[steve-]

For example a manual Subaru. Is it as simple as getting a mounting plat on the transmission with a coupler for the e-motor or is there some ecu mods needed? Would this be different if it was automatic all wheel drive? Are there more important factors to be consider for awd?

I'm thinking of converting my 06 Subaru legacy.

Alternatively for making an awd ev, is skipping the Axel and differential by having 4 smaller motors on each wheel viable or worth while?

And if awd really isn't the way, then it seems rwd is a common design, why?

Sorry, so many questions xD

 

[steve-]

So it seems that having a bigger motor means a bigger rotor and a less of a gap between the stator and rotor. This significantly helps the output. Also guessing more weight of the added motors. Not exactly sure how much weight savings in comparison to removing the axel. I'm getting the info based on the Tesla rear wheel drive motor. I'd still like to test 4 lesser motors at each wheel rather than 1 large motor at some point.

 

[Duncan]

Think in terms of "Torque" not power and for an electric motor Torque is roughly proportional to weight

So if you drive a wheel directly you need a lot bigger motor

Tesla use a 35 kg motor and a 10:1 reduction gear to drive two wheels
Two motors that drove each wheel would be about 175 kg each

 

[brian_]

For example a manual Subaru. Is it as simple as getting a mounting plat on the transmission with a coupler for the e-motor or is there some ecu mods needed?

Fundamentally, yes, it's no different from 2WD... because one engine is replaced by one electric motor. Of course most modern AWD systems are managed by a computer, which expects to talk to the engine controller to reduce power as required for traction control... but again, that's the same as 2WD.

Would this be different if it was automatic all wheel drive?

Yes, but again it is the same problem as with 2WD: an automatic transmission is unsuitable for use with an electric motor, because it is designed and programmed to keep the engine turning at some minimum speed or higher (by allowing slip in the torque converter), and that is unnecessary and unwanted with an electric motor. If you allow the transmission input (motor output) to stop, the transmission loses hydraulic pressure and so it can't stay in gear. Conversion of an automatic transmission to work well with an electric motor is expensive... and pointless, since the whole mess can be avoided by starting with a manual transmission (or not using the car's original transmission at all).

Alternatively for making an awd ev, is skipping the Axel and differential by having 4 smaller motors on each wheel viable or worth while?

Actually mounting motors with the wheels on the hubs is terrible for suspension performance (ride and handling), and packaging them in there is quite difficult. This is true whether the motor directly drives the wheel (which means the motor is way too big and heavy, as Duncan explained), or there is both a motor and a reduction gearbox (so the motor is smaller but there is also that gearbox to fit in).

Some trucks and buses mount the motors (with reduction gearboxes) on the axle beams (which is still bad for ride and handling but acceptable in heavy trucks and buses), but that means beam-axle suspension... not the independent suspension of a modern car such as any Subaru. It could work for a mostly low-speed off-road 4X4.

Keeping the motors on the vehicle's frame and driving through axle shafts - but with one motor per wheel - is a good design... but not done by do-it-yourself builders because they don't want to pay for four sets of motor and controller, and don't want to (or don't know how to) coordinate their management. This is the design of some prototype vehicles (which are "about to go into production"), such as the Bollinger and Rivian SUVs and pickup trucks.

No AWD production electric vehicle uses a single motor and drives all four wheels with it through a bunch of mechanical components. They all have at least separate front and rear motors. Even AWD hybrids generally don't use a traditional mechanical AWD system: they drive one axle through a hybrid transmission, and drive the other axle (usually the one at the far end of the vehicle from the engine) with just an electric motor or pair of motors (one per wheel).

And if awd really isn't the way, then it seems rwd is a common design, why?

Particularly in a do-it-yourself conversion project, the complication of doing AWD well is just too much to be worthwhile for almost any builder. Doing it by just replacing the engine with an electric motor, and keeping a simple non-electronic 4WD system, has been done by a few builders, typically with vehicles such as old Land Rovers or Suzukis. Most conversions are RWD because most of them are of old cars or sports cars, but front wheel drive conversions are not unusual, and have usually been done the same way as RWD (adapt electric motor to the car's original transmission).

 

[brian_]

So it seems that having a bigger motor means a bigger rotor and a less of a gap between the stator and rotor. This significantly helps the output. Also guessing more weight of the added motors. Not exactly sure how much weight savings in comparison to removing the axel. I'm getting the info based on the Tesla rear wheel drive motor.

It's not clear to me that you understand how a Tesla is arranged, or what you are concluding from what you believe the Tesla design to be.

