How to get the most out of a Chinese 2-stroke 50cc moped....cheaply!

Chinese 2-stroke 50cc moped engines are basically a copy of the air-cooled horizontal Minarelli engine as used in Yamaha Jog (basically an air-cooled version of the Yamaha Aerox engine). There are a few differences, all the recent Chinese ones use 12mm gudgeon pin instead of the genuine Yamaha 10mm and they also use a heftier crank with larger diameter shafts (the same size as used on Yamaha Aerox 100 in fact).
So far so good, they are copying a decent engine and have even improved a couple of its weak points.

Unfortunately what can let the Chinese engines down from hereon in is the huge variation in the quality of the components used within them. Without years of experience dealing extensively with different Chinese factories and seeing first hand the problems customers face with them it's impossible for the average customer to know which Chinese scooters are the good(ish) ones.
We know of a certain brand Chinese moped where we can virtually guarantee the main crank bearings will fail within the first year, another brand where the rear gear teeth will always strip within 2 months unless a simple modification is made and another brand where the variator gudgeon pin is 3mm too long causing the engine to over-rev and be at constant risk of seizure.
There are also many brands where the engines are built with every single part being sourced simply from the cheapest supplier at the time, so one batch of engines may be decent, the next poor. The list goes on.

For this blog we're going to be using a Baotian Falcon 50cc 2-stroke. The reason for this choice of scooter is that Baotian has very consistent quality control. It's supplied by a single supplier within the UK with a genuine technical and experience pedigree and they seem to have religiously overseen all aspects of its production during the last few years in which they've been the sole UK supplier. They also don't seem to be afraid to spend a little extra money to get the product right. It only takes a few extra dollars at production level to buy decent crank bearings, decent variator, even a genuine NGK spark plug and Yuasa battery, but not many Chinese scooter importers are prepared to do this with most prefering to opt for the cheapest components to the detriment of the end customer and the reputation of import mopeds.
So whilst they still have their faults inherent to all budget scooters the Baotian brand is generally considered to be of decent quality for the money you pay. It probably should be mentioned that there are other Chinese moped wholesalers selling scooters from the Baotian factory in China. These are not allowed to be openly marketed or advertised in the UK under the 'Baotian' name and are not made to the same specifications as the genuine Baotian branded scooters are.

Ok, back to business. For this blog we're going to look at just mild tuning to get a bit more out of the bike. If you want to heavily tune a moped or you are fanatical about peds you are unlikely to start with a Chinese one, but there are a lot of these Chinese mopeds about and a lot of people would be perfectly happy to pay maybe £50-£150 if something modest but significant and reliable could be achieved.

First of all let's run the bike up to see what it's doing. It's immediately apparent that the moped is restricted to 30mph, but surprisingly it's actually restricted much more effeciently than the Yamaha/ Piaggio mopeds. There is a wire running into the transmission side casing - this wire allows the cdi to monitor the actually rear wheel speed (not the engine rpm) from the clutch bell speed. This is clever because it allows full acceleration and full power right up to 30mph at which point it simply won't accelerate any further. The Piaggio/ Yamaha mopeds don't do this, their restriction method restricts bike power from approx 20mph upwards by prematurely raising the revs (with a variator restrictor ring) until the moped is revving too high to make enough power to rev any higher. Also this variator ring method will not work if you tune the bike (e.g. add a powerpipe). The Baotian method would still work and restrict to exactly 30mph (you could just fit a switch) regardless of the state of tune of the bike.
Curiously the Baotian Falcon doesn't pop or stutter as most electronic limiters do at least to some degree when they cut in. We investigated this and instead of cutting the spark (as say a rev limiter would) the cdi is progressively retarding the ignition timing (sparking later and later) as the moped begins to reach 30mph. By sparking later the power produced is reduced. This is more sophisticated than we were expecting, the Italians could learn from this!

Ok, with the wire coming from the transmission casing disconnected the restriction is removed and the Falcon made it all the way up a speedo reading of 48mph. As with all mopeds the speedo reading was a bit optimistic and the dyno showed the genuine top speed to be 42mph.
There was also a variator restrictor ring fitted to the Falcon similar to those fitted to Piaggio/ Yamaha mopeds but it wasn't actually doing anything. It was too thin to be restricting the variator in any way and made no difference to the performance whether it was removed or not. We removed it anyway on principle :)

Ok, here is how the Falcon performed in standard form but de-restricted.

click to enlarge...

