VCI logo Archive Index | Current Issue
All About Ignition
Phil Singher

There's a tired old saw about motors: "If it's got compression, fuel and spark, it'll run." That's generally true, but it's the way those factors work together that makes a motor run powerfully and efficiently or not. This article explains the basics of ignition and is intended for the "emerging" (a.k.a. "shade tree") mechanic, so I've simplified the complicated parts.

How spark is generated:
An ignition coil is just a step-up transformer. That's a device which trades current for voltage -- we feed 12 volts from the battery through the ignition switch to the coil's primary circuit (which draws appreciable current), and we get 15,000 volts or so out of the secondary circuit (but with very little current). By cycling the primary current on and off, we simultaneously cycle the high-voltage output.

The old-fashioned way of cycling the current on and off is by using breaker points in the distributor, and a rotating cam to open and close the points. When the points are closed, they ground the return side of the coil's primary, and current can flow, generating an electro-magnetic field around the coil's core. When the points open and primary current stops, the field collapses and the energy built up in it discharges as high voltage through whatever path offers the least resistance to ground. The path we like is out the coil wire, through the distributor cap and rotor to whichever spark plug wire the rotor is pointing to. High voltage will arc through air, and the small gap of a spark plug electrode offers a fine discharge path. It's also perfectly willing to arc through deteriorating cable insulation, your hand, or anything else that offers an alternative path -- we like that less.

When the field collapses, a little bit of its energy feeds back through the primary circuit, which we can see as a small spark on the points as they open. The condenser is there to absorb that energy and minimize points sparking and burning. By the time the points close again, the coil's high voltage has hopefully expended itself through the spark plug, and the condenser discharges whatever's left harmlessly through the points.

Whatever coil you happen to have in your car, it will only produce a finite amount of high-voltage energy. At high rpm, there's little time for the field to build and output voltage can drop. It's important to set the point gap right, if you have points -- we want them closed as long as possible to build up the field, but we need the output voltage to be completely discharged before the points close again. Sometimes specified as dwell angle, it's controlled by setting point gap. This can be done adequately with a feeler gauge, but a dwell meter is more accurate.

Aftermarket ignitions like the Crane or Pertronix simply provide an electronic, instead of mechanical, means of applying a cycling ground to the coil primary. They use magnetic sensing or light interruption that keeps dwell more stable than points, particularly on a worn or wobbly distributor shaft, but they do nothing magical to boost coil output. Their main benefit is that you don't have to mess with points and condensers. Some high end models are programmable for dwell angle and can improve ignition timing across the power band -- these are significant performance improvements.

Each spark plug has to discharge at just the right point in each cylinder's cycle, and that's surprisingly complicated. The broad concept is simple: we've got a cylinder charged up with fuel/air mixture, and when the piston reaches the top, we spark the plug, the mixture burns, and the force of the bang pushes the piston back down, thereby making power. But we don't want a quick explosion; we need to apply pressure with a quick, smooth and complete burn. We want to time this so the burn starts making good pressure just as the piston reaches its high point (Top Dead Center, or TDC) and continue to make it as the piston is pushed partway back down. Because it takes a while for the whole charge to light off and make maximum pressure, we need to spark it before the piston gets to the top.

Now consider this: the cylinder is already full of mixture when the piston's at the bottom. When it gets to TDC, that mixture has been compressed into a fraction of its earlier volume (if a motor has a 10:1 compression ratio, that volume is 1/10th of what it was with the piston at the bottom and the pressure is 10 times as high). When a gas is compressed, its temperature increases in direct proportion to the rise in pressure, so it's plenty hot and close to igniting even before we supply spark.

A fuel's octane rating is a measure of how well it resists exploding suddenly (detonation) -- not how hot it burns, not how much energy is in it -- simply how much pressure it can stand before going bang. High-compression motors need stable high-octane fuel because they heat the mixture more than low-compression motors. Detonation can crack or break pistons and rings. Its extreme heat can burn holes in pistons and head gaskets. It is bad for motors and bad for power.

