Model Engine Ignitions

[74Sprint]

Today I'm going to fix my power supply problem by converting an ATX computer power supply. As I mentioned earlier my old faithful power supply I made when I was a private TQ4 died. I wouldn't mind fixing it for old time sake but, I can build something bigger and better today. I have a bunch of power supplies from wrecked/gutted computers so I might as well use them. The ATX power supply has all the voltages that I need (3.3, 5, 12) with plenty of current. My current little power supply that I have been using can't putout enough current for the coils and even though it says it can it can't and I'm getting voltage droop or over current shut downs, it's a ZK-J3X.

It's nice for low power stuff and it has constant current and constant voltage adjustments, besides other little features.

To build the ATX bench power supply I'm going to follow a build off one of the channels I follow. The fellow that runs the dronebot workshop has some pretty good tutorials and stuff, actually a lot of stuff. Everything from simple circuits to programming micro-controllers. I have the meters needed to complete the build on order but, I can still work with the ATX power supply with out them. The ATX power supply can take in either 120 or 220 volt and has a 550 watt capacity which should good for most of my builds. I also have 350 watt supplies which, would work but, what the heck go big or go home. I'll let you know how it turns out.

The links:
ATX-bench-supply html
ATX-bench-supply PDF
ATX-bench-supply YouTube

Ray

 

[74Sprint]

Well lots to talk about today.
First I want to say I broke some golden rules when doing R&D.
1. Don't work on things if your tired.
2. Don't let the grandkids bother you while working (Distractions.)
3. Don't be in a rush.
4. Do the math first.

So on Friday I received a CDI ignition and coil from China like the ones on the model-engine-cdi-easy-and-cheap thread. Problem is the CGI box they sent me is a 5 pin AC instead of a 4 pin DC. I'm still going to try it later with DC when the other coils arrive. I also found a FDP8433 MOSFET, and after checking the specs decided to use it. It is a N-Channel PowerTrench MOSFET 40V, 80A, 3.5mΩ with gate voltage from 4.5v - 10v pulse capable of 1,500 amps. So I would like to say that my circuit works but, with a few small bugs.

So being in a rush and the grandkids wanting to see what I'm doing I forgot to change the dwell capacitor back from a 1uf to a 0.1uf. So the dwell pulse was 10X wider than it should have been. I used the 1uf cap so it would be easier for my meter to show peak values. I measured the resistance and inductance of the coil fields and decided that 1.2k primary should be fine with the +12volt 15amp supply ATX power supply. Ding ding ding, wrong, it should have 10ma but, remember what I said about time constants? In the first time constant the primary coil will look like a dead short almost. This is called current inrush. Voltage is instantaneous and current will fall to 10ma once the coil is charged. So what happened? Well the coil fired 3 times but, then out of the corner of my eye I saw a spark where it should not have been, the coil quit sparking after that. No magic smoke but I also noticed that the ATX power supply was shut down. So started checking everything and I found that the secondary coil was shorted to ground and the case had burst.

Well that sucked, brand new coil only had it for 1 day & 3 sparks, damn R&D.
As I mentioned inductors will draw as much current as they can get. On GM/EDM locomotives use 2 48 volt starters along with 8 - 8 volt batteries with over 2,000 amps of true cranking power. There is a warning sticker by the start station that warns you not to crank the engine for more than 20 seconds and then wait about 15 minutes before trying again. Well some people are impatient and crank it for too long and the starter motors catch on fire. You would think they would put some sort of device to keep the starter motors from catching on fire.

Anyway, I still have my Accel coils and a new GY6 coil coming. Once I get the rest of my power supply parts I'm going to put current limiting circuits in it for all voltages.
The nice part is, the circuit works just as I planned well very close, I had to add a 220 ohm resistor to drive the MOSFET properly.

Ray

 

[74Sprint]

Ok, the next thing that is bugging me is the Hall-Effect/s that people are using and how they use them. I'm not going to go into great detail about how Hall-Effects work, you can do that yourself. Oh, only use a switching Hall-effect, unless your really big into electronics.

When working with Hall-effects there are 4 things you need to pay attention too.
1. working voltage.
2. The Bo & Bp points, sometimes listed as (Bops & Brps.
3. The gauss required.
4. Current operating limit.

1. Make sure that the working voltage (max voltage) is well above the voltage you intend to use. And the lower turn on voltage is 2 volts below your power supply/Battery voltage. Simple right!

2. The Bp & Bo points are the points that the Hall-effect will operate at. In other words the Bo point is the value in gauss that the Hall-effect will turn on. The Bp is the value in gauss that the Hall-effect will turn off. These are import because they tell you the minimum strength of the magnet you need to use. They can also if you have the right equipment to tell you the minimum distance between the magnet and the Hall-effect. The distance between the magnet and the Hall-effect and the strength of the magnet will affect your ignition timing. Lets say the you line up the Hall-effect and magnet at say 8 degrees BTDC and you used the LED to do this and then you ran the engine. So you check the timing with a light but now you see that the timing is at 12 degrees, what the heck! What happened?

Well that LED is more of a trigger indication than a timing light. Now you and me included when we eyeball something we are not very precise. If you look at the Hall-effect datasheet you will see just how fast they are and precise. If the magnet is too strong our the Bo too low then the Hall-effect will turn on before the magnet even gets over the Hall-effect. To correct this you will need to move either the magnet or Hall-effect further way from each other. In some cases this can be over an inch/25mm. In some circuits the hysteresis or the difference between Bo & Bp points are used as the dwell and can be too much or too little, usually with a pnp transistor. Usually Hall-effects when they turn on the signal out goes to ground. So to set this up properly you would have to move the components away from each other or change the magnet and/or Hall-effect or use a different circuit. Of course you can also use to weak of magnet or too high of Bo gauss and it just won't work.

3. So how much gauss do I need? Good question. I'll start by saying that every Hall-effect datasheet I looked at has the test conditions listed. But this still doesn't say what size of magnet and the distance. The places I buy my magnets from always have the strength in Tesla's listed which, is a good start. Magnets aren't listed in gauss instead they use Tesla's. There is a way to help with selecting a magnet and Hall-effect by using an online calculator but, you will need all the specs of the magnet you are going to use beforehand. It is on the Infineon web site Magnet to Strength Calculator. The conversion between gauss and tesla's is an easy one: 1 tesla = 10,000 gauss or (1mT=10 Gauss).
The effects of the magnet can also be changed depending on if the magnet is imbedded in iron/steel or aluminum, or just glued to the surface. One alternative is to use a steel blocker with windows that rotates between the magnet and Hall-effect. Other than this the only thing I can say is don't make anything permanent and just do trial & error. So using rare-earth magnets usually are not the best choice, use cheaper magnets instead if possible.

