Multiple motor & controller approach?

General BVS related area
User avatar
Posts: 2089
Joined: Fri Apr 27, 2007 10:26 pm
Location: Dorset

Postby qdos » Sat Aug 23, 2008 5:31 am

Hi Del,

We are all thinking along similar lines including Stuart from my discussions with him.

A monocoque would be great and I may be able to get something done on that front but as with the Kv question it rather depends on what people want to achieve from the vehicle, ie what sort of transportation are they after? It could be anything from a pedelec to a sports machine.

I'm thinking on the lines of 3 different types of machine really and it's often best to take the middle ground comprimise so to speak. You're welcome to give me a call sometime I'm over in Dorset not too far away.

Posts: 60
Joined: Sat Oct 06, 2007 1:35 pm
Location: Chesterfield, Derbys.

Postby hyve » Sat Aug 23, 2008 8:25 am

A lot of the tech. above on these motors went over my head. Can anyone suggest a link to provide me with the basics on brushless motors please ?

Apart from that: the one comment which seems obvious is, why use a high performance motor if you're going to halve it's power by using 24 instead of 48V ? Sure it simplifies the gearing, perhaps, but you're giving away a lot.

Since spotting the parallel with a multi-piston ICE, I'm near convinced this has to make a lot of sense. The one remaining unanswered question is, why hasn't a big motor mfr. taken this design and scaled it up ? If that is feasible then building a multi-motor is a waste of time because with the pressures now building for a switch to electric propulsion, larger units cannot be far off.
If on the other hand scaling up, as with pistons, does not produce worthwile results then the multi- idea is a winner. I note your latest motor example at 6,5kw is twice the output of the one quoted earlier, but is also 3 times the price. Perhaps there's a clue here.....
Peter Ph

User avatar
Posts: 305
Joined: Fri May 25, 2007 5:17 pm
Location: Luton

Postby EVguru » Sat Aug 23, 2008 10:27 am

The one remaining unanswered question is, why hasn't a big motor mfr. taken this design and scaled it up ?

They have.

Most production EVs from major manufacturers have used brushless motors (Pugeot and Citroen excepted). Some have been induction motors, some have been permenent magnet.

Technically all motors are AC (the homopolar motor excepted). A brushed motor uses the commutator as a mechanical inverter, whilst the brushless designs use electronic circuitry to commutate the motor.

Permanent magnet motors have part of the magnetic field built in and tend to have higher peak efficiencies, but that very field causes drag (cogging, eddy current, and hysteresis losses) at zero load. Induction motors have to generate (induce) their own field and tend to have lower peak efficiency, but when powered down have only mechanical drag and windage losses. The precise aplication may

These RC motors are cheap because they are produced in large numbers. They sell lots of motors becuase they are quite cheap and the motors are quite cheap becuase they produce so many. A common rule of thumb in high volume production is that you can estimate cost by simply weighing the materials involved.

I made a comment earlier about life expectancy. This is completely different from reliability. If a bearing was expected to last 100 hours and only did 25, then that's a reliability issue. If a batch of bearings were all exceeding the 100 hour life, then they would be considered 100% reliable.

If a device is pressed into service where it is operated beyond its life expectancy, then the bearing would BECOME a reliability issue.

User avatar
Posts: 472
Joined: Thu Jul 31, 2008 6:35 pm
Location: Salisbury

Postby Jeremy » Sat Aug 23, 2008 10:31 am

My only reason for dropping down to 24V was to do with the ready availability of controllers that will work at this voltage. There are few RC motor controllers available that will work over about 30V and those that are available are very expensive. This is the reason I've been researching the possibility of adding sensors and switching to the more readily available, sensor type, non-RC motor controllers.

There are some good introductory articles on brushless motors on many of the RC model web sites. Here's a link to some basics on BLDC motors: that you might find useful.

In essence they are very simple machines with only one moving part and nothing, other than the bearings, to ever wear out. The secret to getting them to work effectively is in the electronics that drive them. This can get quite complex, but is fairly easily handled by a small microprocessor or dedicated control chip.

The advantages of this type of motor are that it can have a higher efficiency over a wider rpm and torque range than many other types. It's also easier to implement things like accurate speed control.

For the controller to work it has to know the angular position of the rotor, relative to the stator, at any time. This is so that drive current to the coils can be accurately timed. There are two ways of doing this. The first is to measure the back EMF (the voltage generated from motion) of the un-powered coils in the motor at any instant and use this to keep the controller in synch. This is the technique that most RC model controllers use. The second method is to have three sensors in the motor that tell the controller what the rotor position is. This is the method that's used in most bigger brushless motors.

The back EMF technique works well when a motor is spinning at a relatively stable speed. Where it gets challenging is at start up, as the motor isn't moving so isn't generating a back EMF. The controllers get around this by pseudo-random pulsing the motor to get it to start to move, then sense the direction, correct the pulse sequence if it's wrong and so kick the motor into life. This all happens in a few milliseconds so isn't really noticeable - all that happens is that the motor gives a very tiny angular jerk before it starts.

