How do electronic ignitions work?

[Ed.Kloonk]

Hello.

I have just installed a new engine in my car. It's a 1979 Ford F350 with a 351 cu (5.8l) Cleveland. It previously had a 'points' ignition with a resistive coil and now I have a new Bosch electronic distributor, and a Bosch MEC717 coil installed but not wired up.

I intend to install an ignition relay because I have been advised that the new ignition requires the full 12v so no ballast resistor should be in circuit. I mostly understand how to connect it all but I have a few questions, I hope you can help me.

1) What is the reason for having ballast resistors feed 9.6v to the coil in old cars? Why don't those coils just run at 12v all the time and not just when cranking it over?

2) The 'module' on the side of the new distributor. What -exactly- does it do to the spark and what components are in it, why is it needed?

Thanks.

[Ajahn Lambda]

I'm not used to the Australian versions of Ford's stuff, but I'm more than familiar with the American counterparts, and general automotive theory.


Ballast resistors limit coil current, nothing more or less.  If you ran constant battery voltage directly to an old oil-filled coil, it would overheat and likely fail, go pop, and maybe BOOM! and smoke.


Ignition control modules, a.k.a. igniters, are, at a very basic level, switches.  They do the same thing the points system did:  control the coil current by switching it on and off at the proper times.  The details can vary greatly from one ignition module to another, but they all perform the same switching function.  Instead of mechanical points to do the switching, they use transistors to do the job.  As long as they stay cool, and your spark plug wires are in good shape, they should, theoretically, last nearly forever.

[Ed.Kloonk]

Ah. I see. Thank you.

[vk6zgo]

I'm not used to the Australian versions of Ford's stuff, but I'm more than familiar with the American counterparts, and general automotive theory.


Ballast resistors limit coil current, nothing more or less.  If you ran constant battery voltage directly to an old oil-filled coil, it would overheat and likely fail, go pop, and maybe BOOM! and smoke.

They do a lot more than that!

When a car starter is operated,the battery voltage drops from the nominal 12 volts ,down to around 8 or so volts.
With a normal 12 volt ignition coil,directly connected to the battery via the ignition switch,the result is a reduced voltage spark at the plugs,just when the best spark is required for starting.
Automotive Electrical system designers devised the ballast system to get around this problem.
A coil designed to produce full spark with a lower primary voltage (commonly called an "8 volt coil") is used with a series resistor,during normal running.
When the starter is operated,the series resistor is bypassed,the coil still produces normal spark,& the car starts more easily.
As soon as the starter switch is released,the circuit reverts to normal.

It works,too!
I had a Mitsubishi Colt Galant back in the late 1970s,which was a brute to start in cold weather.
When I looked in the Workshop Manual & compared it with my car,I discovered that Chrysler Australia, who were assembling them, had seen fit to remove the ballast & run a "12 volt " coil.
I couldn't find a Mitsubishi ballast,so I fitted a Datsun one,along with an "8 volt" coil from  an after market parts place,
"Hey Presto!", Cold start problem fixed.

If you run a coil designed for ballast service direct from your 12 volt battery---Yes,they will get hot!
Coils designed to run directly from 12 volts don't get hot,but they still have  worse cold starting characteristics.

[Ed.Kloonk]

Thanks for that. That makes a lot of sense.

You wouldn't believe the tales told on certain car forums by self-appointed experts who really have no clue.

[vk6zgo]

I think most of the aftermarket Electronic ignitions bypass the ballast so they can run directly off 12 Volts.
If it is a real resistor,it is very easy,but if it is resistance wire,you may have to do some circuit tracing.

If you use a relay as you originally suggested,you wouldn't have to mess around with the original wiring much,but the relay would have to able to operate OK with the ballast in series during the start function.
The less you have to change things,the easier it would be to revert to standard ignition if you needed to.

PS: Have a look at the thread "Inductors & Transformers" in this forum for picture of a Transistor switched ignition coil.

[Ed.Kloonk]

I think there is a full 12v IGN wire in the engine bay somewhere. Was planning on finding it and using that to trigger the relay.

PS: Have a look at the thread "Inductors & Transformers" in this forum for picture of a Transistor switched ignition coil.

Beaudy mate. Ta. Will do tomorrow after the wine wears off.