4. The current operating limit is important and depends on how your going to use the Hall-effect in a circuit. First off don't try and use the Hall-effect to drive the coil directly, you will usually make smoke. Most of the Hall-effects I use are good for 20ma and are used in digital style circuits, high/low stuff. Even trying to do what I did with my first transistor. The CD4047 couldn't put out enough current to turn on the transistor, well it was just barely turning on. So match the Hall-effect current rating to the base/gate current of the transistor you want to use.

I did a write-up somewhere on this forum on different ways one can use a Hall-effect, check it out. I have a bunch of different Hall-effects I need to check out.

Cheers
Ray

 

[willray]

remember what I said about time constants? In the first time constant the primary coil will look like a dead short almost. This is called current inrush. Voltage is instantaneous and current will fall to 10ma once the coil is charged.

Would you mind explaining exactly what you mean here? I've seen more than a few people mention "inrush current" with inductors (coils), but inductors display the antithesis of inrush current.

Inductors resist rapid changes in current, generating EMF that "tries" to keep di/dt small. They don't look at all like dead shorts when initially connected to voltage. Instead they look a lot like open circuits. It's capacitors that look like dead shorts.

All that being said, you seem to know your way around electronics, so I have to assume that you're talking about a real physical phenomenon that is unrelated to simple coiled-conductor inductance, but I'm not sufficiently familiar with the actual physical construction of ignition coils to know what that might be. Core permeability/saturation issues? Something completely different? If you wouldn't mind expanding on your explanation, I'd appreciate a better understanding of what's actually going on.

Will Ray

 

[74Sprint]

Would you mind explaining exactly what you mean here? I've seen more than a few people mention "inrush current" with inductors (coils), but inductors display the antithesis of inrush current.

Inductors resist rapid changes in current, generating EMF that "tries" to keep di/dt small. They don't look at all like dead shorts when initially connected to voltage. Instead they look a lot like open circuits. It's capacitors that look like dead shorts.

All that being said, you seem to know your way around electronics, so I have to assume that you're talking about a real physical phenomenon that is unrelated to simple coiled-conductor inductance, but I'm not sufficiently familiar with the actual physical construction of ignition coils to know what that might be. Core permeability/saturation issues? Something completely different? If you wouldn't mind expanding on your explanation, I'd appreciate a better understanding of what's actually going on.

Will Ray

Sure, but I must first apologize because I have a bad habit of jumping ahead and thinking that others are following. Your thinking of the di/dt is correct for a simple coil. Well to start the formula for time constants for inductors is for open air inductors and not for cored ones. The type and construction of core material can change the characteristics of how the coil reacts. Also by now you would think that they would have a global standard when calculating inductance. Some use 50hz, some use 100hz, and some even use 1khz or more. But the good manufactures will state the frequency used to give you a better picture. Anyway, the core is there to concentrate the flux and improve the flux of the coil. I'm building this up, so stay with me. Next the we have a primary coil wound over the core and the core strengths it's field. We then have a secondary winding on top of that which should be in phase as the magnetic field builds outward. Now if you remember Lenz’s law 'the change in current changes flux, inducing an EMF opposing the change'. So you can't have back EMF until you have a change in current first and this includes charging up a coil. Next, the greatest change in current flow happens in the first time constant 63.2% .

Side Note: Some electronic devices are sensitive to the rate of change, especially those huge capacitors for battery backups. They can be damaged if the force (joules) of change is too great for them to handle.

So going by Lenz’s law you would think that the back EMF would slow down the current change the most in the first time constant but it doesn't because the magnetic field is lagging the current change and IMHO it is because of the iron core. I have seen air cores work exactly the way Lenz’s law states. If you look at a series wound DC motor as compared to an AC motor, the DC motor has a lot more torque at start than an AC motor has. If you take Lenz’s law at face value then the DC motor should work just like an AC motor but, it doesn't. If we throw in Faraday’s Law you would have to say that this back EMF is instantaneous but, this starts to become a chicken & egg thing. If we are talking about just DC current then yes the greatest total amount of current flow will be in TC6, we only use 5 though. But in TC5 the current change is very little and the back EMF is at it's greatest because the magnetic field is greatest. Yes I know they say the back EMF is based on change but, that big magnetic field wants to collapse and it is the current holding it out there, just like a solenoid. But the 2 are fighting each other. So what you may have been taught is most likely correct. I really hope I'm not confusing you. But I'm talking about the rate of change in TC1 in a DC situation.

So what happened to my ignition coil? Remember what I said about iron cores and how the ignition coil was wound? Okay, ignition coils can fire twice, once when charging and once when discharging. See European Patent EP1298320A2 below a major screw up by STMicro by patenting it and charging licensing fees. Because the back EMF follows the current flow and not the other way around and when using DC, an inductor will take as much current as it can get. I basically overloaded the insulation between the windings because as the coil was discharging when I hit it with current again. So instead of 12 volts across the primary coil there was more and the magnetic field built up higher than the secondary could handle and shorted out, blowing the coil. In fact some coils used in racing are rated for their max output voltage because their insulation breaks down at that voltage. I have a Crane racing CDI coil with a max 65,000 volts output which is ok with a MSD 490 volt CDI but, not an MSD 8 CDI with 590 volt primary voltage. Also if I had used a 12 volt car battery the coil may have survived because batteries have internal resistance which would have slowed the charge up.

In rush current is mostly related to capacitors charging than inductors charging but it is that first time constant that is in-rush current. The most extreme example of in-rush current I have seen is when we use to start diesel locomotives with a DC welder wired up to 550 volts and 76 volts out. When you hit the starter those 1" inch cables literally jump about 6" off the floor and back down. Now according what you and I were taught those cables should not have jumped until full current is built up but, that's were the formula's & laws fail. According to the laws current should start off low and gradually build up which, is correct. But If you go by the graphs and calculations it is correct to say max current flow in an inductor and max voltage in a capacitor occurs at the end of TC5 but, the biggest change of both current and voltage occurs in TC1, the in-rush.

I have argued with electrical and electronic engineers about the effect of turning the current back on as a coil is discharging and the extra push of current but, they say because of Lenz's & Faraday's laws it's impossible. The effect of back EMF would be pushing current in the opposite direction, guess they never pulled a plug out of a wall socket. We are still discovering things today about things we thought we knew. Who knows maybe one day I'll prove my theory. Having worked with RADAR I learned a lot about resonance and impedance.

Cheers
Ray

P.S. AC locomotives that I have worked on get the AC motors to give more starting torque(punch), at least the ones I worked on by clipping the top of the AC signal and using that to give the AC motors more kick, it gets hammered. It's like a dimmer switch but used backwards. On DC locomotives they start off in series-parallel until the back EMF gets to high, they then make what is called forward transition and going full parallel lowering the overall resistance(Xlr) of the traction motors which lowers the back EMF and they can go faster. There is no need for AC locomotives to make transition.

