LSJ running problems

[Fozda]

Hey guys, I need some help. I feel like I've tried everything but can't seem to figure out what's happening. I'll try to explain it as clearly as possible without rambling on forever.
So first, my engine blew up during an autocross which was diagnosed as piston ring gap too tight.

Then I bought a "new(100,000ish miles)" engine locally, we swapped the stiffer valve springs and neutral balance shafts over, put head studs in it, Cometic MLX head gasket, and a new timing chain set in and tossed the engine in the car. Bolted everything back up just how it was when it was running perfectly before it blew up. I think the only part external of the engine we changed was the PCV valve.
When we went to start it it wouldn't start so I spent a whole bunch of time diagnosing that to discover the MAF sensor died somehow and I lost the coil pack ground. Got all that fixed and then it fired up and ran pretty decent.
This is where it gets FUN. As soon as it would see boost the car would basically just die. If the car was moving I would leave it in gear to keep it running but it was incredibly rough and wouldn't accelerate or even maintain speed until it would "clear up" after about 10-15 seconds and then it still wasn't great until I'd shut it off. Did some data logging but couldn't seem to see anything crazy so I ended up replacing the knock sensor and clocking it in the recommended orientation and that seemed to help.
At that point, for some reason, I was convinced that it was something in the turbo setup causing the problem so I swapped everything back over to the supercharger setup and loaded the tune that the car ran perfectly on before the swap but that didn't fix it.
So here's what I have right now...car runs and idles just great and I can drive it all over town no issues but if I do even a mild pull to about 10lbs of boost it throws a cylinder 1 misfire code, pulls all timing and goes full lean. It'll stay in this mode for a good 10 seconds or so before it will recover but only if I'm moving and keep it in gear, otherwise it just dies.
It also popped the dipstick out of the tube at about 9lbs of boost so I did a compression test and it's at 210 across the board. The data log seems to not show any signs of anything going wrong until it's just but I'll try to get a log attached for you to look at as well.
I'm at quite the loss and have been messing with it for a couple months now so if you have any suggestions I'm pretty much open to trying anything to get it fixed. I considered just replacing the engine but I'm not convinced just yet that it's the engine causing the problem.

 

[Gtstorey]

And no codes other than misfire, and all the dtc’s are all are reporting in the tune? I would still guess a wiring issue /loose connector but most of that would show up with an error code but not all). Have you checked fuel pressure, maybe it ran lean instead of ring problem when the previous engine failed? Tried moving the ignition/injector on cylinder 1 to see if misfire follows? Are you logging boost control solenoid and whether it’s in closed/open loop modes to see if it’s running in ol-fault mode? What is the misfire count when this happens?

 

[escapepilot]

I’m no expert, but the two things that stand out to me is the dipstick and only having problems under boost. If the compressions are good, how is the crankcase getting pressurized? Is that pressurization connected to the cylinder 1 misfire?

Another item - could the stiffer valve springs be causing an issue under boost? I don’t see how yet, but that is the only hardware transferred from the old engine that possibly could.

 

[Ross]

Are you running forged pistons? They are a bit smaller in diameter, usually paired with high boost, so then it has a larger ring gap. All these things lead to more combustion blow by, and higher volume of blow by getting into the crank case.

 

[Desert Sasqwatch]

Is the valve cover port vented to atmosphere or is it ported to the inlet of the turbo housing? It's important to have the crankcase under constant vacuum, one port may not be sufficient with a full forged setup, as Ross indicated. A second port into the valve cover/crankcase may be needed to get this under control.

 

[Fozda]

I am almost completely certain that this is a new issue since the old engine ran great and made a ton of power. What I believe happened was the intercooler pump stopped working and I pushed it too hard when everything was too hot.
I ran a leakdown test:
Cyl 1: 25%
Cyl 2: 16%
Cyl 3: 22%
Cyl 4: 10%
So not the best results and I'd say that's enough to cause the dipstick issue but I still can't figure out if/how it's correlated to the other issues it's having.
The engine is stock aside from those small bits we put in it.
The boost control solenoid has been disabled and it does it with the turbo setup or the supercharger setup on the car.
The wideband doesn't indicate any issues with fueling but I think my next route is to pull all of that apart and inspect/clean everything just to be sure.

 

[Gtstorey]

I would think 25% leakdown would be enough to cause your misfire. A high misfire count could possibly throw it into open loop fault mode like Ghostknife has happening. I've never found much explanation of what it does in OL-F mode though.

 

[Fozda]

I would think 25% leakdown would be enough to cause your misfire. A high misfire count could possibly throw it into open loop fault mode like Ghostknife has happening. I've never found much explanation of what it does in OL-F mode though.

Is there a way to disable the fault mode to test it?

 

[Ross]

Normally our engine runs in closed loop. Closed loop is when the stock narrow band O2 sensor is sniffing the exhaust, and telling the PCM/ECM if the engine is running rich or lean, and then adjusting the fuel trims to correct the fuel mixture.

If the engine is put into open loop (either by a fault, or deliberately when tuning an engine), then the ECM won't be adjusting the fuel mixture based on the narrow band O2 sensor. This is useful for logging the fuel mixture errors, and then making tuning changes to the ECM tables.
This mode will allow the engine to get away from stoichiometric, and could cause rich or lean mixture issues.

 

[Gtstorey]

I guess how much difference open-loop fault makes depends on how well your speed density/VE is tuned. Or is it still running on maf but not making real time adjustments?

 

[Ross]

I guess how much difference open-loop fault makes depends on how well your speed density/VE is tuned.

True.

Or is it still running on maf but not making real time adjustments?

The ECM will use speed density/VE mode when the engine is making quick transitions (sudden throttle changes/ sudden RPM changes)
and use MAF mode when running steady state (2700 RPM going down the highway).
The ECM will calculate fuel for both (VE vs MAF) modes, and then choose a blend between them when transitioning between modes.

Both of these modes can be run in open or closed loop, stock they are run in closed loop.

Some tuners set the ECM to only run in VE mode, as they have reprogrammed the ECM to handle a lot of engine changes (race parts), and they don't care about fuel economy, just want race mode.

 

[Ross]

Nate, there are ways to put our ECM in open loop or closed loop.
I haven't heard of a open loop fault mode before today.
I'm not sure how or why the ECM would put itself in open loop, when it is programmed to be running in closed loop.
Need to learn more about this... Maybe it is trying to protect the engine when the stock O2 sensor has gone bad?

What I believe happened was the intercooler pump stopped working and I pushed it too hard when everything was too hot.

The ECM will pull a bunch of spark timing out if it detects too hot of intake temperatures, which will pull out a bunch of power to protect the engine.
Many LSJs are out there running with non-functioning intercooler pumps... so I am pretty sure the ECM is good at protecting the engine from damage due to this fault, but it might be a contributor to other issues, and in combination, managed to cause engine damage.

 

[Gtstorey]

I never heard of open loop fault mode either until I saw it on Ghostknifes logs and I looked it up. Didn’t find a lot of information. It’s what the ECM does when it doesn’t trust the sensors are doing right. It may be just a different way of labeling limp mode.