Responses to past posts:

Tomas, good point on the effects of increased exhaust residuals for turbochargers run outside their efficiency range. It could be interesting to try to quantify the effect. On the positive side, this is a sort of Exhaust Gas Re-circulation (EGR), which reduces NOx and lowers detonation sensitivity.

Jeremy, I truly appreciates everyone contribution, but find it perfectly appropriate present arguments against explanations I cannot fully agree with, although my responses/arguments (due to stress) might have been a little terse and not inviting further discussion.
With respect to detonation, it is my contention that weak detonation (inaudible under low nose conditions) has no detrimental effect.
As you can see from below, I now believe that “hydro-lock” (while not being the full explanation in the traditional meaning of the word) was pointing in the right direction.


Likely cause of fracture:

I have had time to ponder on why my cylinder cracked and now have a reasonably good explanation, but first some more factual background.

I had an erratic combustion at WOT above 4700rpm (graphically explained in the upper right hand region of a load vs. rpm graph). At the time where the “incident” happened I was running at WOT at 5500 rpm for a longer period than done previously (maybe 5 seconds) and I would judge the combustion “success frequency” to be 50%, but have no strong clue on its distribution among the cylinders, except that there was no back-firing or popping, which would indicate even distribution among the cylinders.
I had also previously noted, that this combustion missing was a function of the amount of water injected and how hot the engine was (clogged water filter, downsized nozzles and hot weather were situations showing this).

At WOT and 5500 rpm the approximate flow of injected fuel into each cylinder of the engine is 8 cc/sec (assuming 325 FwHP and water is 0.7 cc/sec (0.7 gph nozzles). With a 8:1 compression ratio and 625cc cylinder displacement the combustion chamber volume is about 89 cc (@ TDC), which in that situation would take 10 seconds to fill up with liquid if none was exhausted.

It is also worthwhile to note that our engines are slanted and the lowest point is part of a squish area. At the same time if water is injected as liquid, it takes some finite time to evaporate. Another “forensic” observation was missing thermal coating (on piston top) at the squish area on pistons 1-3 and #4 had it missing at the center. You have probably already guessed what I am aiming at.

I theorize (and believe) that with the amount of water injected, not all of it had time to evaporate and some got trapped and slowly accumulated in the squish area and eventually “spitting” on the sparkplug and newly formed combustion, essentially extinguishing a significantly number of combustion events. With less heat generated, more water and likely fuel as well stayed liquid with the potential or generating a hydro-lock. However since the damage happened only after 5 seconds (of erratic combustion) it was not a hydro-lock in the traditional sense. Very likely the amount of water needed to be removed from the squish area was large enough to create a very high twisting moment on the piston with a resulting cylinder crack from the piston top pushing on the power (exhaust) side of the cylinder. The missing thermal coating in the squish is an indicator of this extreme pressure and flow.
It is also interesting to note that the “cylinder cracking event” could not be felt in any way (jerk, knock, bang, etc), very likely since the piston speed close to TDC is almost zero.
While the water very likely was the cause of the missing combustions, it contributed only very little (about 10%) to the partial hydro-lock itself. The fuel was the main contributor to this.

I strongly suspect that the mechanism behind other peoples cracked cylinders also can be traced to missing combustion with continuous fuel delivery providing the piston twisting partial hydro-lock.


So what can be learned and done to avoid this in the future?

For port water injection: At high rpm and boost, reduce the mount of water injected, keep the atomization high and make sure that the engine is thoroughly warm.
As a consequence it may be a disadvantage to thermally coat the combustion cambers since the accumulated heat on the piston dome will aid in water evaporation (at least for awhile).
It is also quite counterintuitive that the engine can tolerate less (liquid) water in the high rpm-boost region.

For all of our engines: Take missing ignition (combustion events) at high rpm-load very seriously and don’t let it happen for more than a couple of seconds.


With that, my faith in the mechanical strength of the engine has been restored. It just has to be managed appropriately.

 

Then, Laust, why ask us?

 

Kevin, what's your point?
As I stated before, I don't think it's possible to increase the flow through a turbo beyond a certain point. I also agree that Laust was a bit agressive in his posts, but so would I probably if I just blew my engine.
I do however think that it's a good thing when someone dares to go beyond the mainstream thinking and push the limits. This forum is full of people who would buy any junk, without any forethoughts, that has a semi known tuner brand and a loose promise of a 30 hp gain. It also seems that if enough people repeats an opinion it becomes the truth regardless of facts.

Tomas

 

This is maybe what Kevin means:

A fellow lister posts his plans and/or asks a question.

Joe Blow replies.

The fellow lister can:
a) totally ignore Joe Blow's reply
b) base his whole life on Joe Blow's reply
c) ask Joe Blow about the content of his post

a) and b) are not suitable in the quest for learning

 

You asked what my point is, Tomas?

When somebody actually takes the time to filter through the wordiness and plagurism of Laust's current and prior posts, they will come to the same conclusion, some based upon personal experiences.

And just stay at the same level of analytical sophistication as you present Laust, here is the root cause for the “disaster”: this block cracked due to 25psi from a k26/6 on 89 octane fuel, with an "o" ringed mls head gasket, and more than likely a lean condition. Obviously the water injection "band-aid" didn't suffice.

"Weak detonation inaudible under low noise conditions not detrimental"? Are you kidding me? Come on guys...

