Direct-to-diff conversion possibility?
 

Hey folks. The car I'm looking to convert has a 4.44 ratio diff in the rear, and weighs 1500kg. I would prefer an AC conversion. Would an AC-50 motor handle that kind of weight? I'm not looking for massive acceleration as such. I'm also open to using a second motor to drive the front wheels :P

Just trying to firm up my plans here.

Thanks :)

No response in 24 hours. You might try adding a little detail.

• 1500kg is 3300lb. Before or after conversion?
• what are the torque and power ratings of the AC-50?
• Direst -to-diff? No transmission? No clutch?
• front-engine rear-drive or what?

Sorry, yeah. Last minute posts before bed are not so good hah.

The car is a Subaru SVX, factory curb weight of 1,590 kg (3,510 lb) according to wikipedia.

The AC-50 is rated at 120 lb-ft and 71 hp (peak power at about 3,500rpm according to EV West's chart, which according to http://www.angelfire.com/fl/procrastination/rear.html will be at 58 miles per hour, which is fast enough for me!)

By direct to diff I mean no gearbox, yes. I've seen it before once with an Impreza, and they had two motors. The first to the diff between the rear axels, and the second to another diff mounted where the gearbox used to be. They also had a good budget, though!

Ultimately I want to keep AWD, but the gearbox is automatic and I think it would be better to get rid of that dead weight (subaru gearboxes are stupidly large) hence the "direct-to-diff" thing. I suppose that means it would be a rear wheel drive car initially at least. I'm just not sure if such a motor could handle it. It is a fairly heavy car, after all. Though for the price I'd certainly hope so! A second motor would definitely mean good performance but would also mean two controllers (which would be good for Paul). My aim is to have "close to stock" performance, and at least 15km range - which should be easy with lithium cells. This is not a quick weekend project of course so I'd rather do it right.


Thanks!

SVX are awesome. I looked at one for sale a while back. My favorite Subaru alongside the 2nd generation BRAT. What color? How long have you had it?

I had a look at https://www.google.com/search?q=Subaru+SVX+cutaway. So, two motors back-to-back in the middle facing the ends? That might could work.

I'd look at the overall ratios of various single-speed motors.
Tesla -- 9.37:1
Leaf -- 7.94:1
Toyota MGR -- 6.86:1
GNK eAxle -- 12.5:1

I don't own it yet! But this is the one I'm looking to pick up. So damn pretty.

Yeah I'm hoping to make use of the driveshaft tunnel, or the gearbox space with the driveshaft heading aft. Probably best to do that, since there are already strong mounts there. Not sure yet if the forward motor would be behind the axle or in front of it, yet. Depends on what fits best!

http://d2ojs0xoob7fg0.cloudfront.net...k-1024x791.jpg

I dunno... For direct drive, I'd say that AC-50 would be good for a lightweight (say ~800kg) car, but for a 1500kg car it'll be pretty anemic without a gearbox, especially at takeoff. Assuming your ratio puts the hp peak at 60mph, you're looking at what, 6hp at 5 mph?

If you had a 5-speed in there, you could be seeing around 40-50hp at that speed, much more fun :).

Look at the orange line. The 'knee' I was speaking of is at ~3600rpm.

Torque for acceleration. Horsepower for top speed.

Interesting. If you plan on driving around 58mph, then at that gear ratio you'll have the most wheel torque possible at that speed. The problem is that you'll have the same amount of wheel torque clear from start. If you had a 2:1 ratio before the differential, or an 8.88:1 diff, then you'd have about 2/3 the total wheel torque at that speed but at lower speeds you'd have up to double the wheel torque.

Another "option" would be to use the smallest rims and tires possible. That would effectively increase the force exerted on the pavement by the wheel. Just be aware that your brakes would be more touchy if you did that.

Doing the math, you have about 110ft-lbs of torque up to 3500RPM. That would be 488.4ft-lbs of wheel torque. With 225/50 16 tires you get 573.8lbs of force on the pavement. That minus about 130N of rolling resistance equals to a little over 3.5miles per hour per second. But as you speed up air drag takes over until 48mph when your air drag will slow you down to about 3.3mph per second, assuming that you have a .29cd and an area of about 2.32m^2. At that speed acceleration will slow you down considerably from loss of motor torque since I calculated 3500 RPM to be about 48mph with your gear ratio and the tires I mentioned above. Your maximum grade you could climb theoretically would be less than a 14% grade, which isn't that bad, but you wouldn't be able to accelerate on it.

