[Panamera]

Hello all,

I'm trying to devise a system to convert my vehicle that runs on a distributor to an independent coil system.

I have no clue on how to accomplish this but I am looking for some insight from the minds that be.

The original system has 3 sensors(along with many engine inputs) that the computer uses to determine the correct "dwell" used to energize one coil to fire to each cylinder as the distributor rotor comes in physical contact with the distributor cap.

1.Cam angle
2.Cylinder position
3.Top dead center

Once the computer determines the correct timing a signal is sent to the igniter which in turn energizes the coil let's call it the "trigger" wire.

The basic jist anyway is the "trigger" wire is hit 4 times per engine rotation but I need to develop an electrical system that can distribute that signal to four independent wires that energize 4 separate coils (one per cylinder)

I understand this may be a tad confusing but I'm really trying to figure it out, any help would be much appreciated.

Let me know if wiring diagrams for the distributor or anything else is necessary.

Thanks all.

 

[Sunnysky]

what makes you think you can make it work without an understanding of what is required?

Hall sensor, PLL, vacuum phase control, SS coil drivers,,coils until you know how it works.

Study existing ignition systems 1st.

 

[duke37]

The rotor arm does not make a contact with the distributor cap.

There is a long running thread which deals with a two stroke motor. It is now up to 141 messages and seems to nowhere near getting to a resolution. Do you understand this?
What is wrong with the existing system? Capacitor discharge systems can be made as powerful as required.

 

[Alec_t]

What advantage would a 4-coil system have over a single-coil one? I would expect the greater number of components to increase the probability of failure.

 

[duke37]

What advantage would a 4-coil system have over a single-coil one? I would expect the greater number of components to increase the probability of failure.

I agree with this sentiment. I do only a few miles a year but have had to replace two 4-coil systems of different makes. It is much cheaper to replace a single coil.
There is also the effect on the catalytic convertor of one cylinder not firing.
Keep the electrical and electronic components off the hot engine.

 

[hevans1944]

What advantage would a 4-coil system have over a single-coil one? I would expect the greater number of components to increase the probability of failure.

Probability of failure is a manufacturing and quality control problem. A 4-coil system should "work real fine, last a long time, cost a little, and do a lot." Recite that mantra to your auto parts supplier and see what they recommend the next time you need a replacement part! The biggest advantage of a 4-coil system is it obviates the need for an electro-mechanical spark distributor, driven from the cam shaft gearing at half the engine RPMs. One less moving part to wear out or break.

Another reason I can think of for having multiple coils is higher RPMs from the engine. It takes a certain amount of time to build up the magnetic flux in a single ignition coil before that flux can be rapidly collapsed to induce a high voltage in the secondary. With a 4-coil engine you could increase the RPM beyond the speed at which a single coil could be effectively fired, perhaps by a factor of four if the engine didn't fall apart first. Red-line at 12,000 RPM?

My 2002 Ford Taurus with a 3.0 L V-6 engine now has about 250,000 miles on the odometer and uses an electronic distributor with three ignition coils, "firing" the six cylinders in pairs since only one of the pair will be on its compression stroke when the two spark plugs "fire" simultaneously. This arrangement completely eliminates the need for a high-voltage electro-mechanical distributor, not that these devices are much prone to failure. There is always an air gap between the rotor electrode and the spark plug electrode, and the rotor electrode extends to either side of the round spark plug electrode to allow for advanced or retarded timing. Not much to go wrong there, although cracked phenolic distributor caps were a common failure mode.

My Taurus has never had ignition problems (OEM platinum spark plugs) until recently when I allowed my son-in-law to replace the valve cover gaskets. Looks like he broke one of the carbon fiber-resistance plug wires when he put everything back in place. I put the wire ends back together with a short steel pin removed from a bulletin board tack sharpened on both ends. Shoved the two broken ends together with the pin in between and covered the whole thing with shrink tubing. Should last long enough to get the car to a shop to change spark plugs, replace the spark plug high-tension wiring, and get that annoying "check engine" light turned off again.