 

[Steamchick]

Wow! - I'm boggled! - I manage simple coil and points ignition, but reading this maybe I'll stick to steam engines? ....
I thought I understood some of it - like WillRay's query:
"Would you mind explaining exactly what you mean here? I've seen more than a few people mention "inrush current" with inductors (coils), but inductors display the antithesis of inrush current.

Inductors resist rapid changes in current, generating EMF that "tries" to keep di/dt small. They don't look at all like dead shorts when initially connected to voltage. Instead they look a lot like open circuits. It's capacitors that look like dead shorts."

But Ray, You sound like my revered hero, Mr. Tesla! Except I can't really follow your explanations... (I have no experience or knowledge of these things - so who am I to judge?).
But sounds interesting, even if I don't understand the words...
K2

 

[willray]

Sure, but I must first apologize because I have a bad habit of jumping ahead and thinking that others are following.

Thanks Ray,

It'll take me some time to parse and digest your response, but I want to really thank you for not taking the question the wrong way.

Professionally I am occasionally a physicist, so I'm familiar with the definitions and behaviors of the idealized components (though my physics is rather distant from this, so I do have to work for it). However, I'm woefully undereducated regarding both the typical physical implementation of things like ignition coils, and of the practical application of the idealized models to their less-ideal physical counterparts. Couple this with domain-specific languages that repurpose words such that they have subtly different meanings than their fundamental physics definitions, and I'm often uncertain where something lies on the spectrum between confusing-to-me and wrong :-/

You're the first person I've seen mention inrush with inductors, who didn't seem to either be parroting something they read on the interwebs, or making the mistake of thinking that inductors "obviously" have inrush because induction motors or transformers are affected. I really appreciate you taking the time to try to explain.

As an initial question, and demonstration of my ignorance - it surprises me that ignition coils would be made with susceptible cores. I would have assumed that a susceptible core would work against the rapid collapse of the field that is desirable for creating the high EMF in the secondary "when the points open". Is this choice a balancing act between stored energy and rate of field collapse, or, am I misthinking something?

Thanks again,
Will

 

[Steamchick]

Hmmm. More interesting comment... I understood the bit about susceptiblec core... rapid collapse of the field....
So am I right in thinking you are effectively tuning the resonance of the primary to be very high frequency, to maximise dI/dt in the primary? And consequently the voltage developed in the secondary? And part of your explanation is that larger inductance leads to lower resonant frequencies....?...
I wonder if using the automotive "coil on plug" would give you a suitable coil .... if you could design the right trigger? There must be thousands of good parts in scrap yards awaiting recycling..... so cheap-to-buy compared to your "3 sparker"?
Or maybe you should ignore my comments as simply wrong? - feedback please... I won't spoil you thread if I am way off understanding this subject...
Thanks
K2

 

[74Sprint]

Wow! - I'm boggled! - I manage simple coil and points ignition, but reading this maybe I'll stick to steam engines? ....
I thought I understood some of it - like WillRay's query:
"Would you mind explaining exactly what you mean here? I've seen more than a few people mention "inrush current" with inductors (coils), but inductors display the antithesis of inrush current.

Inductors resist rapid changes in current, generating EMF that "tries" to keep di/dt small. They don't look at all like dead shorts when initially connected to voltage. Instead they look a lot like open circuits. It's capacitors that look like dead shorts."

But Ray, You sound like my revered hero, Mr. Tesla! Except I can't really follow your explanations... (I have no experience or knowledge of these things - so who am I to judge?).
But sounds interesting, even if I don't understand the words...
K2

LOL, don't worry in electronics engineering school there was quite a few people that had a hard time getting their heads around it.
What you see below is what we were all taught. It shows the charge curve in Blue and the voltage drop across the inductor in Purple. As you can see in the first time constant the current reaches 63% of max flow. Don't worry about the voltage drop for now. Remember this is DC here going from zero to max. And we are working with the change not the max current.

Next, transformers can be wound in many different ways but, for ignition coils they generally use the auto transformer style. This is where one winding is on top of the other and with one end of each coil is tied together and keeping them in phase. Now the voltage drop shown in Purple is basically your back EMF voltage. Remember this is DC so we don't really have to worry about inductive impedance for now. Note: the Purple line can also represent the coils discharge. Next, because the ignition coil is wound auto transformer style anything that happens on the primary also happens on the secondary and in phase. But in simple terms because there is a step-up ratio the current gets converted to voltage, in a step-down it's the voltage that converts to current. Well not all of it but most of it. So in TC1 you get a huge voltage rise/spike in the secondary coil. The current flowing in the secondary is rising at the same rate as the primary and it also sees back EMF because of the change in current flow. Now if the winding ratio is say 100:1 then according to the laws that back EMF is going to be 100 times higher than in the primary. So now we have an inductor that will take as much current as it can get, a power supply with lots of unhindered current, and a 100:1 winding ratio, the back EMF in the secondary got high enough to over come the insulation and short out my ignition coil.

In-rush current, that current flow in TC1, does look like a short to current because TC1 has the greatest change. Every TC after that has less change, just like if someone was increasing the resistance and slowing down the change. Remember that Purple line represents back EMF so when the current reaches a steady state there is no more back EMF. But we still have that magnet field that wants to collapse back onto the primary coil. So lets do a collapse and make a spark. When the magnetic lines of flux collapses back down and cuts through the secondary coil it induces current flow in the secondary which gets turned into a high voltage spike/spark. But that magnetic field also wants to go back to where it was created and that creates a voltage spike in the primary, this is where the ringing effect comes from and shows up on the secondary and can be seen with an oscilloscope.

Let's look at the discharge side of things and we will use points. Current flowing is just like water and according to Newton an object in motion wants to remain in motion. So both current and water flowing is happy until it hits a wall or open circuit point, it piles up and the pressure (voltage) rises up. In the case of current that pile up can raise the voltage so high that it can jump across an opening (points gap). That spark/arc is just like plasma and takes some material with it when it jumps across. To temporarily catch that spark with use a capacitor which is like a short for the current flow, when the points close again that energy is returned to the circuit. But wait, what if we don't use a cap and let the spark happen. Well the high voltage spark is going straight to the primary coil, so now instead of the primary coil being a ground/negative potential it is now at maybe -1000 volts on the negative side and +12 on the other side. Now turn the current back on at just that moment what do you think will happen with current flow instead of having a potential of 0 to +12 volts you now have -1000 to +12 volts. It took me 9 months to figure how to make that one work. Oh that in-rush current can and will blow MOSFETs and IGBTs if the rate of change is to high, some manufacturers list max rate of change.

Cheers
Ray

 

[74Sprint]

Thanks Ray,

It'll take me some time to parse and digest your response, but I want to really thank you for not taking the question the wrong way.