Hrm. Thanks for that! I do have a fairly steep hill to climb on my short commute. How do those numbers go with the 2:1 gearbox before the diff? Alternatively having the two motors should mean roughly twice the performance. (and twice the expense!)

Just off of the top on my head, you'd have twice the hill climbing power and acceleration at speeds below 25mph. That doubled torque would start to wean down after 25mph to about the same torque as before at around 40mph and continue down to about 60% of original (so about 2mph per second acceleration instead of 3) at around 50mph. It would be much less and limit at higher speeds and limit you from going much faster. But all in all, I think that if you were wanting a more practical vehicle the 2:1 ratio before the differential would be best. You'd be limited to a top speed around 55mph but at slower speeds it would feel more like a normal car.

So optimal would probably something less than 2:1. Maybe 1.75:1. Or 2:1 and bigger tires.

Hrm. The acceleration numbers for the 2:1 setup are slower than the standard, but not that much slower. (as per this: Subaru SVX (1995) - acceleration from 0 to 100 km/h ) The car would only be driving in town so 100km/h is not really a requirement, it'd just be embarrassing to not be able to go highway speeds hah. I'll have to look into more powerful motors. I want to go AC, for the regen, and want to keep stock-like performance. With the budget tradeoff being range.

Thanks for your help! Much appreciated.

Yes. Acceleration and hill climbing would be worse than with the original engine. I think it would still do 100km/h though. But you'd get better overall acceleration and performance with a 2:1 gear reduction than without it.

If you could put on two motors (one front and one rear) you could do quite nicely. Especially with about a 7:1 total gear reduction. The only problem is that the gear reduction boxes I've seen are very expensive. I wonder If you could get some sort of a differential with 7:1 gears. Then get two of them and put two motors in.

All IMHO of course, but...

The size of the motor isn't the only consideration. Using what I learned about a single-speed 10Krpm motor as an example; the stress on the system at low speeds is in the power circuits. It must deliver 1+ Coulombs at times. This means cooling on the inverter as well as (maybe more than) the motor and big, fat cables.

On the low end of rpm you are stressing drivetrain components that are used to forces applied in milliseconds being stressed in microseconds. Like clutches. On the high end top speed is limited by the back EMF. The FOC software needs to be tuned.

Hrm. AC is much more complicated than DC. EV West's graphs for their dual AC-34 motors show a nice sounding amount of power and torque. Certainly makes two (not joined) motors look like a good idea, but as you say the diff ratio isn't really high enough. If max continuous rpm is matched to top speed (say 70mph/112km/h, at 5000rmp, I get a 5.28:1 ratio diff. This will mean I get a decent speed range, and ~ 633 foot-pounds of torque at the start(?) per AC-50 motor. So 1266ft-lbs for a two motor setup.

I dunno. I picked the AC-50 because it seemed good. I'm open to other motors, of course. And I'd much rather use Paul's motor controller than the Curtis one that EV West would bundle with the motor. But that's also just a personal preference. Those controllers are $1900 each on their own O_o

This is an expensive hobby. :P

That's why I've changed my mind about building a car and decided a used Nissan Leaf was a better choice for me. I've seen perfectly fine Leafs for $7,000USD. And I've heard that if you look around enough, there are some used car dealers that are trying to get rid of them for as little as $5,000. To that off with living in Colorado, as in my situation, and I can get a 24% tax rebate putting a used 2011 or 2012 Leaf at less than $4,000. I'm actually looking at a low mileage 2013 for around $8,000 right now since the 2013's have a more efficient heater.

Yeah I've been tempted, but I've also been talking about doing this for nearly eight years now. I don't think I'd be satisfied with an OEM car (other than a tesla, but they're out of my budget! haha). Also we bought my wife a new car at the beginning of last year. So I figure that if it's spread over a few years I can push my budget up towards 20k without feeling too guilty. As long as the end result is fun.

Yes ... yes it is ... :(

there was a post about toyota suv hybrid diff/motor combo. for the rear wheels to make it 4wd i think it was about 100 pounds motor/diff with good power, if you just had one front and one rear. you would just have to make custom half shafts, and find a way to control
them

Here's a reduction box that goes for $4495 each. Torque Trends EV Reduction Box.

More than just a post, there's a whole thread: http://ecomodder.com/forum/showthrea...mgr-29878.html

To further Thalass's point, the part in question #G105048010 (used) was selling for $300-600 a year ago. Now they're $1200-1500.

