Spark Plug Tips

Originally Posted by 1978elcamino
reduces flame quench and lowers required voltage


Flame quenching is due to a combination of unburnt exhaust gases and the cooler combustion chamber walls.
 
Last edited:
I probably shouldn't have used the term energy. I realize instantaneous power is different from power over time. What I meant was, there is no way an ignition coil at 18KV is pushing 200A through 10kohms worth of plug and wire resistance, that would require far more voltage. That current can only be supplied by electric charge in the immediate vicinity. If instead that current is coming from some chemical reaction, then something beyond my understanding is occurring.

My impression is that normal combustion is not a sudden bang that forces the piston down, but rather a sudden increase in force after ignition which, while cruising, is only strong enough to keep the engine turning and overcome road resistance. As such it doesn't have to be a strong force, only strong enough to overcome drag.
 
Originally Posted by keantoken
I probably shouldn't have used the term energy. I realize instantaneous power is different from power over time. What I meant was, there is no way an ignition coil at 18KV is pushing 200A through 10kohms worth of plug and wire resistance, that would require far more voltage. That current can only be supplied by electric charge in the immediate vicinity. If instead that current is coming from some chemical reaction, then something beyond my understanding is occurring.


We're talking about Transient Electrical phenomena which occurs over short periods of time. Peak values of power and currents are dependent on short-term snapshots verses Average values due to longer periods.

It makes me wonder if you have read any of my responses to you and the example of the radar system I presented or if you have read any of my responses to other posters, but anyway I'll let you make the calculation:

Voltage: 40,000 volts; SP Gap Resistance = 200 ohms; V = IR; I = V/R; This specific situation occurs within a 10 nanoseconds (10 X 10^-9 seconds) window. What is I in amps within this 10 nanosecond period?

This Gap Resistance resistance quoted above is toward one of the later phases and actually varies from 0.005 to 500 ohms or more, depending on the spark phase.

I takes about 2.0 ms to charge the coil; The complete duration of the spark across the gap happens in a window of 1.3 ms - this 1.3 ms window ENCOMPASES all the three phases within the spark window described above.

The AVERAGE primary coil current over a period of 1 ms is about 8 amps; the AVERAGE secondary coil current is approx. 300 mA. The AVERAGE spark current is about 105 mA.

Originally Posted by keantoken
My impression is that normal combustion is not a sudden bang that forces the piston down, but rather a sudden increase in force after ignition which, while cruising, is only strong enough to keep the engine turning and overcome road resistance. As such it doesn't have to be a strong force, only strong enough to overcome drag.


Originally Posted by MolaKuleasQuotedAbove
There is no explosion. This is a combustion process, not an explosion; there is a difference.

Let me explain: In an explosion the volume of gases is confined until the pressure in the vessel exceeds the material strength of the walls of the constraining vessel and fails and erupts into particles of solids (shrapnel).

In combustion, the expansion of gases is not confined as above but the piston moves downward to increase the volume preventing pressure spikes and subsequent grenading of the head and cylinder walls.

Now as for rails and nitrous powered vehicles using high compression engines, sometimes the gases will pre-ignite due to hot spots in the combustion chamber or the valves and will explode because the pressure spike comes at the wrong time (wrong crank angle).
 
Last edited:
This topic is fascinating to me, and sorta relevant. I've recently tried fine wire iridium (NGK TR5IX and TR55IX) plugs in my track car. However, it didn't like them at all. Above about 5500 rpm, spark blow-out (I'm assuming that's what is going on) was common and mis-fires frequent. This happened at both .060" and .040" gaps. It was less frequent at the narrower gap, but still present. However, I went back to conventional nickel alloy V-groove plugs and even at .060" gap, it runs flawlessly with no misfires. I'm not complaining as the nickel plugs are considerably cheaper, but perplexed as to why that would be the case. This is a rather mild, naturally aspirated drag race application with coolant temps in the 160-180°F range.
 
