Ignition tech and troubleshooting

[mikenixon]

Here we'll explore little-known ignition system parts facts, as well as uber-practical troubleshooting techniques.

Let's start with the knowledge base. First, transistors and pulsers. An ignition system has to do three things to produce a spark: switch, time, and amplify. The switching and timing is combined in just one part on points-equipped systems -- the points, of course. Electronic ignitions however split the duties among two separate parts, the transistor (igniter) which switches, and the pulser (trigger) which times. The pulser tells the transistor when to switch. The transistor then switches the coil off, and this results in a spark. The transistor is a hard-working part, and eventual failure is inevitable, with the failure happening most often when the bike is electric started with a low battery, which makes the transistor overheat. Remember this fact. The pulser on the other hand has a very different failure mode. The pulser's tiny embeded magnet weakens over time, reducing the part's already meager output even further, to the point that it can't keep up at higher rpm. The giveaway is an engine that suddenly has a 4,000 rpm redline. It just won't rev more. You confirm this diagnosis by closing up the pulser's air gap to below its normal 0.020". If the artificial rev ceiling raises, you know what the problem is.

The ignition coil. The ignition coil performs the ignition system's voltage amplification duty. Usually called simply “the coil,” an ignition coil is actually made up of two coils, a primary winding that attaches to the battery and a secondary winding that attaches to the spark plug. Ignition coils are typically constructed with a ratio of primary to secondary windings of, say, 50:1 (example only). The result is increased voltage, just as in a transformer. For cooling, old-time coils had oil inside their thin steel cases. The plastic encased coils found on powersports vehicles however are cooled by embedded wax. Though ignition coil failure is not all that common, they do overheat; remember, they are "on" much more than they are "off". Even leaving the key on after using the kill switch can burn up an ignition coil. Signs of overheating include cracks in the plastic housing and the presence of wax at the coil ends.

Plug voltage. Most street bike coils are rated at about 30,000 volts. Interestingly, the plug's spark requires far less. Why? The answer is that the voltage the spark plug needs to fire is not constant. The voltage required to bridge the plug's air gap varies in direct proportion to the changes in resistance inside that gap, which itself changes constantly because the plug is at the center of an always-changing environment. Engine rpm, road grade and conditions, temperature, gear position, throttle opening, and more -- all vary the voltage the plug demands. If 5,000 volts at idle, it may be 7,000 as soon as the throttle is used, and more yet with each gear change. So maybe 10,000 volts for a bike in a state of cruise, and 13,000-15,000 when accelerating. Quite a bit more than that 5,000 volts at idle, huh? But, wait, 15,000? That's it? That's still only half the 30,000 the average coil is capable of. What's the deal? The deal is reserve. Reserve voltage for extreme conditions such as bucking a headwind in top gear with a passenger, low tire pressues, loose battery terminals.... And also the reserve needed to overcome the inevitable carbon deposits at the same time.

High output coils. The traditional high performance ignition coil is simply one that has a significantly greater primary/secondary windings turn ratio than stock. So instead of the aforementioned 50:1, the ratio might be 70:1, resulting in a higher voltage output. A very effective approach. But some high output coils are made in a different way, by reducing the primary turns, rather than by increasing the secondary turns. Mathematically it's the same result, the same voltage step-up. Sounds like a free lunch, doesn't it? But you know better! A smaller than usual primary winding has less resistance, which permits higher current flow and thus puts more stress on the switcher and timer parts if points, and the switching transistor if your system is electronic. Honda's very high quality contact points can withstand the added strain, but the transistor boxes that later replaced those points don't fare as well, and can overheat and fail. This is precisely why both high-ohm (larger secondary) and low-ohm (smaller primary) model coils are available. The high-ohm coils suit electronic ignitions, the low-ohm are for the more tolerant points ignitions.

Carburetion. A high performance coil potentially improves carburetion. What? Yes. Remember, fuel and ignition work hand in hand. When fueling is less than perfect, more vigorous ignition can compensate. In fact, that's the real-world benefit of a high performance coil. And it works in reverse, too: better fueling makes ignition seem to work better, by making fewer demands on it. This is one advantage of a professionally rebuilt carburetor. Not only is carburetion good, combustion also improves through more efficient ignition utilization.

Okay. Troubleshooting. Start by laying the plug on a good ground and simply cranking the engine. Make sure the battery can turn the engine over adequately, that the keyswitch and engine stop switch are both turned ON, and that all but the test plug' wires are defeated (ground to the engine using screwdrivers or paper clips). We don't want to stress the ignition coils or the engine to accidentally start. Look for a purple spark, and one whose sound is snappy, crackly. If no spark, this is actually a good thing as a no-spark condition is always easier to troubleshoot than is a weak or intermittent spark. A weak or intermittent test spark, if that's what you get, requires a second, special spark test. Take a test plug and open its gap to 1/8". Yes, 0.125". This large gap simulates cylinder compression. Putting a load on the system, to stress it, is what we're doing. Checking spark with this plug and it firing either more intermittently than before or not at all will confirm that the system indeed has a fault, and it's not just bad lighting in your shop or weak eyes.

