[LifeIsOnTheWire]

Hi everyone. I have a stripped 1999 Honda Civic hatchback chassis, and I am planning on converting it to a mid-engine RWD electric car.

I've been reading up on EV conversions for a while. I've pretty much decided that I don't want to use a transmission. I have a small amount of experience with designing parts for CNC milling and laser CNC cutting, so I'm going to be fabricating much of this on my own.

My question is, what kind of transmission-less 2WD drivetrain layouts have people had success with?

Is it typically wiser to use 1 motor, and connect it to a differential? Or would it be better to use 2 motors, and connect them each to their own wheel?

What kind of concerns do each of those configurations have?

 

[brian_]

Hi everyone. I have a stripped 1999 Honda Civic hatchback chassis, and I am planning on converting it to a mid-engine RWD electric car.

An interesting choice... do you at least have access to the rear hubs (or preferably complete rear suspension) of an AWD Civic, so you can drive the rear wheels?

I've been reading up on EV conversions for a while. I've pretty much decided that I don't want to use a transmission.

Since almost no one considers it reasonable to run an electric motor at the speed of the wheels, some reduction drive is required. I assume that you mean that you don't want to use a multi-speed transmission... only a single speed transmission (or gearbox), just like most production electric cars. The motor you choose will determine the reduction ratio, and that will determine what gearing configurations are reasonable.

I have a small amount of experience with designing parts for CNC milling and laser CNC cutting, so I'm going to be fabricating much of this on my own.

Are you up to building a gearbox? That's not easy, but that's what you ideally want. Otherwise, you're using something intended for an EV, or something intended for a somewhat different purpose in an engine-driven car.

Is it typically wiser to use 1 motor, and connect it to a differential? Or would it be better to use 2 motors, and connect them each to their own wheel?

Be prepared for absolute declarations of the complete stupidity of using one motor per wheel. Also, note that this is how Honda and others are driving the electric-only end of their latest AWD hybrids (Honda Pilot, Acura MDX, Acura NSX).

What kind of concerns do each of those configurations have?

Concerns with one motor and a differential:

• You don't get independent control of drive torque to each wheel, so for traction control or stability control you need to use individual brake application or an insanely complicated torque-vectoring differential... or you just live without those features that are now common on even cheap ordinary cars.
• Weight and complication of a differential.

Concerns with two motors:

• two sets of reduction gearing needed.
• Two motors and two controllers (inverters) needed.
• Control system must manage two motors in a coordinated way.

 

[LifeIsOnTheWire]

An interesting choice... do you at least have access to the rear hubs (or preferably complete rear suspension) of an AWD Civic, so you can drive the rear wheels?

Yes, I'm using the entire front-suspension system from a 2003-2007 Honda Accord. Control arms, wheel hubs, and I'm going to fabricate a fixed-steering rack to keep the wheel yaw aligned.

Its going to involve cutting out the sheet metal for the strut towers, and transplant them into the rear of the Civic.

Since almost no one considers it reasonable to run an electric motor at the speed of the wheels, some reduction drive is required. I assume that you mean that you don't want to use a multi-speed transmission... only a single speed transmission (or gearbox), just like most production electric cars. The motor you choose will determine the reduction ratio, and that will determine what gearing configurations are reasonable.


Are you up to building a gearbox? That's not easy, but that's what you ideally want. Otherwise, you're using something intended for an EV, or something intended for a somewhat different purpose in an engine-driven car.


Be prepared for absolute declarations of the complete stupidity of using one motor per wheel. Also, note that this is how Honda and others are driving the electric-only end of their latest AWD hybrids (Honda Pilot, Acura MDX, Acura NSX).

This has been one of my concerns. I'm having trouble figuring out what an acceptable reduction gear ratio could be. If I mated a single electric motor to a Ford 8.8" differential, there are dozens of ratio options available to me, ranging from 2.26 to 5.14. Would something in that range be acceptable for reduction ratio? Or am I not even in the right ballpark?

 

[nucleus]

How about a Tesla? A mid-engine Teslavic should be super-fast!

 

[Karter2]

We need to know what you want to achieve or use this vehicle for .
Daily commuter ?..fully road legal
Weekend warrior/ drag race record attempt ? Pure competition, no legal restraints ?
Project build to learn on ?
Budget ?...$5k ?...$10k ...unlimited ?
Why RWD ? that is a significant complication in many ways on a fwd chassis.
You vould do a lot worse than follow Arlo's project...
http://www.diyelectriccar.com/forums/showthread.php/1988-crx-diy-controller-and-nissan-178169.html

 

[brian_]

Yes, I'm using the entire front-suspension system from a 2003-2007 Honda Accord. Control arms, wheel hubs, and I'm going to fabricate a fixed-steering rack to keep the wheel yaw aligned.

