Last night i ended up putting it back together.
Even though there was fuel coming out of the banjo, there was just nothing noticeable coming from the PRV.
I also changed the fuel filter, just cause, even though the old one was from January, and only had about 5kmiles on it.

This morning I checked rail pressure again, both with and without the FCA connected. (Per the suggestion by CRF450)
still only about 2,000psi while cranking. 7,000 at idle with FCA, 26,106psi with the FCA disconnected at idle.
All the cold starts were good, but just now, after lunch. it definitely still has the problem.

Does anyone know where to get the factor service manuals?
Or have another test for the cascade overflow, or the checked banjo?

I might just order all 3. PRV, Cascade overflow, and a checked banjo. 2 of them are cheap.

Besides this list, is there anything else that can cause this problem?
Checked:
Injectors (Done)
FCA (replaced)
Fuel Filter (Replaced)
Pressure Relief valve. (Kind of verified)
Lift pump (Pressures look good).

Not Checked:
Cascade overflow valve.
Banjo check valve on Injector return.
Injection pump

 

i see you check list of what you have check, but you havent checked the injectors, right? even though they are new, you still could have a bad or they were not install correctly. https://www.dropbox.com/s/rwus86v9aw...al%20.pdf?dl=0

 

Unfortunately, yes.

Think I should start with re-torquing the cross over tubes? or do the total return flow check?

 

Yes, i go over the feed tubes and torque them. You can also cap off the injectors one by one and see if you can find a bad one.

 

I think i will start with a re-torque.
unfortunately, the symptoms are intermittent and heat related.
So every time I make a change, I have to drive it for a day or 2 just to be sure if it really helped or not.

 

I think you should find the source of your low rail pressure during cranking/startup. 2000 is low according to info on this site. Do a search on here for rail pressure during cranking, one thread I saw said that the ECM needs to see 3000 psi before it fires the injectors.

 

Currently I am waiting for a few new parts.
Cascade overflow valve.
The check valve in the banjo bolt on the back of the head for the fuel line drain system.
A fitting that will let me do the flow testing.

I think my current favorite theory is that check valve in the banjo.

I wont have parts till thursday.

This problem is getting rather annoying. I dislike having to do this much troubleshooting.

 

I finally replaced the banjo bolt with the check valve on the back of the head.
When I removed it, it had a black film all over the inside face, including on the needle. This wiped off very easily. (I would think the inside of the fuel system should be very clean) I wonder if this indicates there was no flow through the banjo.
The needle moved freely just like the new one.
I tried to do a vac leak check, but got frustrated getting a good seal with the tubing I had laying around. So no result there.
I did try blowing into it, and could not get it to open, but you can only get 1-2 psi from your lungs.
After I get a good vac leak check, I will apply some compressed air the other way to see if its clogged.



Basically problem is not fixed.
Still very thermally dependent.
Started great 5 times last night, but not exceptionally hot any of those times.
Got it on the freeway this morning, when I got to work, it would immediately restart just fine.
I let it sit for 1 hour, and although no where near as bad as it has been, it had a longer than normal crank time.

If the long crank time is due to injectors, is it usually so thermally dependent?

The dependence on temperature is dramatic.
This last weekend, I had a LOT of really bad issues. Cranking forever would not start it, only generating like 1800 psi during cranking. Like 30+ seconds of cranking and still nothing.
I was at about 6000 feet, it was often near 90s or over 100 degrees.
If I had been on the freeway, it would have to sit for over an hour to even think about restarting. I had to stop for food, and got stuck in a parking lot for an hour letting it cool down.

The behavior is just like when you are trying to purge air out of the system.

I am wondering if my return flow to the tank is blocked....
What should the lift pump pressure be right after shutdown? Mine seems to be erratic, and decays slowly sometimes.

I am going to do the following next:
Cascade overflow valve.
Retorque crossover tubes.
Get the fittings to test return flows. Anybody know where to buy them?
Check all the fuel lines back to the tank for any kind of kink. (I built a flat bed last year, and had to do longer fuel filler tubes, I hope nothing is pinching a line).

