Brian V
USA Retired
- Joined
- Apr 27, 2015
- Threads
- 21
- Messages
- 986
- Reaction score
- 159
- Location
- Native Earthling
- Vehicle(s)
- 2015ecoboost premium 201A Nav Sec Race Red
GM Dexos is a GM Trade mark and has to be a buy in license.
Sponsored
What does that translate into in degF?
When I cracked that Brisk RR14YS plug insulator it was on #4....could tuning strategy possibly be playing a role rather than mechanical design?#1 and #2 show the most knock count, #3 trails those cylinders slightly, and #4 never does get any knock count. Could be that fuel is fed starting at the rear cylinder. The center cylinders will get the most heat as there siamese to the rear and front cylinders, where as #1 and #4 should have better cooling with more water around the cylinder. I would assume that the two cylinders that would have issues would be #2 and #3. It all has to do with knock count, the cylinder that knocks the most is the one that's gonna have a problem, hence why #4 never has any issues. This is just my observation on my own car, other cars could be different. Always over octane people, and dont ask your tuner for to much low engine speed TQ.
That's the only way to detect knock is with knock sensors....it's also where the greatest power and fuel efficiency is made. Most modern engines actually run in very mild knock (more commonly NA and low boost) or just on the very edge of it (high boost engines) even at light throttle (for maximal fuel economy).I agree, I think its cooling issue, but also looking at were the knock sensors are located.
#1 between 1st and 2snd cylinders and #2 by 4th cylinder, so #3 doesn't have one right by it.
Not that you want to rely on knock sensor for timing retard as there always a delay (it has to knock first).
I agree with most of what you say here, my comment on not wanting to rely on knock sensor was at learning stage of ECU.That's the only way to detect knock is with knock sensors....it's also where the greatest power and fuel efficiency is made. Most modern engines actually run in very mild knock (more commonly NA and low boost) or just on the very edge of it (high boost engines) even at light throttle (for maximal fuel economy).
Mild knock won't cause any harm even over the life of the engine. Even severe knock or detonation you can run a couple thousand until the piston either melts or fatigues (or the or the rod / wrist pins could eventually give out too).
There's no reason the tune shouldn't be able to detect knock and scale back timing / fueling / air before it becomes detonation / severe knock unless it's running so aggressively it goes into detonation or even pre-ignition which often leads to mega knock / super knock.
Also the knock sensors operate by detecting the resonance of the block due to the end gas ignition (aka knock). It doesn't matter which cylinder is knocking in terms of detecting it, the entire block resonates when knock occurs in any of the cylinders because of the energy transfer. Think of the block like a giant tuning fork.
The offending cylinder is known by the PCM because of the ignition timing, the PCM knows when it pulsed the plug and expected an ignition, since only one cylinder is firing at a time the PCM knows which cylinder is knock.
Our ecoboost engines have two knock sensors. One sensor is tuned to detect knock and the other is tuned to detect pre-ignition which results in multiple knocks (different frequency range). Pre-ignition leads to mega knock / super knock / LSPI and isn't always necessarily a death sentence if caught early enough.
To my knowledge Ford uses a windowing strategy to help predict mega knock / super knock (aka LSPI) by pre-ignition events. The more severe the pre-ignition the more aggressively timing is pulled and fuel is dumped into the offending cylinder. There are over 480 tables with the factory ECU, some of them can be changed to a degree with out negatively affecting others, but it's a very complex set of control software and that's why I think working for Ford Engineering when tuning is so critical for street car tuning (race cars is a different story, you only need to worry about how the car will run that day which can be done by tuning on a dyno).
Just to be clear there's a lot of different terms for one of two events.
- Knock which is end gas auto ignition which occurs after the bulk of the air / fuel mixture is burned up, can range from mild (not harmful) to severe (harmful over extended periods of time). Typically caused by heat on the piston surface as a secondary ignition. The compression wave from the primary ignition causes some separation or pockets if you will of air and fuel that have not been ignited.