Controllers that use sensors are much better at starting, as they get feedback on rotor position whether or not the motor is rotating fast enough to give a usable back EMF.


Edited to point out that my reply was concurrent with Paul's, not a response to it.............

Posts: 870
Joined: Tue Apr 01, 2008 5:27 pm
Location: Galashiels, Scottish Borders

Re: Multiple motor & controller approach?

Postby GregsGarage » Sat Aug 23, 2008 8:58 pm

Jeremy wrote: - any vehicle that has a power supply that's greater than 50V is supposed to comply with some pretty stringent safety requirements

Hi Jeremy,

Interesting project, if you haven't already seen it here is a link for a microcar conversion done on 48 volts.
The project has been scrapped and components recently sold off after disappointing performance on 48 volts. I am not saying it can't be done but he had set 48 volts as the upper limit and that made his task more difficult with the components he was using. His 48 volt limit was based on already having invested in a 48v thundersky pack and not wanting to purchase any more. His project though, inspired me to do my Fiat 126 conversion which is using 72 volts. My question is what are the "safety requirements" that you mention? Is this a formal regulation or general guidelines? Any links would be helpful.

I also see that you have considered using a Kelly controller, have you also considered some of the motors listed on their website as well. Cost probably isn't as good per kW but if you needed fewer of them you save some money in making fewer couplings. Also because they are designed for land going vehicles cooling should be less of an issue, although still need to be considered.

I just think that the 50 volt limit you placed for your project might restrict your options, for example the Kelly motors and controllers you mentioned can operated at 72, increasing your kW output for not much extra cost.
Greg Fordyce

Daewoo Matiz

User avatar
Posts: 472
Joined: Thu Jul 31, 2008 6:35 pm
Location: Salisbury

Postby Jeremy » Sun Aug 24, 2008 8:43 am

Hi Greg,

The upper limit for a DC power source is 75V (50V for AC), if you want to avoid getting the vehicle approved under the Low Voltage Directive. This only applies to something that you sell, as far as I am aware, but I think that it's inevitable that some jobsworth somewhere will start thinking about applying regulation to home built EVs if they get popular.

All it would take is for one high profile accident and we could find that there would be a knee jerk reaction from government. I've seen the same massive tightening of regulation in aviation from similar one-off accidents, as have sport shooters who faced draconian regulation following the nutter in Dunblane running amok.

Having owned a Prius for three years and seen first hand just how much effort Toyota put in to ensure that the high voltage electrical system was both safe and crashworthy, has convinced me that it would be difficult for an amateur to reach those sort of standards in a conversion of an existing car. Toyota even made sure that the emergency services have been briefed on dealing with the Prius in the even of an accident, including dire warnings on where and where not to put cutting gear etc.

Personally I think that the likelihood of anything draconian being forced on home made EVs is pretty small. However, if the idea gets picked up enthusiastically by the kit car community, as seems very likely, then I think we will see some form of regulatory change. Kit cars already have to go through Single Vehicle Approval, a fairly thorough safety test conducted at a VOSA test centre. If a few electric vehicles go through and start raising questions from testers as to what rules should apply, then it seems quite likely that we'll be faced with some changes.

The idea of sticking to a voltage below 75V DC is that it could be argued that, as this is the voltage below which the LV Directive doesn't apply (which is based on the risk of electric shock), then no special electrical system approval should be required. Making a viable propulsion unit that would work efficiently at low voltages would be key to staying in the safe voltage region.

The problem with making low voltage systems work with the motors and controllers already available is down to efficiency, mainly. As power is simply volts x amps, yet motor, controller and wiring resistive losses are proportional to the square of current, it makes sense to increase the voltage and reduce the current in order to reduce losses.

However, at relatively low power the losses are more acceptable, as it's easier to use oversize wires, windings etc to minimise resistive losses. If you use several, highly efficient, low power motors, to produce a compound motor that gives a high power, then the losses can be less than for a single big motor and controller of the same power. If the motors are linked using over-run clutches, then it becomes possible to switch motors in and out of use as the power demand varies, so keeping the compound motor working in it's most efficient region. It would even be possible to use motors with different values of Kv, so that the compound motor full power efficient speed range was increased.

The idea only looks viable because of the development of reliable, high power, high efficiency, light weight, brushless motors for model aircraft. This makes the weight and cost of a compound motor made up from a cluster of small motors comparable to, or more favourable than, a conventional single motor of similar power.


Posts: 43
Joined: Fri Feb 01, 2008 4:45 pm
Location: Southampton

Postby Delinquent » Sun Aug 24, 2008 2:20 pm

Have to say this has really got my brain going at the moment. I was aiming to build the MEV trike (or at least, a variation on it!) to a point where it would be capable of 50+ mph for 30 miles, and had really decided on one inwheel motor per front wheel, of around 4kw per motor. I was looking at a large diameter (1 ft) brushless to get bags of torque from relatively low wattage. Now I'm wondering if I just ditch that plan for the time being - I can see a very simple way to install 2 x 2kw RC motors per wheel on the opposite side to the brake calliper, with a small serpantine belt drive to give appropriate ratios. A particular up side would be considerably simpler to adjust the gearing on than re-winding a 1ft diameter motor several times!

