Combustion gas in fuel symptom

SteveS · July 16, 2007

I have a 2005 that is driving me crazy. The truck has a Banks 6-Gun that is currently disconnected, so the power level is stock but still has bigger exhaust, CAC and washable air filter. The truck starts and runs well as long as it is not pushed hard. If you floor board it from a rolling start it only gets most of the way through 2nd gear before starting to loose power and misfire. If it idles after that it has one or more cylinders misfiring. If it is shut off or idles about 10 minutes it smooths out again and is OK until run hard.

OK, Bruce has been convincing me that we need to step up to a better scan tool and this truck is really driving the point across; I don't have a good way to cancel cylinders or look at misfire in real time. However, it has set codes for almost all the cylinders at one time or another. Sometimes just one cylinder balance code, sometimes up to 3 different cylinders on the same acceleration run.

To me it feels like it is getting air in the fuel, although I am not seeing white smoke when the symptom occurs. Supply pressure dropped to as low as 49 psi on initial acceleration when the engine was running on all 8, runs about 56 psi at idle. Minimal suction on the inlet side of the HFCM, so I'm not suspecting that there is restriction in the tank. I have clear lines on the supply to the HFCM and the return to the HFCM and don't see bubbles in the supply fuel.

I suspected that the injectors were allowing combustion gas to enter the fuel so I tried GregH's method of filling the secondary filter canister and cranking the engine. I used 15 second crank cycles, on the third time a bubble came to the top. Every crank cycle after that produced a bubble or 2. So I removed the lines from the canister and ballooned them but didn't really see the balloons swell.

I am wondering what drivability symptom you guys see that leads you to the balloon test. Aren't there check valves at the connection to the heads that would prevent this test from working?

TIA

Steve

Bruce Amacker · July 16, 2007

I am wondering what drivability symptom you guys see that leads you to the balloon test. Aren't there check valves at the connection to the heads that would prevent this test from working?

TIA

Steve

Yes, but the check valves are worthless, you can blow through them both ways with your mouth even when they're new. Know anybody with an IDS? /forums/images/%%GRAEMLIN_URL%%/smile.gif I wonder if applying shop air to the fuel lines to the heads would prove anything? I think you're on the right track........

PS I talked to WSAFC today and suggested 16 hrs 6.0, 8 hrs 6.4 and 8 hrs IH Multiplexing.

SteveS · July 17, 2007

Yes, but the check valves are worthless, you can blow through them both ways with your mouth even when they're new. Know anybody with an IDS? /forums/images/%%GRAEMLIN_URL%%/smile.gif I wonder if applying shop air to the fuel lines to the heads would prove anything? I think you're on the right track........

PS I talked to WSAFC today and suggested 16 hrs 6.0, 8 hrs 6.4 and 8 hrs IH Multiplexing.

So this is the sort of symptom that occurs? (I haven't yet experienced the injectors that leak combustion pressure)

I'm hoping that this truck is the motivation to step up on the IDS (at least then I'll get some satisfaction out of it.)

Keith Browning · July 17, 2007

I had two trucks that were doing this. One was hard to find but the other was in my face. The balloon test is rarely effective as cranking speed may not get enough pressure to breach the system. Only one showed anything with the balloons and it was quite noticeable. Typically, you will only see a small pulse from the balloon. I like Greg's idea better too but I haven't had the opportunity to try it, but thats okay with me.

Both of mine were misfires and the longer and harder I drove them the misfire spread to one or both of the adjacent cylinders. Both trucks had a loose injector that blew past the copper seal and burnt away the lower fuel o-ring allowing combustion gasses to enter the fuel rail on that cylinder head.

SteveS · July 17, 2007

Thanks for the help, I'm going to keep going in that direction (and hope there is not something stupidly simple that I overlooked!)

Steve

Bruce Amacker · July 17, 2007

If you're tired of doing diag and want to tear into it, pull all 8 injectors out. The leakers will be discolored near the bottom. My guess is, this won't be wasted time.

Good Luck!

Keith Browning · July 17, 2007

Prudent advice. /forums/images/%%GRAEMLIN_URL%%/thumbup.gif

Here is a picture from the Balloon Test article that shows the failure you are looking for. This injector is from the bad one I encountered and wrote the article from. Arrow A shows the missing copper seal and you can see all the carbon that was blowing into the injector bore. Arrow B is pointing to the failed lower fuel rail o-ring.

