Role of Capacitors in Automotive Ignition, Power Supply & Radio Tuning

[logearav]

dear revered members,
could u people let me the know the role of capacitors in the ignition system of automobile engines? how they eliminate sparking in ignition system
2) how they reduce voltage fluctuations in power supplies
3) how they are used in tuning radio circuits
thanks in advance

 

[mikelepore]

(2) Although capacitor current can jump abruptly to a new value (since it is the rate of charge flow), the voltage across a capacitor cannot jump abrupty (since capacitor voltage is proportional to the amount of charge accumulated, and the movement of charge takes some time.) So if you have a voltage that keeps jumping up and down, adding a parallel capactor will make it smoother.

 

[Bob S]

Hi Logearav-
Here is a simulation of an old automobile ignition circuit.
In the thumbnail is a simulation of an old automobile ignition circuit, in common use before they were transistorized in the 1970’s. V1 is a 12 volt battery. R1 is a 12-ohm series resistor that limits the battery current to about 1 amp. L1 is the ignition coil primary, about 2 milliHenrys. S1 is a switch representing the “breaker points” that is opened and closed by a cam inside the automobile distributor. Across the breaker points is a 0.02 microFarad capacitor (“condenser”) that is shorted out when the points are closed. V2 is just a power supply that I am using to open and close the points.

The red curve represents the voltage across the points. It is 0 volts when the points are closed. When the points are closed, the coil current charges up to 1 amp, limited by the series resistor. At 2 milliseconds, the points open, and the voltage shoots up to about 300 volts, across both the capacitor and the ignition coil primary. The coil and capacitor represent a series resonant circuit with a resonant frequency of about 25,164 Hz (please check my math). Because the ignition coil has a 1:100 turns ratio, the voltage on the coil secondary would be about 30,000 volts. At 4 milliseconds, the points close, and the capacitor is shorted out until they open again at 6 milliseconds. If you look carefully at 4 milliseconds, and again at 8 milliseconds on the trace, you can see a 12 volt step when the points close.
Bob S

 

[fleem]

There is always some capacitance in any circuit. In the case of an ignition coil, most of the stray capacitance is between various windings. Without the external capacitor, there would be many different resonant frequencies (because of those random stray capacitances) and because those capacitances are of rather low values, the resonant frequencies are so high that the other parts of the circuit can't respond as they should AND there would be more unwanted (and energy-wasting) EM radiation. So an external cap is added to force a single resonant freq and at a more suitable low freq.

 

[DeShark]

3) how they are used in tuning radio circuits
thanks in advance

They are used in conjunction with an inductor and make use of resonance in the circuit. This is known as an LC circuit. The resonance is set up by the continuous charging and discharging of the capacitor through the inductor. As the charge flows through the inductor a magnetic field is set up. Once the charge is equal on both plates of the capacitor, the current continues to flow, taking energy from the magnetic field. This means that a charge builds up on the opposite side of the capacitor. The frequency at which this happens depends on the values of inductance and capacitance. A variable capacitor is usually used in radios to change this resonant frequency. Because all frequencies other than the resonant frequency are quickly killed, selection of a specific frequency is made possible.

 

[Integral]

In general you can say that a capacitor blocks Dc and passes AC. To AC a cap is like a variable resistor, the value of the resistance depends on the size of the cap and the frequency of the AC. In a power supply caps are sized to be a low resistance to 120 Hz, This serves as a low resistance to ground for the AC ripple voltage created by the source voltage and the rectifier circuit.

 

[Bob S]

Hi Logearav-
I attach a url to a description of an old vacuum tube regenerative AM radio circuit:
http://www.schmarder.com/radios/tube/1-30.htm
There are several pictures, and a schematic at the bottom. The thing on the left in the first picture is a parallel phate tuneable capacitor. In the last picture, a tuneable parallel plate capacitor (about 30 picofarads to 365 picofarads) is used with an RF coil (inductor) to create a parallel-resonant RF circuit (0.55 to 1.60 MHz) that is coupled to the antenna to create a grid drive signal. The round thing in the middle is a vacuum tube. Instead of an emitter, base, and collector, it has a cathode, grid, and plate.
Bob S

 

[logearav]

Thanks to all the members who helped clarify my doubt

 

[Naty1]

In old style ignition systems with mechanical points, unwanted voltage and current spikes cause unwanted sparking at the mechanical contact points while delivering charge to the spark plugs where a spark is required. In modern electronic ignition systems manual contact points are elimnated.