Any other ideas?

 

OK,
Finally got the new rail pressure sensor and installed it tonight.
Fired right up, and immediately got trouble code P0193 FUEL RAIL PRESSURE SENSOR HIGH.

I also got P0477 EXHAUST PRESSURE CONTROL LOW.
(I think P0477 has been showing up a lot and I am not sure if it will light up the check engine light. Only thing that might be related to that is I do not have a muffler, but I do still have the cat.).. Or do I have a failing wastegate or variable geometry actuator?

After a few good cold starts in the shop, I cleared the codes, and now those codes do not come back.

So far, all starts are cold and I have not tested a hot start, on a hot day, 1/2 hour after shutdown off the freeway. (primary conditions to cause the starting symptoms).

SO,
Based on the fact that I got a "RAIL PRESSURE HIGH" trouble code, the sensor swap OBVIOUSLY made a significant change.
AND, this would fit the model with the problem i was having, assuming the old sensor had drifted downhill by a couple thousand PSI on its output signal.

I think this is a really good indicator that I HAVE fixed the problem. But, hopefully tomorrow's test drive to work will be the real test.

Interesting thing to note.... the old sensor has blue oxidation on the threads, indicating that it has gotten REALLY hot. The new one is clean fresh machined stainless steel. The oxidation does NOT extend to the hex, it is really only on the threads, and obviously none on the fuel side of the sensor, as no air is in there.
There is a chance this is left over from a manufacturing process, but I dont think that is the case.
Because the oxidation is only on the threads, and not on the rail itself, that indicates the hottest places were in the sensor. Also because there is a distinct cutoff where the fuel seal is, I think it happened in service, not manufacturing. Only thing that makes sense, is a short inside the sensor was drawing too much current, and over heated the sensor.
More indicators it was the rail pressure sensor causing the problems.

Thanks for all the help and suggestions guys!
I will report again tomorrow.

 

Drove to work.
Post freeway, let it sit 40 minutes.
Start was a little long, but not terrible.
Hard to say if i fixed it buy most likely not.
It is only 60 degrees out and overcast right now, so the conditions will not really let me test it.

Hopefully when I go home this afternoon it will be >85, and be a better test.

Next.
Re-torque cross over tubes.
Cascade overflow valve.
Injector leak down test.
Pressure limiting valve replacement.

Any suggestions on what order?

 

Did you note the rail pressure while cranking? Is it still low? Did you have your injection pump tested?

 

The injection pump has been tested a couple ways.

The shop i initially took it to used a fancy scanner that let them step the pressure way up and down, and it responded correctly.

I have also unplugged the FCA, and the pressure readings were good... like 26,500psi + at idle. (But not done since the new pressure sensor).

If the start is relatively good and normal, the pressure is climbing so fast there isnt a problem.... I need a REALLY bad start to see a stable cranking pressure.
When it DOES have the problem (Worst was a couple weeks ago in 100+degree heat) it will be typically 2,000psi, and some of the worst was 1800psi.

 

The rail pressure sensor did not fix it.

 

I put a pressure limiting valve in it last night.

Just tried a restart... its 60 degrees out.

Problem is still there.
Pressure limiting valve didn't fix it.

So,
Replaced: Injectors, cross over tubes, Pressure limiting valve, Rail pressure sensor, Check valve banjo, FCA.

Not replaced:
Cascade overflow valve, CP3.

Not checked: Return flow from anything... yet.

 

Hey, I'm a newbie with a 2007 5.9 and was suffering similar ihot start ssues. Brand new injectors , raptor lift pump, better filtration, etc. Found that when it was hot and hard or no start I could pour water over my CP3 pump and it would miraculously start. Picked up a good used CP3 and that solved it. Fast starts, light up at around 3700. Before it was around 2000. Not sure if this will help, I just know it solved my similar problem.