In older engines (prior to engine control modules and computers) there was a gee-whiz mechanical arrangement, adjustable and sensitive to rotor rotation speed and manifold vacuum, mounted between the ignition points and the rotor shaft cam that opened the points to create a high-voltage discharge in the single ignition coil. You could adjust the relative timing between the cam lobe and the points to advance or retard the spark with respect Top Dead Center (TDC) of the piston in its cylinder by loosening a screw and rotating this contraption, which also set the "dwell" or the angle through which the points remained closed to "charge up" the primary of the ignition coil. You needed a stroboscopic "timing light" triggered by pulses sent to the #1 spark plug to illuminate some scratches on an "harmonic balancer" pulley, located near the bottom front of the engine, to see if you got the adjustment right. And a "dwell meter" of course. Some of these distributors could be adjusted externally with the distributor cap in place and the engine running. You just needed long arms to reach the adjustment screw, hold the timing light and squeeze its trigger, and bend over close enough to see the marks. Good riddance! The last car I owned with this arrangement was a 1968 Mercury Cougar XR7 (the one with sequential turn signals in the rear). Haven't needed to use a timing light since then. Or replace "points and condenser" which were always sold in pairs with a tiny bit of cam lubricant thrown in for good measure.

As for @Panamera's original question... yeah, go learn about ignition systems. Is your four-cylinder engine a 2-stoke or 4-stroke engine? If the latter, which is more common in modern automobile engines, each plug fires once for every two revolutions of the engine crankshaft. During each engine revolution each piston goes up and down once, but in a 4-stroke engine it has to do this twice for each power cycle: once (going down) to take in air and fuel, second (going up) to compress the air/fuel mixture, third (going down again) to provide a power stroke after the fuel/air mixture is ignited by the spark plug, and fourth (going up again) to exhaust the combustion products to the atmosphere. A 2-stroke engine produces a power stroke with each downward movement of the piston, exhausting the combustion products near the bottom of the power stroke, and bringing in a fresh fuel/air mixture and compressing same during the upward movement of the piston. The problem has always been how to separate the intake from the exhaust efficiently, but modern technology (especially electronic fuel injection and timing) is solving that problem.

Two-stroke engines are making a comeback with modern technology to assist them, and several automobile manufacturers have offered cars with 2-stroke engines in the past. What automobile do you have? The last 2-stroke engine I had was on a lawn mower, or maybe a weed-whacker or chain saw, and required a gasoline/oil mixture for piston/cylinder lubrication. That's not a mechanical requirement, just simpler to build that way. If you are interested in the 2-stroke automobile comeback revolution, go visit the links on this Google results page.

 

[Panamera]

What advantage would a 4-coil system have over a single-coil one? I would expect the greater number of components to increase the probability of failure.

The rotor arm does not make a contact with the distributor cap.

There is a long running thread which deals with a two stroke motor. It is now up to 141 messages and seems to nowhere near getting to a resolution. Do you understand this?
What is wrong with the existing system? Capacitor discharge systems can be made as powerful as required.

You are definitely correct about the rotor not touching the distributor cap the spark jumps or arcs the gap.

My issue is my 4 stroke engine hits revolutions of 10,000 revolutions per minute. Making my mechanical distributor inefficient at those speeds.

 

[Panamera]

Probability of failure is a manufacturing and quality control problem. A 4-coil system should "work real fine, last a long time, cost a little, and do a lot." Recite that mantra to your auto parts supplier and see what they recommend the next time you need a replacement part! The biggest advantage of a 4-coil system is it obviates the need for an electro-mechanical spark distributor, driven from the cam shaft gearing at half the engine RPMs. One less moving part to wear out or break.

Another reason I can think of for having multiple coils is higher RPMs from the engine. It takes a certain amount of time to build up the magnetic flux in a single ignition coil before that flux can be rapidly collapsed to induce a high voltage in the secondary. With a 4-coil engine you could increase the RPM beyond the speed at which a single coil could be effectively fired, perhaps by a factor of four if the engine didn't fall apart first. Red-line at 12,000 RPM?