Posted Image

SteveS · July 17, 2007

DUH!!! I forgot that the balloon test article was there! And I forgot to look in Articles! This is what happens when the weather gets above about 75* in the Pacific Northwest, the heat gets to me (I grew up in Pennsylvania, so I have experienced HOT & HUMID and didn't like it).

I had the empty secondary filter symptom also. Even with cycling the key 3 times it did not fill up. I found a piece of crud in the air purge orifice and cleaned that but still didn't get it to completely fill until I connected the return line to the HFCM to the line to the tank and bypassed (& plugged) the HFCM on the return side. Then a couple cycles of the key, remove the filter lid and it was full. I guess the recycle valve in the HFCM was sending the air back to the secondary filter.

I don't know the whole history on the truck, so I'm not sure if the injectors may have been touched before. Looks like the next step will be to get in there and find out.

Bruce Amacker · July 18, 2007

You're looking for really small leakage, right? How about removing the feeder lines from the secondary FF that go to the heads, and sliding a small piece of clear vinyl line over them. Fill them with fuel and crank the engine (FP disabled). Even the tiniest leakage from the injectors would show up immediately as raising fuel in the hoses or bubbles. If you're lucky it would be blasted all over the hood. This would be a more sensitive version of the balloon test or "bubbles in the FF test".

Eh?

Mekanik · July 19, 2007

I've been down this road before. The truck I had was an early 04 and it would loose power and stall only after driven really hard. I got it to happen very consistently and if I came to a stop before it stalled, I would be dropping cylinders in the right bank, but if I let it idle, it would smooth out in a little time.

What was wrong was the oarnge plug in the fuel pick-up inside the tank fell out and was sitting on the bottom of the fuel tank. On hard accel all of the fuel would go to the back of the tank and that little hole in the fuel pick-up would take a gulp of air. I confirmed this by putting clear hose in the line going to the fuel pump and on the return line.

When it first came in, I suspected combustion gas in the fuel system and replaced all of the injector O rings on the right bank. Then I put fuel in the tank and drove it really hard down the highway. It ran great and I thought it was a confirmed kill. If I didn't put fuel in the tank, I would have realized that I did not fix anything. /forums/images/%%GRAEMLIN_URL%%/shrug.gif

Anyway, I hope this helps

Brad Clayton · July 19, 2007

Excellent repair, definitely one I will keep in the back of my mind on future possible air in the fuel system problems. How much hair did you have left after that job?

Mekanik · July 19, 2007

That truck got me pretty good. I was treating it like a no start, and the funny thing is I had a fuel pressure guage on the truck and I had good fuel pressure. I was looking right at when it would crank and not start after it stalled. I was really fixated on combustion gas getting into the fuel system and that baloon test obviously didn't turn anything up.

SteveS · July 20, 2007

Thanks for the responses, guys.

Mekanik, that is a good one to keep in mind, I can see that being a real headache. I ruled that out without thinking about it; we have a dyno and I duplicated the problem while it was strapped down, no acceleration G-force. I ran it hard until the problem occured (didn't take very long) and then went underneath to look at my clear lines. I considered being under it while someone else ran it but I didn't have the guts. Never saw any air in the clear lines after it started running bad so I pretty well crossed off poor pickup at that point.

I got a little more background on the truck and apparantly it has had injector(s) replaced, so my guess is that it has loose injector syndrome. At this point I have it on hold while the customer contemplates what he wants to do (like there is a choice?) I think he does not "need" the truck and doesn't mind if it takes a while.

I'm reasonably sure that the next step will be removing the valve covers, I'll keep you posted.

Steve

SteveS · July 20, 2007

Even the tiniest leakage from the injectors would show up immediately as raising fuel in the hoses or bubbles.

I like it! The next time I do this test I am doing it like that!

Steve

SteveS · July 30, 2007

Well, we removed the injectors and found all the bottom O-rings wasted on all the driver side injectors. 3 appeared to have combustion leakage getting past the washers, although none were as dramatic as Kieth's picture. Another tech got the job and reported that none of the hold-downs were loose. Turns out that all the right side injectors were replaced previously. Since hind sight is 20-20 I wish now that I'd just pulled them out rather than spending as much time as I did trying to prove that was the problem by testing. It would have been interesting to have an IDS and see if the relative compression test showed the leakage.

Thanks a lot, guys, I really appreciate the help!

Steve