Professionally I am occasionally a physicist, so I'm familiar with the definitions and behaviors of the idealized components (though my physics is rather distant from this, so I do have to work for it). However, I'm woefully undereducated regarding both the typical physical implementation of things like ignition coils, and of the practical application of the idealized models to their less-ideal physical counterparts. Couple this with domain-specific languages that repurpose words such that they have subtly different meanings than their fundamental physics definitions, and I'm often uncertain where something lies on the spectrum between confusing-to-me and wrong :-/

You're the first person I've seen mention inrush with inductors, who didn't seem to either be parroting something they read on the interwebs, or making the mistake of thinking that inductors "obviously" have inrush because induction motors or transformers are affected. I really appreciate you taking the time to try to explain.

As an initial question, and demonstration of my ignorance - it surprises me that ignition coils would be made with susceptible cores. I would have assumed that a susceptible core would work against the rapid collapse of the field that is desirable for creating the high EMF in the secondary "when the points open". Is this choice a balancing act between stored energy and rate of field collapse, or, am I misthinking something?

Thanks again,
Will

Your welcome, no your not miss thinking something, your right it is a balancing act. The best way to think of an ignition coil is to think of them as transformers and not inductors even though they are technically inductors. Everything is a cause and effect so which comes first, what creates the other. I studied deflagration in IC engines, from spark to exhaust and when a cylinder fires the air/fuel mixture doesn't explode right away, it burns first and progresses to an explosion. It doesn't explode first and the burn. Remember that laminate core is there for the primary because it is closest to it so you design for the primary. Also the fields expand past the secondary coil and the core becomes magnetized helping the primary. But when the field collapses it doesn't need any help because it wants to resume it natural state. The only thing that could hinder the collapse would be core that is to large. The remaining magnetic flux would slow down the collapse. Ignition coils still behave according to the laws of physics just slightly modified. Ignition coils still are subject to resonance & impedance frequencies. I find it is the way the 2 coils are coupled that is the main standout from other transformers. In the old days magneto high tension coils were called transformers. Magnetos only create current and no voltage, the magnetos rely on the resistance of the wire in the primary to create the voltage. Not too low and not too high of resistance.

Most of what I talk about comes from experience. 40 years ago when I was in engineering school some of our instructors when they couldn't explain something would say "shut up and take it for what it's worth and besides it works by FM" in other words just except it and it works by F..king Magic. Of course today we understand more about what is going on because of quantum mechanics. Oh I still keep my formula books handy because my mind isn't as good as it used to be. LOL Retired and Retarded they say.

Do you understand what I mean by the greatest change that is allowed looking like a short in TC1 for an inductor? Because of my work with CDI ignitions I can say that capacitors do work like the formulae. But still with caps the current rises very quickly and the then tappers off as the cap charges. The current isn't at max right away it too has a ramp up just that it is done very very quickly. So where did I get all this? Well from RADAR circuits. With RADAR circuits everything has to be taken in to consideration. A run on a circuit board can very quickly turn into an antennae or an inductor, solder pads can become capacitors, board edge connectors are really bad for that. RASARs are almost always pulsed they can go from 0Hz to GHz and 0 watts to Megawatts in a nano-second. You want to talk about weird Sh_t try RADAR.

Oh remember if the current is not changing in a transformer then nothing is happening except current flowing through the primary. I have a question for you "how come physicists don't understand how radio waves travel through space? after all radio waves will not cross a perfect vacuum"

Cheers
Ray

 

[74Sprint]

Hmmm. More interesting comment... I understood the bit about susceptiblec core... rapid collapse of the field....
So am I right in thinking you are effectively tuning the resonance of the primary to be very high frequency, to maximise dI/dt in the primary? And consequently the voltage developed in the secondary?

I tune for resonance of the primary only in my 48 amp ignition. It works by first opening a window so to say, by window I mean that I allow the resonant frequency of the primary through but only for a set time. It works like a tank circuit that gets big time amplified. Also as the RPM goes up the window gets smaller, so at idle it may spark 6 times and at 6,000 RPM only 4 times and each time the spark gets boosted higher and higher per sequence. I have seen the spark go over 100Kv but not by much. Usually the sparkplug fires before that and levels out around 80Kv or it comes out the side of 8.8mm sparkplug wire to ground somewhere. You can't maximize the di/dt in the primary, that value is fixed when the coil is made. The voltage in the secondary or the quality there of is tied to the ignition transformer design and after that it follows what happens in the primary. What I'm actually doing is taking and using what we normally are trying to get rid of by placing a varistor or MOV, same thing or even a diode on a relay, MOSFET, IGBT to clamp the back EMF which can cause a relay or solenoid to chatter and MOFETs or IGBTs to get punch through with an inductive load.

And part of your explanation is that larger inductance leads to lower resonant frequencies....?...

True that is correct but, bare in mind that the core if there is one and it's effectiveness will modify the resonant frequency.

I wonder if using the automotive "coil on plug" would give you a suitable coil .... if you could design the right trigger? There must be thousands of good parts in scrap yards awaiting recycling..... so cheap-to-buy compared to your "3 sparker"?
Or maybe you should ignore my comments as simply wrong? - feedback please... I won't spoil you thread if I am way off understanding this subject...
Thanks
K2

Have to remember that my China coil blew is mainly I had way to much dwell and just happened to get the timing just right to wack it, my bad luck. Yup I'm going to the auto-wrecker this weekend to get some coils Ford, GM, and Mopar. Either COP or CNP coils. Also GM stand alone coils that were used in the late 80's and early 90's before they went to coil near plug style. These coils have a very high output and are very fussy about the cap and rotor material. A lot of mechanics get fooled by them breaking down and getting spark scatter. No your not far off, also I don't mind the discussion, isn't that what this forum is for. Besides I'll be asking a lot of questions when I finally start making chips.

Cheers
Ray

 

[Steamchick]

Thanks Ray, This is education for me:
I know the bit about "deflagration in IC engines, from spark to exhaust and when a cylinder fires the air/fuel mixture doesn't explode right away, it burns first and progresses to an explosion. ( I thought this was the "knocking" pre-ignition condition...). It doesn't explode first and then burn." - as I read a paper on it very recently! = Not what I was taught 5 decades ago - that the spark ignites the fuel air mix with a small explosion (The H and O ions first) as it ionises the molecular mix... followed by a flame front (of HC, C and O2 and N2 molecules) that accelerates to sonic velocity (faster than the pressure rises) until the piston has travelled far enough down the cylinder so the expansion extracts enough energy, or the flame front meets cool gases near metal surfaces, so that the flame front is cooled so the combustion stops (CO stops burning in atmospheric air around 320deg.C. but cylinder pressures are much higher, so combustion "stops" at a different temperature), OR the exhaust valve opens, when further expansion down the exhaust pipe stops the combustion before all the fuel has burned properly.
I did understand that the spark energy (actually voltage?) needs to be adequate to ionise the gas at "compression" pressure in order to "strike" the arc, that then needs enough current to ionise adequate gas to the "molecular temperature" required so the ignition of the gas occurs. I.E. "weak" sparks won't strike the arc, or ignite the fuel-air mix. - Hence, the coil needs to have a secondary winding with very rapid rise of voltage (due to engine speed?), of a high enough voltage and energy so the spark strikes at the plug contacts and stays long enough to ensure correct ignition for a self-sustaining burn of the mixture. (All happening in micro-seconds).
What I don't really understand is how to trigger the coil primary, nor what sort of impedance is used, to store adequate energy in the "coil charging" phase (contact-breaker dwell?) and release it during the arcing phase (coil discharge). I seems to be a balance between a low impedance for a short "half-life" during charging and discharging, but a high enough impedance to store enough energy for an adequate spark energy to ignite the mixture at all conditions (avoiding mis-fires at higher revs or poor cold-starting, etc.). - A different engineer sorted "ignition" at work. I just had the joy of determining the best commercially practical HT cables... and I gained some small understanding of Radio suppression in the process. (Resistive or inductive HT cable systems).
Thanks for the education,
K2

