Ignition polarity

[Keith Clements]

There is some misconception about ignition coil polarity and how important it may be (or not). This article is a discussion about electrical polarity of the coil and spark plugs.

The ignition coil is essentially a low voltage to high voltage transformer with about 100 to 1 ratio of windings and voltage. The coil case is not grounded, and both primary and secondary windings inside are "floating" or isolated from the case. The only thing the windings have in common is one end connected to the same primary terminal, and it really doesn't matter much which one. Being a transformer, it must have pulsating or alternating current to work. An initial pulse is created by connecting and disconnecting the primary circuit ground connection. Alternating current then comes into the function by electrical "ringing" in the condenser, similar to plucking a guitar string. A transformer is not affected by polarity, since it is an alternating current device, so it matters not to the transformer what the input or output polarity may be. Any polarity on the primary side and any polarity on high tension side will produce the same quality of spark.

Why then do we worry about coil polarity? Because the spark plugs do care which way the electrons are flowing in the high tension circuit. The spark plug has a thermally insulated centre electrode surrounded by ceramic. The centre electrode runs substantially hotter than the body of the plug acting as the other electrode which is in contact with the metal of the head. Design of the ceramic insulator determines how hot the centre electrode will run, leading to the designation of hotter or colder spark plugs. As electrons go, they love to jump away from a hot surface and fly toward a colder surface, so it is easier to drive them from hot to cold rather than from cold to hot. There is a difference of 15 to 30 percent in voltage required to make a spark "initially" jump the gap on the plug depending on which way the current of electrons flows. So, the spark plug prefers to see a voltage potential that is negative on the centre electrode and positive on the end electrode for the very first hop of the spark. Oddly enough, this has nothing to do with polarity of the vehicle electrical system, but it is influenced by the common connection inside the ignition coil.

Electrons, being negatively charged, are repelled by the negative terminal of a battery or power source and attracted to the positive terminal. This flow of electrons constitutes the electron current.
Historically, before the discovery of electrons, the flow of current was assumed to be from positive to negative. This convention is still widely used in circuit diagrams and analysis, even though it's the opposite of electron flow.
If you reverse the cable connections on the battery the current flows in the opposite direction through the vehicle wiring. For most original functions on a Jowett this does not matter as most original equipment in a Jowett is not polarity sensitive (except the radio and clock). Note that many more modern components are polarity sensitive so if you have changed to a new petrol pump which has electronic switching , rather than points, you will have to change or modify these parts.
The ignition circuit, however, is sensitive as one end of the primary winding in the ignition coil is connected to one end of the secondary winding, reversing polarity of the coil primary side will reverse the drive direction of the spark current on the output side.

image001.gif

So, reversing vehicle electrical system polarity will reverse direction of spark current. The engine still runs either way, but the spark should be more reliable under marginal conditions if you get it right. The simple fix for this is to reverse the two primary wire connections on the ignition coil. The voltage of output spark ( about 20000v ) is not significantly changed if the 12v battery adds or subtracts from the spark voltage.

So how do we know which way to connect the ignition coil for best results? Original production coils were generally marked on the primary terminals "SW" for Switch and "CB" for Contact Breaker. This was assuming the vehicle wiring was connected for positive earth (positive battery cable grounded on the chassis). If you reverse battery polarity (going to negative earth), then these old coils need to be connected with "CB" to the ignition switch and "SW" to the distributor points wire. Later issue ignition coils are marked "+" and "-" on the primary terminals where you only need to match the terminal markings to the battery posts. For positive ground the "+" terminal goes to the distributor (to be grounded on the engine block). For negative ground the "-" terminal goes to the distributor (to be grounded on the engine block).

If you are still sceptical about all this, there is a quick way to check directly which way the current is flowing in the high tension circuit. Disconnect a spark wire from a spark plug (or the coil wire from the distributor cap). Hold (with some insulated pliers and wearing rubber gloves) this HT wire near a grounding point (or near the connector end of a spark plug), and position the tip of a graphite pencil in between. When you crank the engine (no need to start or run) you can observe the resulting spark jump between wire and pencil, and between pencil and ground (or spark plug). A flare (hard to see) toward the plug (or ground) shows correct polarity while a flare toward the coil shows reversed polarity. If the flare goes toward the coil, just switch the primary wires on the coil and make note of the connections for future reference.

image003.gif

Another test method.
Using an analog (moving needle) voltmeter is probably safer. Do not use a digital meter which is likely to be destroyed by the spark pulse unless you use a 100:1 attenuator and voltage scale 400v so that it can handle 40kV!

image005.gif

Hook up a voltmeter with the negative lead to the plug terminal and the positive lead to the block. Set the meter on the highest volt range. Crank the engine over (no need to start it), and you should see an upward swing of the voltmeter needle (don't be concerned with taking a reading). If the needle swings down off the scale, your coil is hooked up wrong. To correct it reverse coil primary leads.