A Tesla is like every other common production electric car. It has one motor for each driven axle (so one motor for RWD, and two motors for AWD). Each motor is mounted on a reduction gearbox, to adapt the high speed (up to about 12,000 rpm) of the motor to the much lower speed of the wheels. The gearbox includes a differential, just like the differential of any car, to split the power to the two wheels and allow them to turn at different speeds in a turn. Axle shafts (or halfshafts) take the power to the hubs, which are carried on an independent suspension. The motor and gearbox are mounted between the wheels, leaving the floor of the car between the axles clear for battery space.

The first AWD Tesla Model S (which they call "Dual motor") used two smaller motors (one per axle) instead of the one larger motor of the RWD. Two smaller motors means more weight than one larger motor, because the smaller motors are more than "half sized" (so the total power is higher), and just because there are more parts (due to twice as many gearboxes and drive axles). Later they offered "Performance Dual motor" versions, which use the big motor at the rear and the usual smaller motor at the front, just to provide more performance.

 

[MattsAwesomeStuff]

Keeping the motors on the vehicle's frame and driving through axle shafts - but with one motor per wheel - is a good design... but not done by do-it-yourself builders because they don't want to pay for four sets of motor and controller,

I wonder...

Can't you just drive all motors in series or parallel off of a single controller?

Electrically it's no problem. So what would happen to the handling if you did that?

I've got a pair of ~60lb 7" motors I'm half tempted to attempt an AWD system with, by leaving the RWD intact (well, intact in terms of transmission) and somehow repurposing an FWD from another vehicle's front assembly but, CVs mated to the two motors.

Not really seriously considering it, just idly considering it since I have the extra two motors.

 

[brian_]

Can't you just drive all motors in series or parallel off of a single controller?

Electrically it's no problem. So what would happen to the handling if you did that?

I've got a pair of ~60lb 7" motors I'm half tempted to attempt an AWD system with...

If you're talking about brushed DC motors (and I assume that you are)... yes, you can probably do that, but it doesn't work well. If you connect the motors in series with each other they get the same current, and so would apply about the same torque torque to each wheel (which is good) but it's not clear to me what would happen when the two tires have different traction or the speeds are different due to turning a corner. Connected in parallel if one slips it will take off, possibly to excessive speed.

With AC motors it doesn't work at all: when a car turns a corner every wheel turns at a different speed, so induction motors will produce very unequal torque and synchronous motors... well they won't be in synch. It would likely be as bad as trying to turn on dry pavement in part-time 4X4 with locked differentials.

I've got a pair of ~60lb 7" motors I'm half tempted to attempt an AWD system with, by leaving the RWD intact (well, intact in terms of transmission) and somehow repurposing an FWD from another vehicle's front assembly but, CVs mated to the two motors.

This actually makes a lot of sense to me, to make a hybrid which normally sends the engine power through an efficient mechanical path to the rear axle, but can drive the front when desired and regeneratively brakes with the front axle.
Most current AWD hybrids essentially do this, although in most cases the electric-only is the rear, because the engine is in front and driving the front axle in a typical FWD configuration. Of course it's not a viable hybrid system unless there is a motor-generator in the system of engine and rear drive system as well; if this is a pure battery EV, then the only challenge is coordinating the front and rear motors.

All you need from a vehicle with front wheel drive is the driven hubs and shafts, as long as the suspension can accommodate the parts. A big challenge is fitting motors and reduction gearing (for each motor) in around the engine which is presumably in the same compartment.

 

[MattsAwesomeStuff]

it's not clear to me what would happen when the two tires have different traction or the speeds are different due to turning a corner.

I'm thinking that the outside corner would just rotate faster.

But then, the inside corner would try to shove itself straight, wouldn't it?

I kind of want to rig it up just to try.

This actually makes a lot of sense to me

Oh, let me be clearer then.

What I meant was to replace the gas engine on an RWD with an electric motor. But then add an electric FWD too. The motor being smaller than the engine might leave enough room for the FWD motors. So, 3-ish electric motors, no gas.

Like I said, thinking, but not planning.

 

[steve-]

Very insightful, thank you everyone. I'm a lot less confused lol.

I'll keep the legacy a combustion, and start on a lighter front or rwd Honda down the road. To get my feet wet I'll start on a moped with an electric motor kit, restore/make a battery pack. Cheers 🙂

To persue AWD EV wouldn't a rear engine car be easier, in terms of making space for a front engine?

Also if a wheel slips, isn't there a way to apply the brake and match the other wheel? I'm pretty sure that's what my Subaru does.

