This blog is intended to discuss the advantages and disadvantages of fitting an inner-rotor kit together with what they do, what they don't do and how they affect moped performance.

Why do scooters have flywheels?

An auto scooter is fitted with a flywheel which serves two main purposes. The first purpose is to contain the magnets which rotate around the stator which generates the AC power source for the scooter electrical systems, and the other purpose is to provide significant inertial mass to the crankshaft.
Basically this means that by having a 'heavy' flywheel fitted to a spinning crankshaft the rotating flywheel will want to keep spinning due to its rotational momentum and any forces causing the crank to slow down will be resisted due to this additional rotational mass more so than without it.
Imagine spinning a playground roundabout- it is quite heavy to turn initially but once you've got it turning it's quite difficult to stop quickly. Compare this to say spinning a DVD on your finger, the DVD has far less inertial mass and can be easily spun and stopped almost instantly.

Why is a heavy flywheel useful on a scooter? Firstly it smooths out the crankshaft quite significantly. A single cylinder engine does not produce smooth power but develops it in a series of continual 'thumps' each time the piston is fired down the cylinder a bit like a pneumatic drill. Multi cylinder engines are far smoother because different cylinders fire in different crankshaft positions which whilst still applying the power in a series of "pulses" is is applied much more smoothly (mulitple pulses per crank rotation) whereas a single cylinder 2-stroke just produce the one big thump once per crankshaft rotation. 4-stroke single cylinder engines only fire once every other rotation, hence the term 'thumper' associated with some of them because of this effect.

This 'thumping' effect can take a toll on belts and bearings and cause premature wear and for this reason some single or twin cylinder high power race engines can require a significantly more substantial transmission system than a multi-cylinder engine producing the same power, especially where things like rubber couplings and belts are concerned.
By adding mass to the crankshaft this thumping effect is dampened and the engine becomes smoother. The additional rotational momentum provided by the flywheel will also prevent the crank speed from stalling if for example there was a brief misfire. On geared bikes and cars a heavy flywheel makes the engine much more difficult to stall if for example the clutch was let out a little too quickly when pulling away.

So how is an inner rotor kit different?

The main purpose of an inner rotor kit is to reduce the inertial mass of the orginal flywheel and it does this simply by being far lighter and by concentrating it's mass much closer to the crankshaft centre of rotation (inside the stator coils) rather than on the outside of a large diameter flywheel (outside the stator coils).
The downside of having a heavy flywheel is that whilst that heavy spinning mass requires a (relatively) significantly higher force to slow it down it also requires a (relatively) significantly higher force to spin it up (accelerate it) in the first case. This means therefore that a lighter flywheel on an otherwise unloaded engine will allow the engine to initially spin up much more quickly. In other words if you rev something like a motocross bike which uses lightweight flywheels as standard it will rev up and down almost instantly in neutral as you play with the throttle. If you compare this to something like a cruising bike the cruiser will rev both up and down more slowly (may take a couple of seconds to rev from idle to say 7000rpms). The advantages to this on a race engine are clear, and the disadvantages of the lighter flywheel are tolerated as an acceptable consequence of having a high performance bike (as usual in tuning everything is a trade-off compromise).

A further advantage of an inner rotor kit is that since they also replace the stator/ ignition system they will often be designed specifically to run reliably and with very accurate sparking without restrictions at high revs where a standard ignition may begin to struggle. Additionally some inner-rotor kits come with custom or programmable CDI's which allows the ability to easily modify or alter the timing of the ignition or even to program a custom timing curve although in reality this is not particularly useful on the 50-70cc 2-stroke auto scooters we are discussing here.
A further disadvantage of an inner rotor kit is that there isn't normally room to incorporate the charging system necessary to charge the moped battery, although some inner rotor kits have a small AC output which is capable of running basic lighting. This is not normally a problem on a race engine though and can have an additional benefit... the cost of the battery charging system on a standard flywheel/ stator is that through the magnets it applies braking force to the crank, with much lower power coil(s) this load is reduced.

We mentioned previously that it is important that the mass of an inner rotor kit is concentrated close to the centre of the crank, this is very important. 
On a standard flywheel the mass is concentrated relatively far away from the crankshaft centre of rotation, this has a far higher inertial mass than a flywheel the same weight would have if it was concentrated around the centre of the crank in the way that an inner rotor is (this is why they are called 'inner' rotors because the rotor sits on the inside of the stator coils in order get the mass as close as possible to the crank centre as opposed to a standard flywheel where the 'rotor' is on the outside of the stator coils.)
The reason for this far higher inertial mass is that the outside diameter of a large diameter flywheel will be spinning at a far greater circular velocity than a smaller diameter one even although they are spinning at the same rpm. The power required to rotationally accelerate a mass increases by a factor of 4 if the distance from the centre of rotation is increased by a factor of 2 (the square of the distance from the centre of rotation), if we decrease the weight of a flywheel the power required to accelerate it only reduces by a 1 to 1 factor of the percentage of weight saved... clearly we can see that reducing the diameter of the flywheel has a far greater affect than reducing it's mass.