If we light the mixture too early (the timing's too far advanced), we build too much pressure before the piston reaches TDC. If this exceeds the stability of the fuel, we get detonation, not a smooth, power-producing burn. We hear this as "knocking" or "pinging." If we set it off too late (the timing's retarded), the maximum burn happens as the piston is already dropping and opening up the volume too quickly, decreasing pressure faster than the burn can build it. This doesn't make good power either.

You can see that the "best" timing varies with fuel octane rating, a motor's compression ratio, combustion chamber shape (and a dozen other factors we won't go into right now). That's part of the reason we can't tell you exact specs for setting the timing on your particular car. But it gets worse yet

Retarded timing is desirable when cranking the car in order to prevent the burn lighting off so early it makes the piston "kick back" in the wrong direction (if you've ever hand-cranked an old car or tractor without manually retarding the spark, you'll understand this perfectly). We also need to overcome another problem: at high rpm, the piston takes much less time to get to TDC, and spends much less time near TDC, than at low rpm. Therefore, we can't just start the burn at the same piston position for any rpm -- we need to start increasingly early as piston speed increases from idle through mid-rpm (how much and to what rpm varies greatly from motor to motor) In cars with breaker points, this is accomplished by a set of weights and springs in the base of the distributor. As the distributor spins faster, centrifugal force pushes the weights outwards against spring tension. This turns the cam that works the points so they open earlier until maximum advance is reached -- simple.

Not so fast. At moderate throttle openings, we're putting much less mixture into the cylinder than we are at large throttle openings, so the heating due to compression is much less. The amount of mixture also depends on how much vacuum the motor produces, and that varies with load, throttle position and rpm. The optimum timing advance should track how much compression heating we're getting -- and all the centrifugal advance mechanism described above cares about is rpm.

Some distributors therefore use a secondary advance mechanism, running on carburetor vacuum, which rotates the points relative to the distributor cam. It works pretty well, but it's a nightmare when it comes to performance tuning, because anything you do to the valves or carburetors affects vacuum, and therefore also the timing advance curve. It's already hard enough to get the thing to do what you want by tinkering with just the centrifugal advance, so we don't recommend vacuum advance for performance applications, even if your carbs happen to have the right ports to run it (most SUs don't).

While there are vacuum advance distributors for older Volvos, most of the ones you'll find are actually vacuum retard units introduced as a smog control measure. These hook to the manifold vacuum, and they actually decrease performance. For many years, these were the only replacement distributors available from Volvo in the U.S., so they're common. If you have one of these and it's not part of a smog system, try disconnecting the vacuum.

How to tell? Look at the distributor from the top so the vacuum unit is to your right. If the vacuum is on the side away from you, it's advance; closer to you, it's retard.

Spark plugs:
At last, we get to something simple. We only have two factors to contend with: heat range and gap. Heat range has little to do with how "hot" the spark is (whatever that means); it's a measure of how much heat the plug body will dissipate. For NGK plugs, the higher the number in the plug's identifier, the "colder" the plug, and the more heat it will handle. An NGK BP7HS gets rid of more heat than a BP6HS, for instance [See footnote]. While we need residual heat in the combustion chamber to help the burn along, too much will aggravate detonation. For performance motors that make a lot of compression heat because of a high compression ratio and/or because big carbs and cams put a lot of mixture into the cylinders, we recommend plugs a step or two colder than standard.

So far, we've only talked about timing in terms of when to start the spark. Now, realize that spark isn't an instantaneous event any more than the burn is -- it takes time for whatever voltage the coil is providing to discharge. We want the spark to help the burn along until the piston reaches 10 degrees ATDC, regardless of when we start it (it's conventional to measure piston position in terms of degrees of crankshaft rotation Before or After Top Dead Center -- BTDC or ATDC).

However much voltage the coil is supplying, it will discharge more quickly across a small gap than across a large one. The plug's gap, therefore, has a profound effect on when the spark ends. If the gap is too small for the voltage available, we won't get spark all the way to 10 degrees ATDC; if too large, the spark will "drown" and discharge through the mixture instead of across the electrodes (which doesn't ignite anything).