This is actually quite promising, for comparison a standard Aerox would make slightly more (up to 4.5hp) but not a lot different. You can see that the transmission is not holding the revs constant and they are creeping up from 6500rpm to 8000rpm until it gets to 35mph at which point the variator is fully moved out and the only way the moped can go faster is to increase the revs (in the same way as a geared bike or car).
The variator should be holding the revs constant really up unto this 35mph point, a new Piaggio or Yamaha moped would have shown an almost horizontal graph line here but in their standard form most mopeds have a fairly wide powerband so it might not be worth spending time tuning the transmission to straighten it out and we'll leave it as it is for now.
To have a better idea of what's going on we need to see an engine power curve which shows what power the moped is actually making at any given rpm. Because of the way the variator works by constantly change the ratio between the engine rpm and rear wheel rpm scooter dyno graphs normally only show what power the scooter is making at different road speeds.
In order to get a fixed gear ratio graph (power curve) we need to modify the transmission so as it is fixed drive (like a geared motorbike). For this we use special hubs to replace the variator and rear torque drive which connect together using a toothed belt (like a car cambelt).
This gives us the following graph...

click to enlarge...

From this graph above we can see the engine has a good wide powerband until it starts to exceed 8000rpms at which point the power starts to drop off quite quickly. In the first graph we can see that 8000rpm comes at 35mph and as you would expect the power starts to drop off quite quickly in the first graph after 35mph. Because the variator is fully changed out at this point this means that it is impossible to exceed 35mph without exceeding 8000rpm regardless of any change of rollers or other transmission tweaking. This means it will be impossible to improve the top speed from 42mph (48mph clock speed) unless we either alter the power curve of the engine above 8000rpm or change the gearing. Although Stage6 have now started to produce a range of gearup kits for Chinese scooters (the Aerox/ Jog ones won't fit), at this stage it's unlikely the scooter is making enough power to benefit from this so we can dismiss this option for the moment.

The first option which normally offers a substantial performance benefit for an otherwise standard 50cc 2-stroke moped is of course to change the exhaust.
The standard Baotion Falcon exhaust looks like a sport expansion powerpipe already. But it isn't !
It is in fact incredibly restrictive. It's weight gives a clue as to the large amount of internal workings contained inside, and some forums have mentioned that you can cut into the can, remove the internals, re-weld it back up and it then becomes a modestly decent sports pipe. Unfortunately this isn't true. Whilst it is true the exhaust contains a large amount of baffles, internal plumbing and a catalytic converter the actual shape of the pipe isn't right for a sports pipe. The angle and design of the cones are made purely to aesthetically mimic a sports pipe but are in the wrong places to actually work as an expansion pipe. To prove the point we cut open and completely removed the internals of the standard pipe and then rewelded it.
The exhaust became significantly louder but offered no performance increase whatsoever as shown in the next graph below.
In fact by the end of the day it was clear that the standard Falcon exhaust was its single most restricting factor and it is virtually impossible to get any significant gain out of the Falcon without ditching the exhaust in favour of something better.

We tried a few different exhausts to see what sort of power curve would result. It was quickly clear that some of the higher revving pipes were too much for the standard 50cc cylinder with it's fairly mild timings and consequently the mild sports pipes where the ones which worked the best.
An unexpected surprise was how well the Motoforce standard replacement Yamaha Aerox exhaust performed on it.

The Motoforce exhaust (pictured above) is a small, discrete quiet 'standard' replacement exhaust normally sold for Yamaha Aerox for £49.95. It is however different to a standard exhaust in that it actually contains a half-wave expansion pipe within. A half-wave pipe is similar to a sports pipe and can be contained within a much smaller volume than a full wave expansion pipe. Consequently they can be made to look like standard exhausts. The disadvantage over a full wave sports pipe (e.g. Leo Vince ZX etc.) is that although they normally make good power this power is normally made at fairly low revs. Because the Falcon standard cylinder is designed to make power at low revs also they seem to compliment each other well. In fact there was no pipe which would make a higher peak power when fitted to the standard Falcon cylinder than the Motoforce one.