I'd also like to note that while a millisecond doesn't sound like a lot of time, that's 12 degrees of crank rotation at 2000 rpm and 33 degrees of crank rotation at 5500 rpm.
 
Originally Posted by RDY4WAR
I'd also like to note that while a millisecond doesn't sound like a lot of time, that's 12 degrees of crank rotation at 2000 rpm and 33 degrees of crank rotation at 5500 rpm.


For sure and that's why spark duration and spark timing are critical, especially with the feedback from the various sensors that help determine that spark timing. The sensors are continuously feeding info to the ECU and calculations are being made on the fly as to the timing of the spark for each SP.

Spark count F = rpm × number of cylinders

For example: 4-cyl. 4-stroke engine, rotating at 3,000 rpm Spark count = 3,000 × 4 = 6,000 sparks / min = 100 sparks/second = ~ 10 milliseconds between each spark.

For a driven distance of 30,000 km with an average rotational speed of 3.000 rpm and an average speed of 60 km/h, that works out to 45.000,000 sparks per ignition plug!
 
Originally Posted by RDY4WAR
This topic is fascinating to me, and sorta relevant. I've recently tried fine wire iridium (NGK TR5IX and TR55IX) plugs in my track car. However, it didn't like them at all. Above about 5500 rpm, spark blow-out (I'm assuming that's what is going on) was common and mis-fires frequent. This happened at both .060" and .040" gaps. It was less frequent at the narrower gap, but still present. However, I went back to conventional nickel alloy V-groove plugs and even at .060" gap, it runs flawlessly with no misfires. I'm not complaining as the nickel plugs are considerably cheaper, but perplexed as to why that would be the case. This is a rather mild, naturally aspirated drag race application with coolant temps in the 160-180°F range.



This may answer your question as to why your V-groove plugs performed better:

http://www.ngk-sparkplugs.jp/english/techinfo/qa/q09/index.html
 
Last edited:
Originally Posted by MolaKule


...You are forgetting that each phase represents "short-lived" transient phenomena and waveforms that occur for nanoseconds to milliseconds. For example pulsed radar may have output powers of 50 Megawatts but that power is delivered in microseconds.I.e, the Peak power and the Average powers are different.
...


Let me if I may present another example of Peak Power verses Average Power:

In the older radars a vacuum tube "pulse" tube modulated an RF magnetron.

One of those pulsed tubes was designated as 6293 which was a Military version of the 6146 in which many of you Amateur Radio operators may be familiar. it was essentially operated as a Class "D" switch.

A voltage of 3000 volts on the plate and 1.0A plate current resulted in a Power input of 3,000 Watts for 150 us On-Time. That is, the tube was turned on with a rectangular pulse (at the grid) for only 150 millionths of a second. So it produced 3,000 Watt of peak "pulsed" power when on for only 150 us. In Average Power terms this meant a plate current of 0.03A or an Average input Power of 90 Watts.

If the tube was turned on for grater than 150 us it could literally "melt" down. The glass envelope would literally "suck in" and loose vacuum quickly.

In Amateur radio service this same tube is operated at a plate voltage of 750V and a plate current of 0.12A for an input power power of 90 Watts in continuous "average" power service.




...References for further Information and Study:

1. W. Mitianiec, Factors Determining Ignition and Efficient Combustion in Modern Engines Operating on Gaseous Fuels,
INTECH, 2012.

2. N. Kawahara, et. al., Plasma temperature of spark discharge in a spark-ignition engine using a time series of spectra
measurements
, 18th symposium on the Application of Laser and Imaging Techniques to Fluid Mechanics, Lisbon Portugal,
2016.

3. A. Joshi, Effect of spark advance and fuel on knocking tendency of spark ignited engine, Michigan Technological
University, 2017.

4. A. Franke, Diagnostics of Electrical Phenomena in Gases for the Monitoring of Spark-Ignited Combustion, Lund Institute
of Technology, Lund University, 2000.

5. I. Andersson, Cylinder Pressure and Ionization Current Modeling for Spark Ignited Engines, Division of Vehicular
Systems, Department of Electrical Engineering, Linkopings Universitet, Linkoping, Sweden, 2002.