That, however, is not the issue. Anybody that offered their opinion (which mine seems to be the consensus) was subject to belittlement, and responses/arguments being quite terse and not inviting further discussion.


That. Tomas, is my point.

 

I do belive that it's possible to run 25 psi with a correct set up water injection but I also beleive that it's a pointless idea to do it with a K26/6.
I also believe that there could be other reasons to the cracked block than detonation.
Yes, I agree that Laust was way to agressive in his responses.
What I object to is the "monkey see, monkey do" mentality here on RL. After a couple of hundred posts which (correctly) says that lean conditions causes detonation which blows head gaskets, then Joe Blow (not you Tommy) logic gives that a blown head gasket must be caused by lean condition, there can be no other reason. If someone tries to come up with other reasons they are made fun of.
It seems that people can make hundreds of posts regarding the performance gain from monkeys without a negative comment from anyone.
In that perspective I find it unfair that Laust gets bashed here and in other threads, just because he dares to challenge mainstream knowledge.

Even if I don't agree with everything that is said, it still is these technical threads that I find interesting here at RL, and it's the threads about the performance gain from monkeys that makes me stay away for a couple of weeks.
Tomas

 

ouch thomas what did the monkies ever do to you.

I take offense to that because it's just a joke and a throwback to a old saying that me and my friends say often. If you can't read the goofy fun threads and laugh then I don't know what to say. Not every thread needs to be the technical be-all of existance.

BTW I still have 2 dozen monkies for sale if anyone wants one. Mine haven't been working lately cause the car has been sitting. As soon as she get's back on the road I will be putting these damn monkies to good use again.

peace

 

lol you got 2 dozen monkey's in the end? Hell, your AF's may be way out if you didnt get 1200cc injectors.

 

I didn't realize that Laust's blown motor was a goofy fun thread. I guess some of us were too serious about the whole thing.

 

Boy this thread keeps getting more interesting. More monkeys?

Laust, while I think we'll have to agree to disagree about inaudible detonation not being detrimental, I do think that the Hydrolock theory is reasonable and on the right path if you're against the detonation theory. I'm not sure if that much water can accumulate from WI but your calculations are better then my "theories" at this point. I would think much of that low volume of water would be scavenged/forced out of the cylinder with each exhaust stroke and the remainder would be evaporated by the heat. Considering it should be a vapor at that point, I would think it would be gone or close to it at the end of each cycle.

The standard 2.5L block has no reinforcement in the water jacket area which makes it at risk when any non-compressible material enters the CC as oppose to just bending a rod (I've done this when I had an injector stick open using a cast rod on a block that had a reinforced deck).

Another "thought" and possibility is the air temps caused by the 26/6 at that boost level raised the cylinder pressures enough to cause some head walk/lift and allowed coolant to enter the # 2 CC on the exhaust stroke. That amount of coolant would crack the block in a heartbeat. It may also explain the polishing on your HG. Was there any indication of head movement on the head at the washer area?

I've seen ARP studs stretch enough on FI applications to allow coolant to gush from the front of a cylinder head under high cylinder pressure applications (this wasn’t on a 951). The funny thing was once the head reseated off of boost, it ran perfect. The only sign was that there was a problem with the motor is there was a slow coolant loss and took quite some time to find.

Anyway, my initial objection to this thread was the assumption that the block had a low tolerance for TQ and caused this tragedy as opposed to detonation, hydrolocking, foreign matter, etc. At the end of the day, I would love a definitive answer to this failure but without blowing up multiple motors with controlled changes in between each one we may never know. Good luck.

 

Detonation, gauranteed. 911 Turbos do the same thing when you get them knocking.

I had a customer engine bend a turbo when the turbo catastrophically failed and flooded the engine with oil at high rpm. No cylinder damage.

Laust's recipe is the perfect mix for detonation, WI or not.

Chris Cervelli
Premier Motorsports

 

Chris,
Could you be more specific.
Do you mean by boosting 25 psi on a 951 or boosting 25 psi with a K26/6?

Tomas

 

I'm sure Chris was referring to boosting with a 26/6. His GT 25 set-ups were running 25 psi with race gas.

Chris, I agree with you as well but it seems to be a very unpopular response.

 

I guess I just don’t get it.
I have run into several folks that are trying to get the ‘max’ out of a K26/6, some to the point of spending well over $5k in engine management and other mods (I am not talking about the original poster). It just doesn’t make sense to me, why not just get the right turbo for the job? I am not trying to be a pain here….really, I just don’t get the concept.
As I mentioned earlier in the post – It was detonation. A K26/6 running at that pressure has to be cranking out some extremely high air temps (my guess is that you can exceed 300F but I have never tried to push a K26/6 that hrad), possibly far beyond the range of what a water injection system can deal with. I have a block that looks just like that – it was due to an accident with the timing the added 15 degrees (causing some very brief nut very serious pinging!). It looks exactly the same.

Chris White

 

Yes, there would have been very high temps; more like 400F.
But water has HUGE heat reduction capabilities. It also dampens the burn. A properly engineered WI system (can be very simple) can deal with anything you throw at it and practically gaurantees never any detonnation.
The problem here is that WI cannot make a turbo flow more air or fix a huge back pressure problem.
Unevaporated water and high amounts of hot residual exhaust gases are a bad mix.