My 2002 Ford Taurus with a 3.0 L V-6 engine now has about 250,000 miles on the odometer and uses an electronic distributor with three ignition coils, "firing" the six cylinders in pairs since only one of the pair will be on its compression stroke when the two spark plugs "fire" simultaneously. This arrangement completely eliminates the need for a high-voltage electro-mechanical distributor, not that these devices are much prone to failure. There is always an air gap between the rotor electrode and the spark plug electrode, and the rotor electrode extends to either side of the round spark plug electrode to allow for advanced or retarded timing. Not much to go wrong there, although cracked phenolic distributor caps were a common failure mode.

My Taurus has never had ignition problems (OEM platinum spark plugs) until recently when I allowed my son-in-law to replace the valve cover gaskets. Looks like he broke one of the carbon fiber-resistance plug wires when he put everything back in place. I put the wire ends back together with a short steel pin removed from a bulletin board tack sharpened on both ends. Shoved the two broken ends together with the pin in between and covered the whole thing with shrink tubing. Should last long enough to get the car to a shop to change spark plugs, replace the spark plug high-tension wiring, and get that annoying "check engine" light turned off again.

In older engines (prior to engine control modules and computers) there was a gee-whiz mechanical arrangement, adjustable and sensitive to rotor rotation speed and manifold vacuum, mounted between the ignition points and the rotor shaft cam that opened the points to create a high-voltage discharge in the single ignition coil. You could adjust the relative timing between the cam lobe and the points to advance or retard the spark with respect Top Dead Center (TDC) of the piston in its cylinder by loosening a screw and rotating this contraption, which also set the "dwell" or the angle through which the points remained closed to "charge up" the primary of the ignition coil. You needed a stroboscopic "timing light" triggered by pulses sent to the #1 spark plug to illuminate some scratches on an "harmonic balancer" pulley, located near the bottom front of the engine, to see if you got the adjustment right. And a "dwell meter" of course. Some of these distributors could be adjusted externally with the distributor cap in place and the engine running. You just needed long arms to reach the adjustment screw, hold the timing light and squeeze its trigger, and bend over close enough to see the marks. Good riddance! The last car I owned with this arrangement was a 1968 Mercury Cougar XR7 (the one with sequential turn signals in the rear). Haven't needed to use a timing light since then. Or replace "points and condenser" which were always sold in pairs with a tiny bit of cam lubricant thrown in for good measure.

As for @Panamera's original question... yeah, go learn about ignition systems. Is your four-cylinder engine a 2-stoke or 4-stroke engine? If the latter, which is more common in modern automobile engines, each plug fires once for every two revolutions of the engine crankshaft. During each engine revolution each piston goes up and down once, but in a 4-stroke engine it has to do this twice for each power cycle: once (going down) to take in air and fuel, second (going up) to compress the air/fuel mixture, third (going down again) to provide a power stroke after the fuel/air mixture is ignited by the spark plug, and fourth (going up again) to exhaust the combustion products to the atmosphere. A 2-stroke engine produces a power stroke with each downward movement of the piston, exhausting the combustion products near the bottom of the power stroke, and bringing in a fresh fuel/air mixture and compressing same during the upward movement of the piston. The problem has always been how to separate the intake from the exhaust efficiently, but modern technology (especially electronic fuel injection and timing) is solving that problem.

Two-stroke engines are making a comeback with modern technology to assist them, and several automobile manufacturers have offered cars with 2-stroke engines in the past. What automobile do you have? The last 2-stroke engine I had was on a lawn mower, or maybe a weed-whacker or chain saw, and required a gasoline/oil mixture for piston/cylinder lubrication. That's not a mechanical requirement, just simpler to build that way. If you are interested in the 2-stroke automobile comeback revolution, go visit the links on this Google results page.

Thanks for your reply. I am speaking of a 4 stroke engine, I build them for a living. Mechanically I understand them even electronically but this is not a matter of purely understanding a concept within the make and model of a car but an idea specific to this scenario.

I'm not sure how to distribute the SINGLE output to the one coil into 4 separate coils.

I don't have that much electronic knowhow whatsoever.

