Andy Hollis

Well, she done blowed up.

This is my K-swapped EF track car and it has been dead-nuts reliable right up until it let go. No missed shifts, nothing unusual. This was at a part of the track where the motor just touches the rev limiter (stock @ 8600) before lifting for the next turn. Poof! White smoke and wet oil out the exhaust; puke tan coolant/oil mix with metal bits out the intake. I pulled over quickly (motor had already shut down almost immediately).

History: ITR K20A was installed as delivered from hmotorsonline. Bolt-on mods only (CAI, Hytech, KPro, 3" exhaust). Made 220/150 on Dynojet. Nothing out of the ordinary. Probably 5K street miles and 10-15 hours of total track time. Ran this car in the One Lap of America to a 10th place overall finish, and most of the other time was at Harris Hill Road near Austin. Lived between 4000-8600 rpm on track.

KPro had conservative base map, with fuel only tuned by me (street tuned to get reasonable AFRs; did each cam map separately). Never any knock on datalogs.

Tear-down revealed rod broken in two places. Bearing in the big end showed normal wear, so oiling problem is doubtful (stock pan and pump, Mobil 1 @ "full" mark on stick). I'm thinking that the small end broke first, otherwise there would be no downward force to pull it off the pin and split it. Valve stems were bent like spaghetti. Piston disintegrated into small bits, save for the two large chunks in the pic.

Would love to hear from engine builders or experienced K-powered racers what would cause this? Obviously, I need to avoid this with the next go. I was under the (mistaken) impression that these motors were bullet-proof so long as you stayed within OE rev limits, kept it oiled, and tuned it conservatively.

So, where did I go wrong?

--Andy

PS: Thread is also on k20a.org, but figured I might get more track folks here. Oh yeah, and let's stay focused on real info...I know it sucks, and I know y'all feel sorry for me.

KIWI

It's always going to be a guess which went first? I hate running stock rods and pistons!
We have never had a problem with a properly built K-20 motor.
We've blown a few Stock K-20's, usually after a mis-shift. My guess is - Valve touches Piston, Piston Cracks and flies to pieces, (The K-20 piston is not a very strong design anyway) Rod flails around and destroys it's self and everything else. Unfortunately the couple of seconds worth of carnage after the initial failure usually destroys any real clue of what caused it!
Sorry, I know this is not much help, and I too would like to hear some theorys.


Kiwi

night

My guess is that 'free wheeling' lack of load at redline. As far where it gave.. I'm going with the small end. You can see the dent in the rod where it hit the bottom of the cylinder and then sheared it off. Nothing is blue, and therefore not an oiling/heat issue. Maybe bad metal, maybe bad luck.

Also is there where I say how much I hate certain rules that are supposed to have the intention of keeping racing 'cheaper' while in turn punishing racers with reliability issues and costing more?

Andy Hollis

No doubt.

That said, in this particular case, I'm not working to any rule set. Car is totally a track toy. Thus the desire to figure out how to avoid this next go round.

--Andy

sander

Did you have any data from the incident?

Due to the RPM that the event occurred at I would suspect something like valve float leading to contact, leading to valve breakage, leading to destruction. Or, some variety of rod failure.

However if there were any loss in oil pressure, that's all thrown out the window.

I had a lot of (track) miles on my K20A and never had a lick of trouble related to internal issues. I did however use a different VT combination and never shifted above 8300 for the most part. (And rebuilt it rather compulsively "just because")

-Sander

Andy Hollis

Unfortunately, no.

I'm usually pretty religious about resetting the KPro logger before a session, and thought I had done it. But the data present was clearly from a previous session.

If you look at the head pic, none of the valves broke off. Just bent in place.

Thanks for the input.

--Andy

sander

Oh, sorry then. I thought that looked like a missing valve head.



Apologies.

unusual71

Judging by the amount of damage the rod took but then appears to have suddenly sheared with little bending, tells me it banged around quite a few times before failing.
I would say then that failure had to have come from above.
If the cylinder wall that took the damage is the heavy thrust side, i would say that the aluminum behind the sleeve failed, allowing the sleeve itself to fail
I know it might seem strange but judging by what little evidence is there I honestly think the sleeve went first. If it was the piston or rod i think that there would be more head damage and that the piston would have "escaped" the bore sooner. If the sleeve went first it would give the piston and rod time to get a thrashing like that.

that said I think these blocks are great and the weakpoint is usually the piston? i haven't done my k series homework yet so this is all speculation really

edit: the root of the problem could have been from the rev limiter inducing some sort of block destroying harmonic but i guess that theory would apply to the piston as well.

eflove

Have you checked to make sure the crank pulley was tight? I have seen it come loose causing play then braking the pin on the T-chain pulley making results like this.

Andy Hollis

Oh yes, the pulley was tight. It was a bitch to get it off when disassembling the motor.

And all the timing chain/gearing was tight also. No evidence of chain tensioner issues.

--Andy

JW racing

I like the guys idea of the sleeve cracking...However I know the weak link is in the pistons wrist pin area. It is always fun trying to see which failed first.I followed the heard(and my tuner) to keep revs at 8300 due to this fact and never had any engine failures in 3 years.
It reminds me of the RTR cars with 8 expired K20's in a season(even with aftermarket rods and pistons).

I have seen cases where a wrist pin bushing binds and eventually fail,and once the bushing cracks,the wrist pin can move and will tear out what ever it wants to of the small end or sometimes the piston.

edit - Do you have a baffled oil pan?

