DIY Electric Car Forums banner

In wheel hub electric motors??

11695 Views 39 Replies 13 Participants Last post by  Duncan
Im new to this forum.
I have a intrest in EV cars and looking to do a budget build.

But firstly doesn anyone know of in in wheel hub electric motors?
I recent read a article about a university that converted a frwd car to awd and had two in hub motors in the rear tires. Supposedley only costing $3000. $3000 sounds very low for this type of build but i was intrested in the in wheel motors.

Anyone know where to start looking for one?

Or maybe even a motor that would connect to the drive shaft of a rwd.

Im intrested in keeping the engine in place and having some type of hybrid build.

Has anyone done this successfully?

Any belp would be appreciated.
1 - 12 of 40 Posts
Going by a formula I was given of ~7hp per 1000lbs of vehicle needed for decent performance (more if hilly terrain expected), these look like a possibility.
7 continuous horsepower per thousand pounds of vehicle is okay to keep the vehicle moving at a constant and moderate speed on level ground. To accelerate well by current automotive standards, 70 horsepower per thousand pounds (for as long as it takes to get up to speed) is closer to a reasonable power budget.
The pictured thing from MTSU is an old dead-end project. An earlier discussion of this particular device can be found starting at post #473 of the open source hub motor/wheel motor thread, but there are lots of wheel motor discussions in the forum.
Concept: Using two BLDC motors, one to drive each driveshaft.

1. Two motors are twice as powerful as one and BLDC can be wired in delta and Star configuration meaning that a virtual (electrical) gearing is available for low and high speed rpm.

2. Disposing of a gearbox will save space, weight and resistance.
There seem to be two separate ideas being mixed up here.
  1. Separate motors for each wheel
    This eliminates the differential, not the transmission.​
  2. Changing wiring configuration of whatever you consider to be a "BLDC" motor
    This might help reduce the need for a multi-speed transmission, but has nothing to do with separate motors for each wheel.​

As already noted, this really isn't about in-wheel motors; the only connection is that in-wheel motors are inherently one-per-wheel.

What isn't clear is whether the intention is to use the motors to drive the wheels without any reduction gearing. Again, this is unrelated to in-wheel motors, which can be used with or without gearing. I don't any sense in using motors without gearing... and that's why essentially no one does it.

Two motors are obviously twice as powerful of one of the same motors, but more obvious solution to getting twice as much power is to simply use a bigger motor.
See less See more
2. If the motors are connected (wired) in parallel. Will this work as a differential.
Meaning that; When turning corners will the wheel that is moving faster transfer the extra electrical energy into the motor that is trying to turn slower or will the motors soak up the difference in energy and naturally let more energy go to the motor that is trying to spin faster through the corner.
This depends on the motor type, but seems unlikely to me to work acceptably for any type of motor. At a first approximation motor torque output is dependent on current, so maintaining equal current between the motors would provide the equal torque split provided by a mechanical differential; however, parallel connection will provide equal voltage (not equal current) and if one wheel slips due to inadequate traction the result would not likely be good.

Managing the speed difference is a well-known issue. I noticed that Curtis Instruments has a standard solution: Dual-Drive. It is mentioned as a feature in many of their controllers, and a typical application is described in an article on their site.