Yet another method.
Using a light emitting diode to return the spark to the negative terminal of the coil.

image007.jpg

image008.jpg

image009.jpg

Note that an LED is in series with the spark gap! Apparently, the spark current is either too low or too brief to blow it up; it gives a nice red flash with each spark. Its presence helps to verify that with the current circuit connections the HV lead goes positive, and that conventional current (ignoring electrons) flows from the positive on the right, through the spark gaps, the pencil lead, and the LED to the negative on the left at one of the coil terminals".

[David Morris]

Hi Keith, nice article on ignition systems. I would like to add a couple of observations. If you can manage to view the spark pulse on an oscilloscope, ( and in a previous life I had ready access to these useful instruments ), you could see the effect of lowering the battery voltage, down from 12v. The energy in the spark waveform decreases very rapidly and shows that at 12V, the spark generated is, at best, pretty marginal. When you add-in that the 12V at the coil will droop due to the drain on the battery from the starter motor and even worse on a cold morning, it gives strength to the argument for having a ballast resitor arrangement. Here the coil is wound not for 12v, but for 6v. On starting, the full 9v or so heads to the coil, boosting the spark energy. Then, when you release the key, the ballast resistor is introduced to the circuit, to reduce the voltage from the battery, now the engine is running, to the required 6v. Before electronic ignition systems came in, many manufacturers adopted this system. I suggest it's re-introduction for our classics? Six volt coils are still available, and I have this system on my Jowett Searchlight Generator. I think they are still normal for motorbikes?

My second point is a gadget which I seem to remember was called a 'Norton Spark Enhancer'. I once, being super curious, sectioned one of these and found it consisted of two poles ( like nails) to screw-connect into the king lead from the coil. There was an air gap of about 5mm between the poles. Working in an electronics laboratory at the time, we soon fashioned replicas from short lengths of glass tubing and a couple of nails, one in each end. I don't remember if this gadget, either the real one or our replicas, actually boosted the spark, but we convinced ourselves, being young boy-racers, that it certainly helped cut the mustard! A group of five of us all had Javelins and work often involved travelling up the M6, which didn't have a speed limit at the time and was almost empty. Javelin's speedos would 'go around the clock' on night journeys!

All the best and thanks,

David

[Keith Clements]

David,
As you may know I have a new toy viewtopic.php?p=49297#p49297 which indeed shows different voltages generated by the coil on different cylinders.
Yes if you introduce a gap anywhere in the circuit from the coil to the body of the spark plug you will get a higher voltage, but I am fairly sure that will not improve starting or running. In fact it is likely to cause breakdown of the insulation somewhere and lack of energy in the spark which is required to ignite the mixture.

6.png

The yellow trace is the voltage of the primary circuit measured at the CB or -ve terminal on a negatively earthed Jupiter.
The blue trace is the voltage of the secondary circuit measured using a capacitive probe on the high tension ignition wire to distributor.

I am investigating my Amy's Jup which has two cylinders not firing correctly. I made up some new HT leeds yesterday and fitted them to the two offending cylinders. I am doing the electrical substitution first before looking at the carb which is more suspect as it is on cyliners 1 and 3.

As you see from the two traces the cylinder with the large overshoot on the primary has the weakest voltage on the secondary. I am about to compare after fitting the new leads.

I also have a current probe so perhaps I can glean more information using that to measure the current in the primary circuit.

[Keith Clements]

21.png

These are the waveforms with Ch1 as before measuring voltage on the primary but this time with Ch2 measuring current on the primary.
The car idles reasonably well at 850 rpm, with good pick up on acceleration with DellLorto carbs. The advance measured with the strobe at this speed is 10 deg with max advance at 2200rpm of 23 deg. Static advance is 0 deg. For an unworn dizzie the advance should be 2 deg at 500 rpm, 8 deg at 1000rpm, 14 deg at 1500rpm, 19 deg at 2000 rpm, 23 deg at 2500rpm. The vaccum advance should give 0 deg extra up to 3 inches of vacuum and then 10 deg above 5 inches of vacuum. Points gap is nominally 16 thou to give dwell of 55 deg or 61%. There should be less than plus or minus 1 deg variation in the advance for each cam lobe which is hard to achieve.

I will continue investigating why there is inconsistency. I think the dizzie is a reasonably new reconditioned one from the Distributor doctor.
My first test yesterday was to remove the plug leads in turn and this showed 1 and 3 were very weak. Carb balance as measured by a volume meter was 3.5 on both carbs.