The following graph show the power graphs for the Baotian Falcon with standard pipe and then with the Motoforce pipe and then Leo Vince ZX (most of the sports pipes we tried e.g. Stage6 Pro Rep gave a similar graph to the ZX).

click to enlarge...

As soon as the standard pipe was gone the engine would happily rev a little higher which resulted in a slightly higher top speed of 45mph (52mph clock speed) with both the ZX and Motoforce pipes. All the powerpipes made a peak of around 6.3hp on the Falcon standard cylinder which is a substantial improvement over the standard 4hp.
For comparison a Yamaha Aerox or Piaggio Zip would make around 7hp with a ZX or similar sports pipe fitted, so this really isn't bad at all for a Chinese moped.
It's clear however that the Motoforce pipe is the out and out winner for this bike for the following reasons...

1. Cost... the Motoforce pipe is far cheaper than any other sports pipe
2. Fitment... the Motoforce was pretty much the only pipe we could find which would bolt straight on. The Falcon has a slightly unusual undercarriage arrangement whereby the stand gets in the way of fitting pretty much any sports pipe. Pipes like the ZX will bolt up to the engine for testing but to actually fit the pipe securely we would have had to tailor make some brackets for it.
3. Powerband... the Motoforce has a wider powerband than the other sports pipes when fitted to this cylinder. Normally this results in a lower final topspeed because wide powerband engines do not normally rev very high (which determines ultimate top speed) but in this circumstance the Motoforce pipe curve matches that of the sports pipes at high revs meaning top speed is identical. A wider powerband also means that the exhaust is much more tolerant of a poorly setup transmission (e.g. wrong rollers) or other running problems/ engine wear.
4. Discrete... the Motoforce looks completely standard, this of course could be good or bad depending on your opinion!
5. Reliability... the Motoforce pipe makes peak power at 7000rpms with the Falcon standard 50cc cylinder, this is 1000rpms lower than even the standard pipe! As comparison, although an Aerox would make a slightly higher 7hp with a ZX or Stage6 Pro Rep pipe it would need to be running at over 9000rpms.
6. Plug and Play... because the Motoforce pipe makes its power low down the rev range it can simply be bolted on with no adjustment to the variator rollers or clutch springs necessary. It runs perfectly well within the rpm range within which the standard variator is already setup to run with the standard pipe and in fact the power available is doubled even from static pullaway. If we were to fit the ZX we would have had to change rollers/ clutch springs. The powercurve shows the ZX pipe making less power than standard at low revs so particular care would have to be taken so as not to loose initial pullaway. It should be noted that because the ZX has a narrower powerband the transmission would have to be set up perfectly to maintain it within its powerband as the power drops off very quickly with the ZX pipe as you move away from peak power at 8,300rpm. Because we established early on that the standard variator was not particularly efficient at holding the revs steady this could cause problems meaning it could be necessary to fit a higher performance aftermarket variator. This would not be an issue with the Motoforce pipe due to its wider powerband.

Below are the final results. We also fitted a 70cc Malossi cast iron sports cylinder kit for good measure.
It should be noted that in all cases including the standard de-restricted bike the main jet size was increased to 65 (6mm dellorto jets will fit) in final testing. This is necessary otherwise the moped starts to run a little lean. In summary...
1) Baotian Falcon + Motoforce Exhaust (£49) = *almost* but not quite the same performance as Yamaha Aerox + decent sports pipe(e.g. Leo Vince ZX/ Stage6 Pro Rep)
2) Baotian Falcon + Motoforce Exhaust (£49) + Malossi 70cc kit (£108) = same performance as Yamaha Aerox + decent sports pipe(e.g. Leo Vince ZX/ Stage6 Pro Rep)

As an after-note with both the Malossi 70cc kit and Motoforce pipes fitted we tried swapping the standard Chinese carb with something larger but only very modest gains were noticed so it probably isn't worth changing.

click to enlarge...

Why are my crankshaft splines damaged ?