Its going to involve cutting out the sheet metal for the strut towers, and transplant them into the rear of the Civic.

Wow, that's a project!

This approach certainly does work: it is how Fiat built the X1/9 from 128 components, how Toyota built the first MR2 from Corolla components and the second MR2 from Celica components, how GM built the Fiero from Citation components... and there are other examples of using a complete front suspension and drivetrain in the rear of another model. Of course, those were complete new vehicle designs by major manufacturers...

The track rods (instead of a steering rack and tie rods) are relatively straightforward.

The complexity of the structure is why I was expecting the use of AWD Civic components, which fit the Civic structure and accept drive axles. I suspect a sound structure will involve steel tubing, not just the panels.

 

[D a n n y^]

Wow, that's a project!



This approach certainly does work: it is how Fiat built the X1/9 from 128 components, how Toyota built the first MR2 from Corolla components and the second MR2 from Celica components, how GM built the Fiero from Citation components... and there are other examples of using a complete front suspension and drivetrain in the rear of another model. Of course, those were complete new vehicle designs by major manufacturers...



The track rods (instead of a steering rack and tie rods) are relatively straightforward.



The complexity of the structure is why I was expecting the use of AWD Civic components, which fit the Civic structure and accept drive axles. I suspect a sound structure will involve steel tubing, not just the panels.

Brian,

It's funny you bring up the MR2. I've owned 2 1993 MR2s in my life and sold each one after being reassigned elsewhere on military orders. As I disassemble my 2013 Leaf, I keep hearing this voice in my head saying..."buy another MR2!"

But I suppose that's not a true transmission-less conversion concept...


Sent from my iPad using Tapatalk

 

[brian_]

I'm having trouble figuring out what an acceptable reduction gear ratio could be. If I mated a single electric motor to a Ford 8.8" differential, there are dozens of ratio options available to me, ranging from 2.26 to 5.14. Would something in that range be acceptable for reduction ratio? Or am I not even in the right ballpark?

That's perhaps the best thing about the Ford 9" axle: it is very common and was used in many vehicles, so many ratios (and differential types) are available.

It depends on the motor, but to get most EV motors to a suitable speed at the desired vehicle speeds, a reduction ratio of greater than even 5.14:1 is desired, but you're in the right ballpark. The only production EV I've noticed with less reduction is the Chevrolet Spark: it has a very high-torque motor and 3.17:1 (model year 2014) or 3.87:1 ratio (starting model year 2015). Some DIY project builders who retain the transmission in addition to the final drive reduction report using a 1:1 gear (usually 4th on an old 4-speed or 5-speed), so their final drive reduction by itself is sufficient; however, others need lower gears, at least to start. Again, it depends on the motor.

It's pretty straightforward to calculate the wheel rotational speed for any given road speed. If the motor isn't going too fast at your top speed (too fast to stay intact, or too fast to produce sufficient power), and is going fast enough to produce enough power at lower speeds, you have a workable ratio.

If you don't like multiplying tire diameter by pi and dividing that into speed... tire charts often show a spec such as revolutions per mile - multiply that by the speed in miles per minute, and you have wheel RPM. Multiply wheel RPM by the final drive ratio, and you have your transmissionless motor speed.

 

[LifeIsOnTheWire]

The complexity of the structure is why I was expecting the use of AWD Civic components, which fit the Civic structure and accept drive axles. I suspect a sound structure will involve steel tubing, not just the panels.

Yeh, I'm thinking of fabricating my own strut towers from steel tube, and sheet metal up top for a mounting surface. I might just make a whole structure joining the towers with a cross bar, and connecting downwards to a small frame on the floor that would outline the area cut out of the floorpan with steel tube. Might look something like an integrated roll cage, but it would just be to add rigidity to the modified floorpan by joining it to the strut towers.

 

[brian_]

As I disassemble my 2013 Leaf, I keep hearing this voice in my head saying..."buy another MR2!"

But I suppose that's not a true transmission-less conversion concept...

Even if you didn't use the MR2 transmission (which is just a Corolla/Celica transmission anyway), starting with the MR2 gives you all of the right structure and suspension for a RWD car; you could probably even put a complete Leaf drive unit (motor and gearbox) in it.

The biggest problem with an MR2 is probably getting one. The biggest online auto listing service in Canada (AutoTrader) shows only fourteen of them in the entire country, plus four MR2 Spyders.

