Another Surge Issue

[WSnyder]

I've been fighting a surge issue for about 5 months now and i've searched the forum and followed all I can find. I'll lay out how it began and where I am now as simply as possible.



I have a basically stock 2004 3500 w/6 spd and ~ 75,000 mi



My problems began when an injector failed on a road trip. I experienced the typical misfiring and lots of white smoke when the misfire occurred. I found if I babied the throttle I could minimize the misfire and limped home.



At this point I decided to just replace the injectors with a well known vendors 50 HP units for a bit more power. Following sag2‘s post with the factory install procedures in went the new injectors and all seemed fine. The truck was back to idling very smooth, starts easy and had good power. I set off on another road trip and about 20 mi out of town when things settled out on the highway (~78mph ~2100 rpm) there was a very noticeable surge/bucking feel to the truck. At first I thought the cruise was acting up but it did it w/manual throttle as well. At my first fuel stop on the trip I also noticed the mpg had dropped to 11 mpg which was about a 30% drop from the norm on the same route I have taken many times before. I had never experienced the surge with the stock injectors. When I got back home I called the vendor and they felt I had a fuel leak and I needed to re-torque the connector tubes. Fair enough and sounded plausible. I tore the truck partially down and re-torqued the connector tubes. The vendor asked for me to re-do the complete procedure (loosen everything including injectors, basically start over but with a few minor install instructions he had me follow). I did this and the same surge remained. At this point I hooked the truck up on a Snap On Verus scanner and went for a test drive. When the surge happens it is most noticeable on the highway at 75+mph and ~2100rpm with light throttle. On the Verus I noted a rail pressure oscillation of 1500-1700psi peak to peak that coincided with the surge. With the surge going on I get the following data from the Verus:



APPS Voltage steady (virtually no variation)

APPS % steady (virtually no variation)

Fuel PRS (PSI) 1500-1700 psi peak to peak fluctuation & coincides with surge

Fuel Press (V) fluctuates & coincides with rail pressure and surge

Fuel Set (PSI) steady (virtually no variation) (~15K PSI asked for pressure)

Fuel PRS Reg (%) steady (virtually no variation)



At this point it looked to me like the ECM was asking for a steady pressure and I was getting approximately the asked for pressure but a fluctuation in actual pressure. I now pull the injectors and send them in for test to the vendor. Injectors come back good, I was told they test fine and hold full pressure. Inspected connector tubes and visually look okay, same with the injector ports where the connector tubes interface with the injector. In go the injectors again. Install connector tubes and torqued to a bit extra to 40 ft lbs. Test drive and still have surge. Vendor says maybe its not leaking and no hard start issues common with connector tube leaks. Maybe something else is causing the pressure variations. 
Install new FCA, APPS looked great on the scanner so didn’t replace that, new rail with pressure sensor and PRV. Surge still exists. Vendor wonders if I have a CP3 problem. At this point I have a bit of frustration. I feel I’m headed down the wrong path so I search TDR again. Following sag2’s posts and great pictures I buy the Miller SPX fuel test tools. I install the 9012 and get separate lines set up to measure fuel returns. CP3 return is right on spec. At idle I have virtually no return fuel from the head but with the FCA disconnected and 1200 rpm I have a huge return/leak at ~850-900mL a minute. I use the 9011 cap and find every injector is contributing to the leak. What the heck? Talk it over with the vendor and we are puzzled (the vendor has been helpful through this) so I/we decide on new connector tubes. I get the new larger 6. 7 tubes that supersede the 5. 9 tubes. In go the new tubes and no joy, the leak rate is exactly the same. Vendor is now sending me a spare set of injectors to test.



My question is has anybody had this type of trouble? Injector/connector tube installation isn’t rocket science people do it everyday with little trouble. I’ve tested all the tube nuts in the head without the tubes and they thread fully into the head with no trouble. When installing the tubes and the nut just snugged up by hand I can twist the injector and feel the tubes wiggle side to side so they are seating into the injector. Can an injector leak under high pressure in the truck where they won’t on the test stand? I guess I’ll find out if the injectors are the culprit when I get the test units installed after they arrive next week. Any comments would be greatly appreciated.

 

[cerberusiam]

Well, did they check return flow rates on the injectors and validate that is not out of spec? Can that even be done on an injector bench? Too many times we see the same thing, vendor says injectors test good and hold pressure but show high return.



What exactly do they test is the big question.

 

[jgillott]

I fought a very similar surge/bucking situation with my '07 on and off for approximately the last 4 months. Mine was random at first and then became rather regular under all throttle positions and driving conditions.



Making a very long story short, we found that there was an intermittent connection issue in the wiring harness leading from the fuel rail pressure sensor to the ECM. When the surge/bucking would occur, you would see a sudden dip in rail pressure as you described in your post. Now the ECM would not throw a code for the rail pressure sensor, rather it would apparently either momentarily increase or decrease fueling based upon that sudden change in voltage in the pressure sensor circuit. With that harness fixed, the truck runs perfectly.