This short blog is just to cover a simple check which needs to be made every time you fit a variator, whether you are just changing your mopeds rollers or fitting a new aftermarket performance variator !

The splines on a moped crankshaft and on the outer variator pulley/ kickstart gear aren't strong enough on their own to withstand the high rotational forces applied to them during normal use.
The crankshaft pulley nut doesn't just hold everything together but applies the compressive force required to prevent the splines from becoming damaged and spinning on the crankshaft. In fact so long as the variator is spaced correctly and the crank nut done up correctly it does not even really matter if the splines are there or not, the compressive force of the crank nut holding everything together is all that is required.
It's therefore vital that the variator is spaced out correctly, if the crank nut tightens onto the crankshaft shoulder *before* it reaches the front of the variator assembly (see photo below) then it is certain that the splines on front pulley assembly will strip very quickly. As soon as this happens the front pulley is free to spin indepently which affects performance and if not fixed very quicky will lead to the front pulley hole wearing considerably followed by it subsequently wearing a groove into the crankshaft which can result in an expensive and uneccessary premature engine stripdown to replace the crankshaft.
For this reason the crank shoulder must be recessed by approx 1mm as shown in the right hand picture

click to enlarge...

Worn crankshaft splines or worn a front pulley are almost certain to have been caused by an incorrectly spaced variator unless the variator crank nut has become lose (which would be obvious on inspection). If the variator isn't spaced correctly according to the photos below then something is definitely wrong and most likey you have missed out a washer somewhere !!!
Stop and recheck carefully !
It's worth noting that loctite must always be used on the crankshaft nut, one moped manufacturer even has a policy of automatically voiding variator related claims where loctite has not been used.

Moped tyre pressures vs performance.

This is a short blog covering the often overlooked effect on performance of tyre pressures.
Mopeds and scooters are affected more significantly than geared bikes by low tyre pressures and consequently it's common for a moped to be booked into our workshop for poor running where low tyre pressure is a significant factor of poor setup. Sometimes we even see them where there is nothing wrong at all other than the tyres pressures have been neglected!

It's fairly obvious that the more a tyre is under inflated the more power will be lost and the less efficiently the moped will run. A significant part of this is simply that where a tyre sits on the ground, that area of the tyre is deformed. As the tyre turns it consequently has to reform and deform the tyre. This requires energy which in turn heats and wears the tyre, so the flatter the tyre the more energy is required to turn it which is also exacerbated by the weight of the bike and rider.
This can be demonstrated by the much larger effort required when pushing a moped with a flat tyre.

This drivetrain loss is fairly uniform and can be seen in the graph below when you look speeds from 25mph upwards. Each change in tyre pressure results in a fairly predictable change in the power developed at the rear wheel. There is a significant loss in power in this area once the tyre pressure gets down to 20psi but nothing too serious at higher pressures.
This region above 25mph is basically what you would expect to see on a geared bike except that on a geared bike this loss would extend fairly uniformly across the entire rev range. It isn't quite uniform as centrifugal forces tend to 'round' out the wheel as speed increases but it's not far off in these speed ranges.

We mentioned already that mopeds can behave disproportionately adversely to low tyre pressures compared to geared bikes and we can see this happening in the graph below at speeds below 25mph.
At 35mph the 25psi results in only a 3.5% loss in power compared to the tyre being at 35psi (6.75hp vs 7hp)
However at 15mph the 25psi results in a massive 37.5% loss in power compared to the tyre being at 35psi (3.25hp vs 5.2hp)
3.25hp @ 15mph is less than a standard derestricted Aerox would make at this speed, so for this particular bike (with a Leo Vince ZX pipe fitted) just by allowing the rear tyre to drop to 25psi would result in being out accelerated by a standard Aerox !
In fact the resulting losses can be even worse than this because the graph only takes into account the rear wheel being low.

click to enlarge...

