Motor question for small EV

[mhenstell]

Hi all,

Sorry if this has been asked too many times, I'm still researching this. I'm building a small, custom electric vehicle. It will weigh something on the order of 1100-1500 lbs, including the weight of the frame and four people. The motor I'm looking at is a Motenergy ME0909 (http://www.robotmarketplace.com/products/EMC-ME0909.html), a PMDC motor rated for ~4 hp continuous. My top speed requirement is only 5-10 MPH.

My hope is that someone with more experience in the motor department can tell me if a motor of this size is capable of dragging my fully-loaded vehicle around at 5 MPH for extended periods. If anyone can give me some pointers on how to convert the torque specification in the power curves to more useful "I can move this much weight" information, that would be great as well. Thanks!

 

[major]

Hi,

Take a look at a golf cart. Something like you have in mind? Notice golf carts don't use PM motors. You might want to consider that a clue. While that PM motor might technically meet your duty cycle now, you will eventually want to go faster, or longer, or heavier, or up hills, or on a hotter, sunnier day, and you'll cook the motor. If you can deal with that, go for it.

major

 

[mhenstell]

Ah, so PMDC motors aren't ideal for sort of thing, presumably because they get hotter than shunt or series wound motors? I agree that a golf cart motor would probably be more ideal, however it looks like it would take quite a bit of fabrication to get one going:

https://www.youtube.com/watch?feature=related&gl=CA&v=tkmkzr2k77s

It's not out of the question, it just adds some risk because I'm still new at metal fabrication, although I believe I have all the tools to make something like this. It looks like most golf cart motors are in the 3-4hp range, do you think that would be able to pull my vehicle at ~5mph for ~1 hour or so? Does it matter strongly if I find a series vs shunt wound motor? It looks like forklift drive motors may work as well, although information on them is a bit more scarce.

 

[major]

You're going to have to figure all this stuff out. Is it as heavy as the GC? Is your gearbox as efficient? Tires\rolling resistance? GR? Brake drag? Front end push? All stuff which can play hell with the motor load. Series vs shunt(SepEx)? Either. Depends on your controller. Forklift motors? Cool. Check out sticky thread on the subject over at the DIY Electric Car forum.

 

[spinningmagnets]

1000-lbs sounds very heavy but...at 5-MPH that means that there will be significant reduction...you will be fine. Will this run on hills or flat land?

I am assuming you will be using 48V, but...if it turns out that this particular motor is running a little warm under those loads, raise the voltage to 72V-96V, and then with the higher RPMs you get, add more mechanical reduction to arrive at the same wheel RPMs. Doing that will reduce the amps drawn, and it will run cooler. I seriously doubt at 5-MPH it will get hot. The smaller in diameter of the final drive wheel, the less reduction you will need to achieve your goals.

Plan on a jackshaft and dual reductions, so that you can use easy-to-source...and off-the-shelf sprockets...

Here's a belted primary reduction, and a chain reduction secondary:

https://endless-sphere.com/forums/viewtopic.php?f=28&t=57720

 

[mhenstell]

1000-lbs sound very heavy but...at 5-MPH that means that there will be significant reduction...you will be fine. Will this run on hills or flat land?

Thanks for that. I have a lot to learn about motors obviously, I just wanted to know if I was on the right track and in the right order of magnitude, or if I was going to throw $400 down the drain on a PMDC motor that I'll just end up burning out the first time I use it. Ideally I would be able to use it several times during the summer (maybe two weeks of total use, a few hours per day), at which point I would most likely sell or repurpose the motor for another project. I'm still looking at golf cart motors but I may not have time to get the fabrication right. Over-volting and gearing down the PMDC would be much easier to handle. This is primarily made for flat land, unpaved (desert mostly), so rolling resistance and ambient heat will be a factor, but if I can get away with the PMDC then I'll have more time to test everything. Forced air cooling is also certainly an option.

