Cracked Plug Insulator

[TheLion]

Got a cracked plug insulator today, but the interesting thing is this seems to be highly related to the condensate issues with the F-150 FMIC. I'm wondering if this might be the case. Looking for ideas, currenlty I flashed back to stock PCM software until I can get a better handle on the cause. Looking for other ideas I may not have considered.

1. Condensate from the Levels Gen 3 20x14x3.5 (we've had temp swings here of around 30F from day to night), ambient today is 78 and cool.
2. Plug defect or over torquing (I used a torque wrench and torqued to spec, but I torqued to the upper limits of the range at 13 ft-lbs, range is 7-14.5, so this time I torqued to 11, right in between).
3. Tune related (running LMS 93 Stage 3 v7)
4. Fuel related

Originally I was going to run their 91 as added protection in case of fuel quality variances, (aka run 93 octane but a 91 octane tune), but my need for speed got the better of me and I ended up going with their 93 after running their 91 for the first 1k miles, although there wasn't a huge difference, the 93 had a bit more edge.

Car has had the RR14YS plugs in it since about 1500 miles. Car has had LMS tune since about 2500 miles. This is the first issue I've had. Went out for errands and to church today, was having some fun on back roads, didn't notice any issues, car made good power, idle was smooth etc.

Went to start it up after service and immediately noticed a very rough idle / stumbling maybe periodic hesitation. Spark plugs were the first thing that came to mind, so got home, read codes and found P0304 and P0316 which are both misfire codes. Seems the element may have been stressed and failed as it cooled down or I started it back up.

Pulled the back most (closest to fire wall) plug first and viola, insulator cracked at the base and slide down over the electrode. Threw OE plugs back in, started up with stock software and runs great now. It's sunday so I won't hear anything back from LMS until tomorrow.

I honestly don't think it's tune related. There's tons of F-150's as well as other ecoboost engines out there who have cracked plug issues. Certainly knock could cause that, but it seems to be far more commonly related to temperature changes cracking the elements. I'm guessing because of the high boost all these EB engines are running the plugs see more strain, there's nothing different design wise from NA plugs (2.3L EB uses same plugs as the Coyote V8...).

I also bore scoped the cylinder and it looks as clean as a whistle, no evidence of hot spots, contamination or anything of the sort. Plugs do not exhibit any sort fouling either, they are the picture perfect image of proper heat range and ignition. Car mods are in my sig.

And for the love of all, please don't turn this into an LMS bashing thread. It may or may not be related to the tune, I'm sorting that out with them to see what they suggest as the most probable cause, but at the moment I have no idea as I was running a non-standard plug and I"ve read a few people having cracked RR14YS plugs even on stock cars. If it is tune related I will let you know.

Here's an example of the issue on the 3.5L in the F-150s: https://www.fordf150.net/forums/viewtopic.php?t=113962

My question is, for those who have had the cracked plug issue on stock EB's (of any type, Focus, F-150, SHO etc.), what plugs did you switch to that resolved the problem?

Sponsored

 

[jbailer]

I'm not really sure I buy into the temperature change theory. Porcelain is tougher than that. Having said that though, I treat these platinum plugs like fragile light bulbs. It seems in the past few years or so, there's been lots of problems with cracked insulators. It seems you've already taken into consideration the common things.

I've read enough of your posts to think you'd know if it was detonation which would be my first guess.

My second guess would be over-torquing but 13 lb ft is good, I use 12 lb ft.

The next thought would be under-torquing that seems to be a problem mostly from people that don't use a torque wrench though.

Next would be careless install, people drop the plugs into the hole then maybe don't have the wrench and the correct angle when installing. I lower the plugs in using a 3/8" fuel line and start turning them in that way.

I would also consider shipping and handling. I've never used Brisk although heard plenty of good about them. Are they packaged/protected well?

Then finally would be heat range. At 1 step colder, I think you would be good, have you asked LMS? Maybe 2 steps is needed?

If all seems good, I would just replace the set and watch. It may simply have been a hairline crack when you installed it or defective.

 

[TheLion]

I'm not really sure I buy into the temperature change theory. Porcelain is tougher than that. Having said that though, I treat these platinum plugs like fragile light bulbs. It seems in the past few years or so, there's been lots of problems with cracked insulators. It seems you've already taken into consideration the common things.

I've read enough of your posts to think you'd know if it was detonation which would be my first guess.

My second guess would be over-torquing but 13 lb ft is good, I use 12 lb ft.

The next thought would be under-torquing that seems to be a problem mostly from people that don't use a torque wrench though.

Next would be careless install, people drop the plugs into the hole then maybe don't have the wrench and the correct angle when installing. I lower the plugs in using a 3/8" fuel line and start turning them in that way.