Oh wow. Though the AC-50 from evwest is US$3850, with the bundled controller. (which on its own is US$1980, so I guess US$1870 for the motor alone - if that's possible) I suppose the MGR's current asking price isn't /too/ horrible considering it's a decent unit and has the diff built in. And I know Paul's AC motor controller would work with it as e*clipse is doing the hard work for me!

2012 Toyota. #G105048010: MOTOR ASSEMBLY, REAR TRACTION WITH TRANSAXLE. MOTOR ASSY, <-- CAD$1489.40! And it ships from my province! That's rather interesting.

[edit]Of course the MGR is a 650v motor, apparently. That's quite a bit. :/

[edit2] Just found a single MGR on ebay for about ~CAD$1000 including shipping. Hrm.

I met Mitch last week. I think the transmission is AWESOME. The F150 was not quite rolling at the time - it was waiting on software updates.

He used his small transmission in the F150 conversion. It bolts onto the output of the electric motor (I think it was a Brusa motor?) and the output is direct to the stock F150 drive shaft.

It sounds like a strong transmission. He was telling me that the destructive test made it up to 4000 foot-lbs before it failed.

Sounds pretty good. But expensive.

This is going to be a stupid question, but all of my research over the years has been for DC conversions, with the assumption that a 144v motor needs a controller fed with 144v, etc. The MGR needs 650v, so that means I'd need a 650v nominal battery pack, yes? I don't think the controller will do any high power buck conversion or anything. That is a downside to the MGR, though as a bonus I think I could go 100km on a charge with the smallest large-format LiFePO4 pack! But it would be US$11,000. Ups and downs everywhere.

Yes. As I understand it, you need 650V if you want full power and full speed.

There has been some discussion about adding a boost circuit to Paul's controller, but the components are heavy when you get to high current and high voltage. I don't know how much weight difference there is between a 325V pack with the boost circuit and a 650V pack, for example.

Paul has been working on a fixed boost to simplify things a bit.

40 a-h CALBs are about $60 each at 3.3V. 197 cells * $60 = 11,820 .. that agrees with what I've seen.

The battery packs from wrecked Leafs, Teslas, Volts .. are 1/2 to 1/3 the price of LiFePO4. I like the LiFePO4 - they stack together, no engineering required for cooling and structure ... but the cost is high.

Sadly I let myself get carried away with this, then reality hit again. The windows in our house need replacing - it was a mild winter but we felt it! So I won't be able to pick up the car I wanted. But I'm going to keep working on this, so that when I can pick up a car and start tearing it apart I'll be ready!

At least you'll have something to post in Saving@Home.

Yeah!

As it turns out: I put a deposit down on a Tesla Model 3. So I've got a couple of years to save money/get a raise from the boss. In a way it feels like giving up, but in another way I'll get a better car out of it. :D

I've done much the same thing. $100 down on a $12K electric tadpole that will ship by years end. (The company has announced a $50million IPO). Because it fits my transportation needs.

The electrified Baja Bug was on a back burner anyway; and it's still Plan B. There's a lot to learn in the meantime, and prices and parts will only get better.

Tadpole? You mean recumbant trike? Nice! I like those. An electric offroader would be a good project, actually. I wouldn't want to bring the Tesla off road, but if I could find an old 80s subaru hatchback and lift it, that would be a kickarse beast off road haha.

You're right about quality and price. In the ~9 years I've wanted an electric car things have improved dramatically. A few more years won't hurt.

I assume you're talking about Arcimoto.

Yup. Although it's not a recumbent, and short on legroom. When you are seated in one you can put your feet on the upper A-arms of the suspension and get a nice relaxing posture.

I like their product, although the gotcha is if you want a 4-season commuter from them, you're going to be paying significantly more than the $12k quoted because you'll have to buy the body kit.

I haven't ridden one in the rain myself ...yet. But the windshield is 27" wide and goes over your head. So—windy but dry.

I haven't re-read the whole thread but I think this counteracts the topic drift:

https://www.youtube.com/watch?v=WX6Xhl-SP9Q

How do you make the video go to 16.28? That's where they talk about the Torquetrends 2:1 planetary reduction box. It's splined for a Siemens or UQM motor on one end and has a driveshaft yoke on the other (held on by a grade-8 bolt so you can match the driveshaft).

This is about the simplest (though spendy) motor installation I've seen; for a open driveshaft car, anyway.