The old mechanical points were a bit of a hassle because the points pitted (hole in one side, build up of metal on the other) and as they wore away the timing of the current delivery was changed causing a reduction in engine performance. Parallel capacitors served to reduce spikes across the points extending their life. In some applications, like outboard engines, the points were hidden away inside the flywheel and were a hassle to get to for adjustment; much easier to get into distributor caps in automobiles where such points were housed.

 

[Bob S]

In old style ignition systems with mechanical points, unwanted voltage and current spikes cause unwanted sparking at the mechanical contact points while delivering charge to the spark plugs where a spark is required. In modern electronic ignition systems manual contact points are elimnated..

Is there still a capacitor in modern transistorized ignition systems? Where is it? Is the spark energy still stored in the ignition coil as a current, or in the capacitor as a voltage (like the old capacitor discharge systems with a DC-DC converter)? See Mark Ten B capacitor discharge system schamatic here:
http://www.selectric.org/delta/markten24.jpg

 

[fleem]

Is there still a capacitor in modern transistorized ignition systems? Where is it? Is the spark energy still stored in the ignition coil as a current, or in the capacitor as a voltage (like the old capacitor discharge systems with a DC-DC converter)? See Mark Ten B capacitor discharge system schamatic here:
http://www.selectric.org/delta/markten24.jpg

Yes to all above questions. The new ckts are practically the same as far as the business components go. Its just that the contacts were replaced with a solid state switch. In that ckt you reference, the .01uF on the output is the "condenser" for that coil.

 

[negitron]

Yes to all above questions.

Actually, no. Modern ignition systems don't need caps to store energy for the spark; any caps in the circuit are generally for EMI suppression.

 

[Naty1]

Post 4 and the prior post to this explain the different use of capacitors in modern ignition systems...

 

[fleem]

Actually, no. Modern ignition systems don't need caps to store energy for the spark; any caps in the circuit are generally for EMI suppression.

This simply isn't true. A capacitor can't help but store energy if it ever sees any voltage at all. If a circuit has a cap (unless both leads are tied together!), then there is a cap storing energy. You will be unable to find an ignition ckt that does not have a cap across the coil because all automobile ignition coil ckts, both old and new, have a cap across the coil to lower and unify the resonant frequency for the coil. Without the cap, there would be resonances based on the stray capacitance, which would be too high in freq (exceeding the response of other components and causing EMI) and would spread the energy over different frequencies, unpredictably reducing the peak voltage.

 

[negitron]

If a circuit has a cap (unless both leads are tied together!), then there is a cap storing energy.

Please read more carefully. I said the cap does not store energy for the spark. All the energy for the spark is stored in the B-field of the coil.

 

[fleem]

Please read more carefully. I said the cap does not store energy for the spark. All the energy for the spark is stored in the B-field of the coil.

No, the energy for the spark oscillates between the cap across the coil and the coil. That's how a parallel resonant ckt works.

 

[Bob S]

No, the energy for the spark oscillates between the cap across the coil and the coil. That's how a parallel resonant ckt works.

You are correct in stating that the spark oscillates between the coil and the capacitor. But if you look carefully at the circuit diagram in my post #4*, you will see that the energy is stored in the coil while the switch (breaker points) is closed, and released when the switch opens. This is the basis for all the automotive ignition systems from about 1930 to about 1970. This is less efficient than storing the energy in a capacitor, and triggering a solid state switch. This is the basis for the first CD (capacitor discharge) ignition kits that became available in the mid 1960's.

{Edit] * Sorry, Post #3, not #4.

 

[negitron]

CDI discharges the cap through the coil primary; there is still no significant capacitance across the secondary, regardless of what fleem incorrectly claims.

 

[fleem]

You are correct in stating that the spark oscillates between the coil and the capacitor. But if you look carefully at the circuit diagram in my post #4*, you will see that the energy is stored in the coil while the switch (breaker points) is closed, and released when the switch opens. This is the basis for all the automotive ignition systems from about 1930 to about 1970. This is less efficient than storing the energy in a capacitor, and triggering a solid state switch. This is the basis for the first CD (capacitor discharge) ignition kits that became available in the mid 1960's.

{Edit] * Sorry, Post #3, not #4.

Ah I see now. By "spark energy stored in a capacitor" you actually mean "spark energy stored in the capacitor first instead of second". Thank you for clarifying.

 

[fleem]

CDI discharges the cap through the coil primary; there is still no significant capacitance across the secondary, regardless of what fleem incorrectly claims.

If you take a close look at your quote that I included in the top of my post, you'll see that I was responding to your statement that caps in CDI ckts are strictly for EMI suppression.