 

[Steamchick]

Ray: I was taught (in the 1980s) that the coil heating is due to the combined heat from the inrush current (RMS of I-squared-R in the primary during t of contact dwell, plus zero-current during t of contacts open?), plus the "out-rush" current (I-squared-R in the secondary during arc-duration at the spark-plug - this t is very small!). This meant that inductive HT cables (that I selected as most appropriate for our application) actually should have heated the coil more than the (cheaper) resistive HT cables (that the coils had been designed for) but on real testing the temperature increase wasn't measurable. (secondary winding discharge current is in micro-amps for micro-seconds duration). But maybe my memory is in error? - For those that don't know but are interested, but inductive and resistive secondary cables, plug-caps and plugs reduce the peak current of the spark, hence the peak amplitude of the radio frequency noise produced. The voltage of the spark is determined by the breakdown voltage at the spark-plug = An engine parameter, not determined by coils, etc. After the spark strikes, the arc current rises very fast (voltage across the spark-plug contacts falls as the ionised gas resistance drops rapidly). I'm guessing: but I think that how quickly the voltage achieved (just before the arc is generated) determines the maximum frequency of the radio noise (is it dV/dt? = rate of change of voltage?), whereas the arc current (or dI/dt?) determines the "amount" of radio noise emitted. Thus resistors and inductors in the extra bits after the secondary coil help to limit that arc current (reduce dV/dt - hence max I is reduced?) - therefore suppressing radio noise. The HT cables become the antenna for broadcasting the radio noise. (Any radio engineers want to correct this? - or add to the explanation? - It was 40-plus years ago when I was taught this information... but I forget the maths!).
Not machining, but related to "spark ignition engines"...
Cheers!
K2.

 

[74Sprint]

Thanks Ray, This is education for me:
I know the bit about "deflagration in IC engines, from spark to exhaust and when a cylinder fires the air/fuel mixture doesn't explode right away, it burns first and progresses to an explosion. ( I thought this was the "knocking" pre-ignition condition...). It doesn't explode first and then burn." - as I read a paper on it very recently! = Not what I was taught 5 decades ago - that the spark ignites the fuel air mix with a small explosion (The H and O ions first) as it ionises the molecular mix... followed by a flame front (of HC, C and O2 and N2 molecules) that accelerates to sonic velocity (faster than the pressure rises) until the piston has travelled far enough down the cylinder so the expansion extracts enough energy, or the flame front meets cool gases near metal surfaces, so that the flame front is cooled so the combustion stops (CO stops burning in atmospheric air around 320deg.C. but cylinder pressures are much higher, so combustion "stops" at a different temperature), OR the exhaust valve opens, when further expansion down the exhaust pipe stops the combustion before all the fuel has burned properly.
I did understand that the spark energy (actually voltage?) needs to be adequate to ionise the gas at "compression" pressure in order to "strike" the arc, that then needs enough current to ionise adequate gas to the "molecular temperature" required so the ignition of the gas occurs. I.E. "weak" sparks won't strike the arc, or ignite the fuel-air mix. - Hence, the coil needs to have a secondary winding with very rapid rise of voltage (due to engine speed?), of a high enough voltage and energy so the spark strikes at the plug contacts and stays long enough to ensure correct ignition for a self-sustaining burn of the mixture. (All happening in micro-seconds).
What I don't really understand is how to trigger the coil primary, nor what sort of impedance is used, to store adequate energy in the "coil charging" phase (contact-breaker dwell?) and release it during the arcing phase (coil discharge). I seems to be a balance between a low impedance for a short "half-life" during charging and discharging, but a high enough impedance to store enough energy for an adequate spark energy to ignite the mixture at all conditions (avoiding mis-fires at higher revs or poor cold-starting, etc.). - A different engineer sorted "ignition" at work. I just had the joy of determining the best commercially practical HT cables... and I gained some small understanding of Radio suppression in the process. (Resistive or inductive HT cable systems).
Thanks for the education,
K2

Wow you really know your stuff. One needs to remember that there is no such thing as an explosion right away, at least humanity hasn't found a material that does. I have an old paper somewhere that explains the process of an atomic bomb detonation and even that goes through the deflagration process.

I have found a lot of people miss use terms and even myth becomes truth.
For example:
Knock: This is caused by too early of ignition, timing too advanced. What is happening hear is the piston trying to turn the engine backwards. The knock sound is all the oil being hammered out of the clearances such as the rod to crank and crank to block. IMHO this is also what causes main bearing cap walk. I usually see this on engines where a person is trying to get every last bit of as they can out of an engine, that's nuts and big money. But still no explosion.

Pre-ignition: I like this one because people often have it trade places with knock. Pre-ignition only occurs if there is something approaching or is at a temperature that will ignite a air/fuel mixture. This can occur when taking a granny driven car that is all carboned up for a sustained high speed run or using fuels like nitromethane, I've done both. The sparkplugs used in nitro burning automotive engines use some very tough sparkplugs. They are made from harden steel and fancy ceramics. For example setting the gap on these plugs is crazy, you almost need to use some special pliers to open the gap, a simple sparkplug gauge won't do it, at least I can't. But these nice new plugs turn colours and if the tips and ground electrode are burned away a bit then you might have too much timing. If the centre tip is burned away, the ground electrode is half missing, and ceramic is mostly gone then you have too much timing and pre-ignition is occurring. On my racing snowmobile I knew when I was getting pre-ignition because I would get a snapping sound out of the exhaust and if I didn't cut the throttle right away the next thing to happen was a crankcase explosion with a vaporized hole in a piston. Pre-ignition depending on how it occurs in the combustion chamber can have 2 wave fronts that meet and it will sound like either a dull knock under throttle or a rattle when you shut off the engine, this part is known as run-on.

Ping: In all my years of working with IC engines I've only heard this once and yes it sounds just like 'Ping' whenever I gave the car some throttle. It was on my car and I had the wrong plugs, Ford 302.