Other advantages I'm pondering (as in, are they really advantages..) are the ability to "switch off" multiple motors if you get short on juice to run at reduced power, plus of course losing a motor wouldn't mean losing the ability to get home. Overall weight will be considerably less due to the distinct lack of iron typical of larger motors with heavier magnets, plus it would be a lot easier if moving to liths to keep smaller packs dedicated by motor in balance. I think :lol:

Life expectancy of the motors, well I've run one commercially available motor in one of my heli's for 5 seasons, with it getting between 30 mins and 2 hours use per day during the summer. Probably more than I'd spend on the road!

User avatar
Posts: 472
Joined: Thu Jul 31, 2008 6:35 pm
Location: Salisbury

Postby Jeremy » Sun Aug 24, 2008 3:21 pm

Interesting ideas. I've long wanted to build a very light tadpole style trike, hence my interest in looking at these very light motors. I think I'm inclined to stick with RWD, as it's easy to do with a chain or belt drive. Although making a space frame or monocoque frame would be straightforward enough, it's front suspension and steering gear that is harder to get right, which is why I was originally planning on trying to get hold of an old FF race car front end.

As you rightly say, motor longevity shouldn't be a problem, I think. The chap that has built that RC motor powered recumbent bike, Matt Shumaker ( has been using a big brushless RC motor to run his vertical milling machine for over 2000 hours of use with no problems.

I wish that there was more detailed information about these motors, such as proper performance curves. Although the raw specs look OK, I'd be interested to see just how efficient they are over a range of operating conditions. I think the snag with this sort of data is that it would be fairly controller dependent, as I suspect that controller timing plays a fair part in getting the best from a particular motor.

After dithering about for a while, I think I may just buy a couple of the cheaper ones to play with, so that I can do some bench tests and plot out some performance curves. It will mean rebuilding the old drum brake dyno I built many years ago, though!


Edited to add:

I have just come across someone who has fitted one of these big RC motors to a mountain bike. He's using the same controller that I've just ordered and a similar size motor. Interestingly, he's getting a top speed of 54 mph from this set up, running on a set of small NiMH batteries. His range is limited to just 9 miles at 30mph, but given that his battery pack is only small this doesn't seem too bad. Here's a link to a long thread on the ES forum with photos of his build: for those that maybe interested.

At least this shows that these small motors can start from a standstill and provide a great deal of power for their size when used in a vehicle. The potential for an array of them driving extremely light, aerodynamically clean, trike seems excellent to me. If it were possible to build a single seat trike with a weight down at around 100kg or less, then I think it would be a very efficient means of getting about, with a reasonable turn of speed.


Posts: 43
Joined: Fri Feb 01, 2008 4:45 pm
Location: Southampton

Postby Delinquent » Sun Aug 24, 2008 8:39 pm

Sorry, meant to say thanks for the link to that recumbent - had a good long look at that, and was initially very excited by the motor, right up till I saw the cost of it...!

Just realised you are probably not aware of an excellent resource for these motors - a yahoo group called LRK_torquemax. Ron (group owner) is a bit of a god in the world of these motors, joined by enough grey matter to pretty much have every corner of the arena covered - in fact it is through this group that I have gained most of my (comparatively limited :P ) knowledge, which I expanded on through trial and error and good record keeping!

I'll be running over to check out that mountain bike in a second, I know what you mean about the suspension, but one thing I learnt very early on in another project I'm working on (carbon fibre monocoqued "supercar" ) is that it's 101 trade offs. There is no right, just plenty of wrongs that are easier to avoid than you'd think coupled with your choice of compromise! When it's a trike, it's actually a great deal simpler (or at least, less adjustable!) as you don't get the interaction between the 4 corners. A few bits of string and a couple of sheets of card and going on your posts on here you'd be able to produce something usable in no time.

User avatar
Posts: 472
Joined: Thu Jul 31, 2008 6:35 pm
Location: Salisbury

Postby Jeremy » Sun Aug 24, 2008 8:54 pm

Thanks for that link, I'd come across Ron's name a few times on some of the RC groups, whilst researching motors and controllers.

Well, I've taken the plunge and ordered a pair of Tower Pro motors, rated at 2800 watts maximum each. Including shipping from Hong Kong they have cost me £77.10 for the pair, which doesn't seem too bad...............

Controllers are a lot more expensive, it seems. The ones I've ordered were £48 each, but will work up to 48V at 120A maximum, 100A continuous, each, which is comfortably more than the motors will take.

Total weight for two motors and two controllers comes to just over 1.6kg. Not bad for a system with the potential to deliver around 3.5 to 4kW continuously.

I've managed to dig out some rough data on efficiency. It looks as if these motors are around 88% efficient at about 60A, dropping to 82% at 25A, which doesn't seem too bad, particularly for the price.

At least this set up is a great deal cheaper than the Plettenburg set up that Matt Shumaker is using!


Return to “General”

Who is online

Users browsing this forum: No registered users and 8 guests