 

[spinningmagnets]

1,000-lbs is HALF of a 1969 2-seat MGB. I can push that on a flat street with just my hands and feet pushing, and all the bicycle calculators keep telling me I can only put out 200W (Motenergy at 50V X 50A = 2500W). If the Motenergy struggles (which it won't), you can sell the paired motor and controller for near what you paid. It might take a few months, but it will get bought on ebay, no sweat.

Motors typically have a continuous power rating and also a peak power rating, and both are just guesses related to heat-shedding. There are many choices to make. Motor shaft torque and wheel torque (after the reduction) will tell you if the motor will struggle. Not too hard to calculate. Higher volts are a little better than higher amps. (watts of power are volts times amps, more volts raises RPMs, more amps raises heat). Here's some links:

https://www.electricbike.com/motor-tech-learn-the-terms-part-1/
https://www.electricbike.com/diy-mid-drive/

Agni, axial flux, 205mm diameter, brushed
http://insideevs.com/wp-content/uploads/2015/06/Agni-Motor.jpg
http://endless-sphere.com/forums/viewtopic.php?f=30&t=29711
http://www.jozztek.com

Mars, axial flux, similar to Etek, but brushless
http://acuteaero.com/2010/06/16/breaking-and-fixing-the-mars-bldc-brushless-motor/

 

[liveforphysics]

If it's for dusty environment use (you mentioned desert), many open frame type brushed motors struggle.

If you can really deal with that low 10mph speed requirement, you need <1kW output power if it doesn't have some massive aero drag component.

Is this a burning man vehicle by chance? Something cheap and easy like a used wheelchair motor/wheel assembly might be a decent fit for your speed range.

 

[mhenstell]

spinningmagnets - Thanks for the links, I'll check those out, it looks like there's some good information in there.

Is this a burning man vehicle by chance?

Guilty.

If it's for dusty environment use (you mentioned desert), many open frame type brushed motors struggle.

That seems to be the consensus of the burners I've talked to. Even if it lasts the week, I don't want to completely trash it. It was mentioned elsewhere that it would be worth looking into sealed brushless motors - that two 1.5kW hub motors with a belt/chain drive would be the sweet spot in price, plus it would give me some redundancy, so I'm looking at those as well.

If you can really deal with that low 10mph speed requirement, you need <1kW output power if it doesn't have some massive aero drag component.

Good to know what sort of order of magnitude I'm looking at, thanks.

Something cheap and easy like a used wheelchair motor/wheel assembly might be a decent fit for your speed range.

I was looking at wheelchair motors before, but it seemed like they wouldn't be a good fit. The output speeds are approximately what I need on my axle, so the gear ratio would be 1:1, but they're meant to pull one person in a light chair - I'd be worried that one or even two wheelchair motors wouldn't be able to pull four people and a big metal frame. I would love to be totally wrong about this though.

 

[amberwolf]

Take a look at my SB Cruiser trike; it's for heavy loads at under 20MPH, lots of torque for startup, two (mismatched) hubmotors on 20" wheels, two "40a" 12fet generic (mismatched) controllers. Been running almost a year now, with a few changes now and then.

If you either use smaller wheels (which might be problematic for ride quality or traction), or put the hubs where they can chain-drive larger wheels, then you can have greater reduction of speed into greater torque, to make the motors work better at the higher loads but lower speeds, without overheating.

Most any directdrive hubmotor would work fine for it; if it is disc-brake compatible you can bolt the drive sprocket to the disc side.

Geared hubmotors might also work fine, but have more parts in there (meaning more things to fail).

If you can get a motor that is already wound to run at a lower speed, closer to your needed wheel speed, then there is less reduction needed, which makes things easier for you and the motor.

Powerchair motors are actually likely a good bet, though you may need at least two of the high-powered ones (meant for really heavy people). Take a look at my CrazyBike2 thread (and the old http://electricle.blogspot.com posts about it), in the first pages where it was still run on those.