[NiHaoMike]

Some of the newest cars (such as the Hyundai Sonata) actually use a two stage converter for the ignition. The first stage (most likely push pull or active clamp flyback) boosts 12V to 200-600V or so to charge some high voltage capacitors. The second stage (thyristor or IGBT pulse forward) boosts that to the 30-40kV or so needed to reliably fire a spark plug. Doing so allows much higher peak power sparks that can ignite lean mixtures, giving better fuel economy. (I first realized that when I noticed high voltage warnings on some modules in a Sonata, even though it's not a hybrid!)

That idea has been implemented in some 80s-90s era Honda Civics (and probably other cars from that time period) as well. Those cars got really good fuel economy, but the manufacturers didn't think the costs of the electronics were worth it. Nowadays with cheaper electronics and expensive fuel, it's coming back.

I have seen schematics for converting a conventional "flyback" ignition to a "high voltage forward" ignition. It probably doesn't work as well as one that uses a proper forward mode transformer, but some hypermilers have reported good results.

[Ajahn Lambda]

Just another detail, probably of little consequence on such an old design, but the dwell does have quite a bit to do with coil power dissipation.  The idea is to turn the coil current ON just long enough to create sufficient secondary voltage.  In fact this is why Ford TFI modules changed slightly around 1994, to include PCM-controlled dwell features.  The only physical change, other than slight wiring and terminal differences, was a black module body instead of a gray one.  The 22 k? IDM (Ignition Diagnostic Monitor) resistor was also done away with, and the module was responsible for signalling the PCM whether the ignition coil had fired.  The old modules also had a 'Start' input terminal, but since the PCM was controlling everything now, that was eliminated as well.  For ALL TFI ignitions, there were never ballast resistors present; the E-core coils were not oil-filled, but rather an open design much like a mains transformer in appearance.


Older designs, though, often do NOT control dwell very well, or sometimes not at all.  In fact, the Duraspark systems will apply current to the ignition coil if the ignition switch is turned on, and the engine isn't running.  That's one time to be thankful the ballast resistor is there.  From experience.


I'd suggest checking out some old Chrysler ignition systems - they didn't change much from the mid-70s to the late-80s, at least for many of the 318/5.2L setups.  They used a ceramic ballast resistor, mounted in plain sight to the firewall.  There were 2- and 4-terminal models, and it was common practice to keep a spare in the glove box.  Those things would get HOT!  However, like almost all of these older systems, a bypass was used during cranking.  Not that big of a deal most days, but it can really help when it's -20° outside, and your car is carburetted and the spark plugs haven't been changed in quite a long time.

[Ajahn Lambda]

Some of the newest cars (such as the Hyundai Sonata) actually use a two stage converter for the ignition. The first stage (most likely push pull or active clamp flyback) boosts 12V to 200-600V or so to charge some high voltage capacitors. The second stage (thyristor or IGBT pulse forward) boosts that to the 30-40kV or so needed to reliably fire a spark plug. Doing so allows much higher peak power sparks that can ignite lean mixtures, giving better fuel economy. (I first realized that when I noticed high voltage warnings on some modules in a Sonata, even though it's not a hybrid!)

Wow, an O.E. capacitive discharge system in an automotive application?  Neat. 


BTW, this is what MSD ignition boxes do too, i.e. kick up the battery voltage to a few hundred volts before applying it to an ignition coil.  MSD's claim is that their boxes discharge several times on EACH power stroke, but I've yet to see it on any automotive oscilloscope.  Their boxes don't hold up very long in daily-driven vehicles, either.  The voltage at the coil DOES hurt if you rest one hand on the fender, and the other on the terminals.   


Personally, I find timing accuracy to be of greater importance than peak output voltage to fuel economy and driveability.  If everything is running correctly, the spark plugs will usually fire in the ~20 kV range, assuming a 'traditional' engine and stoichiometric air/fuel ratio.  I'm not too convinced the merits of a lean air/fuel ratio will outweigh the additional costs of the controls necessary for such a system to meet air quality standards; lean ratios increase NOx dramatically, as well as heat and wear.