I would also consider shipping and handling. I've never used Brisk although heard plenty of good about them. Are they packaged/protected well?

Then finally would be heat range. At 1 step colder, I think you would be good, have you asked LMS? Maybe 2 steps is needed?

If all seems good, I would just replace the set and watch. It may simply have been a hairline crack when you installed it or defective.

It very well could have been detonation, but that is not always related specifically to the software.

When the problem first started I thought my self oh crap, something just gave out, but I decided to approach it the same manner I do any other problem, start with the easy stuff.

To my knowledge any modern engine is built to withstand knock events, they do happen even with stock PCM software. It take constant knocking or some serious pre-detonation to usually cause catastrophic failure. Running an aggressive tune that's not set up properly could certainly do it, but given LMS track record and the fact they've had software out for these cars for a while now without any known issues I'm looking at other causes.

For the time being I'm going to run the stock PCM software, compared to the LMS software its performance sucks but I'd rather be safe. I did try their 91 tune just a bit ago and drove around for a bit doing various pulls at different RPM's, stopping and starting etc and the car seemed to be doing fine. It made good power, no misfires reported. But again, until I get some diagnoses from them, I'm going to run the stock software to be safe so I again returned it to stock then test drove it again, no issues with either stock or theirs other than an occasion short dip in RPM at idle. The exhaust did sound a tad erratic, but then again I've never really listened to it for rhythm.

My biggest concern is that this point, what caused the cracked plug element and is there any hidden damage from it if there was knock that I'm not seeing (hence why I'm running stock software).

As far as the plugs go, there was no sign of fouling or anything that would cause hot spots etc. Looked clean around the electrode so it's a bit of a mystery at this point. Cars, I love and hate them at the same time lol.

 

[ypena02]

Get some NGK 6510s in there and call it a day. You may never know exactly why the plug failed but I have seen it happen from time to time with these cars and it seems like it occurs with all brands of plugs.

 

[TheLion]

Get some NGK 6510s in there and call it a day. You may never know exactly why the plug failed but I have seen it happen from time to time with these cars and it seems like it occurs with all brands of plugs.

Thats what surprised me is the number of people with f 150 ecoboosts and cracked plugs. I already miss the brisk pugs though because the idle is rougher with the factory iridiums. Thats exactly what brisk told me when i was researching is most of their customers reported smoother idle.

Im currently going to see if they will send me a replacement set. Ive seen one or two other posts for mazda speed 3 owners who cracked them putting them in and brisk sent them new ones free of charge even though that was user error. I dont believe my case is user error.

And boy do i miss the lms 93 tune...going back to stock is hard to swollow, that top end is so anemic (it was even worse without the ic upgrade) compared to the tune. Now having used their 93 stage 3v7 for several months the top end and throttle response is the biggest difference thats immediately apparent. Theres deffinately less low end and midband power, but low and mid were never the weak points of the stock power curve.
.
If i cant get assistance from brisk i may try the ngks or densos which lms seems to like.

 

[Juben]

Usually detonation causes plugs to drop insulators. Improper installation, i.e. over torquing the plugs, can cause problems too. Like ypena02 said, it's happened to every brand of plugs out there. I'm on my 3rd set of Brisk plugs (just on this car) and haven't ran into any issues with them thus far.

You might have gotten some bad gas or something. That's one of the reasons that I love my Accessport. If I'm picking up knock, I'll notice it on the values and know something is going on.

 

[KrisR]

Replace the plugs and log a pull to check to see if timing is being pulled, also check OAR to see if it's still at -1.0 like it should be. You could send the log to LMS to have them verify everything looks good.

 

[TheLion]

Replace the plugs and log a pull to check to see if timing is being pulled, also check OAR to see if it's still at -1.0 like it should be. You could send the log to LMS to have them verify everything looks good.

Can I get timing and OAR values from the Torque Pro app? I have that and OBDIILink MX for data gathering, haven't tried to go in depth with it much yet however as I haven't needed to.

 

[TheLion]

BTW I think bad gas is highly unlikely. The last fill up was at a brand new shell station, their 93 v-power. It's not a small station either as there are 10 pumps, its right off a highway and next to two large car dealerships (a dodge and VW dealership). Ford dealership, also very large, is right under the bridge, so between the 3 large dealerships (each has at least 100 cars on the lot) and all the traffic passing through town they probably maintain a fairly consistent volume.

I really don't think bad gas was the issue, it's not impossible, but highly unlikely given the size, volume and age of the station (literally it's been there for about 1.5 years), unless of course they re-used the underground tanks as there was an aging BP in that location prior to, but it was pretty tore up so I think it's all new.

Also, is there a difference between LOR (Learned Octane Ratio) and OR, or is it just two terms for same variable? Obviously they both deal with detonation resistance of the fuel, but I've seen both terms used but never a clarification if there's a difference.

