PAVGR

iTrader: (2)

Sorry to see what happened to your car buddy.I thought Rrracing sc kit wouldn't have issues like that.I ve been there with my LMS kit and I burned a valve last summer.I m trying to tune my car for 4 months now with OVtuning.I hope I get lucky this summer.I want to ask you guys about the spark plugs.Did you change them to colder ones or you stayed with the OEM ones?I ve found HKS sc kit uses HKS one piece spark plugs colder than what I have now and I don't know if I should change them or not.

Justin727

You're correct. I should've said Wideband gauge because there isn't a way to monitor it without anything else.

RRRacing

iTrader: (1)

If I may share my thoughts on piston failure....

We took apart Clayton's engine, and there are ZERO signs of detonation on the piston crown, plug, or anywhere else.

This particular engine failed *within several hundred miles* of SC installation. We do not have an exact root cause, but the simple fact is that our shop ISF now has 62,000 miles on it, with more track time, dyno time, tuning trial/error time, and over 10,000 miles on the Supercharger. Many folks have seen the car perform in person, on the east coast, and west coast, and the reliability is real. During the course of tuning it for the SC, we ran into numerous fueling issues that resulted in us running it very lean MANY times.... I actually wouldn't have been surprised if we had a problem, but we simply do not. As an example, those of you who joined us at Pocono Raceway for our initial SC track test, and Fontana for our second track session know that in both instances our IS-F shut down after about 15min of track time and went into limp (until I cleared the code and all was fine again). The reason was that we were pushing the DI system beyond its capability, and although everything seemed good on the dyno, the DI system simply could not keep up at the track, under repeated aggressive shifting and fuel enrichment due to high temps. We were actually running 13-14 AFR at full boost at high RPM's until we figured out and solved the problem!

As part of our periodic inspection of the car, we recently conducted a leak and compression test... All cylinders measure 210psi and 2-3% leak. On Clayton's car (with the piston failure), even the good cylinders measured about 180psi, or about 15% below factory spec.

Based on what we learned from the first batch of kits, whenever folks call us asking about the SC, we tell them to at least measure the compression first. So from data we have from numerous IS-F's so far, we are seeing 2 populations of compression numbers.... some guys are at about 180+/-5psi, and some are at 210+/-5psi. So clearly there is a difference in the condition of engines out there, particularly as IS-F engines are starting to see higher mileage. Our advice is that if your engine is showing 180psi, and you are considering an SC, you should budget for possible piston replacement (rather than spending the last penny you have on the SC, as tempting as that may be

There are numerous potential causes of ringland failure:

1. Detonation -- if you do a search on ringland failure you will find many results related to subaru EJ25 ringland failures... typically those failures are detonation related. That said, our ECU is so much more effective at quelling det, that I don't think this is the issue. I am attaching a pic of Clayton's piston below, cylinder #7 which is the one that failed first, and had the worst leak/compression numbers.
2. Weak cast pistons -- perhaps, after a certain amount of abuse, particularly high temperatures, stock cast pistons simply cannot handle the added cylinder pressures of forced induction. This particular engine was extensively tracked. In fact, I have a video of the machine shop decking the block, and it is clear that there is some distortion on cyl #7. The initial pass at 2 mil was not enough, and we had to make another pass to level the deck.
3. Another theory is that banging off the rev limiter, in the long run, can damage ring lands. We all know that the conventional wisdom on this forum is that there is nothing wrong with doing this and that the car is actually faster in the 1/4 mile if you do this. I don't know if the car is actually faster if you bang off the limiter, but it certainly cannot be good for the engine on the long run.
4. The drivers and passenger side pistons on the IS-F are not the same castings, so perhaps there is a reason drivers side are more vulnerable to ring land failure. That said, I had a drivers side catalytic converter failure a while back at the track once on my IS-F, and I gather its because taking CW turns most of the time tends to puddle the PCV oil towards the drivers side bank. This was one of our motivations for designing an AOS for the IS-F... and in fact, there are many in the Subaru community who contend that an AOS is helpful to avoid ring land failure.
5. I suppose running lean could cause excessive EGT and piston damage. We recently discovered that due to the fueling demands of the SC, the stock fuel sump suffers from fuel starvation below about 1/5 tank even in straight line acceleration. Additionally, we have an in tank fuel filter that could potentially become restrictive. While this restriction is not an issue on an NA car, once you start flowing 500+whp of fuel, a partially clogged fuel filter may become an issue, so it is advisable to change your in-tank filter, even though that is not a typical maintenance item.

