) and comment in case any corrections are required.I believe Dale presents the theoretical situation here:
And Wolfie presents the realistic situation here:
Both are correct of course but let’s stay with Graham’s for the moment as it represents the reality of what’s going on. If we look at the shown voltages we can see a loss of between 1 and 2 volts between the fuse box and the positive side of the coils. Let’s call it 1.5V.Wolfie wrote: ↑Mon Mar 18, 2024 6:19 pm Since there is always at least one set of points closed and drawing current, it means that after all the wires and switches it is quite normal to have less than 11V on the open set(s) of points.
Without taking the closed set into account then there would indeed be full battery voltage across any open points.
So where has this 1.5V been lost and precisely where has it gone? This is where the method I previously described comes into play.
Initially, for the sake of clarity, we’re going to state that we know the components on the negative side of the coil are in good order but I’ll address this later.
We take the negative lead of the voltmeter and clip it securely to the negative battery post and leave it there (for specific reasons that I’ll discuss later). With the ignition on and the engine turned so that 2 sets of points are closed, using positive probe from the meter we start at the positive battery terminal and record the voltage. Then the voltage at both ends of the main fuse and record those. We next move to the back of the ignition switch and record the voltages at the white and brown connection. Next are the voltages at the 2 terminals inside the kill switch and finally we record the voltage at the positive coil terminal (in practice we'd check the voltage at all 3 coils and use the lowest value as they each have their own bullet connector) The results will look something like this: We then subtract each value from the previous one to give us the voltage drop between each: A glance at the voltage drop column shows us immediately where the worst of our problems lie: I decided that it would be a good idea to work though the procedure and see how successfully I could pull the situation back towards the ideal. Not wanting to make it easy on myself I dug out the crustiest KH loom I could find – and boy was it crusty! Now this would be normally done on the bike but with all my bikes hemmed in or in bits it was easier for me to do it on the bench. I plugged in an ignition switch, a couple of coils (1 or 2 coils are usually energised but 2 coils draw more current which maximises the voltage drop), a fuse box, a kill switch and a KH points plate. I cleaned up the terminals between the loom and the points plate, connected a good battery then took my readings with the single probe method. 9.51 volts at the coil – oh my, that's not good! Clearly the worst offender was the 0.88V lost between the ignition and the kill switch. This takes us straight to the brown wire in the red plug in the headlight (and its counterpart in the white ingition plug). No surprise to find it so corroded then! Similarly the 0.56V from the kill switch to the coil goes through the same red connector (yellow wire). I extracted the connectors from the block to clean them but ideally they’d be renewed they’re so bad. The yellow bullet connector to the coil wasn't helping either. The 0.15 volts lost at the fuse is worth addressing simply because it’s small and very accessible and all these losses add up. The 0.47 volts lost between the fuse box and the ignition switch is very significant so the connections of the main white wire needed to be checked at both ends, again, no surprise. After a bit of a cleanup we see a reasonable improvement when we run through the checks again but the ignition to kill switch connector (brown wire) is still giving problems and needs further work. The 0.22 volts lost between the battery and the fuse box was annoying me because it is a short cable, appeared to be in good condition, and no amount of cleaning was restoring the voltage loss. This is where the single probe method came into its own. A little broddling and I found that there was a sudden voltage difference between the strands of wire and the actual fuse holder even though it looked perfect. A quick squeeze with the crimps and the voltage drop was gone. It was interesting to note that the voltage drop within the ignition switch had increased from 0.25 volts to 0.33 volts, but given that this is a moving connection not entirely surprising. A quick look inside the switch showed the problem and it’s easily rectified with a bit of ‘wet and dry’ on a flat surface. The same process was used on the kill switch. By the third run through with the probe I was up to an impressive 11.57 volts at the coil but there were clearly a couple of problem connectors still present. I think only renewal would rectify these but I was satisfied that the ideal situation could be approached if never fully attained: If ever a system cries out for a relay to power the coils, this is it. I’m sure such a modification would solve many a KH’s ignition problems. It would be a very simple modification as most of the necessary wiring is already present and it would almost guarantee full battery voltage at the coil: Whilst it's probably fine when new, the main problem with Kawasaki’s points ignition system design is the sheer number of separate connections between the battery and the coils. Remarkably there’s around 40 of them! That’s every crimp, every contact and every male/female. The kill switch alone has 8. Every single one of those connections has the potential to create a resistance and cause a voltage drop. The battery has to do work to force those electrons past every bit of resistance, work that is exhibited as energy loss in the form of small amounts of heat, energy that otherwise could be used to create a stronger spark.