If you hadn't just been working with injectors and found the issues with the return/leak you described, I'd be highly suspect of something similar to what I have been dealing with.

 

[Prairie Dog]

But could a connection issue (bad harness, etc) cause a high return rate?

 

[jgillott]

But could a connection issue (bad harness, etc) cause a high return rate?

No, most likely not. It just sounds so darn similar to what my truck was doing.

 

[TCDiesel]

What is the default setting of the (MAP) sensor if 14-12 its good if not replace it. . Many times speed sensors will cause this also, Did you replace with Genuine NEW Bosch? Flow rates are just a touch Tone,unless the truck has other symptoms I doubt its injectors, Check all wiring, grounds, batteries, charging as suggested,This same issue was just fix by us,it was found to be a short in the cig liter.

 

[WSnyder]

cerberusiam, when the new injectors came back they included a test printout that shows flow return (Back leak flow) at different pressures and injection speeds. Some of the injectors are close to and some are at or slightly over the factory spec limit (30 mL min per injector max spec) at max test pressure (1499 bar, 21735 psi). I have no idea if the bench data is valid 100%. The vendor seems pretty confident in the data. I had another vendor tell me they cannot expect their test bench to run/test injectors the same as in a vehicle and they have had bad injectors that test ok on the bench. A new MAP sensor was installed at the same time as the FCA and rail (rail sensor and PRV). I of course changed one at a time to try to find the culprit with no change in the surge. I have checked, removed, cleaned all harnesses and connectors. This includes the hard to get to one at the rear of the block near the bell housing. I also checked this harness for rubbing and shorted wires around the bell housing area. Batteries are new Odyssey's that went in shortly before the injector change. This truck has no cranking or starting problems at all. Yes, new Bosch w/50 HP nozzles, at least the vendor says they are and many here on TDR use their injectors so I don't think they are slipping used/reman ones over on me. TCDiesel I don't know what you mean by "Flow rates are just a touch tone"??? sag2 has posted that the factory specs say 180mL min for max return and I have 4. 75x that much (850-900ml min) do you think that's not a problem?

Just thinking out loud, I don't think i can get a frequency on the Verus pressure data to correlate it to RPM, if I could it might clue me in on whether it coincides with injector firing. i may have to look into that further. During the surge when the rail pressure is fluctuating the frequency of the pressure fluctuation is very uniform and stays fairly stable. Another note that I forgot is the actual rail pressure at idle varies approximately 400 psi peak to peak. I have no idea if that is normal or not but the vendor seemed to think that was not okay. The truck does surge at other speeds too, it's just most noticeable and easiest to reproduce on the highway at the stated speed and RPM.

 

[TCDiesel]

Yes flow rates will be effect By many combinations,Saga is correct But I have seen engines as high as 250ML running just fine... 850-900 is just way to high. . I would start shutting them down 1 at a time, I would divert the fuel form the low pressure port of the CP3 and measure its return on every Injector at the same time,Also divert the fuel for the LV and keep and eye on it,I also at times will cut the return flow down on the low pressure side of the Pump to see if it increases the return flow of the Low pressure side of the Injectors, also we will change the signal to the FCA to slow down its cycle and see what effect this has on the performance of the engine,these procedures are dangerous and I would only recommended to perform at own risk. If all this checks out We start to look at the PCI & Bus of the truck.







You cannot also simulate the condition on any stand,its the host that counts.

 

[WSnyder]

Update: About 3 weeks ago I installed another set of known good injectors to test with.



Fuel return dropped to about a third of what my suspect set had. On my test drive I noticed the throttle response in town was now back to normal with no funny surges while getting on or off the throttle. Once on the highway I could not detect any surging at any of the speeds I would routinely have surging with the suspect injectors. The next step was to test drive with the scanner. The test drive on the highway with the scanner showed everything was most likely back to normal as far as I could see. Actual fuel pressure varied approximately 300 psi peak to peak maximum (tended to be 100-150 psi peak to peak) at approximately an asked for 15,000 psi. Idle fuel pressure was way down to 20 or 30 psi peak to peak from the previous 400 psi peak to peak if I remember correctly. Fuel consumption (MPG) appears to be back to normal as well.



So it seems I've been vindicated in my feeling that the original injectors supplied by the vendor are bad. Actually looking over the leak rates I would say that 2 of the suspect set are possibly okay and a 3rd might be okay as the leak rates aren't too far out of spec.



It looks like TCDiesel is correct in that injector test benches are not infallable, what matters is how injectors test in the truck. Somewhere along the way in this process I doubted the test bench. Some of you have had your doubts as well and another injector vendor I spoke to told me they have had bad injectors in the truck but couldn't see it on the test bench. They said it's rare but it happens.