On a positive note, a piston ringland failure is one of the better ways an engine can fail because it is not catastrophic. Just ask the numerous BMW E92 M3 owners who suffer uncontained rod failures on NA and SC applications. Once you have a hole in your block, your wallet just got a whole lot lighter

And here is the pic of the top of cylinder #7 piston (they all look identical to this actually):

lobuxracer

iTrader: (2)

If the ring lands cracked, there was detonation. The tops of the OEM pistons in my FJ1100 looked perfect too, but the ring lands were cracked the same as the previous picture. This happened to me immediately after adding 2 degrees of full advance timing as I was seeking optimum ignition timing. Pistons don't break like that without some kind of shockwave, and detonation doesn't always burn a hole in the piston crown.

I don't think you are stating the right numbers from the machine shop. If the machine shop removed 2 millimeters from the deck and the pistons are not sticking out the top at TDC, this engine was a dud from the factory. If you mean they removed 0.002", then likely just fine, but if they removed 0.080", the squish was non-existent and this engine was garbage off the assembly line. It certainly is possible this happened, and I have seen it more than once, but your comment about compression pressure is telling. If the engine has a low absolute number, it's very possible the deck height is too high and there is no squish going on at all. Engines like this have low absolute numbers, have high sensitivity to fuel quality, and make less power than a properly decked engine does. If the deck height is good, there should be no carbon at all on the squish areas of the piston when you disassemble the engine without doing any kind of clean up. If it is not, you'll see carbon deposits and you'll know it's time to do some careful measuring and machining.

I suspect stacked tolerances are responsible for a lot of the variability we are seeing in the field. I also suspect this particular engine wasn't one of the better ones from the factory. Since the entire bank was lost, I really would be inclined to believe the fundamental machine work done by Toyota didn't offer the kind of precision necessary to do this kind of modification successfully. If you are asking customers to measure compression (which is always sketchy as it is battery/starter dependent) to determine if they are low because there are problems with engines producing low numbers, it sounds like you really need to know if the engine was properly machined at the factory. Odds are it isn't optimal, but is acceptable for stock power. In my experience, it's never optimal, but is by default in favor of surviving warranty.

RRRacing

iTrader: (1)

First, there are other causes of ring land failure other than detonation. The stock cast ring lands are thin and weak, and we already know that this engine saw significant track time previously. The kind of heat that a piston will see under track conditions is much greater than under normal conditions, especially if the car sees high oil temperatures as well (which is why I feel supplementary oil cooling is critical for track days in hot climates). The long term effect of high heat exposure results in embrittlement of the aluminium, to the extent that higher cylinder pressures combined with existing intrinsic defects such as casting porosity could cause ring land failure, even in the absence of detonation.

Regarding the decking of the block, I didn't say "2 millimeter," I said "2 mil." 2 mil equals 0.002"

We not only ask customers to measure compression, we provide them with the factory manual procedure, and we have also done in-house compression testing on cars we get in the shop -- we see the same trend in our own measurement. BTW, I have tried compression testing using several methods (open and closed TB), and over varing crank durations, and the results are surprisingly consistent.

that said, I agree with you that deck height is a possible cause of failure, and something we should look into!

-Rafi

Isfast

iTrader: (13)

I finally had some time to do the compression test. Averaged about 200 / cylinder. This weekend I'll data log and send to Rafi. I have already driven 1500 miles fairly hard. Don't want to jinx myself but no issues so far. Car is a blast to drive. More than the total power, the response at low rpm is great.

tas02

iTrader: (2)

Thanks for the update, Isfast.

juanmedina

That sucks about your stock pistons. We know that the IS-F even on the NA configuration likes to pull timing because it sees knock. I think people should run this high horsepower setups on E85 or with meth injection at similar boost level before claiming that the stock pistons are weak.