If you are using the standard ignition and plugs recommended in your owner's manual, set the gap to what the manual specifies. If you have an aftermarket coil or other high-performance ignition system, increase the gap to whatever that manufacturer recommends, and be prepared to experiment on your own.

What goes wrong:
Points erode. Every time they open, a tiny spark blasts off a few molecules of metal. Rough point faces won't make good contact and won't provide a good ground to the coil. Sometimes, points will bridge. The metal blasted off of one face will weld itself onto the opposing face. This doesn't happen uniformly, unfortunately, and the effective point gap will be reduced, sometimes to the extent that the points never open. Inspect the points regularly. If either condition happens when you're on the road, you can clean them up with a point file -- but that's no substitute for a new set. Replace them the first chance you have.

The little phenolic or plastic thingy on one of the points that rides on the distributor cam -- the "rubbing block" -- will eventually wear down and reduce point gap. If the point faces themselves are good, it's legal to simply re-gap them to compensate. Distributor cam lube is available at most auto parts stores. Get some and maintain a tiny bit on the cam -- more will be slung off and get into places you don't want -- and put a dab on the rubbing block and a thin wipe on the cam when replacing points. This also prevents wear and corrosion on the cam surface itself.

Erosion also occurs on the rotor tip and distributor cap contacts -- this is true of any system that uses a distributor cap, even with electronic ignition. The rotor and contacts don't actually touch (we hope). It's another gap for spark to jump, and metal gets blasted off here, too. You can clean this up with very fine sandpaper, which will get you to the store for a new cap and rotor. Bosch caps with copper contacts are better than Brand-X caps with aluminum contacts.

Condensers do go bad, although rarely. This wreaks havoc with ignition and is very tough to troubleshoot. Save yourself trouble and just replace the condenser every time you replace the points; it's not like they cost a lot. B18 distributors have a fragile insulator where the points and condenser connect. It only needs to stand off 12 volts, but if it shorts to the body, you're out of business. If it needs replacing, RPR has a "distributor repair kit" that includes it.

Clean is good. Any sort of dirt or moisture inside the distributor can make high voltage discharge inside the distributor instead of at the plugs. Any sort of resistance in the 12 volt side of the system will reduce performance, so keep all connections shiny and tight.

Plug wires (and the center coil wire) break down with age and heat. The insulation gets leaky and lets the voltage arc out. Watch the motor run in a dark garage and look for unauthorized sparks (spray a fine mist of water around the wires for a better test). The conductors become resistive and both ends are subject to corrosion. Use an ohm meter to measure from the contacts inside the distributor cap out to the connector on the plug end of each wire. It's normal to read a couple of K ohms, which is designed-in to prevent broadcasting radio static around the neighborhood, but the reading on all the wires should be fairly consistent.

Spark plug electrodes erode and the gap increases. Problems with the motor or fuel system can cause deposits to form on the plugs. Neither is good for performance. New plugs with sharp edges on the electrodes are better than dull plugs sandblasted clean. In a modern motor, plugs can last for years. In a badly worn out motor, they can last for days. Take a look at them once in a while as appropriate to your circumstances.

Like anything else mechanical, distributors wear out, especially if allowed to rust or run unlubricated. A wobbly distributor shaft can't provide accurate timing or consistent dwell. We've seen advance mechanisms lock up from lack of lubrication -- there's a little felt pad in the top of the shaft under the rotor that's supposed to get a few drops of light oil once in a while, and there's usually a fitting on the body that lets you oil the internal bushings. Springs break, fall off, or get feeble. You can test a lot of this by twisting the rotor with your hand. It should twist smoothly and snap back firmly, although that in itself won't guarantee a proper advance curve.

Coils, ballast resistors and ignition switches don't have a predictable design life (in pre-electronic ignition Volvos, the ballast is built into the coil). If these make the car run, they can be presumed good, if not optimum. If they don't, it can be hard to check which component has failed, especially if the switch-coil connection is inside an armored cable. One easy test: with the ignition on and the points open, you should read substantial voltage at the coil negative terminal (the one the wire to the points hooks to), although less than 12 volts. If you have little or no voltage there, a problem with the coil or switch is indicated. With the points closed, of course, there should be no measurable voltage at that terminal.