You got deflagration and spark ignition IMHO almost right, there are a few parts missing. In drag racing at least at divisional & National events after you make a run down the track you have to either right there or on the return road stop and have your car tested. They basically do just weight and fuel. When they check the fuel in your tank they want to know where you got it from first, then they check the colour, check for additives, and then and this is the important part, is the dielectric strength. This will change if there is too much water or additives. Air and fuel both have an electrical resistance. Air is an excellent insulator, we use it on high voltage power lines for example. All hydrocarbon fuels that I'm aware of have an electrical resistance also. Thing is when you combine them and squeeze them their resistance value goes even higher making it more difficult for a spark to go through. One last thing about the air/fuel mix is during deflagration the pressure wave as it travels across the combustion chamber and squeezing the unburnt mix is burning more and more mix increasing the pressure of the wave. This pressure is compressing the unburnt fuel mix so much that when it burns it burns even faster. The pressure increase continues until the mixture does what is considered an explosion. We want that explosion to occur just after the piston passes TDC for max power. Each and every fuel and engine behaves differently. For example with my 455 CID 11:1 Olds engine I can run 36 Degree total timing with 87 octane pump gas but, I can run 38-40 degrees with 100 octane fuel. On my small Chevy with 13.1:1 compression you take 10 degrees off those numbers. On a nitromethane burning engine total timing can be anywhere from 45-60 degrees total timing, usually around 60. In racing we want a fuel that is easy to ignite but, not too easy and we want it to expand a lot and keep the pressure up as long as possible as the piston moves down. Max pressure usually occurs from -5 to -45 degrees, after 90 degrees the piston is moving down, increasing cylinder volume very rapidly and yup or explosion turns back into just a burn. My new roller turbo cam for the 455/496 the exhaust opens at 63.5 degrees BBDC, it's going to be loud without the turbos.

Yup you want a secondary to develop a high voltage spark as fast as possible to ionize the mixture but, you also want a magnetic field big enough so when the secondary is shorted out from the ionization you get a long lasting burn, almost impossible to have both. 50 years! It's been 45 for me. I was at first against going to HEI but, I'm a convert. I remember when the hot ticket was running a magneto setup LOL. Before I joined the military I was a mechanic tuning BBC corvettes, 2x4 barrel 426 hemis, doing hop-ups on flathead Fords, and such.

Anyway, trying to decide how to trigger the primary is a good question but not easy to answer because you need to look at the total package. No matter which setup I use it must have very fast switching. For just triggering I prefer Hall-effects with either a transistor/IGBT or a MOSFET. The first 3 things I look at are the switching speed, current capabilities, and the internal resistance. The voltage and coil primary resistance I'm going to use dictates the current handing. Internal resistance and current dictate the heatsink size. The more voltage drop across the switch and the more current the more heatsink I need. The problems: MOSFETs are voltage driven and are susceptible to noise. But they have very little internal resistance and can handle lots of current and GD voltage. Transistors & IGBTs are current driven so they are usually more complex to drive them. But they are better at handling noise. Transistors usually have a higher internal resistance and therefore generate more heat. IGBTs usually need to have their gates driven to ground or even below that to turn them off, there are IGBT drive chips available. I've had some success with driving IGBTs and they can handle a lot of current. On one IGBT circuit I had, I had to add 2n7000 FETs to drive the gates to zero once.

Cheers
Ray

 

[74Sprint]

Next thing is inductance which is where the magnetic field lives and this is always a toss up and a balancing act thing. There are a few good websites on making ignition coils and I'll try to find some active links. Questions that come up are 'how many Heneries do I need for the primary & secondary? How much current do I need? What voltage? and what wire size do I need or I'm going to use? Solid core or laminate?

The core material is easy, for low tension ignition you can use solid core or laminate low carbon steel, the lower the better. Use laminate low carbon steel for high tension, less stray eddy currents at higher frequencies. For TCIs it's always how much current do I want to use, not so much with CDIs. With CDIs it's always the extra cost and complexity of making a high voltage charging circuit but the coils are easier to figure out. To this day I still can't get a simple Royer oscillator to work, nuts to that . I made a pretty simple high voltage charging circuit that is guarantied to work and when the cap hit 1,000 volts it quits charging and goes into hic-up mode keeping the voltage up. I even had custom CCFL transformers made in China. Shipping was more than the transformers!



For plug wires I always try to use solid/stranded copper core wire and non-resistor sparkplugs. When I found out NGK was going to only make my favorite plug with a resistor I bought a case of non-resistor ones, I should be good for years. Yup I make a lot of EMI but, at the race track nobody cares. On street cars that have stereos in them I prefer the inductor wound ones, they are a bit more but also work better.

Cheers
Ray

 

[74Sprint]

Ray: I was taught (in the 1980s) that the coil heating is due to the combined heat from the inrush current (RMS of I-squared-R in the primary during t of contact dwell, plus zero-current during t of contacts open?), plus the "out-rush" current (I-squared-R in the secondary during arc-duration at the spark-plug - this t is very small!). This meant that inductive HT cables (that I selected as most appropriate for our application) actually should have heated the coil more than the (cheaper) resistive HT cables (that the coils had been designed for) but on real testing the temperature increase wasn't measurable. (secondary winding discharge current is in micro-amps for micro-seconds duration). But maybe my memory is in error? - For those that don't know but are interested, but inductive and resistive secondary cables, plug-caps and plugs reduce the peak current of the spark, hence the peak amplitude of the radio frequency noise produced. The voltage of the spark is determined by the breakdown voltage at the spark-plug = An engine parameter, not determined by coils, etc. After the spark strikes, the arc current rises very fast (voltage across the spark-plug contacts falls as the ionised gas resistance drops rapidly). I'm guessing: but I think that how quickly the voltage achieved (just before the arc is generated) determines the maximum frequency of the radio noise (is it dV/dt? = rate of change of voltage?), whereas the arc current (or dI/dt?) determines the "amount" of radio noise emitted. Thus resistors and inductors in the extra bits after the secondary coil help to limit that arc current (reduce dV/dt - hence max I is reduced?) - therefore suppressing radio noise. The HT cables become the antenna for broadcasting the radio noise. (Any radio engineers want to correct this? - or add to the explanation? - It was 40-plus years ago when I was taught this information... but I forget the maths!).
Not machining, but related to "spark ignition engines"...
Cheers!
K2.

Your correct for most of what you said. First there is no radio frequency as per say a frequency. The EM pulse covers pretty much all frequencies. When you look at an ignition coil and take the primary first it is not much more than a big electro magnet. You need and want a big magnetic field of stored energy. The field has voltage and current as components, better to think of it as watts or joules. Now when the field collapses and transfers that store energy to the secondary there is some loss but you are still transferring energy. Your just swapping the current for voltage. If we could have a transfer of energy without losses then we would have just as much energy on both sides. Yes your right the current in the secondary is somewhere between mico & milli amps but, the total energy is still close to what you started out with on the primary. In the early days of radio they used spark gap transmitters and Morse code there was no frequency or sinusoidal wave. I believe the Titanic was one of the first ships to use a tunable short-wave radio I maybe wrong.