 

[LifeIsOnTheWire]

Even if you didn't use the MR2 transmission (which is just a Corolla/Celica transmission anyway), starting with the MR2 gives you all of the right structure and suspension for a RWD car; you could probably even put a complete Leaf drive unit (motor and gearbox) in it.

Yeh, not being very familiar with MR2s, I would say it would be easiest to transplant a Leaf drivetrain into the MR2 completely, and avoid using the MR2 transmission entirely.

The only real challenge would be fabricating engine mounts, and having custom axles made. Oh, and figuring out where/how to mount the battery

 

[Duncan]

Hi Life
I'm using direct drive to a Subaru diff
Works quite well - great fun to drive

http://www.diyelectriccar.com/forum...dubious-device-44370p15.html?highlight=duncan

Cheap cheerful and fast

But if I was doing it again I would be looking for a Leaf unit to go in there

 

[brian_]

Yeh, not being very familiar with MR2s, I would say it would be easiest to transplant a Leaf drivetrain into the MR2 completely, and avoid using the MR2 transmission entirely.

The only real challenge would be fabricating engine mounts, and having custom axles made. Oh, and figuring out where/how to mount the battery

It's too bad that most EV transmissions place the motor ahead of the axle line, because if the motor were behind the axle, a pretty big stack of battery could sit between the axle and firewall... just like a Tesla Roadster.

Of course, with clever packaging the central tunnel (stock fuel tank location) is an excellent place for battery mass, the front compartment has room (presumably by omitting the spare tire), and the trunk could sacrifice some height for a layer of cells in the bottom.

For his Civic, all of these corresponding locations are fair game.

 

[brian_]

If I mated a single electric motor to a Ford 8.8" differential, there are dozens of ratio options available to me, ranging from 2.26 to 5.14...

When I first read this, I thought "9-inch", because that was the classic hot rod component:

That's perhaps the best thing about the Ford 9" axle: it is very common and was used in many vehicles, so many ratios (and differential types) are available.

Fortunately, the 8.8" has also become very common in both beam-axle and independent suspension versions, so my comment still applies.

I was switching between threads and lost track of who was building what, so I didn't notice this was a Ford 8.8" proposed for the Civic conversion to rear-drive. I have four concerns with that:

1. The 8.8" might be heavier than desired because it is stronger than needed. These things are used in light trucks, mid-sized to intermediate cars, and high-power Mustangs. If you know an accommodating auto salvage operation, it might be a good idea to spend some time there with a scale and tape measure, comparing hardware.
2. The pinion shaft may be so long that it will force the motor quite far forward, making more of a packaging problem than necessary; this is one reason a transverse motor setup (like essentially every production EV) is desirable.
3. Like every normal ring-and-pinion set in a final drive separate from a longitudinal engine, this is a hypoid gearset - that means a few percent more friction loss than spur gears or a non-hypoid bevel gear set. That directly translates into less power to the wheels and shorter range.
4. The hypoid gearing also means that the input shaft is lower than the axle shafts, which is desirable to keep the input shaft low under the car in a conventional configuration, but would place a motor directly coupled to it very close to the ground. At the very least, I assume that it would need to be tilted nose-up a few degrees to raise the motor.

I'm using direct drive to a Subaru diff

When the Toyota/Subaru 86/BRZ/FR-S came out, and I assumed that it had the same final drive unit, I asked some of my friends about its durability - after all, it usually handles only the rear-drive portion of an AWD Subaru's output. They said that people still building Datsun 510s (and there are apparently still many of them) can't get the Datsun units any more, and are adapting Subaru bits and putting lots of power through them successfully. I now realize that the Datsun/Nissan and Subaru final drive units are from the same series: R160 (160 mm or 6.3" ring gear) for Datsun 510 and old Subarus, larger (R180 - 180 mm or 7" and R200 - 200 mm or 7.9") for newer larger Datsuns, newer Nissans, and more recent Subarus.

As it turns out, the 86/BRZ/FR-S doesn't use a Subaru (or Nissan) diff; it has a Toyota (Lexus) unit. From another forum (in a post which also lists the ratios available):

The differential is not an R200, it is a Toyota unit. In the past referred to as the G-series (8.0") it is now referred to as the F20SX/F20TX/FD20A/FD20AT (205mm) unit. The same basic differential was used in many Toyota cars and trucks, notably the 86-92 Supra, 01-05 IS300 (with the same Torsen T2 LSD that the Zx6 gets), and the 06-15 IS250. The older models usually have a different outer case (pumpkin), however the internals are still near identical. The newer models like the IS300/IS250 have the same case and internals as the Zx6 in the 86/FR-S/BRZ.