The idea of Dual-Drive is:
  • each motor has its own controller
  • the controllers are connected to communicate with each other
  • one controller is designated as the master (and the other the slave), just so that the control logic runs in one place
  • an angle sensor provides steering position input to the master controller
  • control parameters are set appropriately for the wheelbase, track, and steering sensor calibration
  • the motors are driven to maintain the speed difference corresponding to the current steering angle
... or from the article:
The standard Curtis dual-drive
software allows the 1236 controllers (one master, one slave) to correctly control the dual-drive operation of the truck. This includes varying motor speed on inner and outer wheels during turns to provide true differential control.
Current production EVs generally use a single motor per axle and a conventional mechanical differential, so they resort to using the same methods for individual control of wheel torque for traction and stability control. A trend toward separate wheel motors is starting, to allow this control to be executed by the motors, without dissipating power in brakes. This makes sense, but implies an advanced level of control.
See less See more
1. Might be difficult to fit two motors between the driveshafts in a front wheel drive, might be easier on a rear wheel drive.
This is an issue at front or rear, since
  1. the width of the motor assembly (across the outputs) must fit between the shortest acceptable axle shafts, and
  2. the width of the motor assembly (across the housing) must fit between the suspension mounting points.
The most common drivetrain configuration in modern cars is a transverse engine with the transmission on the end of the engine, and despite the combined length this all fits in the front of the car. The popularity of McPherson strut front suspensions is largely due the amount of space this design leaves for the engine. To address the shaft length issue, the differential is always arranged to be near the engine/transmission connection. Rear suspensions typically leave less space for a wide motor assembly, because they are designed to fit under a floor (rather than around an engine) and only need to accommodate a small final drive unit (differential).

The packaging is particularly difficult - due to the first dimensional limitation - if the motors are placed end-to-end with outputs (including reduction gearboxes) on the outboard ends. Despite the width, this configuration is used in a few cases, such as the Rimac Concept One (which is a wide car) and the front of the new Acura NSX (which has very small front motors). The normal solution is to face the motor outputs inboard, with the gearboxes between them, placing the gearbox outputs very close together near the centre; this requires that the gearbox shift the shaft centreline far enough to clear the motor cases... such as in this Xtrac:
See less See more
That photo looks like a sweet setup, which OEM makes that one?
The transmission is the P1227 from Xtrac, shown with YASA motors. It probably isn't available directly to a retail customer; the target is vehicle manufacturers. This is a "torque vectoring" configuration (two separate motors and transmissions, mounted together). YASA shows some additional configurations of this series of Xtrac transmissions: YASA P400 Series Motors with lightweight gearboxes
I'm guessing that a motor won't always have the same performance in reverse? actually i don't think it's a problem with 3phase BLDC as you can usually reverse any 2 of the 3 windings to change direction. that way you could stack motors and just reverse the direction of the one that is powering the other wheel.
Brushed DC motors routinely have the position of the brushes adjusted slightly for optimal performance in a specific direction. With one motor per wheel, you would just adjust each motor to suit its forward rotation direction... but performance in reverse is still limited.

I agree - it appears that this is a non-issue for AC motors.

... Or perhaps there is enough play in the driveshafts to off set the two motors?
Not likely - if you place two motors transversely, one ahead of the other, the shaft angle required to make up half the motor case diameter over the length of the shaft (between inboard and outboard joints) seems excessive.
Does your truck also have a tranny or is it direct drive from the motor?
Is that with a three phase AC or DC motor?
I'll have to check it out. Do you have some posts on the forum with some more info?
I assume that this is directed to Wolftronix...

He has a Solectria E-10:
Solectria E-10 Restoration

The E-10 uses two induction (3-phase AC) motors (each with its own controller), both driving through toothed belts to the same gear wheel on the shaft to the truck's rear axle, so it has a two-stage speed reduction system - a first stage by belt and a second stage by the pinion and ring gears in the axle - with a single ratio.
If we are talking in wheel motors then that is potentially 4 controllers worth of lost economy when operating outside of the nominal range for efficiency.
Four controllers don't mean any more loss than one four-times-larger controller handling the same total power.
I was just wondering if there might be a way to do something similar with three phase in order to get the best efficiency possible at top speeds. Since you need some sort of controller for three phase the only other option i could think of is to have multiple controllers that are designed to give better efficiency at different speeds.
The difference might not be much but who knows it might be as much as a 5% increase in range or it could potentially decrease the battery size by 5% which could result in a saving of weight (depending on the weight of the extra/modified controller).
I doubt that even a perfect inverter or controller would be 5% more efficient than an inverter or controller at a typical operating condition. Even if it were, that would only correspond to 5% more range if the motor were operating at this alternate operating point all of the time. So if there is perhaps 1% more efficiency to be gained (from running at this specific condition 20% of the time), wouldn't that likely be lost in the extra wiring and contactors needed to run in two different configurations?
... You pull off the rear brake discs and mount a bunch of permanent magnets in a ring bolted to the wheels.
Then mounting a ring of windings to the brake calliper mounts.
Please poke holes in this idea.
The general idea is a parallel hybrid, and that's sound... due to the benefits of regenerative braking and making power available for greater brief acceleration without a larger engine.