So what's happening below 25mph?
Well basically the variator is being delayed in the time it takes to allow the engine to reach peak operating power rpm at these speeds... so the engine is revving lower than it should be in these sections of the graph at an rpm where there is much less power developed until eventually it makes it through to peak rpm and the bike is finally away.
Although ideally the variator should allow the engine to rev straight to peak power whenever you mash the throttle, moped variators (and particularly standard ones) tends to always engage progressively meaning that it requires a fair amount of power to initially get it there.
On a sports pipe and particularly the higher performing ones (and especially in combination with a worn/ standard variator/) this progressive application means that due to the pipe developing lower power than a standard pipe at low revs it struggles badly to pull through to get to the higher revs.
It can therefore be noted that the effect would be lessened if there was a standard exhaust fitted or a quality aftermarket variator kit (e.g. Stage6, Malossi) fitted which is designed to work well with performance exhausts as they can pull through the lower revs much more efficently.

Or in other words, keep your tyres pumped up !

In this short blog we're going to show some very basic porting.
It should be noted that this is for interest only and should not be used as a guide... if you decide to replicate this you do so entirely at your own risk !!!

Porting at a true professional level is a bit of a black art relying on extensive experience and a profound understanding not least of the mathematics of 2-stroke exhausts. Such experience usually breeds only after years being fully encompassed within a 2-stroke racing and workshop environment and even then only after a multitude of experimental trial and error successes and failures.

Without this knowledge keen enthusiasts still frequently modify their cylinders by copying other peoples' or other manufacturers' cylinder maps. By carefully selecting your cylinder it can for example be possible to duplicate the porting of a much more expensive cylinder.

Today we're going to take an air cooled 70cc Piaggio Polini Sports Cylinder Kit fitted to a Piaggio Zip and modify it to squeeze a bit more power from it.

The reasons for selecting this kit are...

1. This kit is cast iron- because of this it is very hard and needs no cylinder plating. Aluminium cylinders are very soft and the cylinder has to be plated with a very thin nikasil coating to prevent it wearing very quickly. More care has to be taken when porting aluminium cylinders in case of damaging the plating. Cast iron is for the most part beginner-proof :)
2. A modified version of this cylinder already exists as the Polini Corsa kit which we can basically replicate.
3. The kit in it's standard form is at a mild state of tune and easily improved
4. The whole modification can be done by a relative novice in less than an hour including removal and refitting of the cylinder and with no special tools and will still retain a wide powerband meaning it will work without additional bike set-up work.

Ok, all we're going to do is slightly raise the height of the exhaust port without making any other changes. This effectively decreases the tuned length of the exhaust which in turn should result in a shift of the powerband slightly upwards in the rev-range and in turn a higher peak power. This is basically copying the main difference between the Polini Sport and Polini Corsa kits so is essentially well tried and tested. As you increase the power of any cylinder this is always likely to come at the expensive of a slightly higher wear rate and we're getting to the point where a standard crank will struggle to be strong enough to be reliable, so keep this in mind ! The corsa kits come with an uprated chrome plated top piston ring instead of the standard cast ones for long-life durability but there is no spectacular difference and many corsa kits end up with standard cast rings fitted as soon as it is time for the chrome one to be replaced.

Referring to the picture below mark out an area 2.5mm above the top of the exhaust port. This new height should bring the top of the exhaust port to approx 21.5mm below the top of the cylinder. Measure this height regularly as you enlarge the port because it's easy to loose track of the pen marking.

(click to enlarge)

At a push you can remove the extra metal with just a half round metal file. Ideally hold the cylinder in a vice using wood or similar to prevent damaging the fins. It will probably take 15 mins or so to file out but access is easy and it will work just fine. If you have a dremel or dremel copy you can remove the metal much more easily with a small grinding bit in just a few minutes but still use a half round metal file regularly to keep everything smooth and neat and in straight line. You don't need to enlarge the exhaust hole itself on the outside of the cylinder where the exhaust bolts on but make sure it tapers smoothly into the enlarged port on the inside of the cylinder barrel (use a flat file or straight edge to check)
Once you're happy with it be sure to wash out any metal filings before re-assembly

The below dyno printout shows the before and after for the kit we used for this blog...

(click to enlarge)

For a 'free' modification this is quite a substantial benefit. There is a significant benefit everywhere but most significantly at higher top speeds where the power is needed most. At 67mph we're up from 5hp to 9hp !
Keep in mind we haven't optimised the setup in any way for the porting mod. We were anticipating higher power at higher rpms so it would make sense that if we now fitted lighter rollers we should see a more significant improvement. It should also be noted that after modding, the engine was running slightly lean resulting in a required increase in jetting of approx 3 sizes due to the porting.