Just a couple of FYI things bouncing around in my head (ADD is such a b!tch):

• The secondary voltage will only increase to the point necessary to ionize the spark gap(s).  So unless it's a heavily-modified setup, a higher-voltage coil does absolutely no good.
• The factory Ford TFI systems are rated at a peak output voltage of 50 kV, so they'll even run a boosted application to a certain level.
• Almost all modern ignition systems are either COP (Coil On Plug), and the PCM directly controls each coil individually, or they use a wasted-spark system, which was very common on Ford's early modular engines, GM's passenger cars, etc.  The wasted-spark systems have half as many coils as the engine has cylinders, and two plugs are fired at once, one on the power stroke and the other on the exhaust.

[NiHaoMike]

Wow, an O.E. capacitive discharge system in an automotive application?  Neat.

Funny part is, a friend of mine made up a joke about what I believe is that technology, without knowing that it was already in use!
http://www.diyaudio.com/forums/lounge/7181-more-engineering-humor-57.html

The Sonata is known for its really good fuel economy if driven right - more than 40MPG is very easy and famous hypermiler Wayne Gerdes got it up to 66MPG, and it's not even a hybrid!
http://www.cleanmpg.com/forums/showthread.php?t=32860

Personally, I find timing accuracy to be of greater importance than peak output voltage to fuel economy and driveability.  If everything is running correctly, the spark plugs will usually fire in the ~20 kV range, assuming a 'traditional' engine and stoichiometric air/fuel ratio.  I'm not too convinced the merits of a lean air/fuel ratio will outweigh the additional costs of the controls necessary for such a system to meet air quality standards; lean ratios increase NOx dramatically, as well as heat and wear.


Just a couple of FYI things bouncing around in my head (ADD is such a b!tch):

    The secondary voltage will only increase to the point necessary to ionize the spark gap(s).  So unless it's a heavily-modified setup, a higher-voltage coil does absolutely no good.

It's not just the voltage that determines the intensity of the spark. The peak current makes a very significant difference. Just like a blowtorch will light a campfire faster than a flint striker will, a more intense spark will result in faster and more repeatable ignition. A longer spark would also take more voltage. Some of the newer engine designs use spark plugs with the ground electrode at the sides instead of over the HV electrode in order to reduce "flame shading".

The factory Ford TFI systems are rated at a peak output voltage of 50 kV, so they'll even run a boosted application to a certain level.

I still have the old flyback transformer assembly out of my 2002 Ford Taurus, so I'll be looking at it the next time I need a low current HV supply. It failed because two of the three secondaries shorted to each other. (Ford, for whatever reason, put all three flybacks in the same unit and then put the entire unit in the back where it gets hot and is hard to get to! And a week after replacing the transformers, the transmission failed. Fix Or Repair Daily...)

[sonicj]

i built this little circuit for driving a single spark plug and well... i was kinda hoping for a little more (ok, a LOT more) authoritative spark. any suggestions? or should i just hack up a eBay taser?
-sj

[Zad]

A capacitor across the drive transistor C-E, otherwise even with a diode it won't last that long. "Stiffen" the power supply with a decent amount of capacitance, and drive the transistor harder.

[vk6zgo]

i built this little circuit for driving a single spark plug and well... i was kinda hoping for a little more (ok, a LOT more) authoritative spark. any suggestions? or should i just hack up a eBay taser?
-sj

The duty cycle of the drive signal is important.
You're not just driving a transformer,the energy is stored in the coil magnetic field while the transistor is "ON" & when the transistor is turned "OFF" the magnetic field collapses,inducing current in the secondary(HT) side of the coil.

The energy available to create a spark depends on the amplitude of the current through the primary & its duration,so the "ON" time  has to be a fair bit longer than the "OFF " time.

[Chet T16]

I did some playing with ignition coils a while back

 

[G7PSK]

Ballast resistors were not always used with ignition coils many types and vehicles did not have them. They were introduced in order to give a hotter spark on cranking and start up. Basically over running the coil at cranking speeds when the battery volts are reduced due to starter motor draw, This system only became prevalent in the late 1970's before that it was hardly seen except on aero engines which had thing like magneto boost. 

[vk6zgo]

Ballast resistors were not always used with ignition coils many types and vehicles did not have them. They were introduced in order to give a hotter spark on cranking and start up. Basically over running the coil at cranking speeds when the battery volts are reduced due to starter motor draw, This system only became prevalent in the late 1970's before that it was hardly seen except on aero engines which had thing like magneto boost.