Anyway, no one was asking about hardware from a 86/BRZ/FR-S, but it does introduce another option. This Toyota alternative is an "8-inch"; it may not be any lighter or more compact than the Ford 8.8".

The common Subaru unit looks very long, which is fine in vehicles with the motor located in the front of the vehicle, but bad for this case.

The other final drive unit for light RWD vehicles which should be readily available is from Mazda, used in the past for the RX-7 and RX-8, but now only for the MX-5 (Miata); it has ratios up to 5.38:1 available aftermarket, and presumably there is some choice (not as extreme) in OEM ratios. There are also the many final drives for the rear of AWD vehicles, in addition to Subaru; however, these often have bulky and unwanted hardware on the input (clutches to control whether or not the rear drives). Any of these might be lighter and/or more compact than a Ford 8.8", but might also offer little choice of ratio.

 

[brian_]

...
I have four concerns with that:

1. .
2. .
3. .
4. The hypoid gearing also means that the input shaft is lower than the axle shafts, which is desirable to keep the input shaft low under the car in a conventional configuration, but would place a motor directly coupled to it very close to the ground. At the very least, I assume that it would need to be tilted nose-up a few degrees to raise the motor.

I have since stumbled across the Sunrise EV2, which has not much to do with the original Solectria Sunrise (just some of the body). It is being designed to use a motor directly coupled to the input of a Ford 8.8" IRS final drive, as is being proposed in this discussion. The Sunrise EV2 project looks old, not promising technically, and probably dead with the website largely not updated for a few years... but relevant.

This project chose a gear ratio of 5.14:1, for use with various motors (Warp 9 in the prototype) and Gross Vehicle Weight Rating of 4200 lb (1909 kg).

The Sunrise EV2 people were mounting the motor behind the axle, and rotated the final drive around the axle to point the input rearward... which also moved the input shaft from below the axle line to above it. That seems like a bad idea for lubrication, but dry sump lubrication (a pump and reservoir) was proposed to address that, and it did raise the motor to a workable height. I assume they did this just to put the motor behind the axle and avoid reversing ring-and-pinion gear rotation, not for the height, but I don't know.

 

[brian_]

Here's the rest of the available detail from Sunrise EV2 FAQ regarding their use of the Ford 8.8" IRS final drive with directly coupled electric motor (question in red text; answer follows in black):

I enjoy following your knowledgeable postings on the evtech forums. They led me to your Sunrise EV2 project. I'm curious about your setup using the inverted differential, and wonder if you’ve posted any details about your dry-sump lubrication?

We begin by cutting a 5" section out of the front center of the subframe. It was flipped upside down, and welded back in. This made room for the differential to be rotated 170° so the pinion now points backward instead of forward, and is now above rather than below the axle line. It's 170° rather than 180° to provide a 10° departure angle (so the rear of the motor is higher off the road).

The donor cars mounted the differential to the subframe with rubber mounts. We hard mounted the differential and motor to the subframe for better alignment accuracy.

The donor cars also had rubber mounts between the subframe and body. We will retain these, but left them out of the steel prototype chassis for now.

A standard spider motor coupler connects the pinion to the motor. Only a tiny bit of machining was needed to mate it to the Ford pinion.

We changed the differential to 5.14:1 gears. Two new holes were added for pressure lines for lubrication; one between the two pinion bearings, and one with a nozzle to spray oil on the ring/pinion meshing points. The axle bearings are lubricated as usual, by flung oil running down the sides. A new drain line was added at the bottom (since the stock drain plug is now at the top).

We lubricate it with Red Line ATF with an EP additive. An external oil pump provides about 1 gal/min. It also circulates the oil through the Zilla controller for cooling it. In effect, the large mass and surface area of the differential serves as the "radiator".

So it looks like the clearance improvement was a desired effect of the orientation change.

 

[brian_]

For those interested in a longitudinal motor mounting, coupled to a final drive with right-angle ring-and-pinion gears from an IRS rear-wheel-drive car...

I noticed that ElectroMotive Force (or ELMOFO) in Australia built an electric version of a Radical SR8, with this drivetrain configuration. It is just a Ford Atlas series IRS final drive, although a version built by Quaife for racing (QAF6Z). This is the "diff" out of a Ford Sierra and similar models, not familiar to those in North America (except fans of the Merkur XR4Ti) but similar to the IRS final drive units from many common vehicles (including the 8.8" in Ford T-Birds, Cougars, Lincolns, Explorers, etc, and similar vehicles from other manufacturers).