It started out with a 300-ish watt version, a 2-pole brushed type, whcih was ok. But it was MUCH better with a 4-pole (more torque, less speed) 600-ish watt version. I used both at 36V, the latter one I started using at 48v, was so much torque I actually destroyed a bunch of stuff over time with each motor when chains derailed due to frame flex. I overheated a couple of them significantly, and still didnt' destroy them.

Personally I think if you used chain drive from the output axle like I did, and the 4-pole versions, two of them would probably do what you want. If you think they'll get hot, you can add fans to blow air across them (or even suck air thru them that first goes thru a dust filter you can wash out).


Someday I intend to try using these types of motors again (I still have some) for a high-torque super-duty cargo trike and trailer.

 

[mhenstell]

Most any directdrive hubmotor would work fine for it; if it is disc-brake compatible you can bolt the drive sprocket to the disc side.

Ah, ok. I took a look at your trike, very nice. So perhaps two hub motors with small sprockets grafted on would do the trick.

Personally I think if you used chain drive from the output axle like I did, and the 4-pole versions, two of them would probably do what you want.

Interesting. I see a few options for 4-pole geared powerchair motors, as well as those direct-drive gearless brushless motors like you used on CrazyBike2 (those look great but driving them looks like a royal pain. Do you know of anyone that's done the sin/cos to hall conversion vs. adding halls?)

My main question then is - I'm probably going to end up with two motors in any case. Can two motors drive the same live axle? How bad an idea is this in practice? I feel like this might work with brushless but I'd be worried that two geared motors would fight each other and torque the axle. I'd like to avoid redesigning the rear to add a differential or separate axles if at all possible.

 

[amberwolf]

Wire the motors in series (and double the battery voltage you'd be using), then they can't fight each other because they'll both be doing the same thing (same current thru both), as long as they are "identical" motors (like using the left and right off the same chair; they should be close enough).

To prevent motor-caused torsion of the axle over it's length, drive it at the center with both motors.

One question: how can you make turns if there is no diff or freewheeling, and only a single live-axle?

On an unpowered pair of wheels, you'd just drag the outer one around, scrubbing it "in reverse" effectively, while the inner would scrub forward some, depending on tightness of turn and loading and speed.

But on a powered pair that are forced to spin at the same rate, both are trying to push the same, fighting any turn you try to make.


BTW, it's no different with brushless vs brushed, geared vs not, for the problme of two non-perfectly-identical motors driving the same shaft; there will always be some difference in their output for one reason or another, but it probably won't make much, if any, practical dfference.

You can test by watching the temperature of each motor under various conditions; if one gets significantly hotter than the other, it is doing a lot more of the work. Otherwise, they're both doing about the same work.

I don't think I ever found a direct solution to convert the sin/cos to 3-hall, but it's really pretty easy to add the halls like I did. (there are better ways if you look up the various threads on adding halls to outrunners, mostly RC motors).

But...it's gonna be just as easy and probably cost less to use out-of-wheel ebike DD hubmotors rather than deal with these brushless in-wheel powerchair motors; I was going to use one because I already had it, and aside from saving money that way, it is always a challenge to adapt things to my uses (which are usually completely unforeseen by the original designer, I expect).

 

[mhenstell]

One question: how can you make turns if there is no diff or freewheeling, and only a single live-axle?

The two rear wheels are very close together. I don't have my axle yet so I haven't been able to test it, but my hope is that it works more or less like a tricycle, with a minimum of skid steering. It's also meant for dusty conditions and very slow speeds, so I'm hoping this won't be much of a problem in practice. I hope.

Two this occurred to me - one, if I understand the problem correctly, it would be pretty trivial to convert the sin/cos into hall outputs on a small microcontroller. Two, it seems that Kelly sells sinusoidal controllers with sin/cos inputs already, so that might not even be necessary anymore. I'm not considering them for this project but they sound interesting.