Test a pulser using an analog multimeter on its picoamp (uA) scale. A rythmic, uninterrupted needle twitch (a very slight, quick pulse) when cranking the engine indicates a good pulser. No twitch means just the opposite, a bad one. To make sure (maybe your meter is funky or you have little confidence in that easy-to-miss needle twitch), follow a bad pulser twitch test result with the 9v battery substitution test. Unplug the pulser's cable and onto the wiring harness side of the connector rapidly connect and disconnect a 9v battery, looking for spark at your grounded test spark plug. If no spark, reverse the 9v battery's polarity and try again. If still no spark, you can rule out the pulsers as the problem, or at least as the only problem, for now. If on the other hand there is spark using the 9v battery pulser substitution test, and there wasn't before, a bad pulser is indicated. The uA twitch and 9v substitution tests are the best options for pulser testing.

Ignition coils. But assuming no spark even with the 9v battery, then we move on to the ignition coils. They are easily tested, in fact the easiest parts of all. After making sure it is "hot" (reads 12v nominally) when the keyswitch and engine stop switch are both turned ON, rapidly disconnect and reconnect the blue wire going to the points or the transistor box. Each time this wire is connected then disconnected, your grounded test plug should spark. Repeat for the other coil with its yellow wire. If no spark, the coil is bad. Simple.

Spark boxes. What about the spark boxes? Here's how that works. If the substitution test resulted in no spark and the ignition coil test did result in spark, then the spark box is the problem. This is the pro way to test the spark box. None better.

This is the real-world, in-the-trenches troubleshooting process. You will notice there are no resistance tests mentioned. Though contenanced by the OEMs and endorsed my many others, professional techs avoid them. They are not best practice, to put it mildly. See my article on this subject, The 95/50 Rule.

[heraldhamster]

excellent post, as per usual, Mike!
thanks for all the great info.

[mikenixon]

excellent post, as per usual, Mike!
thanks for all the great info.

Quite welcome.

[5speed]

still waiting for the release date of the book.

[Track T 2411]

Thanks again for the great write up. The troubleshooting methods will be a great help, especially for those of us who are "electrically challenged" lol!

[77Gowing]

"ignition coils. But assuming no spark even with the 9v battery, then we move on to the ignition coils. They are easily tested, in fact the easiest parts of all. After making sure it is "hot" (reads 12v nominally) when the keyswitch and engine stop switch are both turned ON, rapidly disconnect and reconnect the blue wire going to the points or the transistor box. Each time this wire is connected then disconnected, your grounded test plug should spark. Repeat for the other coil with its yellow wire. If no spark, the coil is bad. Simple. "

Sorry for being pedantic on this but why not discuss what causes the coil voltage to rise...namely
Ldi/dt. Which means the voltage in an inductor (coil) rises proportionally to the rate of change in the inductors current. The current goes from some steady value to zero in an instant causing the voltage to spike up to your 30k volts. It's your transistor that switches the coil current on and off.
Also, resistance measurements tell you that the circuit has a conductive path without which you get no spark.
Ohms law.

Great article just the same though. Keep them coming.

With regard to transistor heat , wait till they perfect silicon carbide switches. I've played with a few in our labs in extreme situations. They handle far more heat. We were working them in 180 kW ac inverters for motor control drives for heavy duty traction situations. Since I've retired it's lekely they maybe readily available.
Ok, I'm down to the bottom of my memory banks, my ram is getting weak. Lol

[mikenixon]

"ignition coils. But assuming no spark even with the 9v battery, then we move on to the ignition coils. They are easily tested, in fact the easiest parts of all. After making sure it is "hot" (reads 12v nominally) when the keyswitch and engine stop switch are both turned ON, rapidly disconnect and reconnect the blue wire going to the points or the transistor box. Each time this wire is connected then disconnected, your grounded test plug should spark. Repeat for the other coil with its yellow wire. If no spark, the coil is bad. Simple. "

Sorry for being pedantic on this but why not discuss what causes the coil voltage to rise...namely
Ldi/dt. Which means the voltage in an inductor (coil) rises proportionally to the rate of change in the inductors current. The current goes from some steady value to zero in an instant causing the voltage to spike up to your 30k volts. It's your transistor that switches the coil current on and off.
Also, resistance measurements tell you that the circuit has a conductive path without which you get no spark.
Ohms law.

Great article just the same though. Keep them coming.

With regard to transistor heat , wait till they perfect silicon carbide switches. I've played with a few in our labs in extreme situations. They handle far more heat. We were working them in 180 kW ac inverters for motor control drives for heavy duty traction situations. Since I've retired it's lekely they maybe readily available.
Ok, I'm down to the bottom of my memory banks, my ram is getting weak. Lol

Thanks for the input, 77Gowing. Sounds like you know your stuff.

[77Gowing]

"Thanks for the input, 77Gowing. Sounds like you know your stuff. "
Thanks for the compliment, but truth be known I'm a jack of all and master of none.
I like your pragmatic and practicle approach much better. All the best.