Coil on spark plug igntions.

[OP800]

I need some insight on specifically what happens at the plug when the gaps cannot be fired for any reason such as a very lean misture, excessive gap or a short.
The plugs have center electrode resistance anywhere from 500 to 5000 ohms.
Does the spark attempt to find ground to the shell in these cases, causing random missfires?
How does the voltage drop accross the resistance affect the outcome?
In the specific situation the coil is right on top of the plug by spring connection inside a rubber boot.
The coil resistance is 7000 ohms, the plugs are in the 3000 ohm range.
I am trying to track down the reason for random light load missfires that go away with a plug changes but return in a short time and go away again with another new set of plugs.
This last set was selected for lower center electrode resistance and still in testing stages.
If the missing returns, I am next going to replace the boots.
It may be the boots have grown in size and donot fit the plugs tight. As soon as the dielectric grease dries hard the leakage begins.
The coils may be pretty hot for voltage.

 

[MikeHalloran]

I suspect you'll have to replace the coils.

Mike Halloran
Pembroke Pines, FL, USA

 

[Metalguy]

Don't be concerned with the ohm readings for the plugs. The "resistance" is FAR higher in an operating cyl. Try gapping the plugs at the low limit, and try fine-wire iridium plugs.

"When the eagles are silent, the parrots begin to jabber."
Winston Churchill

 

[OP800]

What I didn't say was how this all started.
The OEM plugs had 53,000 miles on without a problem.
They were changed only for maintaince and not for running problems.
The replacments lasted about 15,000 miles before missing started to gradually come in.
The next set of the same new type lasted 300 miles until missing began.
I started investigation on what was going on.
So far the original plug, a plat type, has been superceded by the mfger to another designation.
The difference is the center resistence. The originals are 500 to 3000 ohms. The mfgers replacement is 6000 +/-.
The third set for replacment has been selected for the low resistance values in the original range of the OEMs as well as a make change.
So far they are workng without a miss so time will tell.

The next change will be the rubber boots and improved seal at the plug insulator.
Additionally, this missing condition doesnot set PCM codes and in my opinion is a 'lean miss' condition when the throttle is slowly rolled on and off at low load cruising.
Never a miss at heavy loading such as towing.
Sometimes the miss will be violent as thought the ignition were totally shut off for an instant.
I am thinking this is upsetting PCM processing until regained.
The system uses a crank sensor and cam position sensor as reference points to build the ignition control. Plugs on these sensors have been reseated.
A test was done running the engine under the same conditions with the EGR disabled. No difference. EGR is clean and free in it's operation. EVR has had it's filter cleaned.
Throttle position sensor has been tested for smoothness.
Fuel and water has been ruled out. Fuel filter has been changed.
Battery was replaced for age and low capacity.
There is never a starting, idle or power problem.
Engine has no modifications.
I find it hard to point to the coils given how this all has worked out so far. At about $50 each I may have to change them if the problem is not solved by the last remaing items to change. I may even put the old original plugs back in for test to see what happens.
Sorry this is so long but unless all the info is stated this becomes a guessing game with no engine code to give a clue.

 

[MikeHalloran]

My operating hypothesis, okay w.a.guess, is that continued operation on the second set of plugs trashed the coils.

As in, the plugs' effective resistance went way up, for whatever reason. When the coil field collapses, the stored energy is going to go _somewhere_. If the plug resistance is too high, the energy will find a pinhole in the coil wire insulation and arc to the nearest ground, probably within the coil assembly. The insulation's decomposition products include carbon, so arcing becomes more likely on future cycles.

You might be able to partly repair the coils by vacuum impregnating them with green Loctite (#271 ?), the wicking kind that's used for sealing porosity in castings. Probably not.



Mike Halloran
Pembroke Pines, FL, USA

 

[OP800]

Thanks Mike, these are the tech items that make possible sense because there is not much info on this kind of problem.
The coils are totally sealed in a bakelite case with potting that is still sealed and sound but that dosn't mean internal damage has not been done.
The latest plugs have 400 miles now without a problem.
I selected Bosch #6142 for there lower resistance and this test. They are thin embedded wire single Plats with no exposure except the very tip of the wire in the porclain.
The motor is a 4.6L SOHC v8 with deep plug wells.
A simular case was repored on, in Motor Age mag about a customet using octance booster that resulted in red oxide deposits on the plugs and would 'carry on' during the EGR routine only when the cylnders are run lean and the plugs would not fire reliably.
In my case the throttle roll on often is coming from low injector pulswidth which may be the lean miss.
Along with this, the trans is an electric control so I assume one never knows what all effects could happen from this kind of ignition fault from causing down shifting to converter unlock that may be the violent hesitation the happens every so often as a secondary result. This is the time the ignition feels like it is cut off for an instant.
The coils primary resistance is .55 ohm so the drive current is likely quite high. The turns ratio is should be in the order of 70 to 1 so the secondary voltage could approch 40 KV.
This is why my current direction of thinking involves the plug resistance and is somewhat critical as well as the boot seal living in a sealed deep well.

