Forum » General Tuning Discussion » Just blew my first engine while tuning (rod knock), not sure if it is tune related.

Just blew my first engine while tuning (rod knock), not sure if it is tune related.

General Tuning Discussion

Forum Posts

Courses

Blog

Tech Articles

Discuss all things tuning in this section. News, products, problems and results. 

= Resolved threads

Page 1
Author
386 Views

Freshly built up a 2ZZ (all stock internals) and started road tuning (for fuel only, I plan on tuning ignition when I have access to a dyno later). Was concentrating in the low to mid RPM areas for the first day then begin tuning the higher RPMs past the VVTL engagement point on the second day. Everything was fine at the beginning but started getting what appeared to be a misfire above 6500 rpm (the coils were old and I double checked my dwell times later and noted the dwell have been set too high which might be killing the coils). As we tried a few more runs to diagnose the misfiring issue, the engine suddenly started sounding strange (can't exactly remember what happened but me and the driver just felt something strange). We backed down and gradually noted some very faint clicking/knocking noise which got progressively worse. Not knowing what it was then, we continued driving mildly for 10 min to return to the garage and started to lose power on the final stretch. Only then I begin to realize it was rod knock (never heard one first hand before) but it was already far too late.

We tore down the engine and noted the connecting rod bearing on cylinder 3 had spun out of place and was severely worn down, the same cylinder had also been scored. All other connection rod bearings had visible signs of wear and minor scratches could be felt. The main bearings had visual wear marks but was till smooth to touch.

Now my question is, could a bad tune have caused this? I know severe knocking could hammer on the bearings and cause failure, but would it have showed other signs elsewhere? Other than the failed bearing and damage from the resulting lack of lubrication, I could not find any other signs of damage that would point to knock.

Although I was not confident with my ignition map, feedback from others were that it was conservative/safe although the shape isn't ideal with too much timing pulled at higher RPMs and axis scaling was not ideal (I was experimenting with Alpha-N so it was the first time I had TPS as load), fuelling was accurate looking at my logs. I just wanted to rule out tuning problems and narrow down onto the cause for the failure. I do not want to build up the engine again not knowing what went wrong and have it fail again.

Also one more point to note after looking at my data logs; a couple second after it first failed, the VVTi control suddenly failed and the valves went to fully advanced state and stayed there even after the ECU detected the failure and cut all duty cycle to the solenoid. Seems like a ton of debris had been sent into the oil system that the VVTi solenoid got stuck.

Attached is are pictures of piston three, the failed bearing while it is still on the crank with the rod cap removed, the bearing itself and my ignition map. Seems like I am attaching a lot here already so I will post my logs on request.

Hopefully, someone could provide some insight on what happened so I could prevent this after the rebuild. Thank you!

Attached Files

Attached are two logs. One is when the engine blew. Another one is of a run made on the same day.

Note, on the blown log, the knock sensor is already picking up more noise than normal at the start. And you can see VVTi angle at the bottom going crazy at the end once we noted the problem with the engine and noise really started to be picked up by then knock sensor with the bearing falling apart. I was not looking at knock while on the road (new to Syves, not too familiar with their software so did not setup gauges to monitor so many things at once) so I did not see all that craziness going on or I would have definitely stopped the car.

On the normal log, you could see much less spikes in the knock input. There were a little fluctuation in Lambda beyond 6500 rpm due to the possible misfiring I mentioned.

Attached Files

That looks like molten aluminium deposits on the piston edge, hard to tell if it's from the outer part of the piston or maybe the head, or both. It would suggest a problem on that cylinder, rather than the tues itself, if so. It could have been detonation or, maybe, pre-ignition and in either case it can result in very high cylinder pressures that can damage bearings and once they start to fail it's all over - oh, what were the clearances, as if the journals were a bit worn it could promote the hammer affect?

I'm far from an expert, but what else have you noted - anything on the other cylinders such as piston and/or head damage, what do the spark plugs look like - I assume you did run colder ones, check the other bearings for loss of the 'spring' that holds them in place as that's usually the first sign of the problem? Were the injectors new or cleaned, as that can cause a leaned out cylinder?

Anyone else?

