Multi-cylinder ignition timing
Mark tries to explain to us all the basics of the intricacies of the multi-cylinder ignition.
Ihave been looking at ignition timing recently and now I want to look at a few arrangements to provide correct timing on different four-stroke multi-cylinder bikes.
I also want to later explore the concept of dwell angle and its measurement and I shall look only at old-fashioned contact breaker systems here! We are probably all familiar with the way a contact breaker system works, with the points acting as a mechanical switch to interrupt the low tension circuit at the critical moment, this causing the formation of a high voltage at the coil HT output to provide the spark. Life would be simple if all engines had but a single cylinder, as all that would be required would be a single pair of points and a coil. But things got more complicated well over 100 years ago and a number of schemes have been developed to provide sparks for multi-cylinder engines. Before we discuss bike engines I want to briefly cover a typical old-fashioned car engine as we will refer to that later: you will see why. Look under the bonnet of a typical car from before around 1975 and you will see a cylindrical device called the distributor (Photo 1). Inside it at the bottom there is a single set of points operated by an odd-shaped cam. I say it is odd-shaped because it has a number of lobes, corresponding to the number of cylinders so four on a typical four-cylinder engine. The cam is something like a square with rounded corners (Photo 2). The cam rotates at half engine speed so for our four cylinder engine, over two rotations of the crankshaft the points will operate four times. That corresponds to the number of sparks required of course. So now there needs to be an arrangement to distribute the sparks to the correct cylinders in order and that is done by the top part, called the distributor cap (Photo 3). An arm rotates and passes the HT current to connections to each separate HT lead at the right moment. This is not completely irrelevant to us as that system can be found on older bikes, including Triumph twins and ancient Hondas such as the C71, C75 and C76. If you have had experience of distributors you will know that they were a frequent cause of breakdowns or other troubles and bike manufacturers moved away from them in around 1960. The simplest multi-cylinder engines have two cylinders, so let us take a look at how sparks can be produced for these. In Photo 4 we have the points layout from my TX500 and as you can see there are two sets of points, one for each cylinder. The left set does the left cylinder and… well, you get the idea. Looking very carefully, you might be able to see that they are disposed at 90º apart and the cam which operates them has one lobe. That cam turns at half the crankshaft speed, so for every turn of the shaft (with each set of points operating once) the crankshaft turns twice. The crank throws are 180º apart, and 180º is twice 90º. So, apart from the cam, we have two separate ignition systems side-by-side.
Now, looking at Photo 5 we see the points plate from another twin (a Honda CB175) but that only has one set of points. What happened to the other set? To explain, on this bike the cam again turns at half crankshaft speed but we need to fire two cylinders alternately upon every revolution of the crank (360º apart). You will probably be able to appreciate straight away that this appears to be impossible as the single set of points should only separate at every other turn of the crankshaft and a very close look at the cam shows us what is going on (Photo 6). The cam is oval shaped so there are two lobes, thus the points operate twice for each cam revolution. So we get a spark for every crankshaft revolution but as no distributor is fitted, how are the sparks sent to the left and right cylinders alternately? The simple answer is that they are not: both plugs fire at the same time as the coil has two outputs. Photo 7 shows a typical 1960s Honda coil with two HT leads. So that means that on every crankshaft revolution, one plug fires at the correct time (i.e. at the end of the compression stroke) but the other fires at the end of the exhaust stroke, during the ‘overlap period’ when both inlet and exhaust valves are open. It is then that the inlet charge is being drawn in by the rapidly exiting exhaust and you might think that there would be a chance that there would be a backfire through the carburettor as a result, but that does not happen. The spark has no effect at all, and this is known as the wasted spark principle as a spark is produced but it is not used, though of course the aim is to make life simpler. There are a couple of issues resulting from this. Firstly, the plugs are wired in series and thus the ignition system fires both at once, which means that we are in effect making a spark 1.2mm long rather than the normal 0.6mm. That sounds a bit of a job for an old-fashioned coil but the fact is that the spark on the temporarily redundant plug has a much easier time as the resistance to sparking is much lower because the gas pressure is low, so this is not a problem. The second point is that as the plugs are in series, their polarity is different. That means that in effect, in one plug the spark travels from the centre to the side electrode and in the other the opposite is the case. As a plug wears, the spark slowly erodes one electrode and that is meant to be the centre one as this makes adjustment easier and this is why coils are often marked with the correct polarity on the low tension connections. If the polarity is wrong the side electrode will wear instead and this is usually regarded as a bad thing. But one plug will always be ‘wrong’ using the wasted spark arrangement. In fact, it does not seem to make much difference and the effect can be greatly reduced by using modern plugs with platinum or iridium electrodes.
Next: dwell angles.