I Built a 10-Valve Engine

Engine Enthusiast
24 followers
AI Generated·May contain errors
Video Summary
AI GeneratedThe video documents an engineering experiment to create a single-cylinder engine with 10 valves. Drawing inspiration from historical high-valve-count engines like the Honda NR500, the creator uses a 125cc test engine, 3D printing, and modified screws to build and successfully run a functional 10-valve head.
Most modern engines have four valves.
Some manufacturers experimented with five.
But what's gonna happen if you build
a 10 valve head for a single cylinder engine?
Back in the 1980s Yamaha decided
that four valves simply wasn't enough so they developed
an engine with five valves
per cylinder the FC750
But Yamaha wasn't the only manufacturer to experiment with this idea.
and here is what it sounds like.
Maserati took things even further,
developing a 6 valve per cylinder engine.
And then there was Honda,
which took things to a completely different level.
Back in the 70s, Honda wanted to return
to the 500cc Grand Prix World Championship,
the most prestigious racing competition at the time.
There was a problem.
The championship was dominated by Yamaha and Suzuki,
whose 500cc two-stroke engines were
producing around 120hp at 10,000 rpm,
and Honda wanted to beat them using a four-stroke
engine.
Now, unlike two-stroke, which has
a power stroke every revolution, a four-stroke only
has one every two revolutions,
so Honda figured that to compete with these monsters they need
They need an engine capable of revving to 20,000 rpm.
But racing regulations limited them to just 4 cylinders,
meaning they couldn't simply
build a conventional V8.
So instead, Honda came up with something absolutely insane,
an oval piston, 8 valves per cylinder
and 2 connecting rods for each piston.
Essentially, they were trying to squeeze
the breathing capacity of a V8 into a V4 engine.
And that's how the legendary Honda NR500 was born.
And now, almost 50 years later,
here I am in my laboratory trying to take this madness
one step further, by squeezing 10 valves into one single cylinder.
But unlike Honda, I don't have a team
of world class engineers or millions of dollars
of development money,
all I have is a 3D printer and some questionable engineering skills.
So let's see if we can actually make this thing
but before we start designing anything
let's establish what we are actually working with here.
For this project I'll be using my
trusty 125cc single cylinder test engine
which has already been to quite a few experiments.
It features a water cooling system,
a fully functional oil pump and a PCV system to manage
the crankcase pressures.
So instead of building an entire new engine from scratch
We already have a solid functional base,
which means we can focus on the real challenge of
this project.
But before doing any designing, we need some external measurements.
We have a 57.9mm bore,
which is already a tight fit for the 4 valve head I've been
using in the past, and since those valves are 17mm in diameter,
there's absolutely
no way we're gonna squeeze 10 of them into this tiny space.
So I started looking for alternatives, too small, too big.
Weird.
Perfect! And believe it or not,
the solution might be as simple as using some ordinary screw.
Now I know this sounds a little ridiculous
but if you look closely some screws actually
have a very similar shape to an engine valve.
They already have a circular head
and a long narrow shaft which is pretty much exactly
what we are looking for.
But these things were never designed to be engine valves.
And while I'm fairly confident they'll
do a good job mechanically, how well they'll
handle the heat of the combustion is a completely different question.
So to give these little guys
a chance I'm going to try to heat treat them.
I decided to keep things pretty simple.
I started by heating the head of each screw until it
was glowing red and then immediately
quenching it in oil.
Now the idea behind the process is that heating steel
to a high enough temperature and then
rapidly cooling it can change its internal structure,
potentially making it significantly
harder.
And since these screws are going to constantly opening
and closing against their valve seats,
Having a harder surface could help reduce wear.
The rapid cooling process is called quenching
and is commonly used to increase the hardness
and wear resistance of certain types of steel.
Now that we have a set of valves it's time to
work on the cylinder head.
But before we get into that let me tell you about today's sponsor
PCBWay. If you've been following my channel
for a while you've probably already seen some of the
crankshafts PCBWay has CNC machined for my engines.
From my inline twin engine to my flat engine
And even the crankshaft we are using in today's engine
were all made by PCBWay. But CNC machining isn't just one process.
PCBWay offers 3-axis milling for
simple parts, 5-axis machining for more complex geometries
and CNC turning for cylindrical
components like shafts. And the best part
is that you can get an instant price quote right in the
browser by simply dragging and dropping your file,
selecting the quantity and the material you want
your part to be made out of and you get
an instant price quote in a matter of seconds.
So if you need some professionally
manufactured parts for your next project,
make sure you check out PCBWay using the link in the video description.
Once again, a huge thank you to PCBWay for sponsoring the channel.
Let's get back into our project.
Now that we've got our 10 tiny valves,
it's time to design a cylinder head to fit all of them.
I started in Fusion 360, trying to figure out how to squeeze 10 valves,
the intake and exhaust ports
and an entire valve train into such a small space.