 

[TheLion]

BTW I think bad gas is highly unlikely. The last fill up was at a brand new shell station, their 93 v-power. It's not a small station either as there are 10 pumps, its right off a highway and next to two large car dealerships (a dodge and VW dealership). Ford dealership, also very large, is right under the bridge, so between the 3 large dealerships (each has at least 100 cars on the lot) and all the traffic passing through town they probably maintain a fairly consistent volume.

I really don't think bad gas was the issue, it's not impossible, but highly unlikely given the size, volume and age of the station (literally it's been there for about 1.5 years), unless of course they re-used the underground tanks as there was an aging BP in that location prior to, but it was pretty tore up so I think it's all new.

Also, is there a difference between LOR (Learned Octane Ratio) and OR, or is it just two terms for same variable? Obviously they both deal with detonation resistance of the fuel, but I've seen both terms used but never a clarification if there's a difference.

What I'm really interested to see is what explanation LMS provides. Most often knock is related to timing, fuel quality or heat (hot spots). My memory isn't perfect on the events of the day it happened, but it was about 75 and low humidity, perfect day. Was driving to 11 am service and was having fun on the back roads to, did some WOT pulls in 1st and 2nd, partially into 3rd, kids love it, call it the roller coaster car. I did not notice any rough idle on the way there and it's only about 9 miles with mostly back roads, so only a few stops.

But I don't recall any abnormalities until after I came back out of service and started the car. So it must have cracked during one of the last pulls. Either that or it fatigued, had a hair line crack and then the temperature change finished separating it as it cooled down in the parking lot because I really don't recall any odd behavior until it failed. Once that misfire started the idle was very rough, shaking the whole car (that 30lb flywheel really moves things around). I remember thinking to myself "oh crap, what just broke". First thing I checked was for codes and got 0304 and 0316, which is consistent with misfire conditions.

First plug I decided to pull was the rear most (closest to the fire wall), dunno why but my intuition said "that one is responsible" and amazingly enough it was the cracked plug (I believe that's Cylinder Number 4). Insulator separated pretty cleanly at the base and slide down. No blow out at the top, no arching marks at the top either. Plugs also looked clean, no fouling or oily residue on the electrodes or grounding lug. I also scoped the cylinder as I was certainly paranoid of a cracked cylinder head, which is the most common result of constant knock or a super knock event. That is the whole reason I undertook the TST optamization thread, to counteract heat in every area I could before putting a tune on (FMIC, 160F Thermostat, Mishimoto 2" Performance Aluminum Radiator, 1-step colder plugs).

Brisk also has a RR14S, which is a 1-step colder non-projected tip variant of the RR14YS: http://www.jegs.com/i/Brisk+USA/084...=c&gclid=CL_U5-Pv4c8CFYpehgodRCgIUA#reviewTab

Here's what they suggest in their description: RR14YS (projected tip- low boost forced induction app.), RR14S (non-projected tip-good for nitrous or forced induction) Two step colder: RR12S (non-projected tip) – High Compression, Forced Induction and/or Nitrous Three step colder: RR10S (non-projected tip) - High Compression, Forced Induction and/or Nitrous.

I may try the RR14S this time around as the factory plug is also a non-projected tip type. Although this would only resolve the issue if it's heat related, not timing related or fuel related. I also contacted Brisk about this to see what they say. I would think 20 PSI is considered high boost application...most NA engines converted to FI only run around 8-12 unless they are built bottom ends.

 

[TheLion]

I don't think this was pre-ignition, but rather detonation:

https://www.briskracing.com/spark-plug-diagnosis

I did notice the surface of piston #4 looked a bit rough in a few spots, there where several spots which reflected the scope camera's light and appeared to glow, two or three spots at the center of the piston, one towards the outer area, but not at the edge. There's only one reason I would see this in the camera, light is reflecting at a different angle in those areas due to deformation of the aluminum (it's not chunks blown away, but compression divets)

These are from the detonation event or several of them one of which finally broke the camel's back so to speak (the spark plug insulator). Detonation is far less concerning than pre-ignition if you resolve the cause.

So from the evidence I now have, it appears I have experienced a few detonation events and not pre-ignition. Knock is detonation, not pre-ignition which almost always results in cracked pistons, bent rods / bearings or damaged wrist pins.

So way back in my TST optimization thread before I got greedy...I was going to run a 91 tune but use 93 octane as a buffer (probably the most important aspect) to essentially eliminate any risk or pre-ignition or knock (detonation) as neither is exactly desirable. Unless there is some other way to mitigate this risk (aka, LMS may adjust their software on their 93 tune a bit if I present this evidence), I think the most I can reliably run is a 91 tune, given the variations if fuel octane (while it's highly controlled, it's not 100%, there is some permissible variation and running on the ragged edge may cause issues).