 

[MikeHalloran]

I wonder what problem the mfgr was trying to solve with the plug resistance change. I wonder if they changed the coils they were installing at the same time.





Mike Halloran
Pembroke Pines, FL, USA

 

[patdaly]

A couple of things I have observed with the Chevy LT1.

When you change the plugs, change the wires ( Or boots in your case ) For whatever reason, it seems like you get a random misfire if you attempt to reuse wires that are very old at all.

With newer Chevrolets, I have had zero luck with Bosch plugs.

 

[OP800]

The boots are the next item.
I have found that information about this subject is a bit rare but did find a paper written on advances in ignition tranfer efficiency, that deals in part with the same issue, of plug center resistance.
Most plug resistance is thought to be there only for RFI purposes but also really cuts the "conversion efficiency" of a system already very poor, as I think I am having a problem with. The change of plugs may have pushed this over the limits and created a marginal condition.
The authors put forth several ways to increase the efficieny to dramatic levels above the present levels.
The use of 'peaking capactior' right at the plug is one easy way.
In reference to my engine's specific design, the peaking capacitor method makes me wonder if the deep well and the plug connection are indeed increasing the capacitance as a configuration just the same as a peaking capacitor would intentionally do. They are only talking of capacitance in the 80pf range but the voltage dielectrics involved are quite touchy ie possibly the present plug boots.
Maybe that I have now begun to figure out why these engine plug location designs are so touchy for plugs, boots and coils.
The new plugs have been working very well but at about 400 mile there may be signs of missing again but more miles are needed to be sure.
Stay tuned.

 

[Ron07663]

op800:
Sounds like you have a coil-over-plug arrangement - One coil per plug. If you can isolate the miss to one cylinder, you might try moving that cylinder's coil to another location to see if the problem follows the coil.

The coil will fire at higher voltage as the spark gap increases. The insulation in the coil and the boot are highly stressed with this higher voltage and can fail. New plugs can help for a while since the initial gap can keep the voltage below the level where the damaged insulation breaks down.

The resistance rating of the plug affects the RFI emissions that cause radio noise and probably not a factor in your problem.

 

[patprimmer]

Try closing down the plug gap considerably. Like if its 0.040" go to 0.030".

If the miss goes, its a breakdown under load. This would most often happen in a cap, coil post or suppression leads. As you have none of these it must be in the coil or boots as there is nothing else.

Does the coil clip firmly onto the top of the plug.

What happens if you run plugs with no suppression.

Regards

eng-tips, by professional engineers for professional engineers
Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[LordMalaK]

gotta love C.O.P. systems,

Try spraying a fine mist of water onto the coils- look for sparks.
look for carbon tracks on your boots
look for vacuum leaks (the biggest offender, and the most overlooked problem)
move coils around and plugs, see if the problem moves

non-resistor plugs aren't a good idea as they effect the amount of energy required initiate the spark jumping the gap of the plugs- which effects timing (so does your plug gap)

you may also need to put a scope on your coil driver (module)- you may have a flaky transistor., or a carbon tracked module effecting a particular coil.

You may also have an injector spray pattern thats way outta whack (mercedes CIS are notorious for this)- they clog and spray a stream instead of a mist.


...unless you just want to throw money at it until you fix it. I see techs do this ALL the time.

 

[Metalguy]

>"non-resistor plugs aren't a good idea as they effect the amount of energy required initiate the spark jumping the gap of the plugs- which effects timing (so does your plug gap)"<

Do you have any hard data on how this affects timing?


"When the eagles are silent, the parrots begin to jabber."
Winston Churchill

 

[patprimmer]

I would think the knock sensor should take care of minor timing changes.



Regards

eng-tips, by professional engineers for professional engineers
Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[LordMalaK]

metalguy- Sure I can scan a few pages for ya, but it's only a few degrees, and without a nearly identical resistance between each cylinder you may have early ignition on some cylinders compared to others. Besides, with a resistor plug you have a hotter initial spark instead of a trickle building up to full discharge.