That statement is incorrect.

I thought I clearly explained why it is incorrect, but perhaps I didn't. I'll try again with different wording and more detail this time. Please read carefully before responding.

There are stray capacitances in the windings of the coil. The capacitances vary and exist between different windings. Each capacitance and winding is a separate LC ckt and has a separate resonant frequency. If we allow each of those LC ckts to oscillate, the output signal will be a mixture of all those oscillations. The probability that all those oscillations will constructively interfere and produce a strong voltage spike is very low. The solution is to forcibly cause the resonant frequency of the system to be much lower than those separate LC ckts would have, so that all the energy is in phase and you get a strong pulse of voltage. CDI ckts most certainly do have a cap across the coil just as the old flyback ckts did, and for exactly the same purpose.

 

[Bob S]

No, the energy for the spark oscillates between the cap across the coil and the coil. That's how a parallel resonant ckt works.

Please look at my SPICE simulation in post #3. It is a series resonance. I worked on these ignition circuits many times.

 

[fleem]

Please look at my SPICE simulation in post #3. It is a series resonance. I worked on these ignition circuits many times.

Ignoring the limiting resistor (which I'm sure you'll agree is only for current limiting and not so much the point here), that is a parallel circuit, not a series circuit. The battery impedance is zero (theoretically, of course), which means the impedance between the top of the coil and bottom of the cap (except for the limiting resistor), is zero. So the top of the coil is connected to the bottom of the cap, AC-wise. Thus they are in parallel as far as AC goes (after the switch opens and they resonate).

 

[Bob S]

Ignoring the limiting resistor (which I'm sure you'll agree is only for current limiting and not so much the point here), that is a parallel circuit, not a series circuit. The battery impedance is zero (theoretically, of course), which means the impedance between the top of the coil and bottom of the cap (except for the limiting resistor), is zero. So the top of the coil is connected to the bottom of the cap, AC-wise. Thus they are in parallel as far as AC goes.

See example of series resonance in
http://hyperphysics.phy-astr.gsu.edu/HBASE/electric/serres.html

BTW, in some cars, the external series limiting resistor is shorted out when the car is starting.

 

[negitron]

Uh, no. You don't know what you're talking about; that IS a series-resonant circuit that Bob posted.

EDIT: Not you, Bob. I was addressing fleem's prior post. You snuck in while I was posting. Weird that we posted the same reference.

 

[fleem]

Uh, no. You don't know what you're talking about; that IS a series-resonant circuit that Bob posted.

EDIT: Not you, Bob. I was addressing fleem's prior post. You snuck in while I was posting. Weird that we posted the same reference.

Rather than argue whether we should call a "series" circuit with its ends connected (AC-wise), "a parallel circuit" or not, let's back up to the reason bob wants to describe it as a series ckt. He did so as a refute of my statement that the cap is across the inductor--apparently as an argument that the cap is not across the inductor. The cap is most certainly across the inductor (AC-wise) in the ckt bob posted (except for the limiting resistor). So the cap and inductor trade all the energy in that resonant ckt. Call it a series ckt with its ends connected if you like, I normally call an inductor with each of its leads connected (AC-wise) to leads of a capacitor, a parallel ckt, and that is exactly how it should be analyzed--as a parallel ckt. That cap and coil are in parallel AC-wise, which is all that matters in a discussion of AC! They are a "tank" ckt, and if you tell an electronics engineer that a tank ckt isn't parallel, but really its a series ckt that has been shorted out, they'll look at you kinda funny-like, I assure you.

EDIT: Or do you disagree that the impedance between each lead of the inductor to a corresponding lead of the capacitor is not zero (except for the limiting resistor, of course)?

 

[negitron]

The cap is most certainly across the inductor (AC-wise) in the ckt bob posted.

No. It is not. It is series, period.

This talk of series circuits being parallel "AC-wise" is nonsense.

 

[fleem]

No. It is not. It is series, period.

This talk of series circuits being parallel "AC-wise" is nonsense.

Would you mind telling me what you think is the impedance between the top lead of the inductor and the bottom lead of the capacitor (assuming the limiting resistor is 0 ohms)? Its a very easy question.

 

[negitron]

That would be the complex sum of the impedances of the capacitor and the inductor in question. We wold never consider the voltage source to have a zero impedance; that would be stupid.

 

[Bob S]

Would you mind telling me what you think is the impedance between the top lead of the inductor and the bottom lead of the capacitor (assuming the limiting resistor is 0 ohms)? Its a very easy question.