As far as wires go I did find that the carbon core wires did boost the voltage but only by a very small amount. But the resistive carbon core wires do knock the current down quite a bit, I don't like them. Resistive wires and plugs kill the amount of current flow to reduce the magnetic field (EMI) generated by the sparkplug wires at the cost of the sparkplug gap wattage/joules. I have never seen wires get warm from sparking. Yes the field collapse is drawn out dI/dt longer but with no benefit. However a larger inductance on the secondary side can create a longer dI/dt with benefit. Problem there is if you go too large of inductor you will also slow the field collapse. Everything to do with an ignition coil is a balancing act of trade offs. There is a way around all this but I want to patent it first sorry.

cheers
Ray

 

[Steamchick]

Thanks Ray, Most interesting.
But when you got to talking jargon...." I was at first against going to HEI (???) but, I'm a convert. I remember when the hot ticket was running a magneto setup LOL. Before I joined the military I was a mechanic tuning BBC corvettes, 2x4 barrel 426 hemis, doing hop-ups on flathead Fords, and such. " ... - you just lost me there. My English thinks the BBC is the National TV set-up, a Corvette is a small ship used for submarine chasing, a hop-up is some kind of "human" exercise routine... etc. I never got involved in American Racing stuff. Lotus, Brabham Repco, Williams and Renault a little bit of interest in F1, but mostly British and Japanese motorcycles..... And I avoided Hardly-Able-Twos as agricultural 2-wheeled tractors that people use to make a lot of noise ... We are different sides of the ocean with a different view of life.
On "arc explosions": From my experience with HV electrical design, (400kV circuit breakers) I understand a little about the ionisation and energy transfer - and how the sililar process happens in "exploding chemistry"...
Initially, when an pair of contacts see the electrical field rising, the gases between are a simple insulator. But as the field passes a certain point the gases begin to ionise (corona) and this ionisation is "hot" - a few thousand degrees - so as the ions take up electrons (cooling and mixing with un-ionised gas) they emit everything from X-rays to long radio waves, including a bit of light. The collapsing hot gas bubbles cause sonic waves - the crackle sound you can hear when HV electrical equipment is discharging corona under extreme conditions.
The ionised gases increase as the voltage goes higher (electic field at the conducting parts increases) until the "cloud" of ionisation becomes a complete conductor. I.E. the resistance of the gas has switched to that of conductance of ions and electrons. Thus the arc strikes. The surge of current as the striking voltage collapses to the "arc-currrent" voltage across the gap has a decay determined by the whole secondary circuit impedance in a coil ignition system. But as the arc current is flowing there is a rapid heating of the gas, ions and electrons so that the positive and negative ions mix and instantaneously combine as chemical combustion takes place: The chemical energy and arc (electrical) energy stuffed into this tiny pocket of ions and gas bheats other gas very rapidly and that heat accelerates the ionisation to strip apart other large molecules (more fuel becoming elemental ions) which add to the chemical re-combination of positive and negative ions with the release of even more heat energy. This is all a spontaneous chemical reaction inside the arc as the plasme heats and chemistry accelerates inside the plasma... This is a true explosion inside the arc, not deflagration. The expanding gases from that (tiny?) explosion inside the arc then forms a ball of combustion as the transition layer of burning gas heats the surrounding cooler gas - and thus the deflagration flame front is initiated. This pressure wave at sub-sonic speed can compress other zones of gases (by the complex reflections and wave interactions) such that other spontaneous explosions take place - this is knocking and the shock waves are detected by modern "knock sensors" (microphones listening for these explosions above a certain amplitude level). And yes, they are heard from the metal of the engine because they travel through the metal and oil films (which do not have time to break-down when a sonic wave passes). But when the human ear can detect these noises above general engine combustion and mechanical noises then the combustion is SERIOUSLY damaging to the engine! All the combustion-related failure modes can occur at levels below the level that we can hear. Hence the reason why (the best and biggest) productioon car companies have very expensive laboratories to research every possible failue mode and elliminate it before selling millions of cars... A happy customer never experiences failures. (A happy customer drives in to the Dealer's garage for a service - then later a new model - rather than getting it towed to the nearest garage and buying a different make!).
Now if a tiny zone of metal (aluminium mostly) is so hot from the general - but localised - engine temperature, (say, in the corner between cylinder head and cyinder wall, next to the gasket?) that spontaneous local combustion occurs from the pressure wave of the main combustion heating the local pocket of gas, then the spontaneous combustion causes a high pressure wave (the knock) followed by a low pressure wave (all the molecules in the local bubble of gas rush away with the explosion leaving a void = vacuum). This local low pressure wave can "hit" the metal surface, which, if hot enough, can cause molecules to be "sucked" off the surface of the metal. (Just like cavitation on a speed-boat propeller). This can give the surface of the metal an appearance of "being chewed by rats", or lots of pin-holes or small cavities.
As to melting holes in pistons: as they heat-up with "high performance" running - the heat flow from the centre (or hottest zone) of the piston transforms the metal from being lots of crystals bound solidly together, into being a mix of "metal" and intermolecular "cracks" - between the crystals... I.E. the material is in the transition zone (pre-melting). But this no longer can resist the stresses from the combustion and mechanical pressures so it instantaneously fails. But there is an interim point - momentarily between the start of failure and destruction, so if the combustion heat is removed just at the right moment, the piston can be deformed, but not failed. You describe this well. I could create this with a Yamahe 350LC engine - with "stage 3 tuning" for production bike racing... I.E. it needed 105~110 octane rated fuel. Actually, it was OK for short blasts at wide-open-Throttle (WOT) at up to 12500rpm... but when sustained at 55mph in top gear (just a piddling 4500rpm) it would overheat the edge of the piston at the exhaust port and the top-land would collapse, trapping the ring and causing a ring-breakage and the piston top-land to crack/break. After 5 pistons, I reduced the overall compresssion (measured using a compression tester at kick-over) from 17.5:1 to 13:1 by inserting an extra head gasket... This allowed me to run on road fuel at 97 octane + Octane booster".
The pistons showed various stages of failure: The crown showed where the metal had been partly washed-away at the exhaust port, as if by cutting with an oxy-acetylene torch; deformed to trap the ring, parts broken and trapped between the piston and cylinder and sheared-off, etc., so the metallurgucal examination could show the cracks and crystalline aluminium that had been overheated. There was no sound of pinking or knocking when running at a fairly constant 55mph (50~60mph), just a "bang" as the engine seized! This was the case with the first 3 pistons - with varying degrees of damage - but the 4th was not a total failure: I had re-built the engine, checked everything and on a cold day took the bike for a short run at 55mph before stripping to examine the piston = trapped ring from deformed top-land, but nothing broken. Some slight top surface erosion from the exhaust flames, but not as bad as the 3 failed pistons... The 5th piston (after compression reduction) was OK for all running conditions and didn't fail.
Glad your experience of inductive and resitive HT wires is about the same as my "Theo-rhetorical knowledge"!
Cheers.
K2