90_EX_Civic

that's what i remember hearing too, that's why some guys use h22 rods(if i remember right) and that MAX revs would be 8500 on a stock engine.

Race Egr

I don't know much specifically about K20 motors, but I agree that the wrist pin bushings failed. You figure if this bushing fails this is going to allow the piston to get thrown around in the cylinder. This would explain the damage done to the cylinder head as well as the sleeve. This would also explain why your piston looks beat to death. Also, as far as knock sensors. I don't believe they can always be trusted as usually by the time the signs show up the damage has already been done. Not to mention the endless possibility's of noise interfering with them.

exgr

me: https://honda-tech.com/forums/showthread.php?t=2873174 andy is a good guy. you should chip in if u got a minute
Luke: Easy peasy. Sleeve failed first.
me: yah? and that's a common issue?
i thought i heard honda runs k20a (and all their R motors) 24hours pretty much at the rev limiter....
Luke: I dunno why the sleeve failed bro,but thats a sleeve failure first. Its roughly identical to my own k-series failure even.
me: what's the logic behind sleeve failure first?
me: is it because mostly only the inner part failed? outer al. casing still mostly present...
Luke: The deformation. If the rod failed, it would have been a localized stress riser, this damage indicates it got f-ed up from forces smashing/bending/torqueing on it while it was still trying to hang together.
It went up and down long enough to turn the piston into a mcnugget after the sleeve was broken.

unusual71

ah this is what i was trying to say but i didn't want to confuse honda tech with terms like "f-ed up" and "mcnugget" lmao

Andy Hollis

Thanks for all the input folks.

At this point, I'm starting to have impure thoughts about moving on to a K24. I can do it with a built bottom-end for the same money as the hi-zoot JDM ITR K20A and make more power at the same time. Probably have to change my final drive for the lower rev range, though. But that's a whole nuther thread...

Consider this thread done. And again, thanks.

--Andy

night

No thread here is ever done :D.
If it was the sleeve, would posting and/or block guards saved it or only prolonged it?
Has anyone seen a stock bore B do that? (and even stock, perhaps still discounting the B20). I mean the K20 r/s is better than a B18C so this is kind of a surprise for me. Not that I have been fortunate enough to have massive race experience to see these things. Though I guess in proportion, 8200 B vs 8600 K. Myself I do not like touching redlines ever and this will always be in the back of my mind now once I finally get on the track.

JamesL

Hi Andy,

Im taking a wild guess, but did this happen in T1 @ H2R? Even with a baffled oil pan, my oil pressure shows dips thru T1.

And while that is a bit disconcerting, I have about 30 hours on my stock K20A2 to date... Ive been shifting at 8200, which is about 500rpm lower than where I want to be with the 4.7 final drive. But I was advised against going above 8300, especially with the OEM Honda rods.


Off topic, how do you like the repaved surface/how do you feel it compares time wise to the track 6-12 months ago?

spAdam

iTrader: (1)

This was my first thought. I (somehow or another) still have next to zero K experience, so i can't vouch for what their weaknesses are aside from what I've read/seen second hand. But my first thought when I looked at the pic with the "mcnugget" piston was that the rod was about the least beat up part there, which leads me to believe that it was the last thing to hold onto its intended location.

Andy Hollis

Actually it happened just before 1, along the front straight. Typically, with a good run off of 10, I get to the rev limiter in 4th there (99 MPH) before a light brake into 1. That lap, just when I was expecting the limiter to pop, the engine died.

Very nice! Can't compare times for me since I put the old D15 back in during the time the track was being patched up. I'm 6 seconds slower now! I think the only part that got faster for me was 8. The new stuff has more grip, so you can banzaii the entry a little knowing that the "Hand of God" will grab you soon thereafter. Turn in a little more at that point with your foot to the mat, but make sure and turn back out a smidge just before the new pavement ends, or you'll go right off the track when the grip goes away. I am consistently in the low 1:32's with the DPFI D15, with a best lap of 1:31.8. I am matted everywhere, save for slight brake into 2, more into 4, 7 and 10. Kinda like a Spec Miata with a crappy motor. Ran mid 1:26's with the K20A. More shifting, more braking.

--Andy

exgr

you analysis was spot on. but i do love my buddy's mcnugget comment. lol. rev limiter induced harmonics though, that seems a little far fetched. :-) but it's a good one.

unusual71

well regardless of whats happening, the power is cut, and then resumes
The problem exists when the power resumes. All of a sudden the properties like flow and temperture are changing very fast
could lead to (albeit over a very short interval) bad circumstances such as lean conditions. It can also introduce a shock load due to the decelerating crank having force applied to it from the rods, rapidly changing its acceleration. Since these motors make a lot of power all the way to redline these forces are large and directions are changing VERY rapidly.
maybe harmonic was the wrong word to use. but i thought that maybe during this split second the cylinder wall would be loaded more than when the engine is just accelerating normally, and thus it let go.

rash

I had a similar issue with a stock h22a. Got it from Hmotors, drove it for a few months as a DD, with the occasional romp here and there, then boom. Never actually figured out what failed first, but I lost about 3 valves, and destroyed the piston so bad you could see straight through it.

Sr420Det

When u let off at high rpm, id imagine the crank shifting to the side and putting a lot of load on thurst bearings and other areas....just something to consider

night

Also, wouldn't you be far better off with a finer gearing adjustment through tires than screwing with the hard stuff?