And to make things a little easier,
I decided to reuse the valve train from my previous 4-valve engine.
So, instead of designing a completely new system, I had
to figure out how to make the existing one operate 10 valves at a time.
And after a lot of messing around with the design,
I finally had something that looked like it might actually work.
With the design finished, I fired up the 3D printer and printed the
main parts of the cylinder head out of carbon fiber reinforced nylon.
But obviously, we can't just expose plastic directly
to the combustion chamber,
so I CNC milled this 10mm thick aluminium baseplate,
Which will form the bottom of our cylinder head.
Using my Carvera Air I machined
the tapered valve seats directly into the aluminium
giving all 10 valves a metal surface to seal against.
I've also added two water cooling channels
to help carry the heat away from the combustion chamber
and of course a hole for the spark.
So basically the aluminium takes care
of the combustion chamber and valve seats
while the 3D printed parts make up the rest of the cylinder head.
The next step will be the most time consuming part
of this entire build.
which is lapping all 10 valves.
You see, for an engine to build compression, the valve needs to
seal properly against their seat.
Even the tiniest gap can allow compressed air and combustion gases
to escape, resulting in a loss of compression
and serious drop in performance. And that's where
valve lapping comes in.
The process involves applying a small amount of abrasive lapping
compound between the valves and the seats,
then rotating the valves back and forth to gradually
smooth out the imperfections
and create two perfectly matching sealing surfaces.
Luckily, using wood screws as valves actually came
with an unexpected advantage.
I can use my drill and some ordinary bits to make my job easier.
With all 10 valves finally lapped,
it's time to move over to the valve springs.
Now choosing the right spring is actually pretty important.
If they are too weak, the valves might not close fast enough
at high RPM, causing something called valve float.
But if they are too strong,
they'll put unnecessary stress on the camshaft and the rest
of the valve train while also wasting precious engine power.
So we need to find the perfect balance.
But after looking through a bunch of different springs, I settle
on this, which have a wire thickness of 1mm and a free length of 50mm.
Hopefully they'll be strong enough
to keep all ten valves under control without putting too
much strain on our valve train.
Now with the spring sorted out I could finally assemble our
10 valve head and I have to say seeing all 10 valves
in place is pretty satisfying.
Next up it was time to install the head on the engine.
But we still have a pretty important problem to solve.
How exactly are we going to operate ten
valves using the same camshaft that was originally designed for four?
Well, my solution was to
redesign the tappets. Instead of having each tappet operate
a single valve, I designed these
larger tappets that can push down on five valves at a time.
This way we can use the two existing
I am using existing cam lobes on each side
to operate all 5 valves together.
But there was one more thing I had to change.
Because of the increased size of the new valve
tappers I had to redesign the valve train base
making it five millimeter taller than the original.
And surprisingly those five millimeters actually
solved another problem. You see increasing the
height of the valve train
also changed the distance between the camshaft and the crankshaft.
And somehow this gave our timing belt just the right amount of tension.
So now I don't even need
to use a belt tensioner.
Some engineering problems really do solve themselves I guess.
Ok, so the engine is all put together,
I think all that's left is to give this bad boy a spin.
We have absolutely nothing, not a single sign of combustion.
So before messing around with the fueling system
or the timing I decided to check the
spark.
And sure enough I found our first problem.
The metal connector on the spark plug wire
was touching one of the valve springs giving
the ignition current an easier path
to the ground instead of jumping across the spark
plug gap.
So basically instead of making a spark inside
the combustion chamber we were sending electricity
Luckily, this was a pretty easy fix.
I properly insulated the connector,
made sure it couldn't touch any of the springs and
we're
ready for
another
round of
testing.
Well, this time we had a combustion event,
but it was more like an external combustion
event, the cylinder head and the engine block,
so technically we made a working external
combustion engine.
Anyways, I decided to give the engine
a couple more spins to see what's actually gonna
happen.
Hmm, 3 combustion
events, I guess
we're improving.
And somehow, even with the massive compression leak,
Our 10 valve engine actually started
but before pushing it any further I decided to take the
engine apart and check if anything has been damaged
and surprisingly everything looked
pretty good, no visible damage on the valve,
the piston or the cylinder head. However,
I'm pretty sure reusing the head gasket wasn't
the best decision I took so I'm going to remove
the old gasket material,
clean up the sealing surfaces and apply a fresh coat of gasket maker
and hopefully this time we can keep the combustion inside the engine.
That was a great run but I know I can get it
to run even better. All I need to do
is a bit of carb tuning.
Looks definitely better
but I feel like the mixture
is still a little
rich. That was definitely the cleanest run
so far but let's try
to keep it alive a little bit more.
Not bad
but I
decided
to lean
the
mixture
even more
and then
I got
this run.
Well there we
have it guys
a fully
functional 10
valve engine.
Once again big
thank you to
Thank you to PCBWay for sponsoring this video,
thank you for watching guys and see you in
the next one!
