The humble scooter is one of the most accessible vehicles on the planet. Likewise, scooters are easy to learn on, forgiving to mistakes, and usually aren’t very fast. But every now and then, someone comes along and turns a scooter into something entirely different. That’s the case with Vietnamese YouTuber Le Dan, who wanted to have the crossplane four-cylinder goodness of a Yamaha R1, but in a scooter. Any regular crazy wrencher might have just welded a Yamaha R1 engine to a scooter, but Le Dan decided to create his own small-bore Yamaha R1 engine by welding four discarded scooter engines together.
This wild build comes to us from our friends over at RideApart, and I’m in love. Scooters are normally pretty mild vehicles. Most of them have puny single-cylinder setups that can’t even go on a highway. My Genuine Stella (a license-built Vespa) is great, but it’s not fast or particularly mechanically impressive. Some of the fastest scooters out there might have two whole cylinders. The crazy-fast Italjet Dragster 559 – basically a sportbike pretending to be a scooter – has a 550cc twin making 58.33 HP.


So what Le Dan is doing here is simply bonkers. I cannot even think of a single mainstream manufacturer that has built a four-cylinder scooter, including more obscure makers, too. It’s crazy enough that Le Dan built a four-cylinder scooter, but how he got there is somehow even more impressive.
Le Dan begins his video by explaining his rationale. Since 2009, the Yamaha YZF-R1 has featured a crossplane crankshaft, a technology that trickled down from MotoGP racing. I’ll let Yamaha explain:
When an engine produces torque, it is actually producing what our engineers refer to as “composite torque”, which is a combination of the torque produced by combustion, and the inertial torque resulting from the revolution of the engine’s crankshaft.
Although the level of combustion torque is both proportionate to, and also a direct result of the rider’s throttle input, the inertial torque is produced in direct relation to the changes in inertial force caused by the engine’s revolutions – and so this element of the composite torque is not under the direct and immediate control of the rider. The composite torque level varies throughout the engine’s rpm range, regardless of the rider’s throttle input, and the unpredictable nature of this composite torque means that the rider cannot select precisely and immediately the torque level that they desire at any given moment.
Therefore, in order to be able to offer more linear control to the rider, it is necessary to create an engine where the inertial torque can be minimized, and the composite torque optimized.


Yamaha got into the weeds there, so I’ll simplify it. In a typical sportbike engine, the so-called “inertial torque,” as Yamaha’s engineers call it, is caused when the crankshaft speeds up and slows down over the course of a single revolution. Yamaha believes this reduces feedback and traction for the rider.
These forces are caused by the pistons racing up and down, and balance can be improved through counterweights on the crank, a counterbalance shaft, or by adding more cylinders and pistons, which will offset a lot of motion. Let’s get Yamaha back in here:
One of the most effective means of achieving optimal composite torque in a 4-cylinder engine is the adoption of a crossplane crankshaft with the crankpins positioned to give a firing interval of 270º – 180º – 90º – 180º, as opposed to the 180º – 180º – 180º – 180º firing interval of a conventional 4-cylinder engine.
So, while a conventional 4-cylinder engine sees the four pistons and con rods effectively move up and down in the cylinders as two pairs (i.e. on the 2008 YZF-R1 the two outermost pistons and con rods move together, as do the two innermost pistons and con rods), each piston and con rod in the new crossplane crankshaft has its own individual and separate movement. The result is that the inertial torque created by the reciprocating mass in the new YZF-R1 engine is minimized, while the asymmetric firing sequence achieves a strong pulse at low to mid rpm range, together with excellent linearity across the whole rpm range.
In order to further optimize the combustion forces in the new asymmetric motor, the electronic mapping for the fuel injection system provides separate fuel injection and ignition timing for each cylinder, and this results in highly linear torque characteristics that would not have been achievable on a conventional engine with a standard crankshaft and 180º firing sequence. And for smoother running the YZF-R1’s new engine is also fitted with a primary coupling balancer which reduces the vibration produced by the asymmetric crankpin layout.

To simplify that again, Yamaha’s crossplane crank has 90-degree throws and balance is inherent in its design. Thus, a secondary counterbalance isn’t necessary. The result, according to Yamaha, is a finer throttle response and linear torque. It also makes a killer soundtrack.
Now, the “easy” thing for Le Dan to do would be to find a wrecked R1 somewhere and just graft that engine onto his scooter. Instead, he goes for the much harder (but way cooler) option: He starts with a pile of four Yamaha single-cylinder scooter engines, then custom-combines them into one inline-four. We’re not told how big these are, but they can’t be more than about 125cc or so.