Points closed: Z = R + jwL (R = internal coil primary resistance)

Points open: Z = R + jwL - j/wC

R is roughly 6 ohms for a 12 volt electrical system.

 

[fleem]

Points closed: Z = R + jwL (R = internal coil primary resistance)

Points open: Z = R + jwL - j/wC

R is roughly 6 ohms for a 12 volt electrical system.

You seem to have neglected the impedance of the battery. Hint: its zero.

 

[negitron]

No. It is not.

Ever.

 

[fleem]

No. It is not.

Ever.

No, an (ideal) current source has infinite impedance, a voltage source has zero impedance. Practically speaking the impedance of a 12V car battery, at a couple hundred kilohertz, is in the milliohms.

 

[negitron]

That's internal impedance. It's not considered in circuit analysis when you're trying work out the impedances of various current loops.

 

[Bob S]

You seem to have neglected the impedance of the battery. Hint: its zero.

Please look at the impedance equation for a series resonant circuit in
http://hyperphysics.phy-astr.gsu.edu/HBASE/electric/serres.html
It does not include the impedance of the voltage source on the left side.

 

[negitron]

It does not include the impedance of the voltage source on the left side.

Or anywhere else, for that matter.

 

[fleem]

That's internal impedance. It's not considered in circuit analysis when you're trying work out the impedances of various current loops.

And what is the external impedance of a voltage source, then? Infinity?

 

[negitron]

Well, why don't you draw for me what you think a series AC circuit looks like?

 

[fleem]

Well, why don't you draw for me what you think a series AC circuit looks like?

I'll wait until you give me some proof that the impedance of a voltage source (i.e. a battery) is anything but zero (practically very low). By the way, if you google "impedance" with "voltage source", you might find the answer. HINT: its very low--zero for an ideal voltage source. yet you claim it isn't. Why?

 

[negitron]

When I said it was not zero, I meant it was not zero in terms of the way the circuit is connecting when analyzing a circuit loop and determining if it is a series- or parallel-resonant circuit. Your claim that the top of the inductor is connected to the bottom of the capacitor through this so-called zero impedance is laughable.

Batteries and other voltage sources do have an internal impedance (which is ideally zero, but never really is; and sometimes it's not even low), and this fact is important when attempting to transfer maximum power from a voltage source to a load, but it is not considered as a part of such an analysis as this.

 

[fleem]

Well, why don't you draw for me what you think a series AC circuit looks like?

Why should we argue about the definition of "series" and "parallel" before we even agree on whether the impedance of that battery in your circuit is zero or infinity? Or are you trying to avoid the fact that a simple google search shows the impedance of a battery as nothing but very low (ideally zero), rather than high, since that is the question that must be answered in order to even decide whether we want to call that circuit series or parallel?

 

[Bob S]

Why should we argue about the definition of "series" and "parallel" before we even agree on whether the impedance of that battery in your circuit is zero or infinity? Or are you trying to avoid the fact that a simple google search shows the impedance of a battery as nothing but very low (ideally zero), rather than high, since that is the question that must be answered in order to even decide whether we want to call that circuit series or parallel?

Hi Fleem-
Why don't you post a sketch of what the ignition circuit looks like, and run it with SPICE like I did in post #3. or I will run it in SPICE myself. Include the breaker points in the circuit.
Bob S

 

[negitron]

since that is the question that must be answered in order to even decide whether we want to call that circuit series or parallel?

No, it doesn't. There's nothing to argue about concerning the definitions of "series" and "parallel" as they are rigorously defined in the link that both Bob and I provided for you.

 

[fleem]

When I said it was not zero, I meant it was not zero in terms of the way the circuit is connecting when analyzing a circuit loop and determining if it is a series- or parallel-resonant circuit. Your claim that the top of the inductor is connected to the bottom of the capacitor through this so-called zero impedance is laughable.

Batteries and other voltage sources do have an internal impedance (which is ideally zero, but never really is; and sometimes it's not even low), and this fact is important when attempting to transfer maximum power from a voltage source to a load, but it is not considered as a part of such an analysis as this.

Guys, the fact that you are in agreement that the battery is a high impedance in an analysis of that circuit, is giving you both a confidence that is going to result in some notable embarrassment when you realize that it really is a short at any non-zero frequency. Specifically, the battery "connects" any voltage changes on one electrode to the other, just like a wire but for alternating voltages. I suggest you both offline think some more on exactly how you would calculate the waveform on those components, and ask yourself where the AC loops are in that circuit (when the switch is open). I've got to move on now, but my silence only indicates I've given up, not that I concede that batteries are like opens at non-zero frequencies.