 

[brian_]

Also if a wheel slips, isn't there a way to apply the brake and match the other wheel? I'm pretty sure that's what my Subaru does.

Yes, but it would be ridiculous in an electric car with a motor at each wheel to waste energy by using a motor to a wheel against turn a brake, instead of just controlling each motor properly. In a vehicle with one engine or motor a differential needs to be used to split the mechanical power between the wheels, and so designers resort to energy-wasting brake and clutch systems; there's no reason to do that with separate motors.

 

[brian_]

This actually makes a lot of sense to me, to make a hybrid which normally sends the engine power through an efficient mechanical path to the rear axle, but can drive the front when desired and regeneratively brakes with the front axle...

Oh, let me be clearer then.

What I meant was to replace the gas engine on an RWD with an electric motor. But then add an electric FWD too. The motor being smaller than the engine might leave enough room for the FWD motors. So, 3-ish electric motors, no gas.

Okay, I wondered about that, which is why I included this:

... if this is a pure battery EV, then the only challenge is coordinating the front and rear motors.

The clarification also suggests the motor for the rear would be in the original engine location, not in the transmission location or at the rear axle. That means all three motors are "under the hood", which leads to the pure EV version of this:

A big challenge is fitting motors and reduction gearing (for each motor) in around the engine which is presumably in the same compartment.

With three motors under the hood, plus their controllers, plus the radiator for any liquid cooling and likely the charger, there will be little if any room left for battery. That means finding space for all of the battery pack somewhere in the back, in a car which was not designed for that and has a shaft running down the middle of it (from a motor in the front to the rear axle).

To persue AWD EV wouldn't a rear engine car be easier, in terms of making space for a front engine?

Yes, but there is only one rear engine car still in production (the Porsche 911). Old rear-engine cars exist, but have their own issues (such as all of them being piles of rust, equipped with antiquated suspension). Mid-engine cars (with the engine behind the driver) would also work, but they tend to be expensive, even exotic, because it is fundamentally an impractical layout. All that is really needed to make the multi-motor AWD setup more viable is to be able to put the rear drive motor at the rear axle, but Matt may be thinking of his Opel GT plan, and that car has a beam axle in the rear... it would need independent suspension.

 

[brian_]

I'm thinking that the outside corner would just rotate faster.

Yes, but how would the motor respond to that, in terms of current and voltage, and torque produced?

But then, the inside corner would try to shove itself straight, wouldn't it?

Same problem, other side. Nothing tends to make the car straighten, but the reaction of the motors to different speeds determines how the car is affected. Ideally, you would want the motor on the outside of the turn to produce more torque (in proportion to the greater traction it has due to dynamic load transfer), but that's not going to happen.

Equal torque would be fine... until a tire (usually the more lightly-loaded inside tire) spins, at which point you would want its torque to be reduced to keep its speed in proportion to that of the non-spinning tire (depending on the radius of the turn). Nothing like that is going to happen without active management of separate motor controllers in a coordinated way.

 

[steve-]

Exploring alternative electric motor locations, configurations.

Here's an electric wheel motor that can be added to any car and make it AWD. Comments point out that the direct gear drive is susceptible to dirt and wear.
https://m.youtube.com/watch?v=iQ4lTPVR3qc

Alternatively there's a hub motor made into the rotor. Only works on low driving speed.
https://m.youtube.com/watch?v=kiQaCJOWTrY

This motor is mounted on the end of a transmission which then connects to the axle. It might be possible to apply the same concept except after the differential?
https://m.youtube.com/watch?v=pGH2MfCbF0k

One last motor. It doesn't have the possibility to make a car AWD, but is said to improve torque like a supercharger. I'm guessing it could help fuel economy.
https://m.youtube.com/watch?v=4-Vck6oCi3A

 

[SWF]

For example a manual Subaru. Is it as simple as getting a mounting plat on the transmission with a coupler for the e-motor or is there some ecu mods needed? Would this be different if it was automatic all wheel drive? Are there more important factors to be consider for awd?

I'm thinking of converting my 06 Subaru legacy.

Alternatively for making an awd ev, is skipping the Axel and differential by having 4 smaller motors on each wheel viable or worth while?

And if awd really isn't the way, then it seems rwd is a common design, why?

Sorry, so many questions xD

Have a look at this site, which provides a lot of detail about converting a 2006 Subaru WRX.

 

[steve-]

Have a look at this site, which provides a lot of detail about converting a 2006 Subaru WRX.

I really appreciate the link. Its a thought out full feature rebuild, thoroughly documented. So much so that I'm very intimidated to bother with my half azz diy attempts lol.