Next time something a bit more advanced !

This is the first in a series of short blogs we're going to be doing which covers some of the basic queries which we're asked on a regular basis.
This blog addresses the question of whether or not it's important to fit a gear-up kit after fitting a 70cc cylinder kit

Firstly it's important to recap how a moped variator works in it's basic form. If working and set up properly a moped variator allows the engine to run at peak power constantly as the road speed increases. It does this by seemlessly and constantly changing the gearing over an 'infinite' range as the road speed increases to keep the engine rpm constant at a set rpm determined by the choice of roller weight. This means that whenever you mash the throttle the revs jump to the rpm at which the engine makes the most power (provided the optimal weight rollers have been fitted) and keeps them there as road speed increases meaning that the full peak power of the engine is always immediately available. This is in contrast to geared bikes where the revs go up and down in direct relation to the speed of the bike and you are virtually never at peak power rpm but only as close to it as you can maintain by constantly changing the gears.

There is an important limitation to the moped variator though in that it can only keeping changing the gearing in this way until you reach a certain road speed at which the variator is fully changed out. As soon as this speed is reached the only way that the road speed can increase further is if the engine speed also increases... in effect the moped then becomes a geared bike and the road speed can only increase in direct proportion to any increase in the engine rpm.
This 'fully changed out speed' is affected by roller weight but only up to a point (which is why roller weight doesn't affect top speed (see Moped Tuning- ROLLERS)) but it is affected predominantly by the final drive gearing in the gearbox (can be changed with gear kits) and other factors such as tyre or wheel size which do affect top speed.

The following graph shows a dyno printout of a standard Aerox...

Click to enlarge...

You can see here that the variator is holding the revs at around 7000rpms (where a standard aerox makes peak power) *until* it reaches 33mph at which point it can no longer continue changing gear and the revs increase on a 'fixed gear slope' until it gets to 9000rpms at 42mph at which point the standard Aerox is no longer making enough power to continue to accelerate.
If you change roller weight the 7000rpms will change *until* it meets up with this 'fixed gear slope' as in the following graph...

Click to enlarge...

This graph is of an unrelated moped but shows clearly how this fixed gear slope comes into play with different weight rollers. For all roller weights point 'A' shows the point at which the variator is fully changed out and the ped can only accelerate further if the rpm's increase. Although these points are at different rpm's they all meet on the same slope and this slope cannot be moved without changing the final gearing (e.g. gear kit, different size wheel etc.)

Ok, so now let's fit a Polini Sport 70cc kit and a Leo Vince ZX pipe to our Aerox.
The following graph shows how much power the most common 70cc sports kits make at different engine speeds with a ZX pipe fitted.

(click to enlarge...)

We can see that whilst our standard Aerox was making peak power at around 7000rpm's our 70cc Polini Aerox with ZX pipe is now making much more power, but it now makes this peak power at 9,300rpm instead of 7000rpm. This means we have to fit lighter rollers so as the variator runs at 9,300rpm.
If you look back to the first graph you can see a red dotted line graph added at 9,300rpm which simulates where the engine revs will be set when optimised for our new cylinder and exhaust for maximum performance.
However we've already discussed that it can only hold these revs until the variator is fully changed out at which point it hits our 'fixed gear slope line'. We can see on the first graph that this happens at 45mph with standard Aerox gears. We can therefore see it is impossible to exceed 45mph on a standard Aerox under any circumstances without the engine revs exceeding 9,300rpm.
If we refer back to the 3rd graph showing the power curve for our modified engine we can see the power developed by our Polini/ ZX engine very quickly drops off as the revs start to exceed 9,300rpm....
This means that it is therefore impossible to make full use of the power available from a Polini Sport/ ZX Aerox at any speed over 45mph without a gear kit fitted.