 

[hevans1944]

Thanks for your reply. I am speaking of a 4 stroke engine, I build them for a living. Mechanically I understand them even electronically but this is not a matter of purely understanding a concept within the make and model of a car but an idea specific to this scenario.

I'm not sure how to distribute the SINGLE output to the one coil into 4 separate coils.

I don't have that much electronic knowhow whatsoever.

You cannot distribute the high-voltage output of a single ignition coil except with a gear-driven distributor synchronized to the cam shaft. That is what distributors were designed for. To get four ignition coils, one for each cylinder spark plug, to fire sequentially requires four drivers, one for the primary of each coil. The problem then becomes "where do I find the control signals required to fire each ignition coil primary at the proper time with respect to the position of that cylinder's piston?"

Without much electronic know-how this will be an uphill battle to learn what you need to know. Right now, you don't even know what you don't know that you will need to know later. In electronics (and most other things) this is called climbing the learning curve. There are many side tracks and dead ends that you will encounter during your climb. You might even "fall off the cliff" and have to start over from the beginning. So get started now. Join some auto forums where this topic is explored by others with similar interests. Try not to re-invent a solution someone else has spent hours, months, or years to develop. Learn from others who have made the climb up the learning curve before you.

The basic problem seems to be how to make that single, computer-derived "trigger" wire, divide into four wires going to four separate coils that are fired in sequence. You could use this "trigger" signal to advance a two-bit binary counter, that counts 00, 01, 10, 11, 00 ... repeating forever with the decoded outputs of 0, 1, 2, 3 being used to select which of four coils receives the "trigger" but how do you get those counts to correspond with cylinders 1, 2, 3, and 4 instead of, say, 3, 4, 1, 2?

You don't have to do this binary counters... a 4-bit shift-register/ring counter approach where you circulate a single "1" bit through a 4-bit ring counter would also work: 1000, 0100, 0010, 0001, 1000, ... repeating forever. No decoding required, but the synchronizing problem still exists.

Perhaps there is a cam-shaft position sensor that can be used to determine when the #1 cylinder is in position to fire. You crank the engine and wait for this sensor output to tell you when the #1 cylinder is ready to fire before energizing any ignition coils. This allows the counter and/or shift register to be initialized, synchronizing everything that occurs afterward. If it were my engine, I would check this synchronization with every revolution of the cam shaft, just to make sure everything stayed in sync. Hopefully, you can use the "trigger" signal without firing any of the four ignition coils until your counter circuit is initialized and synchronized with the cam position. The devil is in the details, and you probably don't have access to the details of how the trigger signal is derived or what the cam position signal looks like or what the TDC signal (if present) looks like. The TDC signal, if it exists and responds only to TDC for the #1 cylinder, could be used to initialize the counter whose output will determine which coil will fire.

A really simple solution might be to just remove the high-tension wiring from the existing distributor cap and modify the cap and rotor to generate the timing signals you need. An infrared emitter/detector could be installed in place of each of the four high-voltage terminals that originally went to spark plug wires. As the now electrically disconnected rotor spins past these detectors, each would produce a pulsed output as the IR reflected off the rotor tip and back into the detector. Might be tricky to get everything lined up optically, but its simple enough. You could also use Hall-effect sensors with a small neodymium-cobalt rare-earth magnet epoxied or brazed to the rotor tip to activate each sensor as the rotor passed by. You might even be able to detect the change in capacitance as the rotor sweeps past the existing distributor cap terminals if you apply a strong bias field (a few hundred volts perhaps) to the rotor and look at the signal capacitively coupled to the distributor terminals. Point is, the synchronizing "signal" you need is already in the distributor cap. You just need a way to turn it into a gating signal to be used with the "trigger" for your four ignition coils. You will want to use CMOS logic however you do this and place the logic circuits in a small shielded metal box bonded to the system ground to keep ignition pulses from playing havoc with the logic inside.

Sounds like a fun project, but I would bet someone, somewhere, has already done it. Find that person and save yourself a lot of trouble duplicating their work.