The electromagnetic reality is that with these in-wheel non-geared motors running at wheel speed, they will have very low braking torque (too low to be effective) and low driving power (too low to be very useful). There is a reason that every practical traction motor is geared to run at much higher speed than the vehicle's wheels (and so the torque delivered to the wheels is much higher than the motor output torque).

A practical issue of construction is sealing these large-diameter rotating assemblies. A gap large enough to tolerate road dirt would not be efficient.

But the problem which kills the idea is that the motor/generators would be insufficient as rear brakes, because they would be neither sufficiently reliable not be sufficiently effective to be the primary braking system. The assumption that front brakes by themselves are adequate is false.

This sort of scheme has been proposed as an add-on to the brakes, rather than replacing them. That still has most of the same problems, and is less desirable than a conventional parallel hybrid (with the motor/generator mounted between engine and transmission) because it would be more expensive, heavier, and less effective.
See less See more
The brake disk its self could be the rotor in an axial topology AC induction motor. No magnets needed.

You could still have your brake pads and caliper for hydraulic brakes.

Then the rest of the circle is taken up with the stator coils.
I'm sure this has occurred to many of us while daydreaming about future possibilities, but I usually imagine the rotor as PM. Induction helps both cost and temperature tolerance (needed because of the friction brake), but adds a challenge for regenerative braking at very low speed (because there is no torque without slip).

The rotor ends up completely encircled, which is a serious cooling challenge, so it would need to be vented (which is normal for front brake rotors, but not rears on production cars), with airflow radially through the stator.

Windings would normally be a major issue for a rotor which needs to withstand being used as a friction surface for braking; however, the intention would presumably to use the normal squirrel-cage design (rather than the relatively rare wound induction rotor); due to the axial flux the conductors would run nearly radially and be shorted by rings at the rim and inside of the stator face area. To my surprise - although in hindsight it makes sense - these rotors do not have insulation between the conductors and the laminations, so the brake pads could actually slide across the conductors and wear down their surface. Still, I wonder about the compatibility of a mix of iron and whatever the conductor material might be (aluminum or copper).

It may even be practical to use a solid aluminum rotor, although this would presumably be less efficient than a design with conductors across iron laminations, and would be problematic as a brake rotor. Aluminum brake rotors do exist - they're even a regular production item from Wilwood - but they are not ideal (for multiple reasons); they require an anodized, ceramic, or plasma-coated surface, and I don't know how that would affect induction motor performance. Solid aluminum would be simple.

There are two major issues with the motor-instead-of-brake idea: the loss of the brake, and the effectiveness of the motor. WolfTronix's dual-purpose rotor idea addresses the first, but still leaves the second.
If you could manage 5HP continuous per wheel, and perhaps 10-25HP peak, it could work as a bolt on hybrid conversion.
Even if the brake issues are manageable, as a motor the interrupted stator is interesting. With a PM design I can see how this could work, but with induction it is not so clear. Even if three-quarters of a stator is three-quarters as effective as a full stator, this is still a motor running at wheel speed, and not an optimal one at that.

The combination implies (in practical terms) that the motor gap would be open, rather than enclosed.

It might be a "bolt on" conversion, but it would not be a simple one, because the vehicle will have a mounting bracket for a brake caliper... not for a stator assembly. Vehicles designed for drum brakes but available with disks would probably be easiest, because they would accommodate a stator mounting plate similar to a drum brake backing plate. It also looks like a workable stator would be bulky compared to a typical disk brake caliper, so fit may be a challenge.
See less See more
1 - 12 of 40 Posts
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.