6. Technical Information No. 7, All About Ignition Coils, BERU. Division of Federal Mogul.

7. P. Matthew, et. al., Experimental verification of modified Paschen's law in DC glow discharge argon plasma, AIP
Advances 9, 025215, 2019.

8. Y. Raizer, Gas Discharge Physics, Springer, 1997.

9. B. HNATIUC et. al., SPECTROSCOPIC DIAGNOSTIC OF TRANSIENT PLASMA PRODUCED BY A SPARK PLUG,
Technical University, Romania, 2011. Paper presented at the 15th International Conference on Plasma Physics and
Applications,1-4 July 2010, Iasi, Romania.
 
Last edited:
Originally Posted by MolaKule
Originally Posted by RDY4WAR
This topic is fascinating to me, and sorta relevant. I've recently tried fine wire iridium (NGK TR5IX and TR55IX) plugs in my track car. However, it didn't like them at all. Above about 5500 rpm, spark blow-out (I'm assuming that's what is going on) was common and mis-fires frequent. This happened at both .060" and .040" gaps. It was less frequent at the narrower gap, but still present. However, I went back to conventional nickel alloy V-groove plugs and even at .060" gap, it runs flawlessly with no misfires. I'm not complaining as the nickel plugs are considerably cheaper, but perplexed as to why that would be the case. This is a rather mild, naturally aspirated drag race application with coolant temps in the 160-180°F range.



This may answer your question as to why your V-groove plugs performed better:

http://www.ngk-sparkplugs.jp/english/techinfo/qa/q09/index.html


I understand that, but I'm not following why that made it perform better than a fine wire. Wouldn't a fine wire tip be even better at that?
 
Originally Posted by RDY4WAR
This topic is fascinating to me, and sorta relevant. I've recently tried fine wire iridium (NGK TR5IX and TR55IX) plugs in my track car. However, it didn't like them at all. Above about 5500 rpm, spark blow-out (I'm assuming that's what is going on) was common and mis-fires frequent. This happened at both .060" and .040" gaps. It was less frequent at the narrower gap, but still present. However, I went back to conventional nickel alloy V-groove plugs and even at .060" gap, it runs flawlessly with no misfires. I'm not complaining as the nickel plugs are considerably cheaper, but perplexed as to why that would be the case. This is a rather mild, naturally aspirated drag race application with coolant temps in the 160-180°F range.



Try an even smaller gap with the iridium down to .025" if needed. I run 17psi of boost with the iridiums at .027" with no issue. I have used the conventional plugs at a similar gap as well, pretty much the same HP but fuel economy was better with the iridiums.
 
Originally Posted by RDY4WAR
Originally Posted by MolaKule
Originally Posted by RDY4WAR
This topic is fascinating to me, and sorta relevant. I've recently tried fine wire iridium (NGK TR5IX and TR55IX) plugs in my track car. However, it didn't like them at all. Above about 5500 rpm, spark blow-out (I'm assuming that's what is going on) was common and mis-fires frequent. This happened at both .060" and .040" gaps. It was less frequent at the narrower gap, but still present. However, I went back to conventional nickel alloy V-groove plugs and even at .060" gap, it runs flawlessly with no misfires. I'm not complaining as the nickel plugs are considerably cheaper, but perplexed as to why that would be the case. This is a rather mild, naturally aspirated drag race application with coolant temps in the 160-180°F range.



This may answer your question as to why your V-groove plugs performed better:

http://www.ngk-sparkplugs.jp/english/techinfo/qa/q09/index.html


I understand that, but I'm not following why that made it perform better than a fine wire. Wouldn't a fine wire tip be even better at that?


