Occasionally the EGR valve(s) fail. Usually they stop moving through their entire stroke. Sometimes the electrical portion develops a fault. This will set a CAN code (which I don't remember right now, but I don't think it's one you've listed). If they stop moving through their whole stroke, they can sometimes be cleaned. Once the solenoid is removed, the plunger is revealed. Tons of crud can build up inside. It can be scraped out or wire brushed.
If both valves are stuck in a very open position, I could see that causing a low rpm smoke/low power/poor running condition, but I haven't encountered that in person.
The EGR cooler flex pipe crack is an issue. If exhaust gas leaks through, this is energy being lost that would otherwise be used to spin the turbo. More often, the flange that bolts to the exhaust manifold breaks free from the rest of the EGR pipe. After a lot of thorough cleaning, it can be welded back together. I imagine that can be done with MIG and regular steel wire, but since it's stainless steel I TIG weld them back on using a backing flux and stainless filler rod. The EGR coolers (on all models) are of great quality, aside from the occasional flange breaking off. I've yet to see an EGR cooler fail in a way that allows coolant to be consumed by the engine.
P0500-This occasionally pops up on trucks that desperately needed injectors a year ago. When the injectors have utterly, completely failed, the engine's rpm is all over the place. Huge spikes and dips in what's supposed to be a smooth line. The ecu can sometimes interpret that as an engine speed sensor fault, because the data doesn't make sense.
Mostly unrelated anecdote:
A new coworker was doing his first set of injectors and a rocker box gasket. Getting some of the stuff unhooked at the rear of the engine is a challenge because of limited room, and he accidently pulled the two wires out of the camshaft position sensor connector. No big deal, put two terminals on and plugged them in. Later, after the job was done, he fired it up and the engine is running like dog doo. Terrible. Some poking and prodding later, and some head scratching, he switched the position of the two leads at the cam position sensor. The engine instantly ran correctly.
Another interesting footnote: If one was trying to do a backyard version of a cylinder balance test (by using a Cummins blocking tool on the fuel rail, or even by using DX), and the ecu is messing up the results by changing the cylinder correction in an attempt to get it to idle correctly, unplugging the cam position sensor will remove the ecu's ability to adjust individual injectors and a true reading will result.
SCV testing: The SCV, aka suction control valve, aka fuel pressure regulator, is the cylindrical looking thing on the top front of the fuel pump. It has an electrical connector. It's failure leads to weird idling, surging, P0087/P0088. None of this sounds like what you are dealing with. However, this test is also used to evaluate injectors on trucks made from 05-10.
A recording is made of Engine RPM, Coolant temp, injection quantity, Final value of pump current target (target scv current), scv driving current value (actual scv current, but not obtainable on 05-07 trucks), Target common rail pressure, actual common rail pressure.
A completely warmed up engine is allowed to idle for one minute, run at 1500 rpm for one minute, then the accelerator is mashed to the floor and held there for 5 seconds, returned to idle for 5 seconds, mashed to the floor again for five cycles.
What we're looking for is smooth rpm at 750 and 1500 rpm, a relatively smooth scv current value with actual current following target, rail pressure target and actual being very close to each other, injection quantity staying above 5-6 mm3 at 1500 (higher the better, with new injectors around ten, and ultra fragged ones going below 5).
If the scv current is all over the place and target doesn't meet actual, but the rail pressure is pretty stable, then the scv is probably scrap. If scv current is stable, but the pump cannot reach target or is very slow to do so, it's dumpster ballast (provided that there are no fuel restrictions or air leaks in the fuel supply).
So lets say that scv is good, but rail pressure is pathetic. A few things to check before condemning the pump: uninterrupted fuel supply. Occasionally you'll catch me doing laps around the block with a jerry can ratchet-strapped to the truck's steps. In that case, I've got the pump hooked directly to the jerry can, no filter or anything. I'm ruling out every single part of the fuel supply. If the truck runs great on the can, but has problems hooked up correctly, then I know there is a supply side problem. It is important to try things like this when one isn't totally sure where the problem is.
If rail pressure is still pathetic on the can, the rail pressure limiter valve should be checked for leakage. The hard line is removed from it and plugged, then the truck is run at wide open throttle. If anything more than a drop or two of fuel comes out of the rail limiter, it has failed. Isuzu trucks like to blow these, they set something like a P0193 Large fuel leak detected code, or something like that.
Edit: I've never seen a rail pressure sensor fail. I'm sure it happens, but that's never where I've ended up while troubleshooting.
Interesting anecdote: Isuzu NPRs with common rail seem to run just fine with a Hino rail limiter, not that I have any first hand knowledge of that or anything.
I can go on and on and on, but what I really want is that .txt file. Like, before 5pm on Friday.
As far as my location is concerned: PM me.
Last edited: