Archive for BRIT NERF Home of the British NERF Community

       BRIT NERF Forum Index -> Q&A and New members

Starting into brushless AC motors?

Im planning to do a build this summer with Brushless AC motors but I have some questions.
This is what I do know: I want outrunner, I need to get som ESC's, I know how to code. I'll get an appropriate 3s lipo.(going all out for fps)

Can anyone recommend a type of motor? I don't really know what the specs mean for AC's.

I want the motors overall to be smaller, physically than dc with flywheel. But is there a point where this adversely affects performance?

I assume I have to match the ESC's to motor by voltage. Anything else?

I'm doing quad flywheel set up and have lots of space in a centurion shell. So can I run all 4 motors off one big lipo? or would you recommend two smaller ones?

Thanks for your help!
Flint McCrae

Here's my experiences.

I found these motors worked well:

The Hobby King motor finder at is a useful tool.

Outrunners generally have lower max rpm than inrunners, but they tend to be better at torque.  The "KV" number quoted for brushless motors represents RPM, you multiply it by your supply voltage to get RPM.  So for a nominal 11.1v 3S lipo and the 4200kv example I linked above that's 11.1 x 4200 = 46,620rpm.  I think that's one of (if not the) fastest outrunners that Hobby King list.

With ESCs it's important to match current as well as voltage.  They can run hot when approaching their max currents, so best to allow a safety margin.  They are surprising bulky, but that shouldn't be a problem in a Centurion, so you should be able to fit sufficiently beefy ESCs.  You might have to do special coding to get around safety features because some ESCs that are designed for quad rotors expect certain inputs at start-up to convince them that they are connected to a functioning receiver.

I would recommend a single lipo, but for quad flywheels it must have huge discharge capacity.  You'll need appropriately large wire from the battery to the ECSs, I would run a couple of positives and negatives right back to the battery (so for quad flywheel each pair of ESCs would have a +ve and -ve to the battery).

One issue I encountered is heat.  It's not such a big problem for quad rotors because by definition they have four big fans cooling them Smile.  For this reason I would avoid programming the ESCs to do motor braking, and use materials & adhesives around the motors that are good for high temps.

Be careful to protect any sensitive micorcontroller etc bits as much as possible from magnetic and electrical interference, brushless does seem to generate a lot of both.

One useful thing to know - you can change the direction a motor spins by swapping over any two of the three wires.

KV can refer to different things depending on where you see it. It's actually meant to be the back-EMF constant which tells you how fast a motor spins to generate a given back-EMF, as opposed to how fast a motor spins at no load at a given applied voltage, although quite often you'll see it quoted as the latter (sometimes suppliers calculate KV by using the applied voltage multiplied by a "fudge factor" to guesstimate the back-EMF). In reality the two values are pretty close as the back EMF generated when a motor spins will almost equal the voltage applied that caused it to spin in the first place (in an ideal world they would be equal - the only reason they aren't is because, even at "no-load", the motor has to overcome losses and friction). It is is still an important distinction to make though as a 4200KV motor running on 10V will still only spin at (close to) 42,000RPM at no-load - as soon as you apply any sort of load to the motor that will reduce and potentially quite significantly. "True" KV is more useful as it is also equal to the inverse of the torque constant which allows you to calculate the torque produced at a given current draw (they are, in effect, the same constant).

The important thing to remember is that these constants apply to both brushed DC motors and brushless/synchronous AC motors - a 4200KV brushed DC motor will spin at about the same no-load speed as a 4200KV brushless DC motor and both will generate about the same amount of torque at a given current draw. You won't necessarily see a performance benefit from using brushless motors.

Muzzle velocity is determined more by flywheel geometry than motor speed or torque anyway - as long as a pair of motors can maintain the required speed at a given load they will achieve the same sort of muzzle velocities as any other pair of motors that can. Stock Stryfe motors, for instance, can achieve broadly the same velocities with standard flywheels as aftermarket 130s or even 180s - on top of that ultra high-speed high-torque 180s like Wolverines also don't perform any better with most flywheels than lower speed lower torque motors such as Hellcats, fk180SH-3240s or even, with some flywheels, Rhinos. The main benefit of an aftermarket motor is a faster spin up time but even in that respect brushless motors don't necessarily perform better (or even as well) as may brushed motors. Sometimes the ESC is to blame, as they almost always limit the startup current, and sometimes even the rate of acceleration, of the motors. Even if you flash the ESC with custom firmware you can't always get around these limitations (although you could build your own ESCs). Brushless motors, and outrunners especially, also often have a greater moment of inertia due to the larger diameter rotors, which in turn have greater rotating mass, and this also acts to increase the time taken for a the motor to get up to speed. ahalekelly posted this graph on reddit showing motor speed against time from initiating startup for various motors. The brushless motor traces were derived from empirical testing data whereas the brushed motor traces were the results of simulations based on measured motor specs - the brushed motor simulations also modelled the additional moment of inertia of a standard flywheel on the shaft of the motor whereas the brushless motors were tested bare. The inrunners tested took about the same, if not slighlty less, time to reach 22kRPM (approximately critical speed for a stock flywheel geometry) whereas the outrunners tested took significantly longer than most 180s and almost as long as the slowest 130 (i.e. a Rhino).

That said there are compelling arguments to be made for brushless. The obvious one is the increased longevity due to lack of brushgear. Overall reliability isn't necessarily better as the additional electronics add complexity and potential points of failure but the motors themselves should last significantly longer. Deleting brushes from the equation should also reduce (if not completely eradicate) electrical noise, at least the electrical noise caused by arcing, which is a big deal if you want to run a microcontroller for burst fire or ammo counting (the switch-mode nature of ESCs may introduce their own interference however so YMMV). It's also easier to build custom setups with outrunners as they can be mounted to a simple flat plate rather needing a full cage to house them. Similarly, whilst the outer can of an outrunner may not be the best shape/surface to act as a flywheel itself, they also allow you to easily mount custom flywheels to the outside of them and are one of the few applications where 3D printed flywheels (or, effectively, solid "tyres") have been demonstrated to be viable. Brushless inrunners with a conventional flywheel setup would likely be a better option but you'd need a custom flywheel cage as well as custom flywheels to suit (which is probably the main barrier to using larger-can brushed DC motors as well since most motors larger than the 100 series also use larger diameter output shafts).

For the theory behind multi-stage setups in general this blog post is worth a read through although the velocity achieved and RPM required will vary a little depending on the flywheel material, gap and geometry and won't necessarily match the table (the table data was calculated for standard Hasbro flywheels). Whether or not the second stage needs to spin as fast as the first stage will depend on where you place the flywheel stages relative to each other - if they're relatively far apart (in as much as the dart will leave one stage before entering the next) you can use different speed motors for each stage, whereas if they are relatively close together (so that both will contact the dart simultaneously) then the first stage will need to spin as fast as the second stage (or at least almost as fast since the dart will still be briefly accelerated by the second stage after it has left the first stage).

In general a larger diameter flywheel should result in greater velocities (larger diameter results in a larger contact patch all else being equal) so I would avoid going too much smaller in diameter than standard flywheels. You can compensate to some extend with a smaller flywheel gap but that causes more damage to the dart and puts more load on the motor bearings. If you're going multi-stage anyway this might not be as much of a concern.

One LiPo is always preferable IMO. It's usually more space efficient in terms of energy per volume and means you won't end up with wasted capacity if one pack runs flat before the other.

       BRIT NERF Forum Index -> Q&A and New members
Page 1 of 1
Create your own free forum | Buy a domain to use with your forum