Quick and dirty timing:
I prefer to take the distributor out of the car when changing points and condenser. It's a lot easier to do on the bench, and lets me give the whole unit a good look-over. If you're fortunate enough to have access to a distributor machine (getting hard to find in these days of computerized ignition control), it makes it easy to set the dwell perfectly and check the advance mechanism. Here's how to get the distributor back into the car with the timing already close to right.

Before pulling it out, turn the motor so the rotor is pointing at #1 plug wire (that's the forward-most one). Leave it like that until the distributor goes back in. If that's not possible, turn the motor so #1 piston is at TDC on the compression stroke by lining up the zero mark on the crank pulley with the pointer on the timing gear cover. You can tell it's the compression stroke and not the exhaust stroke because both forward valves will be closed -- look through the oil filler cap and see that the rocker arms are even with each other (if your filler is in the middle of the valve cover, pull the cover off).

The pulley is marked at 10 ATDC, 0 (TDC), 10, 20 and 30 BTDC (some also have 40 BTDC). Turn slightly more until your desired timing mark lines up; in other words, if you want to set basic timing at 15 BTDC, the pointer should be midway between the 10 and 20 marks. Now insert the distributor, seat it fully and hook up the points wire. The rotor will point at #1, but we still need to turn the distributor body to the right angle. Turn it counter-clockwise (looking from the top) until you're sure the points are closed. Switch on the ignition. Turn the distributor clockwise until the points just open -- you'll hear and/or see a spark snap across them. Turn off the ignition and tighten up the distributor enough for starting the car. Put rotor and cap back on.

This will get you within two degrees or so of where you need to be; certainly good enough to start the car and permit it to idle. Of course, you'll need to finish up with a timing light to really get it accurate.

Tuning an older Volvo:
Factory tuning specs for older Volvos simply don't apply today. Fuel is different from what it was when they were new, and octane is rated under a different system. Engines have been rebuilt, modified or swapped. Even originally, those specs were designed to get a range of different motors in and out of dealers' shops quickly, not to optimize performance. Beyond gapping points and plugs as discussed above, ignition tuning consists of setting the basic timing and tinkering the advance curve. Assuming that your motor, distributor and other ignition parts are in good shape (and if they're not, what's the point of tuning?), there are a few guidelines.

Remember two things: 1) You must avoid detonation (pinging). 2) The important factor in performance is the total timing (at full advance) rather than the basic timing (at idle). You will generally get the best performance by advancing the timing in increments until you can just make the car ping under load -- usually, this happens in the mid-rpm under full throttle -- and then backing it down just enough so it doesn't. Let the basic timing fall where it may, and make note of it so you can set it that way again next time.

If you don't have a dynamometer or distributor machine around the house, changing the advance curve is an arduous, hit-or-miss business. It requires changing the spring tension for the centrifugal weights, and that means taking the breaker plate out of the distributor for each adjustment. Most home mechanics will want to leave well enough alone, but it is possible to optimize midrange response by adjusting the springs so that maximum advance happens at greater or lesser rpm. Mild adjustments are possible by gentle bending of the tabs the springs hook onto; greater changes require experimenting with different springs.

There is a maximum amount of total advance any particular motor can use, and it varies greatly. If your motor runs happily on regular gas and doesn't ping, you won't get more performance by using premium gas and advancing the timing beyond that maximum point (I once put 102-octane in a '64 VW bus and experimented. Nothing pleasant happened). On the other hand, many performance motors require the highest octane available -- using anything less will have you retarding timing to the point where performance degrades drastically.

Common sense applies to ignition, as it does to everything else. I hope I've given you some knowledge on which to make sensible decisions. Be sure to write in if you have questions or if we can help in some way.

Footnote: In our print edition, I used Bosch plug identifier numbers as an example of how to read heat ranges. Unknown to me, Bosch had changed its numbering system and higher numbers no longer mean colder plugs. I apologize for any resulting confusion. Thanks to Daniel Stern for pointing out the error.

Back to the Top