Le Dan gets started by disassembling the scooter engines before taking a cutting wheel to their cases. The scooter engine cases are integrated into their CVT transmissions, and Le Dan isn’t going to need four of them. So, he cuts off everything until he gets down to a portion of each engine’s block.
Then, Le Dan cleans them all up, welds in additional metal to fill in the blocks’ walls, and then cleans up the rough edges.

What happens next is pretty neat. Le Dan takes the three smaller blocks and stacks them on top of each other, completing the tower of power with the fourth engine, which still has its transmission intact. He welds the stack together and then cuts out holes for the cylinder block.
Add in some more cleaning, and it’s shocking how quickly Le Dan had transformed a pile of junkyard engines into something entirely different.

But he’s not done yet. Next comes a block of cylinders, which is milled out of what appears to be a single piece of aluminum. Le Dan then turns to the four individual cylinder heads.
These also get chopped up and welded together.

As we near the finish line, Le Dan now has a lot of other tasks to do. He creates a long camshaft seemingly by welding the four scooter cams together. Le Dan also makes his own custom intake.
Of course, the whole crossplane crank part is important here, so Le Dan mills out his own version.

From here, there are a lot of little things left like upgrading the scooter’s clutch, painting the new engine, and attaching a quartet of carburetors. Of course, it also has to be bolted onto the scooter using a custom-fabricated engine mount.
But, finally, near the end of the video, we get to hear this beautiful creation purr.

Le Dan doesn’t give any real information about this engine. He doesn’t give any guess as to how much power it makes or how many hours he spent making this thing. Heck, who knows how many hours he spent just coming up with a game plan for how to make this engine?
But I’m not sure any of that matters because the sound alone is totally worth it. Le Dan also gets in some rolling shots, and it does seem to be faster than a single-cylinder scooter.


Certainly, it’s way faster than anyone should ride when their riding gear consists of a T-shirt and sandals. Oh yeah, Le Dan basically made this engine while barefoot, which is hilarious.
I highly recommend watching the whole video because the process of Le Dan making this thing is fascinating. It might be worth subscribing to this channel, too, because Le Dan seems to be a sort of one-man Garage 54, but with cheap scooters. This has given me a bunch of bad ideas, and I’m thankful I’m not nearly as talented as Le Dan clearly is. A four-cylinder two-stroke Vespa, anyone?
This whole thing reminds me of my brother who worked at a bicycle and moped sales and service centre in Texas during the late 1970s and early 1980s.
In 1979, he decided to hot-rod Puch Maxi moped (one of the best selling moped in the US). So, the engine was bored out from 48.8 cc to 90 cc and used the pistons from larger engines (he didn’t specify which “donor”) and custom-made gasket to fit the larger bore. He also made some modifications to the airbox, exhaust pipes, and such. No visible signs that the Puch Maxi had been hot-rodded other than slightly louder exhaust.
One day, he was caught for speeding at 70 mph in his Puch Maxi. The court threw the charge out because it was clearly impossible for Puch Maxi to go that fast and chided the idiotic police officer for erroneously busting my brother.
That seems to be massively underselling him. Garage 54 mostly does very stupid things that don’t require much technical skill. This guy appears to have legitimate fabrication experience.
This is insane.
But, y’know, the good kind.
Cool.
Now let’s get him to fuse two Goldwing motors to make a 3L H12 and three 0.5L V4 Honda interceptor engines to make a 1.5L V12!
Wrenching Hall of Fame for this guy. He’s the Fred Offenhauser / Smokey Yunick of Vietnam.
All hail to you, Le Dan!
My brother had a knock-off Vespa with like 8hp and it was plenty quick and not nearly nimble enough to ever want more power. The thought of riding the scooter above would be even more terrifying than when I nearly killed myself on my built 78hp (estimated) KX250.
Really curious what the incidence of foot injury is in SE Asia vs the western, shoe wearing world.
You may – or may not! – be amazed at how going barefoot reduces foot/knee/hip/back injuries. (Since we are on a tangent here I’m conveniently ignoring the flip flop issue). We (humans) weren’t “designed” to have a rubber/leather/plastic/wood/cloth/whatever barrier deadening all the sensors that tell our body how much force to step down with, yet here we are, doing just that.
In the natural world I would agree, but I would argue that our feet were never meant to come in contact with anything at greater than 10mph and certainly not walking on concrete and asphalt exclusively.
as to 10+ mph firm agree
as to surfaces IMO the harder the surface the more necessary the built-in equipment becomes!
Wow… my very limited wrenching skills only seem more limited now.
One thing is bugging me with his build though: no cylinder head cooling?
Around the 20 min mark of the video he mills out cooling jackets around each cylinder, then welds the top shut, for liquid cooling I assume?
Edit to add: Yep, on the finished build he has a little radiator just in front of the rider’s feet.
sweet jesus
Exactly. Wrencher? This is a whole different category from wrenching, close to miracle-making.
I want one sooooo much.