Diode across coil?

[MrTube]

See this is what I don't understand and is what spawned this conversation.
Why doesn't a solid state ignition need a capacitor in series with the coil if it is indeed an tank circuit?

If it is a tank circuit the cap would be needed no matter what the switching device is, no? Unless of course modern coils have a cap built in which I suppose is possible but I'd expect it to be listed on schematics then as well.

 

[marc hildebrant]

MrTube.....

Lets try another approach :

1. When the points close, the current in the coil's primary creates a large magnetic field.

2. When the points open, this magnetic field collapses since their is no current to keep it present.

3. When a magnetic field collapses, the changing amount of magnetic field creates a current in the secondary and because the number of turns of wire is large, a large voltage is produced in the secondary. The amount of voltage is related to the rate that the field collapses and the number of turns of wire...so you want a fast collapsing field.

4. When the points open, the current that was going into the coil from the battery wants to continue to flow...this is called inductance. Since the current tries to keep flowing when the points open, a spark will form which is the current trying to keep flowing.

5. The resulting spark at the points is wasted energy that could have been in the collapsing field, thus the spark at the spark plug is weak.

6. If a cap is placed at the points, then the current that was trying to flow by jumping across the points is diverted into the cap. After a very short time, the energy in the cap flows back out into the coil and boosts the collapsing field, thus increasing the spark at the spark plug.

7. Moving the cap from the points changes this flow of energy from the points to the cap and back to the coil.

8. It's not really a "tuned circuit" in the RF sense, but it's like a tuned circuit.

9. If a diode is uses instead of a cap, then when the points open, the energy from the current trying to keep flowing (inductance) will flow into the diode and become heat and you will have a poorer spark than you would have with a cap.

Marc

 

[MrTube]

MrTube.....

Lets try another approach :

1. When the points close, the current in the coil's primary creates a large magnetic field.

2. When the points open, this magnetic field collapses since their is no current to keep it present.

3. When a magnetic field collapses, the changing amount of magnetic field creates a current in the secondary and because the number of turns of wire is large, a large voltage is produced in the secondary. The amount of voltage is related to the rate that the field collapses and the number of turns of wire...so you want a fast collapsing field.

4. When the points open, the current that was going into the coil from the battery wants to continue to flow...this is called inductance. Since the current tries to keep flowing when the points open, a spark will form which is the current trying to keep flowing.

5. The resulting spark at the points is wasted energy that could have been in the collapsing field, thus the spark at the spark plug is weak.

6. If a cap is placed at the points, then the current that was trying to flow by jumping across the points is diverted into the cap. After a very short time, the energy in the cap flows back out into the coil and boosts the collapsing field, thus increasing the spark at the spark plug.

7. Moving the cap from the points changes this flow of energy from the points to the cap and back to the coil.

8. It's not really a "tuned circuit" in the RF sense, but it's like a tuned circuit.

9. If a diode is uses instead of a cap, then when the points open, the energy from the current trying to keep flowing (inductance) will flow into the diode and become heat and you will have a poorer spark than you would have with a cap.

Marc

Hi Marc,

I'm understanding that part.
Now I'm confused as to why it looks like modern ignitions do not have a cap in that section of the circuit.

Perhaps either it's built into the coil and we aren't seeing it, or the schematics are wrong?

Or, maybe the coils just have far more turns and the switching device can handle a lot more current than points?

 

[MikeK]

See this is what I don't understand and is what spawned this conversation.
Why doesn't a solid state ignition need a capacitor in series with the coil if it is indeed an tank circuit? . . .

The answer is the waveform from a transistor switch is a very fast and sharp fall, and a mechanical switch that arcs on opening produces a slower, ramped fall. Since the secondary high voltage spark is generated from the rapid FALL or collapse of the coil's magnetic field a simple mechanical switch will not provide this. The widening arc as the points open do not allow a fast enough mag field movement to produce the required secondary voltage. The L-C is actually an 'after the fact' event that produces the necessary rapid waveform transition for a high voltage with mechanical points.

A solid state system without L-C produces a much shorter duration spark, as there is no coil current resonance after the initial mag field collapse. Most transistor switched ignitions do not employ the capacitor because a discharged cap would momentarily present itself as a direct current short when the transistor switches 'on' (points close). The instantaneous initial high current surge into the cap would damage the transistor junction.

 

[MrTube]

The answer is the waveform from a transistor switch is a very fast and sharp fall, and a mechanical switch that arcs on opening produces a slower, ramped fall. Since the secondary high voltage spark is generated from the rapid FALL or collapse of the coil's magnetic field a simple mechanical switch will not provide this. The widening arc as the points open do not allow a fast enough mag field movement to produce the required secondary voltage. The L-C is actually an 'after the fact' event that produces the necessary rapid waveform transition for a high voltage with mechanical points.

A solid state system without L-C produces a much shorter duration spark, as there is no coil current resonance after the initial mag field collapse. Most transistor switched ignitions do not employ the capacitor because a discharged cap would momentarily present itself as a direct current short when the transistor switches 'on' (points close). The instantaneous initial high current surge into the cap would damage the transistor junction.

Ah!

I wish this forum had a "like" button cause that was a perfect explanation.

Thanks Mike!

 

[hoop98]

The schematic is not meant for use in a car

Here is the block diagram of a HEI IC.

Bottom line is that with a mechanical switch on a Kettering ignition you need the condenser (capacitor) to allow the points to get far enough open to prevent arcing as the primary voltage rises to ~150 volts. Introducing this capacitor does effect the circuit but is only required due to the contacts needing a head start.

With an electronic switch no capacitor is typically used.

I wonder if Charles's experience at NCR on electric typewriters might have inspired his ignition. He would have faced the challenge of despiking the solenoids that did the typing work?

 

[MikeK]

. . . I wonder if Charles's experience at NCR on electric typewriters might have inspired his ignition. He would have faced the challenge of despiking the solenoids that did the typing work?

I think you're on to something! Likely a shocking unexpected discovery. Kettering, of course, pre-dates diodes by over 30 years.

Wow, that schematic shows how things have progressed. 1996 tech and a Darlington output! Today the coil drivers are all MOSFET.

 

[johnbuckley]

Not sure if I can agree with that statement at least not in general. Very often a fraction of an inch effects how a circuit performs.

.
I've run my As with the capacitor on the firewall for decades. Electronic/magnetic fields move at very approx 1/10 speed of light . work the maths out for a couple of extra feet of wiring if you're so inclined. ( Ditto HT leads from plugs to distributor!)

 

[MrTube]

.
I've run my As with the capacitor on the firewall for decades. Electronic/magnetic fields move at very approx 1/10 speed of light . work the maths out for a couple of extra feet of wiring if you're so inclined. ( Ditto HT leads from plugs to distributor!)

As I said previously it's these kind of assumptions that cannot be made.
Many many electronic circuits even a thousandth of an inch can matter.

No, it probably doesn't matter in this circuit, but just going by what speed things move at isn't something you can base an assumption on. If you add a half inch to one of the traces in your mobile phone there is a good chance it would no longer work, at all.