Let's examine why misfires might happen (which is what I am assuming happened in your vehicle and assuming your spark energy capability was sufficient), just thinking out load for a moment:

1) High pressure exceeds spark energy capability, (i.e, not enough spark energy to overcome pressure as per Paschen's Law),

2) Mixture too rich for spark energy to overcome fuel/air ratio (stoichiometry related - not enough spark energy to overcome rich gas mixture as per Paschen's Law)

3) Turbulence in combustion chamber extinguishing or impeding expanding plasma outside core plasma,

4) Spark knock sensor retarding spark timing (possibly sensing preignition due to carbon deposits and not enough spark energy to overcome gas Temperature as per Paschen's Law),

There may be others but I think the V-groove geometry has to do with #3. "The flame core is generated near the perimeter of the electrodes and grows" and "Ignitability is improved because the electrodes are interfering less with the growth of the flame core."

I.e, you have two flame cores at two fine points rather than one expanding out in what I assume is a Figure 8 pattern and meeting at their edges to enhance the compression wave.

Anyway, that's my hypothesis.
smile.gif
 
Last edited:
I don't think the spark takes both sides at once, this is not usually how sparks behave (the first spark leader to touch both electrodes fires and shorts the other leaders) and the NGK diagram shows the spark on one side. I would think that there would be a wear-in period with the spark favoring one side and then after wear in it would switch sides erratically.
 
Ok, I get what you're saying. Spark energy definitely shouldn't be a concern with an MSD 6AL-2 box, MSD blaster coil, MSD optispark, and Taylor Thundervolt 8.2mm wires. There's plenty of energy there. The air/fuel ratio is steady at ~12.8:1 above 5000 rpm only going slightly richer to ~12.5:1 below 4500 rpm. This engine doesn't have a functioning knock sensor and is tuned/run in open loop.

If we can expand the topic a little bit... old "tales of the pits" (as I like to call them) commonly say that indexing spark plugs and cutting back the tip of the plugs is worth consistency and power. I've never ventured into it myself as I always thought of it as an old wives tale. Is there anything legitimate about those claims?
 
Originally Posted by keantoken
I don't think the spark takes both sides at once, this is not usually how sparks behave (the first spark leader to touch both electrodes fires and shorts the other leaders) and the NGK diagram shows the spark on one side. I would think that there would be a wear-in period with the spark favoring one side and then after wear in it would switch sides erratically.


Look at the link again: http://www.ngk-sparkplugs.jp/english/techinfo/qa/q09/index.html

It shows two definitive spark columns. Physics says that theoretically anytime you have something like a v-groove geometry with two spire points, the electric field, and hence the ion density, is going to be greatest between the center electrode and those two spire points. With the materials used, the two spire points should wear evenly.

I think you may be allowing lightning studies and photographs to color your perception of what happens here.
 
I see the photograph, but photos of sparks are not reliable; the flame core graphic on the other hand shows a spark on just one side. We don't know how long the shutter was open and how many sparks occurred during that period. I also think there must be a motivation to show the sides as both firing for marketing purposes, as people are bound to worry about lopsided sparking.

It takes a certain amount of energy to generate a given volume of plasma and with two plasma cores at the same time, I think the gapping requirements would be too different for one thing. I am fairly sure the energy requirements for firing two simultaneous parallel plasma cores is very different from the requirements for firing only one.

I guess what someone really needs to do is test it. I have a grill sparker, but I would be looking for a worn in V-groove plug to maximize the chances of seeing a parallel fire. Mine are all new.

This topic has come at the right time as my Taurus with coilpacks needed new spark plugs. The previous (almost new) Autolite platinum plugs had one platinum button pop off. I suspect the base metal fouls more easily and that one bad plug was causing the ECM to limp. So I installed V-groove plugs. I had the opportunity to go for U-groove plugs but they were cheaper and I couldn't envision any great leap in performance.
 
Originally Posted by keantoken
...the flame core graphic on the other hand shows a spark on just one side...
.

That's because there was explanatory text on the left side taking up space that would otherwise show a flame core also eminating from the left side as well.

Originally Posted by keantoken
It takes a certain amount of energy to generate a given volume of plasma and with two plasma cores at the same time, I think the gapping requirements would be too different for one thing. I am fairly sure the energy requirements for firing two simultaneous parallel plasma cores is very different from the requirements for firing only one.
.

