is about finding a more efficient combination of components one can actually buy today ... the theory of operation becomes mute ( for this topic ) compared to the tested performance curves of real devices that one can buy today.



A very good question ... one that can easily span an entire thread / topic of it's own ... and aside from the side note ( already posted previously ) ... is not relevant to this particular thread / topic.



It is already understood and stated that the efficiency will vary ... and is not a set static thing ... and it is agreed that Combination B might have a net operating efficiency that is higher than Combination A under some situations and conditions ... even if combination A has a higher peak efficiency under certain conditions ... if you know of such a combination B ... please post the specific components ... that is the point of this thread.

*SNIP*

We can not possibly hope to cover and go over every topic related to DIY BEVs in one thread / topic ... several of your points are very valid ... I encourage you to further their discussion ... but if we try to discuss everything in one thread ... individual topics can easily get lost in the shuffle.

Ian, if you start treating the design or specification of components in a system in isolation, you'll be running in the dark. You simply cannot continue a conversation about the most ideal/efficient motor for an EV unless you look at the bigger picture:

1. Use pattern - mostly the RPM and torque regions it will be working in to attain the required efficiency.
2. What its driving - selectable gearbox, fixed reduction, etc.
3. The controller limitations - voltage, current, precision, control algorithm, etc.

Sam.

 

[samborambo]

samborambo: I'm talking about 1:1 comparison of induction and synchronous motors, both having the same stator and operational speed. For this case, power density ~ torque density. By freeing the stator from the need of providing excitation to the rotor, one can obtain significantly higher torque.

I understand you'd rather a 1:1 comparision vs speed - of course it simplifies the discussion. However, the decisions you will make on other components such as gearbox or differential may be strongly affected by the limits imposed on the speed range of the motor.

By "pullout torque limit" I meant limited span of constant power region, due to falloff of pullout torque. For synchronous machines, this region is somewhat extended.

I can't see how that's possible. BEMF for a BLDC will keep increasing above base speed. Once the controller reaches its maximum voltage (hence constant power region) the torque will tail off sharply because the allowable volt drop (and therefore current) through the windings is reduced. There are phase advance schemes used to overspeed a BLDC motor but they are limited in range due to voltage limits of the power components and the motor windings.

Induction motors on the other hand, combined with FOC drives can vary the field current (Iq)independent from the drive current (Id). This esentially means the induction motor can function the same as a shunt wound DC motor - reducing the field current as speed increases to keep BEMF within a manageable range.

Sam.

 

[samborambo]

Ehm we were talking about induction motor vs wound rotor synchronous one. This makes the difference.

Sorry, when you said synchronous machine, I assume you inferred BLDC. Excuse my rant then.

 

[samborambo]

It isn't nit picking when from the beginning I have been referring to specific components to still be doing so... it is the very concept this thread was started on.

So you were hoping to get an answer at the end of this thread that, for instance, the most efficient motor/controller combination is a Solution 1 with a 9" Netgain. Many posters (before I chimed in) identified the flawed logic of the value of efficiency vs battery capacity, peak efficiency vs wide band, etc. That's probably the reason why we're still debating over the type of motor, let alone make and model.

As already stated ... we know the context will have significant influence ... we also know that each tiny piece of the context can each by themselves be an entire thread on their own ... to try and discuss the whole bigger picture all at once , ends up including too many topics and guesses to be usefully confined to one thread / topic.

Most ideal and most efficient are entirely different things... for a wide variety of reasons... this thread is not about most ideal ... which is a very different topic.

The 'bigger picture' is only relevant if the threads topic of 'most efficient motor' were exclusively for a specific predetermined EV context... which is not what I was looking for ... that would be more useful in that one specific context ... but would be just as much less useful in any other EV context.

Put in other terms ... I am looking at it from a 'component level' design of the system not the components themselves ... instead of a 'higher level' design of the system... Yes I understand the 'higher level' design method is far more common to identify the performance requirements first and then determine the components for those requirements second ... but systems can be designed in a variety of ways ... in this specific context I am looking at one metric out of many metrics of motors and controllers ... as I said this specific topic is not trying to cover all the contexts ... but specifically the one metric of efficiency ... and in the metric of efficiency not all aspects of the vehicle but the motor and controller specifically.

Granted this view / method has its restrictions and limitations ... and I understand that.

I'm not sure that you do understand the restrictions of your design method. Imagine if an automotive engineer started a clean-slate design of an entire car by first designing the alternator. No consideration made of intended market, performance, budget or any other high level question. A "bottom-up" approach like this appears to me to be utter insanity. I've never seen such a method taught as good engineering practice - for good reason.

Many people have asked me why I'm using an induction motor for my current conversion. My short response has always been, "Because it's powerful, efficient and cheap". The long response is, "Since I have chosen (objectively, due to other design considerations) to use a manual gearbox without the remote gear stick, leading to a fixed reduction, an induction motor provides the widest power band while remaining very efficient at partial loading. Also, the motor cost me US$40 second hand."

Do you see the difference? My short answer has no context and mostly irrelevant even to another EV DIYer. Do you see now how counter-productive it is to try and prescribe a specific motor/controller without a design context?

Sam.