So there’s the overview and general principle of the process, but let’s look a little deeper at some areas.
We describe voltage drop with reference to the battery voltage of course but it’s important to understand that there are 2 stages to this. Firstly, when the battery comes under load due to the ignition coils, there is an electro-chemical response from the battery resulting in an immediate drop to a ‘working voltage’. It is this working voltage that we are interested in. Secondly, at any one time it is likely that 2 ignition coils will be energised (2 closed sets of points) and due to the nature of coils, it takes a few minutes for them to stabilise. Initially they draw about 2.5 amps each but this current falls a little as the coils reach working temperature. We need to wait until the current (and hence battery voltage) has stabilised.
There are many ways to use the voltmeter but I’m quite insistent that the primary reference for zero voltage should be negative battery terminal in this investigation. There are a few of reasons for this. As the battery is our voltage source, the negative terminal is the only true zero. Voltages are relative, but in a localised system it makes sense to assign a datum at true zero and to stick with it. It also means that the negative connection to the meter can be soundly established and not disturbed. In this way we only need one probe to investigate the voltages in the system, leaving us with a free hand to hold the item we’re trying to measure. The principle is to record the voltage at both sides of a connection and then to subtract one from the other to find the voltage drop. The voltage drop could be read directly by using both probes but that’s quite a fiddly operation without a free hand and can be prone to error. The single probe method is a surprisingly quick way to map the whole system and gives a very enlightening picture. A further reason is that if you choose an arbitrary place for the negative such as an engine screw or the frame, it’s not immediately clear whether any voltage discrepancy is upstream or downstream of the meter. If you’re unaware of a downstream issue, you may get a misleading reading without even knowing it.
I demonstrated the voltage audit starting at the battery because it works well with the concept of voltage loss, we can literally see the voltage descending. In practice it actually makes more sense to start at the other end and watch the voltage ascend, because the first method is reliant on the negative side of the coil having no problems. By starting at the negative battery terminal we can check the earth lead, the frame, the engine cases, the points and the connectors with the positive meter lead. Anything other than zero (or negligible) volts at all these (with the points closed) indicates a problem that needs to be addressed before proceeding to the positive side of the coil. This method will easily pinpoint a bad engine earth, dirty points or bad connection etc.
Ironically, a VERY bad engine earth will manifest as a HIGHER voltage at the coils (or open points) and the situation gravitates more towards the open circuit situation I was grappling with before, and its resulting misleading figures (a fully open circuit/earth would show FULL battery voltage at the points but not in any useful sense as there is no current flow). This is why it is necessary to confirm that this side of the circuit is sound first, otherwise a bad earth could suggest a very healthy voltage at the coils unless the negative meter lead is placed topside of the bad earth, but that comes with its own issues as previously discussed.
The above 'bad earth' situation can be easily simulated on the bench using a rheostat or other resistance in the negative connection. As the resistance is increased, the apparent coil voltage (or open points voltage) is pushed upwards towards full battery voltage. In this instance I used a twin filament bulb to demonstrate it. I think the takeaway from this exercise is the emphasis of the inherent weakness in the system (especially in the UK climate), the simplicity of the method to diagnose the faults (it’s much easier to do than to explain), hopefully a deeper understanding to the wider audience and a refresher course for myself.