Are you referring to the silver part near the top edge of the piston? Let me look at it again when I head down to the garage tomorrow. I suspect it might just be the exposed bare surface of the piston itself that remained clean and free of deposits. Speaking of which, the pistons were sparkling clean/silver when they were installed and I am actually quite surprised it got so much carbon and golden/brownish deposits on it after such a short run. The deposits were present on all other pistons too.

Don't have the clearance off my head but all journal bearings were new and spec'ed according to stock clearance. However, I cannot remember whether I have reconfirmed the clearance on the journal bearings as they were installed. I know I have checked the main bearings with plastigauge but cannot remember whether I have done so for the connecting rods as I was really in a rush while putting the bottom end together.

Edit/update:

Took another look at the piston (it's actually cylinder 2 that failed not 3 as mentioned previously), the silver parts is actually the bare piston surface and definately not aluminium desposit.

The clean section of piston near the exhaust valve reliefs is from the piston contacting the head after the bearing has failed, as when the bearing has worn/ground/hammered, it has become thinner, allowing the piston to travel further up, and the piston has contacted the squich pad in the combustion chamber.

Check the other rod bearings. The failed one won't give you any information, but the other bearing can give you indicators, particular if it was a clearance issue across the big end bearings.

Detonation generally needs to be fairly severe to result in a rod bearing failure. The massive cylinder pressure from detonation is enough to break the film barrier of oil between the crank and the bearing causing it to hammer. The results can vary, but a peened bearing surface on the top shell is common, as is the bearing losing its "spring" that retains it in the rod.

normal knock indicators on the piston are fine pitting near the spark plug and eroding of the edge of the crown. it will usually show up as a shiny aluminium spots.

Attached Files

Yes N8B, I was just about to report back on the head contact issue you described. I took a close look at the head today and noted slight markings which matched the factory stamp marks on the piston and figured it was making contact with the head due to the added play.

Very interesting read on the bearing failure brochure. Sadly we drove a little too long after the bearing failed and it was too far gone to give a diagnosis. All the bearings just showed symptoms of foreign particle in lining and oil starvation but we do not know whether this happed before the cylinder 3 rod bearing failed or was it just a result of the loss of oil pressure after the bearing have failed.

There are no other peened or loss of springiness in the other bearings or any pitting of the piston observed so I guess the chance of failure due to knock is low. However, I doubt I will ever find out what actually went wrong mechanically and just have to hope for the best when I rebuild/replace this engine.

We had no oil pressure gauge and the stock oil warning lamp had been removed by the previous tuner when installing a digital dash. Will definitely be installing an oil pressure gauge this time around (standalone only as the ECU inputs are full) to at least rule out the lack of oil pressure.

Not that it couldn't happen, but I'd be suprised to see enough knock on what seems like a test drive to take out a bearing whilst on a naturally aspirated engine.

If the tune was enough out of wack to cause enough det to ruin the bearing then I'd expect to see the piston in much worse shape. I know the 2ZZ is fairly high compression but with the timing you were running you should have been in the clear. Check your base timing is accurate!

My money is on engine build contamination causing debris to take out the bearing or assembly error. Normally bad tuning errors end in melted pistons way before the bearings come into question.

Syvecs knock control is pretty good, but for the future I'd advise that you log at 1000HZ for CylXKnock, log KnockThreshold & log cylXretard if you have concerns about det at least until you have the knock system configured.

Also never trust the ECU to begin with, always confirm what you are seeing by listening to the engine with a set of det cans or electronic device like a Plex. Many times I have heard other issues with engines other than knock using my det cans and allowed me to stop the session before further damage was done.

In your normal log it seems like your motor is leaning out quite a lot after the lift cam engages. Between lift, lots of VVTi & ~0.94 lambda on the high end, there is potential for knock, so I'd address your fueling in this area on the next engine.

Yes I understand that knock should be be logged at a high frequency but my problem lies with the measly 1MB logging memory of the S6GP and the fact that at this stage, there are still a ton of parameters that I need to monitor so I had to skimp on frequency.

The leaning out of you see is most likely due to misfire that I am experiencing at above 6K rpm where air/fuel remain unburned. I know boost in a turbo engine could cause the mixture to ignite but in an NA engine, what could be the usual cause? The coils from the used engine might be worn down and old but what could cause it to misfire only at 6K plus RPMs? It's not high enough where there is insufficient time for dwell yet.