 

[Alec_t]

Couldn't the 'waste spark' principle be extended so that all 4 coil primaries are triggered simultaneously? That would eliminate the need for a sequencing circuit.

 

[Panamera]

Couldn't the 'waste spark' principle be extended so that all 4 coil primaries are triggered simultaneously? That would eliminate the need for a sequencing circuit.

A waste spark system works on two "paired" cylinders only that fire simultaneously. One cylinder will be completing compression stroke while the paired cylinder will be completing exhaust stroke so not to cause any misfires during the cycle.

 

[hevans1944]

Any luck on adapting your existing distributor cap and rotor with optical or Hall-effect sensors?

 

[Panamera]

Any luck on adapting your existing distributor cap and rotor with optical or Hall-effect sensors?

No sir. I feel there is a way to solve this without the use there of.

I may be naive to think but by using the sensors already equipped I feel it may be possible.

 

[hevans1944]

No sir. I feel there is a way to solve this without the use there of.

I may be naive to think but by using the sensors already equipped I feel it may be possible.

It may be possible, after all the computer manages to figure out when to fire a single coil four times for every two engine revolutions of a four-cylinder four-cycle engine. So, what sensors do you have to work with? You mentioned in your first post that there were three: (1) cam angle, (2) cylinder position and (3) top dead center. What sort of signals do these sensors produce, and are each of them accessible for connection to auxiliary electronics that will be used to fire four ignition coils?

This isn't a problem with a distributor to route the high voltage ignition to four spark plugs in four cylinders, but how do you determine which of four coils to "fire" without a distributor to sort them out? Are there sensors that will give you piston position for each of the four cylinders independently? I doubt it. The mechancial distributor will have four lobes and ONE sensor for all four lobes that the computer will use to determine when to fire the single coil, four times for every two revolutions of the four-cycle engine crankshaft. With a single coil, the computer doesn't care which cylinder gets a spark. The rotor sorts that out.

 

[Panamera]

It may be possible, after all the computer manages to figure out when to fire a single coil four times for every two engine revolutions of a four-cylinder four-cycle engine. So, what sensors do you have to work with? You mentioned in your first post that there were three: (1) cam angle, (2) cylinder position and (3) top dead center. What sort of signals do these sensors produce, and are each of them accessible for connection to auxiliary electronics that will be used to fire four ignition coils?

This isn't a problem with a distributor to route the high voltage ignition to four spark plugs in four cylinders, but how do you determine which of four coils to "fire" without a distributor to sort them out? Are there sensors that will give you piston position for each of the four cylinders independently? I doubt it. The mechancial distributor will have four lobes and ONE sensor for all four lobes that the computer will use to determine when to fire the single coil, four times for every two revolutions of the four-cycle engine crankshaft. With a single coil, the computer doesn't care which cylinder gets a spark. The rotor sorts that out.

So the vehicle uses two main sensors (hall effect)

Cylinder position(cyp) and Top dead centre(tdc)

TDC : 4/360° rotation (four times per camshaft rotation. This is when each cylinder is at the point necessary of spark)

CYP: 1/360° (once per rotation. Used to reset the firing order)

Firing order for this vehicle is cylinder 1-3-4-2.

There is an ignition output signal from the ecu which is used to energize the dwell of the the coil before the distributor rotor jumps gap.

This signal wire would need to be "distributed" to each individual coil in the sequence of 1-3-4-2 at the count of the TDC then the sequence reset when the CYP sensor is detected. On and on and so forth.

 

[hevans1944]

If I understand you correctly, you have three signals to play with: TDC, CYP, and IOS (Ignition Output Signal). The IOS currently fires the ignition coil four times for every two revolutions of the crank shaft. All you need to do is farkle with IOS to make four sequential firing signals: IOS1, IOS3, IOS4, and IOS2. You should be able to do this with a 4-bit shift register (SR) that is reset to 1000 by CYP and clocked by TDC to yield outputs 0100, 0010, and 0001, 1000 ,,, etc. These four SR outputs are then ANDed with IOS using four AND gates to create IOS1, IOS3, IOS4 and IOS2.