Hey!,Maybe that's why British cars were difficult to start in the winter!

They were common on Australian & US vehicles in the 1960s.
I remember leaving my car keys at a friends place & "hot wiring" my 1963 Holden so I could drive to his place & pick it up.
NOT a good idea,I cooked my distributor points!

[NiHaoMike]

i built this little circuit for driving a single spark plug and well... i was kinda hoping for a little more (ok, a LOT more) authoritative spark. any suggestions? or should i just hack up a eBay taser?
-sj

For such low voltages, use a MOSFET instead of a bipolar.

[Ajahn Lambda]

i built this little circuit for driving a single spark plug and well... i was kinda hoping for a little more (ok, a LOT more) authoritative spark. any suggestions? or should i just hack up a eBay taser?
-sj

You're going to have to drive the switching transistor(s) a LOT HARDER than that to get a powerful spark.  Ignition coil primary current is measured in amps, and it's happening pretty quickly.  Remember that V = L * (dI/dt), so if your switching speed is slow and/or the current is small, that results in much lower voltage.  You need to be slamming that current on and off as fast and hard as possible; this is why 'weak' ignition control modules often do the same thing, i.e. they get weak, lazy, switch slowly, and usually die under stress & higher temperatures.

For such low voltages, use a MOSFET instead of a bipolar.

Hmmm....  I'd think twice about that.  The primary voltage is low, granted.  However, the inductive 'kick-back' may be too much for a standard MOSFET to handle.  This can be alleviated somewhat with a capacitor, as in an old points-type condenser, but an NPN or IGBT is the modern preferred method.  I used a Darlington configuration when I developed an ignition circuit for an automotive electronics class about a decade ago -- I can't remember the exact details, nor the schematic at the moment, but if I find it I will post it here.


More useless/useful information! 

This page has some background infos:  http://hyperphysics.phy-astr.gsu.edu/hbase/magnetic/ignition.html

Here's a decent pdf on the basics of ignition system theory (I really like its explanation of dwell and controls):  http://www.autoshop101.com/forms/h39.pdf

Another page with actual values for L, dI, and dt:  http://home.earthlink.net/~jimlux/hv/igncoil.htm

And here is a [Wikipedia] schematic for a basic ignition control circuit:




Note the ratings for C2.

Check out the attachment for a screenshot of an example circuit using Freescale's MC33812.  Not so much for the IC itself, but rather the ignition coil control setup.


I'm really enjoying this discussion, btw.

[Ed.Kloonk]

I'm really enjoying this discussion, btw.

So am I. Thank you all for this. It's going to take me forever to absorb it all, but I'm glad I asked.

 

[sonicj]

thanx SigEpBlue, NiHaoMike, vk6zgo & zad!
i'll try some caps across the transistor and supply. i think i may have already killed it though... the circuit has gone intermittent. might be using the wrong plug.

hmm....  might be a good time to start from scratch with adjustable frequency & duration controls and a more experiment friendly layout.

So am I. Thank you all for this. It's going to take me forever to absorb it all, but I'm glad I asked.

 

sorry for the thread jack!  im trying to light a jet engine, so my interests are slightly off topic. 
-sj

[Ed.Kloonk]

thanx SigEpBlue, NiHaoMike, vk6zgo & zad!
i'll try some caps across the transistor and supply. i think i may have already killed it though... the circuit has gone intermittent. might be using the wrong plug.

hmm....  might be a good time to start from scratch with adjustable frequency & duration controls and a more experiment friendly layout.

So am I. Thank you all for this. It's going to take me forever to absorb it all, but I'm glad I asked.

 

sorry for the thread jack!  im trying to light a jet engine, so my interests are slightly off topic. 
-sj

Pics or ban.

 

[SeanB]

Funny enough to start a jet you do not need a high spark rate. Mostly you need a bloody big spark, firing about 4 times a second, across as big a gap as possible, preferably with 2 or more plugs firing. You would be better off finding an old or new oil burner ignition transformer, and then using long reach plugs and cut off the front of the threads to expose the centre electrode more, and add a pair of ground electrodes on again. Much higher current, and will burn out the plugs in a few hours of continuous operation, but fine for a few seconds each time starting.