If you model the tips and gaps as two coequal but very low resistive electrical branches, the currents divide for the two but we have essentially the same potential across the gaps in order to maintain conservation of energy..

Originally Posted by keantoken
....So I installed V-groove plugs. I had the opportunity to go for U-groove plugs but they were cheaper and I couldn't envision any great leap in performance.


And you may not see any great improvement unless you are racing.
 
Last edited:
Originally Posted by MolaKule
If you model the tips and gaps as two coequal but very low resistive electrical branches, the currents divide for the two but we have essentially the same potential across the gaps in order to maintain conservation of energy..


Except this is a highly nonlinear system and a linear model is unlikely to give accurate results. Case in point, as current increases, the gap voltage decreases. This is the definition of negative resistance and is a product of the self-feeding nature of the discharge. Negative resistances in parallel are not stable and the one with slightly higher energy will act as fast as loop inductance permits to snuff the other one. Resistance starts high and transitions to a low value, with a period of negative resistance in between, but a linear model presupposes that resistance is constant.

Transistors in parallel have the same problem, one will hog the current from the other due to positive thermal feedback. In the case of transistors we can design the circuit to share equally, but in the case of plasma streamers in air there is very little we can do.
 
Originally Posted by keantoken
Originally Posted by MolaKule
If you model the tips and gaps as two coequal but very low resistive electrical branches, the currents divide for the two but we have essentially the same potential across the gaps in order to maintain conservation of energy..


Except this is a highly nonlinear system and a linear model is unlikely to give accurate results...


Nowhere have I stated that a spark ignition system was a linear system (that is a another strawman/side topic argument you brought up) for reasons unknown.

In fact, what I described in Physics of the Gap I and II is a highly non-linear system, but I guess you may not have recognized the implications.

If you have ever done simulation in PSpice you would know that you first draw schematically an electrical model of your system and then for each device you describe that device's behavior. Except for a few passive components, most electrical/electronic device behaviors are non-linear, so within the PSPice listing you attribute a behaviorial model to each component.

All of the author's in the papers I referenced model non-linear transient electrical phenomena for spark plug systems and you really need to read these scientific papers if you really want to discuss this topic further, especially study reference #8. All of the other references are available on-line.


Quote
Recommended References for further Study:

1. W. Mitianiec, Factors Determining Ignition and Efficient Combustion in Modern Engines Operating on Gaseous Fuels,
INTECH, 2012.

2. N. Kawahara, et. al., Plasma temperature of spark discharge in a spark-ignition engine using a time series of spectra
measurements
, 18th symposium on the Application of Laser and Imaging Techniques to Fluid Mechanics, Lisbon Portugal,
2016.

3. A. Joshi, Effect of spark advance and fuel on knocking tendency of spark ignited engine, Michigan Technological
University, 2017.

4. A. Franke, Diagnostics of Electrical Phenomena in Gases for the Monitoring of Spark-Ignited Combustion, Lund Institute
of Technology, Lund University, 2000.

5. I. Andersson, Cylinder Pressure and Ionization Current Modeling for Spark Ignited Engines, Division of Vehicular
Systems, Department of Electrical Engineering, Linkopings Universitet, Linkoping, Sweden, 2002.

6. Technical Information No. 7, All About Ignition Coils, BERU. Division of Federal Mogul.

7. P. Matthew, et. al., Experimental verification of modified Paschen's law in DC glow discharge argon plasma, AIP
Advances 9
, 025215, 2019.

8. Y. Raizer, Gas Discharge Physics, Springer, 1997.

9. B. HNATIUC et. al., SPECTROSCOPIC DIAGNOSTIC OF TRANSIENT PLASMA PRODUCED BY A SPARK PLUG,
Technical University, Romania, 2011. Paper presented at the 15th International Conference on Plasma Physics and
Applications,1-4 July 2010, Iasi, Romania.
 
Last edited:
Back
Top