Without a gearkit you are very quickly completely out of the powerband over 45mph. At 50mph the engine is doing almost 11,000rpm and making around 6hp which is way down on peak power and although high enough to maintain 50mph, it isn't enough to significantly increase the speed much further. This means that the bike will spend most of its life over-revving at 11,000 just over 50mph. Because these sports kits aren't designed to run so high they are easily damaged by continual over-revving and the crank is put under more stress running at 11,000rpm under 6hp than it would have been running 9hp@ 9,300rpm.
For these reasons it's imperative that a gear-up kit is used with Malossi or Polini Sport 70cc kits.!!

The addition of a gear kit (up to 20% for this level of tuning) would shift the fixed gear slope significantly to the right allowing approx an additional 8 or 9mph of being held at 9,300rpm. This means we now have the full 9hp available right up to almost 55mph whereas without the gear kit we're down to 6hp (and falling very fast) at just 45mph.
Because once you get up around 55mph you need *significantly* more power to continue to accelerate than you do at 45mph the sports kits tend to limit themselves nicely as the power drops off when the revs finally do increase over 55mph with a gear kit fit and it's therfore not really possible to over-rev them

It's interesting to note that the 'mildest' sports kits i.e. malossi sport and polini sport are the ones where it is most important of all to have a gear kit fitted
The reason for this is that high power kits do not drop off in power at such low revs - e.g. if you look again at the power curve graph for the sports cylinder kits (our 3rd graph) you can see the Stage 6 one continues to make peak power all the way from 9000rpm to almost 11,000rpm. For this reason these sorts of kits will always destroy the Malossi and Polini sports 70cc kits on final top speed although at the price of running at higher revs (they should not however be run on a standard crankshaft as opposed to the Polini and Malossi kits which will happily run on a standard crank).

There is a widespread belief that you will loose acceleration when fitting a gear kit which is a trade off for the higher top speed but this isn't entirely true. This would be true on a single geared bike, if you had to choose only one gear to use on a geared bike then obviously 1st geared would accelerate best but have the lowest top speed... most people think therefore that fitting a gear kit is like selecting 2nd gear instead of 1st gear where you have slightly worse pull off but higher top speed.
Because of the variator mopeds don't work quite like this... the revs are still held at peak power, so in theory if your bike is making 9hp peak power and the bike is set up correctly then when you will still have 9hp available at 10mph when you mash the throttle regardless of what your final gearing is because the final gearing is being compensated for by the variator to have the same overall ratio as is necessary to keep the engine speed at our 9,300rpms. A bit like fitting a smaller gear on the back of a BMX only to compensate it with a smaller gear on the front also - net result is the same only you have the advantage of the higher overall top gearing.
In reality though the variator works less efficiently on initial pull away and struggles to be efficient until the moped is up and moving. Because of the higher final gearing the moped has to initially accelerate through a higher difference in speed until it is fully engaged and working efficiently which can affect pull away initially especially if the engine is not particularly powerful or well matched to the final gearing.

This however means that as soon as you exceed approx 15mph there isn't really any perceivable acceleration difference between a bike with or without a gear kit fitted. So unless you're frequently stopping and starting and in contrast to popular belief you can even quite happily use a gear kit on a standard 50cc bike with a decent powerpipe fitted.

This is a blog to show how restrictive the standard 12mm carb is as fitted to a mildly tuned Yamaha Aerox 50cc moped, and subsequently explain why standard Aerox's sometimes struggle to pull away after a powerpipe has been fitted

The following dyno graph is for Yamaha Aerox which was standard apart from a PM Tuning 360 exhaust, Polini 70cc sport cylinder big bore kit, and gear up kit.

click here to enlarge...

Everything else on the Aerox was completely standard apart from the variator restrictor ring being removed.

The standard carb was of course a standard Aerox 12mm carb, the 17.5mm carb was a budget but very capable 17.5mm carb , and the 19mm carb was a budget 19mm TNT carb
The standard air filter and airbox was used on the standard and 17.5mm carbs and an open filter was used on the 19mm TNT carb. The only reason for using the open air filter was that the standard airbox doesn't fit onto the TNT carb. There is however no performance benefit whatsoever of using any air filter other than the standard airbox at this level of tuning (see this article for more details)
The jetting and transmission was tweaked with each carb to ensure the mixture and revs were identical for each carb for a fair comparison, and the runs were done on a load cell dyno which allows a load to be programmed to accurately recreate wind resistance, driver weight and other real world circumstances to ensure the comparison would be identical to that experienced in real life use.
We could have professionally optimised the transmission to improve all the results all round but instead left it basically untouched to better replicate real life user experiences. Also it wouldn't have made any difference to how the graphs compared relative to each other

As you can see, although there is substantial benefit in increasing from 12mm to 17.5mm there is no performance benefit in this instance from increasing thereafter. This relates in part to the fact that as the carb inlet area increases the velocity of the airflow signicantly decreases to flow the same volume of air unless the bike is able to utilise a higher volume airflow (which it can't at this level of mild tune)
The higher revving kits e.g. Malossi MHR rep or Stage6 Race benefit more from a 19mm carb but realistically 21mm is the maximum optimal size even at this higher level of moped tuning.
It's interesting to note that Piaggio mopeds already come with this larger 17mm carb size fitted as standard, and as such there is no real performance benefit in changing from the original Piaggio carb unless you are tuning past the level of 70cc sports cylinder + sports pipe.

This graph is also interesting in that it shows insight into a problem which is common in standard 50cc Aerox scooters which have a slightly revvy sports pipe fitted e.g. Leo Vince ZX.
This is probably the most common problem we see in the workshop with Aerox mopeds where the customer has fitted a sports pipe to a 50cc Aerox and involves the bike no longer pulling away properly.
The customer will complain that whilst the bike worked fine with a standard exhaust, it will no longer pull away properly with a sports pipe fitted. When cold the bike will often be fine, but once warmed up the bike will struggle to pull away (especially on hills). On the flat the bike will often pick up speed very slowly until it eventually suddenly picks up and takes off all at once at around 20mph+... until the bike comes to a stop again at the next set of lights at which it struggles to pull away again
We can see this starting to happen if we look at the graph above at point A. You can see that the green graph for the standard carb is consistently almost exactly 1hp lower than the 17.5 and 19mm carbs at all speeds *except* below 25mph. Below 25mph the power is very significantly down on the standard carb, at 10mph the 17.5mm carb is allowing the engine to develop a massive 70% increase in power of the standard carb which is completely out of proportion with the 1hp (approx 12%) increase seen at higher speeds but why is this ?

The answer is that it's mostly due to the fact that the original Aerox moped variator is particularly poor when used with anything other than a bog standard Aerox.
The function of the variator is of course to hold the revs constantly at the revs at which the scooter makes peak power, thus ensuring peak power is always on tap instantly whenever you mash the throttle, but we can see from the lower point A on the graph that the variator is initially struggling to get the engine up to peak power (9000rpm in this case). This lower revs obviously explains the lower power until the revs match that of the other carbs at 25mph. If we look at this graph here which is for a 50cc standard cylinder...
click here to enlarge...

and look at the graph for the standard Aerox exhaust you'll see that at anything below 6800rpm the standard exhaust makes more power than a Leo Vince ZX, TT, and Stage6 Pro rep, or indeed any other half decent sports powerpipe which has a high (but narrow) powerband.
Although ideally the variator would allow the engine to rev straight to peak power whenever you throttle, the standard aerox variator tends to always engage progressively meaning that it requires a fair amount of power to initially get it there. On the high performing pipes, especially in combination with a worn standard Aerox variator this progressive application means that due to the pipe developing lower power than a standard pipe at low revs it sometimes simply can't pull through to get to the higher revs.
The problem is also particularly prevalent on Aerox mopeds because of the additional power restriction from the 12mm carb. As you can see from the Point A section in the top graph, even with a 70cc Polini sport kit fitted the slightly lower power of the 12mm carb can be seen to substantially increase the spin up time of the variator and consequently the whole performance up to 25mph is significantly affected as it struggles to make it through the progressive application of the variator
If we were to repeat the runs with a decent variator fitted such as a Stage6 Sport Pro Variator we would see Point A significantly improve and the rpm graph for the 12mm carb come much more into line with those of the other carbs.

So in conclusion...!!! When tuning an Aerox it is particularly beneficial to ditch the standard carb and standard variator even at the most basic level of tuning even although these rules would not be such an early priority when tuning other mopeds such as Piaggio/ Gilera