Polini produce a lightened conventional style "outside rotor" flywheel/ stator for Piaggio which does include full charging coils albeit at slightly reduced output. It's interesting to note that in line with our above observations in order to achieve a lower rotational momentum with what is essentially a standard style stator/ flywheel setup they have concentrated on significantly reducing the diameter of the flywheel in order to concentrate its mass more closely to the crank centre of rotation.

So how much more horsepower can an inner rotor kit give me?

Despite a common misconception to the contrary an inner rotor kit or lighter flywheel assembly cannot give you any measurably increase in hp at the rear wheel on an auto scooter.
We can demonstrate this with dyno graphs from our blog "Fitting an Artek Inner Rotor Kit to late Piaggio Scooters".
The below graph shows power curves of a scooter with first a standard flywheel/ stator fitted and then an Artek inner rotor kit.

click here to enlarge...

The dyno runs were done under virtually hermetically identical conditions and with the dyno programmed to accurately replicate actual road conditions and loading.
The graphs are identical with any differences being due to inevitable marginal discrepencies between any two runs.
There were no modifications in the ignition timing on the Artek inner rotor kit. Both Artek and standard ignitions run at exactly 23 degrees BTDC dead static across the entire rev range once above idle.
We were in fact expecting a very slight increase with the inner rotor kit due to the lower magnetic braking force of the inner rotor kit not have charging coils (rather than it being due to any weight difference) however this did not bear out in any convincing manner and the graphs are identical for all intents and purposes.

So what is going on? At first this seems illogical, after all it surely requires an amount of power to accelerate or maintain constant rotation of a 'heavy' flywheel compared to a lighter one and we have established the rotational mass of the standard flywheel is significant enough to affect the crankshaft.
In fact this much is true, but we have so far not considered something very important.
When the engine is at tickover and before the clutch is engaged the rotational mass of the standard flywheel is very significant compared to that of the crankshaft and other moving parts. If we were to remove or significantly reduce it, this would be a substantial percentage change in the entire rotational mass of the crankshaft assembly. However what we have not considered so far is that as soon as you pull back the throttle and the revs have picked up and the clutch and rear transmission have begun to engage, the entire rear transmission and rear wheel effectively become a huge flywheel with far far greater rotational intertia than that of the actual rotor or flywheel. Because of the way the torque drive and variator operate, the rear wheel, once engaged, remains engaged to the crankshaft assembly even if you come off throttle right down until the road speed drops very low (maybe 10mph).
This means that on an auto scooter we can effectively consider the rear wheel as if it were an enormous flywheel bolted directly onto the crank in all circumstances when the engine is running *except* when intially revving the engine from idle to the point where the scooter starts to move. The huge relative diameter and sheer mass of the rear wheel now render the rotational momentum of the flywheel (or more accurately the difference between the rotational momentum of the standard and a lighter flywheel) as an entirely insignificant percentage of the overall rotational mass whereas before the clutch was engaged this percentage difference was significant enough to allow the engine to rev up significantly more quickly. Once the bike is moving any flywheel weight difference between the standard flywheel and an inner rotor would clearly be completely insignificant than compared to say fitting a lighter tyre or wheel.

But the flywheel still must use some horsepower? Why is there no saving at all? On a scooter a standard flywheel requires very little power to accelerate it, it simply doesn't have enough inertial mass, and once spinning at a constant speed a rotating mass consumes very little energy (the earth would have stopped spinning long ago otherwise!). Don't forget that on a well set up scooter the engine revs stay the same all the time so whilst the scooter is accelerating the flywheel is remaining at the same rpm and consuming no appreciable engery. A standard scooter flywheel requires less than 0.1hp to keep it spin it at 8000rpm, there simply isn't any power there to be recouped by slightly reducing this, you would have to throw the flywheel away completely to recover less than 0.1hp.
Interestingly the fan on an air-cooled scooter (which is bolted to the flywheel) can require in the region of 1hp to spin it at the same 8000rpm (although obviously has no useful rotational momentum).

Conclusion

At standard or sports level tuning we can safely conclude that the advantages of an inner rotor kit are most significantly that the engine will spin up more quickly and aggressively when you initially throttle to pull away from idle, however this is at the expense of a slightly less-smooth engine which could be more susceptible to stalling in the event of any small misfire type situation.
Because the main disadvantage with an inner rotor kit is that you are likely to loose your battery charging circuit and keeping in mind it offers no actual measurable power increase advantage it needs to be considered carefully whether it is a worthwhile modification on a road going scooter.
At race level however, where everything no matter how small counts it becomes much more interesting.
We don't care about the likely lack of charging coils and the advantage of a high quality unlimited and possibly programmable aftermarket ignition system is very attractive!
It should also be noted that the power required to spin up a flywheel increased by a factor of the square of the rpm. This means that if you double the rpm the power required increases by a factor of 4. Therefore the benefits of the initial faster spin up (before the drive engages) is of most benefit (and the most improvement will be seen) on very high revving highly tuned engines.