.png "Powered by vBulletin")
I seem to be often asking for advice on here so I thought it time that I gave something back!
I have made these same comments on several forums and sites in answer to the same questions for many years so I have now typed this out in Word so that I can paste it into any post at any time without redoing it.
I have been heavily into electronics, lecturing, writing books, writing conference papers etc. for many years (www.rttonline.com The Team) and also building road and successful race bikes (Ric Molnar rode my GSXR400 in 14 races and was never off the podium). I wire my own bikes to my own circuits and build my own rec/regs when it suits.
Questions are often asked about the resistance of coils, ignition, stator assemblies etc. There are a few things to be aware of (excuse the grammar). In the case of generator coils the resistance to be measured is often below five Ohms and tenths of an Ohm are usually quoted e.g. 4.3 Ohms. The problem here is that the connections to the meter, the probe connections and the probe leads themselves can be several tenths of an Ohm. Firstly, use good quality leads and probes and keep them clean. Secondly, before measuring anything touch the probe leads together and note the reading, do this several times. If everything is clean the reading should be the same each time. Then when you make a measurement, subtract this figure from your reading to get the true measurement. As a coil is a passive device i.e. simply a length of wire it does not matter which way round you connect the probes (positive or negative), the reading should be the same.
The resistance of the coil is stated in the manufacturer’s service manual and should be in spec. However, this is not the whole story. We are usually checking the coils because there is little or no spark. There is another effect which can cause this problem – the shorted turn. This is not so easy to measure, the resistance check may well still be correct. A shorted turn occurs when one single turn on the coil shorts to the next turn on the coil (due to heat, aging, vibration etc.) A coil may have a couple of thousand turns so it may seem that the loss of one turn is neither here nor there. After all, 1999 turns is just as good as 2000, BUT, this one complete turn, or loop as it now is round the coil will have a disastrous effect on the coil performance. It forms a short circuit to the magnetic field and dissipates most of the energy created by the coil, leaving little or none for the spark or voltage generation. It can only be detected by measuring the inductance of the coil; this requires a complex bridge measurement device – not usual, even in a workshop. The best way to handle this is if it is a two coil bike swop the coils over to see if the problem swops over or if it is a one coil bike borrow your mates to substitute.
Measuring rectifiers/regulators. These contain active components e.g. diodes. It is not possible to measure them using the ‘ordinary’ resistance setting on a multimeter. The problem is that they will not exhibit a measurable resistance until they are made to conduct by applying a small voltage across them. This voltage comes from the multimeter when it is switched to the ‘diode’ position. This may be shown as a symbol rather than the word diode (a triangle on its side with a vertical line at the pointy end). When buying a multimeter for bike/car work it is important to get one with a diode test function. Connecting the leads one way round will show a very low reading, connecting them the other way round will show a very high reading – this means that the diode action is working correctly.
Multimeters tend to be priced according to the application. If it is to be used professionally and thrown in a tool box when not in use and then expected to survive then it will be expensive. If it sits on the shelf at home when not in use and only used occasionally I would not pay more than £20 at the most. The accuracy of both will be about the same and certainly sufficient for what we use it for. If buying a cheap one it will usual be fitted with a cheap zinc battery that will corrode with time and its acid will eat the battery contacts. It is better to replace this type with a good leak proof alkaline battery at the start.
When reading the service manual it will often give a specification figure for the multimeter of something like 20k/Volt. This is a measure of how much the meter itself affects the reading when being used (20k/V means very little). Modern digital meters often quote a figure much higher than this. That means they affect the reading even less – which is better.
And the last point is going to get me some stick! Probes are usually clip types or sharp points. Both are useful. I take a pointed one and grind the point to a very very fine needle point. If I am trying to measure a voltage on a wire that goes into a connector where I cannot get the point into the housing to make contact I carefully pierce the wire insulation until it makes contact. Massage insulation afterwards to ensure the hole is closed up. In all my years of bike work I am not aware that I have ever caused a problem by doing this.
