280Z Requires Starting Fluid for Cold Starts

[Captain Obvious]

Dave, I was thinking the same thing, but the OP said this:

On ‎10‎/‎17‎/‎2019 at 12:20 AM, S30Grit said:

During a failing cold start, what is known:

• Fuel pressure is correct/consistent at the fuel injectors during priming and starting sequence

So it sounds like he already confirmed that the check valve is holding and the fuel pressure is OK even while the problem is occurring.

[Dave WM]

yep, I see that. Oh well...

If it was me I would have a spark indicator on it, a fuel pressure gauge on it, a stethoscope on the injectors, and have someone else cranking while I watched for everything else.

[S30Grit]

3 hours ago, Captain Obvious said:

So, I got a question.... When you say it won't start without a spray of starter fluid, how long are you cranking it for before saying it "won't start"? Reason I ask is that I removed my cold start valve completely and it definitely takes longer to start without it than it did before I took it off. Maybe five seconds of cranking? I'm wondering if you just are giving up too soon.

What happens if you crank the engine for fifteen seconds straight? Sounds like a short amount of cranking, but with your hand on the key, it doesn't feel like a short time.

I've definitely tried some starts over 5 seconds (and yes, my hand gets tired with just that due to the small key!).  15+ seconds hasn't been tried, but is that extra strain on the starter motor? I've always used starters sparingly (i.e. <5 seconds) my whole life due to concern of heat/wear on them for long starts, but that may be a completely unfounded concern!

I've also tried some ~5 second starts, waiting 10+ seconds, then repeating. But I'll try some other approaches as well.

[S30Grit]

1 hour ago, Dave WM said:

yep, I see that. Oh well...

If it was me I would have a spark indicator on it, a fuel pressure gauge on it, a stethoscope on the injectors, and have someone else cranking while I watched for everything else.

This does seem ideal.  Right now, my main focus will probably be tracking down the injector signal pulse issue since I know they aren't firing (and checking CSV).  After that, I'll be re-confirming those previous things like fuel pressure and strong spark, etc..

[Dave WM]

what are the engine temps when you are having problems with starting please, maybe you covered it already.

I just tested my spare engine on the test stand, with the CSV disconnected and the fuel pressure at zero, it started almost instantly. this was at 82f which should not need the CSV and even connected it prob would not have fired since the temp was too high for the thermotime to activate.

 

Edited October 18, 2019 by Dave WM

[S30Grit]

6 hours ago, Dave WM said:

what are the engine temps when you are having problems with starting please, maybe you covered it already.

Dave WM - Ambient temperature has been around the 80's previously when trying to start, so I'd guess coolant/engine temp wouldn't be too much lower than that, maybe 70F if I had to guess. 

Thanks for the video.  Definitely good to know that it should crank right up without issue in warm weather regardless of CSV!  I'll check mine just out of curiosity

[Zed Head]

I have a new thought.  Maybe the AFM blade is sticking shut during the low air flow of starting.  That would lower the injector open time, delivering less fuel.  Maybe pop the cover off the AFM and watch it, or prop it open.

Also possible that you have a vacuum leak that also allows air to bypass the AFM.

[S30Grit]

This is one vacuum hose that isn't OEM and doesn't seem to be sealing properly.  Might try clamping it before replacing to see if that helps.  As for the AFM, not sure how that guy works yet.  But I think the FSM has some basics checks that I can do on it.

[Dave WM]

Smoke test is a good way to find vacuum leaks. plug up the throttle body with a stryofoam cup, get a cheat stogie, I like to use a clear tube connected at the brake booster fitting, puff on the stogie and blow smoke  in the tube. You should feel resistance, but as it fills with smoke look for leaks. I recently found a few using the method, one was a missing injector o ring and the other was a split on the valve cover vent pipe.

[Dave WM]

2 hours ago, S30Grit said:

Dave WM - Ambient temperature has been around the 80's previously when trying to start, so I'd guess coolant/engine temp wouldn't be too much lower than that, maybe 70F if I had to guess. 

Thanks for the video.  Definitely good to know that it should crank right up without issue in warm weather regardless of CSV!  I'll check mine just out of curiosity

Well if its in the 80's or upper 70's its def not a CSV issue. Forcing it to fire at higher temps may help the starting but that's really just covering something else that is wrong.

[Zed Head]

31 minutes ago, S30Grit said:

This is one vacuum hose that isn't OEM and doesn't seem to be sealing properly.  Might try clamping it before replacing to see if that helps.  As for the AFM, not sure how that guy works yet.  But I think the FSM has some basics checks that I can do on it.

Take the black cover off of the side of the AFM.  It is just a press fit with some light sealant to hold it in place.  There is a damper weight that moves with the AFM vane.  If you move the weight the vane will move, and vice-versa.  It should start moving as soon as the engine turns over.

[Captain Obvious]

21 hours ago, S30Grit said:

I've definitely tried some starts over 5 seconds.  15+ seconds hasn't been tried, but is that extra strain on the starter motor?

Yes, I would agree that 15+ seconds of continuous cranking is disconcerting and not something that you would want to be necessary. But I'm trying to make sure I have an accurate handle on the problem.

You said that the engine refuses to start on it's own when it's cold, but with a quick spray of starting fluid it starts up and RUNS instantly.

I'm wondering if the problem is more accurately described as "when the engine is cold, it takes a lot of continuous cranking (like fifteen seconds) to get it to light off. But once it does start, it runs fine".

I would apply a different approach to troubleshoot those two different scenarios.