2011 GMC Acadia weirdness - ROUND 2... FIGHT

wrote:

I thought it was interesting that this go-round, the misfiring starts up NOT when it started idling down, but when the high pressure fuel pump started tapering off. :huh: Can't figure why that'd cause a miss on #4 and #6 and nothing else... It's also meeting desired rail pressure at all times. Cam timing is on. Fuel trims are a bit richer on bank two, but nothing crazy.
 

Both the earlier scan logs showed the same thing for #6. A group of misfires right after what appears to be the PCM switching from base OL map to controlling fuel. Still at high idle. Appears to be taking fuel away when the misfires start. But doesn't start adding any back when they stop???

 

FWIW the graph top right of your HPTuners screenshot looks like the misfires happen after the engine idles down because the PID you're using is the total, history, misfires PID instead of the current misfires one. This screenshot shows current counter getting written to history and cleared.

 

I agree Matt, the fuel control strategy doesn't make sense. :huh: But, I've also never looked this closely at rail pressure during a cold start.

Yeah, I should have mentioned not to read too much into that chart. :silly: That was my first attempt at making a misfire chart in HPTuners. Just playing around to see what works/what doesn't.

Also FWIW, the number of misfires has gone down since it came in last week? It also didn't flag a P0300 or P0306 on the last startup. Dunno if there's anything to read into there...

Hmmm...

I am, still, seeing #4 as the lowest contributor. But, looking at the Crank Case pressure, I would expect to see a repetition of 6 pressure events. I see two. I see pressure spike #6 compression, and #3 compression.
 

 
 

Yeah I noticed the first scan log was a good bit worse on #6 but didn't run long enough to show anything on #4. Was that one taken prior to warrantying out the injectors or swapping ignition components?

wrote:

Yeah I noticed the first scan log was a good bit worse on #6 but didn't run long enough to show anything on #4. Was that one taken prior to warrantying out the injectors or swapping ignition components?
 

All data and scope captures are after moving the coils and changing the injectors.

So crankcase pulses during #3 and #6 compression... I didn't leak down test #3, but #6 had less than 5% when I tested it last week. I think I'll pull the intake off (again) and compression/leak down all cylinders. And borescope all cylinders, just for giggles.

wrote:

 I think I'll pull the intake off (again) and compression/leak down all cylinders. And borescope all cylinders, just for giggles.

 

I'd like to see Cranking Compression on all cylinders. (I'd REALLY like to see Running Compression, on all cylinders. But,....)

wrote:

Hmmm...

I am, still, seeing #4 as the lowest contributor. But, looking at the Crank Case pressure, I would expect to see a repetition of 6 pressure events. I see two. I see pressure spike #6 compression, and #3 compression.

 

The #6 spike drops off before TDC. The #3 spike extends past TDC which makes me wonder if #4 is also contributing to it??

 

wrote:

 The #6 spike drops off before TDC. The #3 spike extends past TDC which makes me wonder if #4 is also contributing to it??

 

I suspect the same thing.  Hey, Tyler. When doing cranking compression, grab crank case, too.
 

We need to have a live video chat going when I do this stuff. :silly:

 

Emailed Chad the cranking compression capture. PV350 in the cylinders, intake off, injectors unplugged. Did it in the reverse of the firing order, so I might catch whatever is wrong with #6 and #4 sooner. Readings are in kPa, so you'll need to do the math for freedom units.

Rough measurements put the very lowest clean tower (#6) at 158 PSI, and the very highest tower (#3) at 167 PSI. That's being pretty nitpicky. Not much of a spread, IMO.

 

Don't sweat the noise between cranking events - that's me handling the PV350 while swapping cylinders. Shout out to Juergen and his suggestion of externally grounding the PV350. Didn't get rid of the noise, but definitely took it down a notch.

Zoomed in on #6:

 

I do you have all the spark plugs pulled. I see a crank case pressure spike follow your in-cylinder.

 

The spike is not the same for each cylinder.

 

Yep, all plugs out. Are we looking at the higher initial pulse on the first tower of #6? 

wrote:

We need to have a live video chat going when I do this stuff. :silly:

 

 

Fire it up!
 

wrote:

Yep, all plugs out. Are we looking at the higher initial pulse on the first tower of #6? 
 

That is something to look at. I don't know what to make of it, though.  Is it repeatable? Besides that one, cyl #3 pushes into the crank the most. #2 the least.  I don't want to spend too much time barking up the wrong tree, but can you post a Cranking Crank-case, with all the plugs in, please?


 

Can do. I've got the plugs back in and intake back on. After your cranking waveform, I'm gonna run it again and set it up for Monday morning.

Cranking crankcase pressure in red, #6 trigger in blue, RC in green.

 

I think this is an ugly intake waveform, too. These are new heads, correct?

 

wrote:

I think this is an ugly intake waveform, too. These are new heads, correct?

 

 

The long block is a 'reman' from Midway.

I did eyeball the ports with the intake off. No significant buildup of anything.

Started it up again this morning. Only four miss counts. No codes set.

 

Caught it setting P0300/P0306 on Sunday. I got a little bit smarter and decided to watch HPFP angle during the cold start:

 

It's got the angle dialed up during the miss, then scales it back. Miss disappears. There's enough of a correlation there to make me suspect our issue is still fuel related.

I got desperate and put a request into the Identifix Hotline. The gentleman on the other end correctly pointed out a GM PI that I'd missed early on. :blush: PI1296.

 

Talks about the orientation of the injectors being incorrect after service. Well, we've had these rails off like, 20 times now. Sounds promising! Nope, they're correct. :silly: