Massimo Quattrini; digitally detailing future 2-stroke kits

In my work for various scooter publications I’ve visited many Vespa and Lambretta kit manufacturers and tuners. Each of them adds pieces to the 2-stroke jigsaw puzzle, but few provide as much clarity as Massimiliano (‘Max’) Quattrini from Parma.

SLUK Racing’s outright victory in the last BSEC 6-hour endurance race was with an 11-year-old Quattrini M1-L reed cylinder. It’s now completed four six-hour races, taking a trophy in every one.

Standing on the shoulders of giants

Racing improves the breed. That’s an old truism which, for 2-strokes at least, largely ended when Grand Prix bikes switched to 4-stroke propulsion. You can blame Honda for that.

Prior to this moratorium, Aprilia were one of the factories that had done the most work on racing 2-stroke engines. As a result you’ll find that many current 2-stroke kits are based on the port layouts and angles of various Aprilia engines. Also many ex-Aprilia or former Grand Prix 2-stroke specialists have found their talents in demand in the remaining areas of 2-stroke use such as minibikes, motocross, karting and also scooter kits.

Two-stroke technology has not stopped; companies like KTM, Sherco, Rieju and TM are still producing 2-strokes and buyers still expect them to improve with every generation.

To learn the behaviour of gas inside a 2-stroke requires an epic amount of effort, not least because the 2-stroke engine is always doing more than one thing at a time. Wasted gas is still leaving the combustion chamber at the same time as fresh gas is entering through the transfer ports. Compression occurs at the same time that fresh fuel and air mixture is sucked into the crankcase. Every action has an overlap with some other function.

VIDEO: Quattrini minimoto cylinders being cast in a steel shell mould

Explaining 2-stroke operation is like describing the actions of a talented juggler

In the scooter world, Max Quattrini is one of the few tuners continuing to push the boundaries of the 2-stroke engine by utilising new methods of design and testing.

In the past he showed me how he used 3D design and modelling software called Flow to not only shape ports prior to production, but also to predict how they would flow gas. This software can be used to reduce turbulence in the ports which would otherwise act to limit gas-flow.

The advantages of conducting the vast majority of design and development within software are numerous. Doing so saves time, and lots of money, by eliminating the production of functioning prototypes to physically modify and test.

This is the point that Quattrini was at two years ago when I visited him to learn about his M210 TV kit for my Lambretta Kit Book. You can buy that book here.

If a new kit works great the first time you make one, because it was optimised in software, then surely that’s ‘job done’ and you won’t be able to improve it.

You’d think so wouldn’t you?

Then, Max added another weapon to his tuning armoury…

max quattrini at work

Quattrini’s new M200-S kit for Vespa smallframe

Come, Quanto, Quando


While those three words sound like an Englebert Humperdinck song lyric, they are three important questions in Italian.

• Come (pronounced kom-eh) means How
• Quanto means How Much
• Quando means When

The computer is good at calculating the first two. It will tell you How efficiently a port flows at a certain moment and How Much turbulence there is in the design. However, software is not so good for knowing When it starts to flow efficiently.

The core problem of mapping 2-strokes in software remains the same as describing a juggler juggling. The computer takes a static snapshot of something that is dynamically changing as the piston moves and the gas pressure changes.

If you look at a photograph of a juggler then there may be two balls in the air and one in his hand. You don’t know when he is catching the next ball or where he releases it, or even which direction the balls are rotating. A snapshot is only part of the story when complex dynamic movement is at play.

To understand the function of the engine, or the juggler, you really need a series of snapshots detailing the timings and forces at many different positions in the cycle. But how do you measure that?

A plastic model of an exhaust port design used for measuring gas flow characteristics when bolted to a flow bench.

Enter the Flussometro

The latest tool that Quattrini has bought with a view to improving his port layouts is called a flussometro in Italian, or better known as a ‘flow bench’ in English.

The flow bench measures airflow efficiency through a given part using a high-pressure pump system. Flow benches have a long history in 4-stroke applications but their use by privateers in 2-stroke engineering is less common.

Quattrini’s application has been to make a 3D printed mock-up of an exhaust port in a cylinder. Into this is fitted a matching 3D printed plastic piston. With the flow bench it is then possible to map the flow of the port, not just as it opens, or when it is fully-open but also at every position in between. The plastic piston sits on a threaded rod so that one full turn moves the piston 1mm to open or close the port.

With this system it is possible to test and modify the port shapes entirely in plastic prints, with each generation being used to reduce turbulence in the ports which normally limits gas flow.

After conducting this work using plastic models of cylinders with exhaust ports, he then prints different 3D models of cylinders only with transfer ports to test the transfer layouts.

Max says that he can tell how well a model cylinder flows just from the noise it makes on the flow bench. Transfers that work well give out a metallic note. By contrast a good exhaust port is quiet but emits a very deep bass tone.

Inside the 3D-printed model of Quattrini’s M200 S exhaust port design used for flow bench testing.

How much difference does the flow bench make?

Like all tools, the flow bench makes no difference on its own. However, if you have the ability to interpret the results and use that information to improve port shapes then the results can be massively better than designing in software alone.

A comparison of the gas flow (vertical scale) compared to port opening (horizontal scale) for the old M200 (green) versus the new M200 S (red). This improvement in port flow rate is matched by massive power gains across the whole rev range.

Quantify ‘massively better’

So far Max has reworked two of his most popular Vespa kits using what he has learned making new prototypes for the flow bench.

His already powerful M200 smallframe engine (external stud layout kit to fit Quattrini casings) makes around 32hp on his dyno.

The new M200-S kit makes 44hp without any change of port timings or capacity.

Power is improved over the whole rev range and the kit revs on further, but peak power remains at the same rpm. That is an incredible increase in efficiency over what was already a powerful kit.

For the M1L 60 kit (with 60mm diameter piston and using the normal smallframe Vespa cylinder stud layout) the change has been similarly remarkable. Power of the 150cc engine has jumped to that of the ‘old’ M200, not just at peak power, but with a similar width to the power spread.

Essentially, his engines have made a ‘great leap forwards’ thanks to what he has learned with the flow bench, and how he can now optimise the port layout within the cylinder to improve gas flow.

This sliced-through M1L-60S cylinder shows that the kit now has five exhaust ports in total.
A new bolt-on alloy flange accelerates the exhaust gas out of the cylinder on both kits.

So what exactly has changed?

In order to make the 150cc smallframe Vespa kit perform like the ‘old’ M200, the exhaust ports and transfers were completely redesigned.

Rather than a single exhaust port and two sub-ports (called ‘boosters’ in Italy) the new M1-L 60 kit now has four boosters. The inner pair flows into the main exhaust port directly, but the outer pair flow around the outside of the cylinder studs.

Designing a single exhaust port to flow well is relatively simple, but making sub-exhaust ports flow into the main port outlet without the gas streams colliding is much more complicated. If you get the angles wrong then total flow can end up worse than a single port.

The M1-L 60S also uses a completely new exhaust port shape where the top edge of the main port – where it meets the cylinder – is now generously radiused. If you look at the central port it looks much higher than the sub ports either side, which would work out to much longer timing if measured conventionally, but the radius is so close to the piston that there is no real flow until the piston has travelled down as far as the tops of the sub ports.

Notice the radiused top edge (upside down in this image) of the main exhaust port of the M1L-S and how the sub exhaust ports divert around the cylinder studs.
Max explained that the exhaust port timing must be taken from when the port really starts to flow, and not from when the piston edge first opens the port window. In this case it is much like the small decompression drillings above the exhaust port on kits like his M210TV and the SST from Casa Performance. You do not take the exhaust timings from when these little drillings open as they are too small to really affect the flow.

According to some other tuners I’ve spoken to in Italy, this radiused port shape is not unique in the world of 2-stroke tuning, but it is certainly unique in the world of production scooter kits. It took Max’s foundry and the plating company several goes to get exactly the right mould dimensions and thickness of plating to make the port shape exactly as per his design in Solidworks.

The M200-S has a new exhaust outlet shape with much bigger sub-ports.

M200 S

When it comes to the 200 kit for Quattrini’s own cases the exhaust port layout is much more conventional: still one main port and 2 boosters, so it’s much harder to see where the massive performance improvement in the ‘S’ version comes from.

One obvious area is the new, cast aluminium stub manifold that bolts between the cylinder and the exhaust. It’s a bit like the standard one used on smallframe Vespas except it’s designed to improve performance, not ruin it.

The outlet of the M200 S cylinder is not circular, it’s more of a rounded letterbox shape as required to get the boosters to flow more efficiently. What the new aluminium manifold does is continue the flow optimisation even further away from the piston. The casting prepares the gas direction to meet a round tube exhaust with the minimum of turbulence.

Naturally, this design means that you must use an exhaust which follows Quattrini’s preferred smooth route around the rear shock. My only difficulty with this design is that in a steel bodied scooter it is a real struggle to get the exhaust off to change a tyre. For the Italian Vespa race community, who mostly have removable fibreglass rear ends, this layout is not such an issue.

The aluminium manifold nozzles the gas and aims it correctly for a round-the-shock exhaust pipe.

What Next

Max’s next mission is to revisit his other Vespa cylinders with the help of the flow bench to try to increase their efficiency in the same way. You can expect revitalised Vespa largeframe cylinder kits next.

Given a free hand with the moulds – as Quattrini had on his 2020 Championship-winning Minimoto cylinder – then he can reach remarkable flow coefficients. The limiting factor of improving existing kits is the external design of the steel moulds for casting the cylinders. If there is sufficient space in the design then I can see all his kits making similar leaps forward in terms of performance.

Perhaps after that he will turn his hand to a kit for the Lambretta large-block engine. Anticipation and demand for that one will be quite high, I expect…

Words and Images: Sticky
Thanks to: Max Quattrini and Dario Capatti for helping with translation.

SLUK strives to bring you the most in-depth features and news on the latest scooter developments from around the world. Please use the SLUK Shop and the services of our advertisers, it’s how we pay the bills.