 

[OP800]

Lets regroup and begin again.
In researching this problem, I found a paper addresed at SAE on 'Automotive Ignition Tranfer Efficiency'.
The paper addresses all the items I have been thinking about and seeing in fact, in the chase of the problem.
Series resistance in the secondary circuit kills power tranfer efficiency to the spark kernal in the plug's gap.
There are three phases to the spark generation.
a. the inital trailer breakdown.
b. the channel heating resistve phase that ignites the mix.
c. the residual spark.
This all is said to occurr in about +/- 5 ns.
Adding series resistance causes a voltage drop due to the current flow when b. phase occurrs. This power is waisted when dropped accross the plugs upper series resistance and or feed wire. This resistance is usually incorporated there for RFI reduction and serves no better purposes.
Adding a capacitor (SAE paper) at the plug's top connector on a plug with no series resistance increases the tranfer efficiency.
IMO the capacitor takes a charge on the inital rise time, then when the voltage reaches gap breakdown, the stored energy in the cap adds to the b phase of the spark even lengthening the spark time a small amount, increaseing the total efficiency (power). At heat phase time the low resistance of the gap breakdown becomes a short to the cap, discharging it into the resistive phase of the spark.
Back to the problems.
If something happens that causes power to be reflected back into the primary; this could propagate all the way back to the PCM upsetting processing to everything for instants in time. The results each instance could show in any number of different ways depending on what routines were being addressed at that point in time.
For example, wrong/late/early/no firing of the next or other cylinders. The affecting of the electronic transmission control signals causing such things as the converter lock up to falsely function, fuel injection pulsewidth changes etc.
What is felt while driveng is a random feel of missing that sometimes is violent in nature.
This type of action cannot very often be diagnosted by the PCM for fault coding and CEL lamp indications.
Question I would be asking if I were the PCM is what cylinder/s are at fault or is it all of them. How can the PCM deal with this kind of fault that is never static long enough for the PCM to count up enough hits to decide on the codeing to indicate the fault?
I am about 500 miles into the third set of plugs selected for their lower center conductor resistance and the fault has not returned as of yet.
Starting, idle and power pulling is never affected.
I am assuming the theroy put forth in the ASE paper that has touched on all these areas to be showing many of the same symptoms I think I see, as I have no way to measure the performance on these type systems. It take specialized insturments.
A mode 6 scanner may be able to look at system history but that is about out of the question because IMO a dealer tech will not go that far in most cases because they don't often do that. Just replace parts based on codes. For my case they will say no codes -no troubles.
I will be replacing with new boots in the near future if the missing returns even lightly in case it is a breakdown and seal problem.
Normally a special dielectric grease is used on the boot to plug seal. There may be a chance the fault returns when the grease hardens from heat and no longer can hold off leakage spark jump to the plug body.
Something is going on in the plug 'wells' that needs to be pinpointed because it sure won't just go away by it'self.

 

[patprimmer]

I guess it's to basic to ask if the plug well is clean and dry and there is no sign of oil or water on the plug main body porcelain.

I have seen similar problems after engines have been cleaned or when oil leaks into the plug well.

Regards

eng-tips, by professional engineers for professional engineers
Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[Compositepro]

One problem I see with all suggested solutions so far is that they suggest that the misfires are due to the spark occuring at other places than the gap in the plug. This tends to happen under high engine load, not no load, because under load there is higher pressure in the cylinder which then requires higher voltage to break down to make a spark. So that can't be the problem. I would guess that the problem would have something to do with timing or fuel mixture which changes between load and no load. Something about the plugs are changing so that the spark is not intense enough to ignite the mixture at no load conditions.
Perhaps the gap is too small?

 

[Metalguy]

Compositepro,
Agree, now that his problem doesn't happen under heavy load/high KV demand conditions.

I used to run MSD ignitions, which would fire the plugs for a relatively long time (~10-20 deg. of crank rotation, IIRC). The MSD would make the idle and low-load cruise smoother, but have no real benefit at full throttle.

A long time ago I ran a SBC in a Camaro with various combinations of added oxygen or nitrous, plus water/methanol. I put a set of "surface-gap" plugs in from my collection of Mercury outboard plugs. After ~5,000 miles the MSD had great trouble firing them, so I had to take them out. The Merc. "Thunderbolt" ignition would fire them easily--no wonder Merc. had a decal inside the engine cover that warned about "Possible LETHAL voltages"!

"When the eagles are silent, the parrots begin to jabber."
Winston Churchill

 

[LordMalaK]

could be just a bad or internally shorted/carbon tracked coil. I've seen a good coil produce a spark that jumps a 4 inch gap. maybe yours is so weak (but tests fine with an ohmmeter0 that it can only generate a spark under no load that only jumps a gap of .030 inches.

Or maybe just a cracked ported vacuum line.