- Detonation is a severe form of knock and is just another term used to describe a high intensity damaging knock, can result in cracked spark plug insulators and pitting of the top of the piston. It is only permanently harmful if it's not corrected and aloud to continue as it can over a few thousand miles of constant detonation crack pistons or damage rods etc. Modern engines are built to withstand multiple repeated detonation events without permanent damage.
- pre-ignition is the auto ignition of the air / fuel mixture prior to spark ignition. Again this is not necessarily a death sentence, pre-ignition could occur just a few degrees before spark ignition and acts like slightly advanced timing, it often results in knock or even detonation that ranges from mild to severe depending on how advanced the auto ignition occurs. Caused by hot spots or more commonly from heated deposits of sufficient mass to retain enough heat to ignite the air / fuel mixture during the ensuing compression stroke. Those deposits can come from fuel / oil mixing, thus lowering the flash point of the fuel or from solid deposits that detach from the rings / lands / grooves of the piston, valves, plug or injector etc.
They become heated during the power stroke from the primary ignition and because all of the oxygen and fuel is used up their thermo-chemical reactions are suspended until air and fuel is present on the next compression stroke in which they resume.
- after pre-ignition it was found that most commonly large quantities of deposits are dislodged from the rings, grooves and lands which then retain heat (you can see them glowing in images taken from bore scope cameras), these will often lead to LSPI / Mega Knock / Super Knock (all terms for the dreaded rod breaker event) if the PCM doesn't dump fuel / cut air and retard timing on the next compression stroke to "douse" or "choke out" the ignition sources. The windowing strategy for predicting LSPI is part of how Ford does that.
That's why I think anti-LPSI Oil formulations are so critical to highly tuned cars. While it's not a 100% gaurantee (it doesn't protect against defective or sub-par rods aka metalurgical defects, bad tuning, really bad fuel, improper modifications etc.), oil has been found in every study I've seen to be the primary offender.
Why? Because it dilutes the octane rating of the fuel or stated otherwise it lowers the octane rating of the fuel = bad. Now you have fuel that you thought was 93 and the PCM was programmed to expect 93 but is really 91 or even lower.
Some one mentioned an octane buffer and I think that combined with an oil cooler and more importantly a Dexos 1 Gen 2 rated oil. Gen 2 specification specifically tests the oil for LSPI inducing properties, aka does it highly affect the octane rating when mixed with fuel and cause pre-ignition in their test engine.
Eventually the GF6 spec will come out which is a more universal standard that's industry wide instead of just GM, but GM's Gen 2 spec is pretty decent and far more stringent than GF5 which does not even address LSPI related issues.
So oil can significantly contribute to normal knock as well as pre-ignition and LSPI.
Obviously it's not always the cause, but I think there's a fair bed that it is a significant factor in some of these failed engines. It's entirely possible that some of them are experiencing multiple pre-ignition events that lead to a mega knock event which fractures or bends the rod. In the testing I've seen, the second ring is usually the one that breaks off from LSPI, not the top, so you could fail a piston and run for a while until that chunk of piston starts wreaking havok in the crank case and the rest of the piston eventually gives way (this last part is only my speculation).
Here is the source of my information regarding LSPI, pre-ignition and knock, your welcome to disagree with 82+ automotive engines from some of the most respected engine and engine component designers in the world which contributed to the information therein including BMW, Mercedes, MAHLE, Toyoda, Honda etc.: https://www.iav.com/sites/default/f...ons/34242_iav_ottomotorisches_klopfen_ans.pdf
Here's an example of some things to think about:I had the FP tune installed on my 2015 Premium PP auto in April and will be off their warranty in December of this year. I have a Premium Ford service contract 7 year 60,000 miles. With the Ecoboosts having problems, I might have to rethink keeping the car. I just turned in a claim for heavy wear on the left bolster of my leather seat on the drivers side and was turned down by Ford! The tune is the only modification to the car and I am retired and drive the car pretty easy. Can't afford to replace a $7,000.00 motor! The car has 31,500 miles. I was surprised to see a FP car go boom!
Thanks Lion for all of your posts, great info!
