CAT HEUI issues, failures and fixes

[Ronmar]

Yea it could be...

the ECU uses a Pulse Width Modulated signal to control that pressure reg valve Based on what it sees on the pressure sensor. So if the system doesn’t see a sensor, it will either not fire the injectors same as if it does not see adequate pressure, or it will detect that no sensor is connected and feed a basic PWM signal to the regulator valve that will deliver an adequate if not ideal injection pressure to allow the engine to at least function with a check engine light…

If you can read ECU sensed pressure with cat ET, you could pretty easily rig a test source with an old grease gun, oil a T fitting and a pressure gauge to compare cat ET sensed pressure to that applied by the test rig to the gauge and sensor…

 

[GeneralDisorder]

Yep have to get 870 psi or no love making between ECM and injectors.

Check the fuel delivery system. The lift pump on the back of the HEUI pump, the fuel hoses, the seals on the various filters, etc.

Engine should start and run with the sensor unplugged or faulted out but will have no power - no boost without proper injection control - the fueling will be in a low power state to prevent the engine from spooling up the turbo. Assuming it's getting fuel.....

 

[87cr250r]

That is quite possible. The HEUI system operates the same as the newer common rail systems. There is a minimum pressure required for a start. No dribbling allowed.

 

[Skyhawk13205]

So, "...injectors are not fired until..." pressure is sufficient. I wonder if we just replaced a pump that needed a new sensor—

View attachment 946879

Does the value change on ET? My experience with bad sensors is a hard failure, were the sensor produces no computed data usually outside of range or unreasonable data range, intermittent failure give correct/incorrect reasonable values and unreasonable data range. Then the unannunicated failure of incorrect values with reasonable data range. The last 2 are the hardest to diagnose. Usually to rule out a sensor you have to compare the another data source that is related or use a stand alone measuring tool, like a high pressure gauge. I don’t think there is another source of data for the high oil pressure so you might need an external gauge to value the high oil pressure to verify the sensor failure.

 

[Skyhawk13205]

Does the value change on ET? My experience with bad sensors is a hard failure, were the sensor produces no computed data usually outside of range or unreasonable data range, intermittent failure give correct/incorrect reasonable values and unreasonable data range. Then the unannunicated failure of incorrect values with reasonable data range. The last 2 are the hardest to diagnose. Usually to rule out a sensor you have to compare the another data source that is related or use a stand alone measuring tool, like a high pressure gauge. I don’t think there is another source of data for the high oil pressure so you might need an external gauge to value the high oil pressure to verify the sensor failure.

The troubleshooting steps are not much help

 

[hike]

Yep have to get 870 psi or no love making between ECM and injectors.

Check the fuel delivery system. The lift pump on the back of the HEUI pump, the fuel hoses, the seals on the various filters, etc.

Engine should start and run with the sensor unplugged or faulted out but will have no power - no boost without proper injection control - the fueling will be in a low power state to prevent the engine from spooling up the turbo. Assuming it's getting fuel.....

New fuel line from tank, new primer/fuel separator, new line to lift pump, old line to primary, fuel at tee after primary. You can hear fuel return to tank when pumping a hand primer. Primer stays hard until after cranking, still firm, just not hard.

Maybe the pressure valve at the rear of the block needs to be replaced? I recall @Ronmar saying the lift pump provides 50 - 60# of fuel pressure, though the hand primer generated only about 10#.

Does the value change on ET? My experience with bad sensors is a hard failure, were the sensor produces no computed data usually outside of range or unreasonable data range, intermittent failure give correct/incorrect reasonable values and unreasonable data range. Then the unannunicated failure of incorrect values with reasonable data range. The last 2 are the hardest to diagnose. Usually to rule out a sensor you have to compare the another data source that is related or use a stand alone measuring tool, like a high pressure gauge. I don’t think there is another source of data for the high oil pressure so you might need an external gauge to value the high oil pressure to verify the sensor failure.

Was thinking to pull the sensor connection to see what CAT ET reads then.

Before pulling the sensor CAT ET read 283#, last night it was 289#. Both times though it read 0# or a static reading. No fluctuation. Pulled the sensor connection.

Success: Truck fired right off once we pulled the sensor connection. CAT ET 'read' injection actuation pressure to be 2583#.



Reading through the TM and CAT's HEUI testing mentions attaching an external gauge with the sensor, (@Ronmar idea). At this point I think I'll just get a new sensor. For a shade tree DiY mechanic like myself it may be best practice to just replace pump, o-rings, sensor all together and have a newer system in place.

We'll pickup a sensor tomorrow and hopefully be fully back up and running.

Thank you all for your help—

 

[Skyhawk13205]

New fuel line from tank, new primer/fuel separator, new line to lift pump, old line to primary, fuel at tee after primary. You can hear fuel return to tank when pumping a hand primer. Primer stays hard until after cranking, still firm, just not hard.

Maybe the pressure valve at the rear of the block needs to be replaced? I recall @Ronmar saying the lift pump provides 50 - 60# of fuel pressure, though the hand primer generated only about 10#.





Was thinking to pull the sensor connection to see what CAT ET reads then.

Before pulling the sensor CAT ET read 283#, last night it was 289#. Both times though it read 0# or a static reading. No fluctuation. Pulled the sensor connection.

Success: Truck fired right off once we pulled the sensor connection. CAT ET 'read' injection actuation pressure to be 2583#.



Reading through the TM and CAT's HEUI testing mentions attaching an external gauge with the sensor, (@Ronmar idea). At this point I think I'll just get a new sensor. For a shade tree DiY mechanic like myself it may be best practice to just replace pump, o-rings, sensor all together and have a newer system in place.

We'll pickup a sensor tomorrow and hopefully be fully back up and running.

Thank you all for your help—

I am guessing when you pull the sensor the ECM just assumes the pressure based on engine RPM and spool valve current. It must use a trend line or a table to estimate the values. Glad you are making progress.

 

[Ronmar]

I am guessing when you pull the sensor the ECM just assumes the pressure based on engine RPM and spool valve current. It must use a trend line or a table to estimate the values. Glad you are making progress.

Yep, basically a limp mode so the truck is not completely disabled. As General said, it is a bare minimum amount of power, not enough spool the turbo, maybe just enough to move the truck off the highway or idle it around the repair yard