From what I see on Brisk's diagnosis, knock is not caused by the plug, but by the fuel octane rating variances and the sensitivity of the knock sensors (or in the case of PCM software, how it manages knock based on the knock sensor feedback).

Modern engines in order to reach the maximum fuel economy run on the edge of pinging. They use knock sensors to "listen" for the pinging and retard the timing accordingly. However prolonged and severe detonation will hurt the engine and the spark plug, and is probably the most common cause of failure due to inconsistency of the octane rating of gasoline

Here are the plugs, they appear to be in the proper heat range as far as I can tell.

 

[arghx7]

uhh dude, this is classic LSPI (Low Speed PreIgnition) symptom. Just because it didn't blow up anything else (melt a piston, whatever) doesn't mean it's not pre ignition.

 

[Juben]

There's no doubt to me that the culprit was more than likely detonation. Like I said initially, that's usually always the root cause with plug failures of this type and nature.

I'll be honest with you about your fuel choice. I'd try another brand. I've recorded hundreds (by now) of datalogs and Shell has ALWAYS been a consistently poor performer. Some people think that it's attributed to their additive package in their fuel, but for me, I'd run no name gas before I'd put Shell in my cars. Do as you please, but I'd recommend trying a different brand. If you don't believe me, ask Adam about Shell sometime and wait for his response.

 

[TheLion]

There's no doubt to me that the culprit was more than likely detonation. Like I said initially, that's usually always the root cause with plug failures of this type and nature.

I'll be honest with you about your fuel choice. I'd try another brand. I've recorded hundreds (by now) of datalogs and Shell has ALWAYS be a consistently poor performer. Some people that it's attributed to their additive package in their fuel, but for me, I'd run no name gas before I'd put Shell in my cars. Do as you please, but I'd recommend trying a different brand. If you don't believe me, ask Adam about Shell sometime and wait for his response.

Here I was thinking Shell was the premium fuel due to their additives. From testing I've seen, their fuel does keep things cleaner as they have nearly double the additives, however it would seem their additives may influence the octane rating or burn characteristics of the fuel enough that higher performance TDI engines have issues. In fact, it would seem they increase chamber temps (it's possible they influence heat transfer).

I know Ford endorses BP with their additive package, which is quite a bit less than shell's. I suppose there might be something to it and it may not necessarily be LMS's fault, but I'm still waiting to hear back from them.

No, it's NOT LSPI. LSPI is pretty much always catastrophic, LSPI or Pre-ignition will typically result in immediate failure (piston can take at most a few cycles before failure), aka cracked piston heads or broken lands usually result in EB engines as they use hyperutectic pistons instead of forged.

Once that happens you'll get constant misfire, smoking exhaust, excessive heat...the engine won't run for long in those conditions. It's going to be "knocking" very loudly as well due to piston slap (which is what I noticed when it was misfiring while idling, it was very clanky as the rings weren't being expanded in #4 due to failure of detonation, you know the hollywood dying car engine sound).

Diesels are notorious for that metallic knock sound (aka piston slap), especially when cold. I obviously drove to work this morning, drove around yesterday both with LMS 91 and Stock doing some WOT pulls and cruising, watching temps, oil pressure, boost etc. I'm pretty sure if there was a hole or fracture in the piston I would have experienced complete failure, especially after running their 91 octane tune and doing a bunch of WOT pulls...and then doing it again with stock.

 

[TheLion]

uhh dude, this is classic LSPI (Low Speed PreIgnition) symptom. Just because it didn't blow up anything else (melt a piston, whatever) doesn't mean it's not pre ignition.

In fact this is often something that's misunderstood. Exhaust temps actually decrease with detonation / knock, not increase. People assume because it's running cooler it's always better...the most common failure of detonation is broken insulators as we have all stated.

Basically the remaining fuel residing close to the piston head detonates after the bulk of the fuel has detonated. The piston is still traveling downward at this point, so some of the force can be absorbed by transfer to the crank and to the wheels. Essentially the blast of the second smaller detonation happens very rapidly and "hammers" the top of the piston head. This can be caused by hot spots or inadequate thermal management OR inconsistent octane ratings. Run 87 octane with a Tune+ or LMS 93 tune and see what happens...

On high boosted or high power density engines it can result in damage if it occurs continually over just a few minutes and timing isn't adjusted accordingly. So far I don't think we have had a single 2016 fail, I have a may 2016 build and certainly have taken quite a few steps many others have not to mitigate heat (160F thermostat, 2" Mishimoto radiator, 1 step colder plugs, FMIC). Most people do just plugs and a FMIC but leave the radiator and thermostat stock, so I think Juben may be right, it's more fuel related and from what he stated about Adam's testing, there may be something to it.

Sponsored