Commanded fuel pressure stuck

[windowrattler]

Have a ISX 2250 that I'm having a weird issue with. If you monitor fuel rail pressure commanded and actual it will get "stuck". I do have video but it's shakey. I'll try and describe it as best as possible.

The commanded FRP will all of sudden just stop changing. Tends to get stuck at three different numbers. 650 bar, 1056 bar and 1475 bar and will randomly change between the three values. When this happens the rail pressure tends to stay right around 900 bar +\- a little bit no matter where the commanded is "stuck" at and no matter the driving condition.

When this event happens with the oem cal it doesn't pop a check engine light. With a modded cal it trips a CEL sometimes for rail pressure. When it's doing this it can last a minute or two. It's not just a second or two thing.

I lean away from a physical problem like the fuel pump actuator because once commanded fp starts acting like it should everything is fine and it builds rail pressure like it should. I know the engine terminating resistor by the aft treat might not be hooked up but I don't think a datalink problem would present itself this way. So that kind of leads me to ECM being flakey. My other thought is its some type of protect mode or default mode. Thoughts?

[tree98]

(11-10-2023 )windowrattler Wrote:  I know the engine terminating resistor by the aft treat might not be hooked up but I don't think a datalink problem would present itself this way.

The fuel pump actuator is on J1939. So if your missing a terminating resistor the data is getting clobbered, that could explain why the commanded is sticking on certain pressures, maybe theres default pressure settings in the program that the ECM uses if it isn't getting the information it needs to calculate commanded rail pressure correctly (just a guess) ??

This could also explain why it won't throw CEL with the stock program, when all the components are operating on the J1939 the signal might just get clobbered and not make it where it needs to go. The modded cal eliminates all the traffic and the signal makes it through.

Those are just the first things that come to my mind after reading what you posted. If I were you I would take the 10 minutes and replace the resistor, rule it out atleast so you can move on instead of guessing and wondering.

[Rawze]

It could also simply be the number not updating on your screen. == And yes.. it absolutely could be due to communications issues if you have an open ended J1939 (reading roughly 120 ohms or higher with power off).

If that is the case then ... FIX THE DAMNNED J1939 AND STOP GIVING YOUR ENGINE EXCUSES TO TEAR ITSELF UP!!!>


An open ended J1939 can and will cause damage to every single component connected to it.. if not burn something out due to excess electrical noise!!!>. I HAVE SEEN PEOPLE GO THRU MULTIPLE TURBOCHARGER CONTROLLERS.. MULTIPLE ECM'S,.. AND A LOT OF OTHER PROBLEMS WHEN THEIR j1939 NETWORK WAS HALF-ARSSED AND NOT IN GOOD SHAPE due to some complete moron unplugging the harnesses (and therefore the terminating resistors) to the DPF/SCR systems.

[windowrattler]

Perfect. Its honestly the best answer I could get. I will pass on the info. Last night I couldn't dig too deep into the issue as I was already a few hours late to leave out of state. My other half wasn't impressed with me last night. Lol. I will update on the results good, bad or ugly.

[windowrattler]

Figured I'd update this post. We did fix/tidy up the can network. Still was having an issue. Was able to catch the fuel pressure going into the negative reading when the hiccup happened in the drivability. Well there is no way that the negative reading was correct so we put a fuel rail pressure sensor in it. Still had the issue going on with commanded staying at three levels and actual staying constant at 13k no matter what was commanded.

So what I was seeing the symptom not the cause of the issue. I was assuming that the difference between commanded and actual when "stuck" was tripping the 1911 code. What was going on was when the under full load fuel pressure would be fine at 37k when you go to shift you get a small increase in fuel pressure to about 39k when the injectors shut off for decel. The fuel rail pressure relief valve was opening at that 39k dumping the fuel pressure too quickly down to about 13k. The ecm would then go into a protect mode which keeps fuel pressure constant and commanded at the few different levels. After a little bit the ecm will "reset" the fuel pressure which causes a short engine cutout, fuel pressure sensor will cycle (where my negative numbers were coming from) and it will start operating 100% correctly again until the next time the relief valve opens. After the replacement of the relief valve fuel pressure doesn't dump at a shift like it was.

The code we were getting was 1911 and eventually a 3727 (which helped us look at the data more correctly). When 1911 is active that is what makes the fpr "stick" in some default values/limp mode until it resets itself and returns to normal.

[JimT]

So, if I'm reading this correctly, your issue was a faulty fuel rail pressure relief valve? Your saying it would open to relieve a spike in pressure then close too slowly allowing pressure to drop too low?

[windowrattler]

Not sure how I missed your response. Yes it was the fuel rail pressure relief valve opening at a little too low of pressure. The ecm would see this sudden drop and go into a limp mode for a minute or so and then self clear until the next time the valve opened too low.