Melted pistons on new built engine - help diagnose please

Hey,

After building my engine by myself and running a few hundreds km for break in I started tuning for boost, very quickly after that the engine had heavy blow by and the pistons were melted - less then 100psi compression on 1-3 and 0psi on 4.

I did not hear any detonation while tuning (my headphones were at home :\)

AFR was 11.5-12 all the time (could be the old sensor is off?)

The timing offset for the ecu was -13deg a little out of the ordinary...

I run 95RON fuel

The first boost runs were to see what actuator spring I got - I found out that the turbo run linear boost upto 20psi at redline, in order to test this I run it like that a few tims while afr was 11.5-12 and timing on the map was 10deg or less. (The actuator needed a good tightening)

Another thing is the marking spots on the valves from the head shop - could it be responsible for pre ignition ?

I ask just two qustions:

1 - What most likely is the error reposibale for the damage ?

2 - What can I do next in order to prevent it ?

Thanks !

Elior

What heat range spark plugs?

6

The engine is showing signs of severe detonation. This doesn't need to be the result of running lean, but too much timing under boost can do this very quickly. I'd be checking the base ignition timing as I'd guess that could be your problem. As a separate issue, it looks like someone has numbered the valves with a punch which is definitely not advisable. This is going to act as a stress raiser and could result in the valves failing. I only ever number my valves with a sharpie marker.

How could I prevent it ?

I did not use my knock sensor headphones but there was no audible sound...

On my previous engine (which is fine btw :)) When timing was too advanced by 2-3deg I could hear a very loud knock...

In this case it died with out making any audible sound.

Thanks

Elior

6 range on an NGK plug? That's not helping you for that much boost. 7 would be better.

Listening for knock by ear is not a viable way of tuning. The problem you face is that even with a quiet exhaust system where you can hear the engine well, knock will be occurring perhaps 2-3 degrees before you can hear it with your ears. Add in a loud exhaust or an external wastegate dumping to atmosphere and I'd never trust my ears alone to pick up knock - If you can hear it then it's almost certainly too late.

My advice is to always use an audio knock detection system.

I'd also agree with Raymond - 6's are probably a bit hot for your application, although the pics show no sign of the plugs melting.

As Andre said, there can be in-audible detonation nibbling away at the pistons until they fail, and those pistons sure were subject to a lot of nibbling. Remember, while we are usually concerned about this under full load, poor timing and/or fuel mapping can result in it at light to medium load, even idle.

Looking at the pic's, I'm not that happy with the spark plug colour as they are much whiter than I would have expected - especially with the supposed AFR and petroleum based fuel -and, while this can, in part, be due to a too hot spark plug, it might warrant further investigation. On that, what did you use to monitor the AFR, as some cheaper equipment may give false reasings and some cheaper fuel pressure regulators are known to give fluctuating pressure regulation. How confident are you that the fuel supply was able to maintain the required pressure and supply at all times?

20 PSI of guage (?) boost seems rather a lot for a (RON?) octane of 95, what was the nominal compression ratio of the engine? Comes to that, what was the engine? What was the quench/piston to head clearance - if this is incorrect, it can actually promote, rather than reduce, problems with detonation.

Re: the punch marks to identify the valves, personally I was appalled by that - one might make an arguement for small punch marks, but those were deep and the displaced material was left proud of the surface which, especially on the exhaust, could be a hot spot for pre-ignition to initiate. that they weren't even removed would be a huge black mark for the machine 'shop'. Good practice is to use boards or racks that keep all the components in order and, if that is inconvenient, the good old marker pens and some care work well. Heck, if you re-used the camshafts and followers, makes me wonder if they might have mixed those up as well.

Oh, also looks like early signs of head gasket failure.

Last thought, did you verify the engine's TDC markings were actually true TDC, and that the engine management was accurately set for that?

The new engine is in the car and all was ready to fire, the engine will not start, got fuel, got spark, crank and cam are good and the ecu "see" them.

The same symptoms like with the melted engine...(very hard start every time)

Eventually I've found out my coils (2 packs) where connected wrong A to B and B to A...

When I switched back the engine fired right up ! :)

Current base timing offset is -6deg instead of -13.5deg on the melted engine.

My question:

Can you help me understand how this mistake (+the very large offset) melted the engine ?

Thanks !

Elior

How did you hook up the timing light and how did you determine true TDC? That could help us diagnose if the base timing wasn't set accordingly.

If you are running COPs then this might be the case. Not enough power to trigger the timing light gun. I am telling this because many MX5 owners including me run Toyota coil plugs.

Plus did you checked your base timing at higher revs also to be sure about it not drifting ?

Some info:

Stock miata NB two coils, 2x(on plug + near) (wasted spark).

ATI damper, when the head was off it was easy to see TDC is right.

Timing light got power from the car power lines and the pickup was on the ignition cable of cyl1.

If Top Dead Centre isn't accurately referrenced by the timing offset, the ECU will reference the incorrect position and so all the timing will have the same offset. For example, if the offset is incorrectly set so the ECU 'thinks' it is at TDC but the crank is actually 5 degrees after, then all the timing you parameters you put into the ECU maps will be advanced 5 degrees more than you think it is.

As you have found out, this is especially dangerous when one is using map values already supposedly optimised for the engine, especially with louder exhausts/good sound insulation/poor hearing, etc, as one may not hear light detonation and think everything is fine - until it isn't.

My question is why coils connected in wrong order (A connector to B coil and B connector to A coil - wasted spark) can cause piston melting.

Why do you think that had anything to do with it?

Base timing was -13.5deg, when the coils are connected right it is just -6deg (as expected)

You said that you are running an ATI Damper

Have you seen the failure on the belt cog?

It advances the ignition....

https://www.miataturbo.net/engine-performance-56/incorrect-tooth-clocking-ati-dampers-96402/

Check it out. If you are sure about AFR's your fuel injectors etc etc then something is wrong with your ignition table What ECU are you running??

I know about that misalignment - I'll check.

I run MS3 Basic

I'm sure the wrong coils order made the base timing very advanced and that cause the whole timing table to be about 7-8deg advanced - I just look for another opinion.

Ah, my mistake, I thought you meant the coil packs and/or leads were swapped over - you mean the sensor?