I have made these same comments on several forums and sites in answer to the same questions for many years so I have now typed this out in Word so that I can paste it into any post at any time without redoing it.
I have been heavily into electronics, lecturing, writing books, writing conference papers etc. for many years (www.rttonline.com The Team) and also building road and successful race bikes (Ric Molnar rode my GSXR400 in 14 races and was never off the podium). I wire my own bikes to my own circuits and build my own rec/regs when it suits.
Questions are often asked about the resistance of coils, ignition, stator assemblies etc. There are a few things to be aware of (excuse the grammar). In the case of generator coils the resistance to be measured is often below five Ohms and tenths of an Ohm are usually quoted e.g. 4.3 Ohms. The problem here is that the connections to the meter, the probe connections and the probe leads themselves can be several tenths of an Ohm. Firstly, use good quality leads and probes and keep them clean. Secondly, before measuring anything touch the probe leads together and note the reading, do this several times. If everything is clean the reading should be the same each time. Then when you make a measurement, subtract this figure from your reading to get the true measurement. As a coil is a passive device i.e. simply a length of wire it does not matter which way round you connect the probes (positive or negative), the reading should be the same.
The resistance of the coil is stated in the manufacturer’s service manual and should be in spec. However, this is not the whole story. We are usually checking the coils because there is little or no spark. There is another effect which can cause this problem – the shorted turn. This is not so easy to measure, the resistance check may well still be correct. A shorted turn occurs when one single turn on the coil shorts to the next turn on the coil (due to heat, aging, vibration etc.) A coil may have a couple of thousand turns so it may seem that the loss of one turn is neither here nor there. After all, 1999 turns is just as good as 2000, BUT, this one complete turn, or loop as it now is round the coil will have a disastrous effect on the coil performance. It forms a short circuit to the magnetic field and dissipates most of the energy created by the coil, leaving little or none for the spark or voltage generation. It can only be detected by measuring the inductance of the coil; this requires a complex bridge measurement device – not usual, even in a workshop. The best way to handle this is if it is a two coil bike swop the coils over to see if the problem swops over or if it is a one coil bike borrow your mates to substitute.
Measuring rectifiers/regulators. These contain active components e.g. diodes. It is not possible to measure them using the ‘ordinary’ resistance setting on a multimeter. The problem is that they will not exhibit a measurable resistance until they are made to conduct by applying a small voltage across them. This voltage comes from the multimeter when it is switched to the ‘diode’ position. This may be shown as a symbol rather than the word diode (a triangle on its side with a vertical line at the pointy end). When buying a multimeter for bike/car work it is important to get one with a diode test function. Connecting the leads one way round will show a very low reading, connecting them the other way round will show a very high reading – this means that the diode action is working correctly.
Multimeters tend to be priced according to the application. If it is to be used professionally and thrown in a tool box when not in use and then expected to survive then it will be expensive. If it sits on the shelf at home when not in use and only used occasionally I would not pay more than £20 at the most. The accuracy of both will be about the same and certainly sufficient for what we use it for. If buying a cheap one it will usual be fitted with a cheap zinc battery that will corrode with time and its acid will eat the battery contacts. It is better to replace this type with a good leak proof alkaline battery at the start.
When reading the service manual it will often give a specification figure for the multimeter of something like 20k/Volt. This is a measure of how much the meter itself affects the reading when being used (20k/V means very little). Modern digital meters often quote a figure much higher than this. That means they affect the reading even less – which is better.
And the last point is going to get me some stick! Probes are usually clip types or sharp points. Both are useful. I take a pointed one and grind the point to a very very fine needle point. If I am trying to measure a voltage on a wire that goes into a connector where I cannot get the point into the housing to make contact I carefully pierce the wire insulation until it makes contact. Massage insulation afterwards to ensure the hole is closed up. In all my years of bike work I am not aware that I have ever caused a problem by doing this.
Comment