I'm going to wager that modern knock sensors are pretty good. There's also TONS of processing power available as well in the ECU which wasn't just 10 years ago. But nothing is perfect and piston engines operate entirely off of historical data and prediction.I agree with most of what you say here, my comment on not wanting to rely on knock sensor was at learning stage of ECU.
AFAIK, with modern ECU, if knock is persistent, ECU will learn and back off whole timing table.
I didn't know on two frequency types of knock sensors used now.
I know in past they are pretty crude sensors (on paper it sounds great but reality is not so).
The early systems were prone to false positive detecting knocks, in fact Ford tried different positions of mounting them on older 2.3 and it never really worked well, but those ECU were fixed, no flash, no adjustments, no datalog off it either.
On detonation though IMO damage can happen fast, I guess it depends on how bad it is detonating.I seen first hand many pistons crack, break top lands.
These were all cast original NA pistons, and no damage to rods shown up.
The problem with rods is they slowly can bend over time, user/driver would not even notice it, of course if really stressed they can bend to point were interference happens and break.
I think most of ecoboost are piston damage first, then rod is unsupported so goes through block.
It would be interesting if it was easy to measure piston deck height with motor intact, to see if piston tops are dropping over time.
Could maybe use sparkplug as access point to measure.
Probably. I cracked that plug insulator on #4 the ONLY time I filled up on Shell V-power gas. I had never used it before (always ran BP or Speedways locally). Yes the plugs were running a bit on the hot side for the application unbeknownst to me, but I don't think it's coincidence that all the highly tuned cars see a lot of knock when using shell gas AND I just happened to have detonation crack a plug insulator after a fill-up on shell gas way back.@TheLion if you are correct about the Oil contributing to the event, isn't it also possible that a bad batch of gas would have the same impact if not worst when combined with the Oil factor? Bad gas could also lower the octane rating = bad. I am not sure if you are aware that the last 3 failures had filled up just prior to the event. 2 just filled up and one within 40 miles. I am not saying bad gas is the only factor but a catalyst.
"Because it dilutes the octane rating of the fuel or stated otherwise it lowers the octane rating of the fuel = bad. Now you have fuel that you thought was 93 and the PCM was programmed to expect 93 but is really 91 or even lower."
I would think that bad gasoline would be a bigger concern than oil.
https://www.knowyourparts.com/technical-resources/fuel-system/bad-fuel-fact-or-fiction/
Forged is way to go on FI, as far as durability.I'm going to wager that modern knock sensors are pretty good. There's also TONS of processing power available as well in the ECU which wasn't just 10 years ago. But nothing is perfect and piston engines operate entirely off of historical data and prediction.
Predictive methods aren't perfect because they rely on averages or expectations which don't always pan out. When you play the odds, eventually you will loose. The real question is, can you play the odds long enough and in such a way that you won't loose within the engine's normal life span....
Piston failure makes sense though. They are not forged, just your standard hyper eutectic pistons, so nothing special material wise. I'm surprised they didn't have forged pistons, but it all has to do with cost savings. We all know you can build a bomb proof bottom end and run these street car tunes until the end of time without failure, but the cost from a factory engine like that would mean massive profit loss for the OE's and of course most of these engines run reliably for their life on the factory tuning so there's no incentive to do otherwise.
Agreed, that's why I was suggesting running a less aggressive tune with higher octane. Fueling has a HUGE impact on safety margin as well. All engine timing strategies are predictive and "after the fact" based, the best hedge against damage is simply a large enough safety margin.Forged is way to go on FI, as far as durability.
I think they stayed with cast this time for tight piston clearance (emissions/MPG etc), just like they use low tension piston rings.
I agree and think it combination of many factors, for daily driver best you can do is hedge your bets and help the % in your favor.
Run 91 tune on 93, run good oil with catch-can, plug gaps etc, anything that might help in small way.
The other problem, no matter its still 4 cylinder with one firing per rev so you have long load time on the piston/rod for given HP and rev, compared to V8, 300+HP/TQ is impressive even in stock form.