 

[brian_]

Have a look at this site, which provides a lot of detail about converting a 2006 Subaru WRX.

I really appreciate the link. Its a thought out full feature rebuild, thoroughly documented. So much so that I'm very intimidated to bother with my half azz diy attempts lol.

It's thoroughly and clearly documented, which I certainly appreciate, but that also means some of the undesirable features and false assumptions are clear. Some examples:

• It now has about 50:50 weight distribution... until you add anything to the rear seat, at which point it becomes undesirably rear-heavy. The AWD system is not designed for a rearward bias.
• All of the battery was placed behind the rear axle, and the hefty motor is in the original engine location ahead of the front axle, so the mass is too close to the ends, hurting dynamic response.
• The rear spring rate was not changed (the springs were just shimmed), so the rate is now too low for the mass of the rear of the car.
• "The car neither understeers nor oversteers"... that's completely implausible, so this indicates that the builder is not a sufficiently skilled and experienced driver to be able to tell what the car is doing. It probably does handle acceptably for typical driving.
• 85% motor efficiency might be achievable at one specific speed under ideal conditions, but most of the time it's just a fantasy.
• He thinks that an automatic transmission would be a good idea.

I think it's fascinating that this writer makes the usual nonsense claim that "Electric motors provide maximum torque at all speeds", then shows a graph which clearly shows that this is not true!

 

[MattsAwesomeStuff]

It's thoroughly and clearly documented, which I certainly appreciate

Wow, is it ever.

I wish every conversion was so well documented.

Really a gold standard for someone sharing a finished project.

He seems like a very precise person, which is odd that he makes some quite basic errors.

 

[SWF]

It's thoroughly and clearly documented, which I certainly appreciate, but that also means some of the undesirable features and false assumptions are clear. Some examples:

• It now has about 50:50 weight distribution... until you add anything to the rear seat, at which point it becomes undesirably rear-heavy. The AWD system is not designed for a rearward bias.
• All of the battery was placed behind the rear axle, and the hefty motor is in the original engine location ahead of the front axle, so the mass is too close to the ends, hurting dynamic response.
• The rear spring rate was not changed (the springs were just shimmed), so the rate is now too low for the mass of the rear of the car.
• "The car neither understeers nor oversteers"... that's completely implausible, so this indicates that the builder is not a sufficiently skilled and experienced driver to be able to tell what the car is doing. It probably does handle acceptably for typical driving.
• 85% motor efficiency might be achievable at one specific speed under ideal conditions, but most of the time it's just a fantasy.
• He thinks that an automatic transmission would be a good idea.

I think it's fascinating that this writer makes the usual nonsense claim that "Electric motors provide maximum torque at all speeds", then shows a graph which clearly shows that this is not true!

No DIY conversion is perfect, partly because you have to work within the original design of the car. This guy did a nice, high-quality and functioning conversion, documents it so that it may be helpful for others, and you review his site and the best you can do is offer pedantic comments. Nice. I thought about contacting him and pointing out this post so that he could come defend himself, but decided it would be a waste of his time.

 

[MattsAwesomeStuff]

It now has about 50:50 weight distribution... until you add anything to the rear seat, at which point it becomes undesirably rear-heavy.

I had a question about this, since I'm about to struggle with the same myself.

Isn't every vehicle's weight distribution whatever it is, until you "add stuff" like cargo and passengers to it?

You wouldn't calculate it's proper weight distribution assuming rear passengers and such, or, I don't see how that's any more typical than an empty seat.

I think there's part of this I'm missing.

 

[brian_]

No DIY conversion is perfect, partly because you have to work within the original design of the car.

Yes, obviously. This one has done about as little as possible to modify the vehicle design to accommodate the conversion.

This guy did a nice, high-quality and functioning conversion, documents it so that it may be helpful for others, and you review his site and the best you can do is offer pedantic comments. Nice. I thought about contacting him and pointing out this post so that he could come defend himself, but decided it would be a waste of his time.

There's no need for him to "defend" himself. He has done a very good conversion by DIY standards, and a wonderful job of documentation.

But if someone thinks that the nice documentation and instructive style mean that this is the "right" way to do a conversion, they should be aware of the reality.

An example would be that simply changing rear springs, instead of just shimming them, would be a meaningful improvement. Unlike many home builders, I don't think that he even suggests that he is not absolutely sure of the best approach, or that there might be a better design of any aspect. The claim that the car neither understeers nor oversteers is similar: it is pronounced with such confidence that people might believe it.