 

[74Sprint]

So what did I get done lately? Well I blew one of my MOSFETs by hooking it up wrong, stupid old eyes and wrong eyeglasses. I played around with the drive circuit of the gate and I am at 6.21 volts. The Vgs(th) voltage gate-to-source threshold (turn on) is 2.8v to an optimum of 10v. I'm taking the gate voltage up in increments to see how things go. Below is the change in the gate driver circuit. The least internal resistance is at 10 volts Vgs(th). D1 & R1 form a voltage divider for the gate voltage, also R1 helps to remove the voltage built up on the gate capacitance for quicker turn off which gives better spark. The CD4047 has no problem driving this kind of circuit. I know I said I wanted to do this without a scope but, I'm having some tuning problems that can only be solved using the scope. So I'm going to have to clean off the other bench and use the scope. Sorry about that. I'll let everyone know what I find.

OLD

NEW

Cheers
Ray

 

[74Sprint]

Thanks Ray, Most interesting.
But when you got to talking jargon...." I was at first against going to HEI (???) but, I'm a convert. I remember when the hot ticket was running a magneto setup LOL. Before I joined the military I was a mechanic tuning BBC corvettes, 2x4 barrel 426 hemis, doing hop-ups on flathead Fords, and such. " ... - you just lost me there. My English thinks the BBC is the National TV set-up, a Corvette is a small ship used for submarine chasing, a hop-up is some kind of "human" exercise routine... etc. I never got involved in American Racing stuff. Lotus, Brabham Repco, Williams and Renault a little bit of interest in F1, but mostly British and Japanese motorcycles..... And I avoided Hardly-Able-Twos as agricultural 2-wheeled tractors that people use to make a lot of noise ... We are different sides of the ocean with a different view of life.
On "arc explosions": From my experience with HV electrical design, (400kV circuit breakers) I understand a little about the ionisation and energy transfer - and how the sililar process happens in "exploding chemistry"...

Glad your experience of inductive and resitive HT wires is about the same as my "Theo-rhetorical knowledge"!
Cheers.
K2

I have worked on Jaguars, Porsches, Mercedes, Triumphs, Opals, and European Fords like the Capris but, I have a liking to the V12 E-Jags and the little Capris otherwise it's American muscle, the more CID the better, I love torque. My boss at the time belonged to several car clubs, sad to say but, they (car clubs) are almost all gone now. A lot of collector cars that were here have been sold to either the USA or Japanese markets. The only European bikes I have worked on were Triumphs or BSA. I drove Yamahas and Hondas, raced Yamaha dirt bikes along with Ski-Doo and Arctic-Cat snowmobiles with my brother.

On RADARs we didn't have switch gear but, we did on locomotives and those had solid silver contacts and when they opened under load (1,200V & 2,000Amps) flames and sparks would fly out of the arc chutes and sound like several 12 gauge shotguns going off in the cab. The biggest plasma arc I ever saw was when we had the RADAR in dummy load and something went wrong. Well we had mega-watts of power blow a fist size hole out the side of the dummy load, scary as hell, good thing the safeties kicked in and shut the RADAR off. As for sparkplugs I wouldn't say the spark is approaching plasma at least with normal ignitions. I have seen what appears to be plasma on a couple of ignitions including my high amp analog ignition. But I think your idea of the plasma during ignition to start the deflagration process is interesting.

I use to tell people "don't get greedy with the performance, something will blow", well wouldn't you know it I should have listened to myself, I vaporized a hole in the #3 piston on my ACR Neon. I was in a rush and put in regular octane gas instead of high test. In my brothers Arctic-Cat 250cc SxS twin he wanted to try alcohol and nitromethane, well at 10% it's ok, at 20% it's ok and going real good, at 40% it was going like crazy at WOT but, after about 15 seconds the engine just quit. Turns out the top of #1 piston blew off in pieces embedding some in the head and split the engine in 2. The piston looked like a chunk of concrete with no top. In my years of racing I have seen all kinds of things, collapsed ring lands to burst cylinders and probably like you I have lots of stories.

One last thing I wanted to say was, with my high amp analog ignition where there is multiple spark discharges for each firing and where each spark is stronger than the previous one we noticed that the exhaust was much much cleaner. Also we could use less ignition timing and the engine revved up faster. With my scope and probe (home made) I can only measure up to 100Kv and I have seen it go above that in one spark sequence. I made a dummy load (shunt) and measured 1 amp at the sparkplug. Took a hit/poke once from it through my little finger and out my elbow. F..K did that hurt, it threw me back about 6ft and the timing light ended up across the garage. My forearm had a really bad cramp for half an hour and my little finger was numb for 2 hours. My local college has a new advanced automotive research centre which, I am an alumni of, I mean to ask them if I can put a test mule engine on their engine dyno and I'll share my research with them. Otherwise I'll have to buy dyno time.

Anyway cheers
Ray

 

[Steamchick]

Well Ray, Sorry to hear of your shocking experience!
I hope my explanation of the ignition is somewhere close to what is really happening... it is a compilation of all my varied experiences of "sparks" - and deducing what is happening...
Curiously, when lightening strikes, the thunder is the sound of the shock wave of the rapid collapse of the fine thread of expanded gas from the arc. And - please correct me if I am wrong? - a plasma is created in every arc where the current flows in the ions? (even in the corona from HV electrical kit). - The ions make-up the plasma (I think?). The temperature of the plasma is easy to see as it is white~blueish... 2000 ~5000 deg.C? I.E. the "visible spark" of light emitted from the electrons leaping to higher energy levels from electrical energy input, then falling back to their earlier lower energy levels while emitting (EMR) radio waves and visible light. (I think a high enough voltage emits X-rays?). And a plasma of ions from air and fuel molecules will explode (spontaneously combust) as the ions re-combine as different molecules (exhaust gases) and release their chemical bonded energy in the process. This is the ignition that starts the flame front that travels in the combustion chamber.
If insufficient arc energy is input, the chemistry of combustion fails...
In circuit breakers, the arc is generated immediately as contacts separate (incredibly high electrical stress when the gap is near to infinitely small.. well molecular dimensions...) and continues until the voltage reaches zero (as in AC breakers) or adequate arc cooling occurs to prevent a continuous path of ions (plasma) between the contacts: I.E. some resistive gas is put in the place of the conducting plasma. This plasma-cooling and substitution with interposing fluid (resistance) is sometimes generated by external means, and sometimes by a gas explosion generated by the rapid heating of the insulating medium. (e.g. oil or gas).
All good for the soul. Using the body as a conductor is never good for the body or the soul!
Cheers!
K2