This reminds me of my subaru days... everyone cracking ring lands left and right for running aggressive maps with improper cooling on pump gas. And later people running E85 on 100% stock long blocks and not having issues with lots of miles. Some making 500whp+ and making low 10sec passes.

530whp is nothing for 8 pistons and this is a toyota/lexus that are known for over-designing their components. If the pistons are really this weak why would anyone bother to modify this car. 10K for a supercharger and another 10k for build engine for 500whp no thanks.

RRRacing

iTrader: (1)

The problem is not the pistons, the problem is the history of the car before the SC was installed. This engine had something like 20 track days in southern California, with no supplementary oil cooling.

Juan, you are comparing seriously tracked and higher mileage ISF to drag strip 500whp STi's... drag racing does not put massive heat into the engine/pistons, there is no comparison.

a 500whp EJ25 may hold up a bit at the drag strip, but I bet you I can grenade it in a single day at the track, e85 or whatever you put into it

BTW, in addition to the 10,000 tracked miles we put on our 62000mile ISF since SC installation, we have a local customer who has already exceeded 5000 miles with the SC, without issue. The engine with the cracked ringlands failed within a few hundred miles, so clearly there are other factors at play here.

Rafi

juanmedina

I see what you are saying. There may have been some previous issues.

By the way that Subaru that run 9sec in 1/4th was a time attack car that run a few track days at TMP and at Watkins Glen on that same stock short block.

https://forums.nasioc.com/forums/sho....php?t=2497782

RRRacing

iTrader: (1)

Using that standard, I could probably start out with a low mileage 2UR-GSE and run it at 800whp, "for a while." Then of course I would have to post on the forums about it to make it a fact

Between 2003-2014 I figure I tuned about 150 Subaru's, mainly EJ20's and EJ25's. I also owned a WRX and STi. All things being equal (mileage, condition, usage) you cannot compare the reliability of a POS EJ25 to the 2UR-GSE! I cannot imagine any REPUTABLE Subaru tuner recommending to run a stock EJ25 to anywhere near that power level.

The entire Subaru aftermarket is based on people modding their Subies, blowing the engine (often times with very low mileage and street use only), and then going to the hundred or so vendors out there that make money rebuilding boxers

Rafi

DaOne54

According to Lexus standards, 203 psi is standard, Minimum compression is 145 psi. It also states that there shouldn't be a difference greater than 14.5 psi per cylinder.
But according to you Rafi, 180 is too low for a SC instal? Better off rebuilding the engine at 180 psi? instead of SC install?
Now, what about us performing compression tests at higher elevation? Im at 5,000 feet above sea level.

Haven't performed a compression test on mine yet, but I plan on it this weekend.

MileHIFcar

I was thinking this as well........ I'm at 5,280+ so what's the variation for psi @ our altitude when checking compression up here?

fwdGTP

It's a c​​​​ombination of because we like this car and didn't know of it's shortcomings at the time. There are not many high HP ISFs out there. I dont think it's only related to overly aggressive timing maps on Subarus, there are very few 300+whp Subarus that last even remotely long, I just can't imagine 500whp on stock bottomend. That being said our cars were only built to be NA which becomes very apparent when you start to make power. I'm not sure on the consistency of our engine's build tolerances either. Since a few are holding up fine and a few are having issues. Also I'm wondering if even NA the motors take "damage" but never really get to a point of actually breaking.

RRRacing

iTrader: (1)

Our cars can take the power but, generally speaking, how many 80,000 mile cars, with long history of hard use, can take a 200hp increase in power?? You guys think that an 80,000 mile LS engine or Coyote v8 with long history of track use would fare any better after?? No way. I wish I kept a log of Vette/Mustang engine failures at the track Most track c5/c6 vettes are 40-80k cars that get converted to track use, and are not modified with big power adders, and yet they fail -- in fact, there is an entire aftermarket performance industry based on re-building these engines!

Street driven LS's and Mustangs with lower mileage and without a history of hard usage will fair well with forced induction, as will the 2UR-GSE..... but let me repeat what I've said... We have 62,000 miles on our shop ISF. with many track days, 10,000 miles on the SC, and many track days with the SC. We are measuring 210psi compression on each cylinder, with 2-3% leak. Cars that saw the ring land failure are measuring 15% less compression to start!

Rafi