How the ignition system works

[Agreen]

This is going to be as basic as I can make it, but I'm going to try and improve people's understanding of how the ignition system works. Here we go.

First of all, we have to understand what voltage and current is.

Voltage is the potential. It's how much attraction it has. It's the measure of how enthusiastic it wants to get somewhere.

Current is the bulk; the flow. It's the measure of how much you're going to get when it gets there.

ex: A garden hose. Open the spigot and watch the flow coming out. It's not under much pressure (voltage), but it's got some good flow of water coming out (current). Increase the pressure by blocking the end off, and it shoots a longer distance, but the flow went down (higher voltage, lower current ) Switch to a fire truck. Mega pressure and volume (equivalent to a main power line when compared to the garden hose)

Now that we have that figured out, where does the voltage come from? Well, it is fairly simple! Just don't try to understand the atomic level kinetics here and just believe that it works.

It's called Faraday's Law. Essentially, when a wire passes by a magnet, it makes the wire produce a voltage. This is as simple as I can make it. Connect the wire to something, such as a light, and it will momentarily light the light because you have given the voltage a path, therefore it makes current FLOW.

So how does this apply to my Chinese 2 stroke bicycle kit?

There's a magnet on the left side of the engine, and it's actually attached to the crankshaft. That means it spins at engine speed. For every revolution of the engine, it makes one revolution. Every time the piston goes up, then down, it rotates one time. So we now have this spinning magnet... great.

Well, turns out that they have a loop of wires packed all around that spinning magnet! Wait... wires? As in multiple? With an 's' at the end?

Yes, more wires, all coiled together help make a stronger voltage! Sweet, so now what. It's spinning, it's making voltage... how does that equal sparky spark?

...stay tuned!

 

[Agreen]

So currently we have a magnet spinning inside wires. It's actually one really long, thin wire, looped around a steel core hundreds of times. The steel core helps, but I won't be going in to detail as to why. Let's just leave it at "it helps, and makes it easy to mount and replace". So if you were able to graph the output of that wire, it would look like this:

The x (horizontal axis) represents time, or one engine revolution. The y axis (vertical) represents voltage. Let's say that, for the sake of this picture, that the piston starts halfway up on the up stroke. When it reaches T/4 (which is one quarter revolution, and also when it reaches TDC) it is creating max positive voltage. That's because when the NORTH pole of a magnet passes by the conductor loop, it creates max positive voltage. Then at T/2 (piston halfway down on the down stroke) it's not making any voltage at all because the magnet is as far away from the wire loop as it can be. At 3T/4 (piston at BDC) it's creating max negative voltage because the south pole is now under the wire loop. The cycle continues.

I know what you're thinking. So it's making this weird wavy voltage thing now, but HOW IS THAT USEFUL!!!

That's where the CDI comes in! So when you put your kit together, you connect the 2 wires from the engine (one blue, one black) to the 2 wires of this little magic box (also, one blue and one black) that has a spark plug boot on it. So as it turns out, this is where all the magic happens! It's got a circuit in there that holds the voltage until the exact right time, then when it's ready, it lets it all go to the spark plug all at once! But it doesn't just stop there, no. It actually uses another set of coils inside the box to make the voltage even higher! If you're really that interested in it, look up how a transformer works. It essentially takes a weak voltage from the "magneto" (that's what the spinning magnet/wire loop thing is called, btw) and ramps it up to be able to shoot a spark across the spark plug's gap.

The reason that little box is called a CDI is because it's a Capacitive Discharge Ignition. Think of a capacitor as a little battery, but it can store current, then blast it out quickly. So it gathers the current from the magneto (charges) then when the time is right (piston reaches within a few degrees of TDC) it blasts all the current through a coil, discharging the capacitor, and making that coil fire at a much higher voltage, but lower current, through the spark plug.

 

[Agreen]

But wait, what about the kill switch?

So you wire the kill switch to the same wires as the CDI and magneto. Ever heard that current takes the path of "least resistance"? Well, the kill switch is just that. It's a near zero resistance path for current to flow. So instead of it going to (and through) the CDI, it goes through the kill switch. Since nothing is making it through the CDI, it makes no spark and the engine dies.

So, if you're having trouble getting your engine running, how can we figure out what the problem is? Well, I like to test for spark first. It's a yes or a no, and you're not worrying about whether you poured gas all over the bike when you were fiddling with the carb (sparks... gas... makes fire... not good). So start with spark. If no spark, check your connections. If all is well, get yourself a cheap multimeter. They can be bought for less than $5. Take the wires loose at the magneto and hook up your multimeter. Remember, it's AC voltage here. (Stands for "alternating current" ... because it alternates between pos and neg...) Crank the engine over and see what it puts out. You may not get a lot, or you might. I got like 30 volts, but it depends on how fast you spin it. If it's pumping out volts, but you have no spark, it's probably the CDI.

THERE IS NOT A GOOD WAY OF TESTING THE CDI WITH A MULTIMETER!!! It doesn't tell you hardly anything, so don't even bother. If the magneto is good, wiring is good, and spark plug is good, replace the CDI.

Well, there you go, I sure hope this helps someone at least.

 

[Kartooo]

does our simple little CDI vary timing according to RPM ?
spark advance ? if so by how many *'s ?
anybody ever put a timing light on one to see ?

 

[Agreen]

That I don't know for sure. I'll have to find the schematic for the stock one. All I've seen so far is the DIY CDI schematic, which is adjustable with a potentiometer. And from what I've heard, they run better with a little timing retard, rather than more advance... which is backwards from what one would normally suspect (from stock components, that is)

 

[Kartooo]

we know it has a reference for crank position because it knows when to fire the plug.
if it wasn't so cooold in my garage i'd run it with a timing light aimed at the magnet/crank to see if it varies with RPM. yrs ago the smaller bikes i worked on with centrifigal advance timing was very critical to squeeze the most out of them. i had a honda CB160 that i never worried much about the idle timing but had the full advance setting dead on. 2 sets of points and a backing plate that rotated.

 

[Agreen]

It uses an SCR, which fires at a specific gate current. Looking at the DIY circuit, it seems like it uses a current divider to keep things proportional as rpm increases (as does the voltage and current). The SCR is the magical component that holds the current until the time is right. Bias it to fire later in its half cycle and it retards timing.