Fortunately this system is very similar to the Denox 2.2 which allowed us to diagnose the fault and gather the evidence but I wanted to go over some of the differences. Firstly let's take a look at the schematic for both the Denox 2.2 and the Cummins.
1. Dosing supply module
2. AdBlue reservoir
4. One way check for the valve and throttle. Throttle is used to generate pressure by restricting the flow
5. Pressure sensor
6. 4/2 flow control valve
7. Motor-driven single direction, fixed displacement pump
8. 2/2 solenoid valve (DEF Injector)
Above is the schematic for the Denox 2.2 where the pump, sensor and filters are all contained in the dosing module. Below is the Cummins version where the key difference is the sensor is now in the injector module and note the extra filters in the tank and the injector module.
1. DEF reservoir
3. 4/2 flow control valve
4. Single direction, fixed displacement pump
5. Pressure relief valve
6. Throttle/orifice to generate pressure by restricting flow
7. Pressure sensor
There are 2 components which are different between the two systems. The pressure relief valve over the pump in the Cummins and the one way check valve in the Denox 2.2. The check valve plays a big part in the purging of the Denox system which we will talk about later.
In the Cummins system we see at marker 5. a pressure relief valve over the pump. Its job is to protect the system should there be any over pressurisation in the pipes downstream of the pump. This is marked as variable meaning it could be adjusted but I would not advise adjusting and although not having opened one of these pump units up, I’d be surprised if it was adjustable. What we found from a case study was that the system is quite fussy when it comes to pressure and monitors it very closely. If it doesn’t like the pressure reading it is seeing it will not activate the injector. This could be both due to the pressure being too low or the pressure being too high.
DEF pump from a DAF CF
The injector module is very different to the Denox 2.2 setup. As seen in the image below it is considerably larger than and has both a supply and return of DEF fluid. What is also notable is the lack of coolant pipe work. This is because the DEF fluid flow is used to remove the heat of the exhaust from the injector. Included in this module is the pressure and temperature sensor as well as an additional filter.
Injector module from a DAF CF
As with the Denox 2.2 the control of the system is very similar. There is a priming phase where the pump builds up the pressure to around 9 Bar (130 psi), dosing phase where DEF is injected into the exhaust stream and the purging phase where fluid is drawn back to the reservoir to prevent the DEF freezing in the pipework.. There are of course heating stages as well but I’ve focussed on the main functions.
If we take the initially priming stage we can see the similarities. The waveforms captured below were created by carrying out an active test from a scan tool.
In step one for both of these systems we system that as the pump is activated the pressure begins to rise. I’ve also included the math channel for the duty cycle of the pump control as seen in Channel C, Green, for the Denox 2.2 and in Channel B, Red, for the Cummins. This helps to show how the pump is commanded during each step. In the Denox 2.2, the pump is activated and pushed to around 90% duty to get the pump to produce a lot of flow, pressure = resistance to flow. The same control is used for the Cummins.
What is interesting with the Cummins system is you would have noticed Channel D is active, Yellow, which is the feedback from the pump speed. By including a freq math channel we can show how the speed changes with regards to the pump command. This could be useful if we have a discrepancies between the pump control and the pump speed. In theory they should follow each other with a slight delay between the control and the feedback. As you can see in the image above, they do follow each other as we would expect.
During the dosing section both systems appear very similar in that the pressure remains stable till the injector is activated where we see the pressure drop.
As you can see whenever there is an injector event, it is shown in the pressure signal. Depending on how long the injector is operated for, depends on how much DEF is injected and so how much the pressure drops. For the active test in the Denox 2.2 the pulse width of the injector is quite long so there is a dramatic drop in pressure. With the Cummins, the active test is for a quantity test and so continues to spray at a smaller rate but for a longer period of time. This means the drop isn’t as significant but it is still clear from within the capture.
What is interesting is the change in the duty of the pump for the Denox 2.2 when the injector fires. As the pressure drop is significant, we see the pump is commanded to speed up to produce more flow and therefore restore the lost pressure.
Some additional information within the pressure signal during the dosing stage is by zooming in on the signal you can see the pressure ripple which is generated by the pump. Should there be a pressure issue, this signal could prove useful in the diagnosis.
In the above captures a lowpass filter of 1kHz has been applied to the two pressure signals on Channel A, Blue. As with a lot of scope diagnosis we are looking for those repeating patterns. Should there be anything out of the ordinary between a known good such as the above and the capture you have it could be that something is going on with the pump itself. Looking at the pressure signal is something we use when looking at variable displacement hydraulic pumps with WPS600C as it allows us to see the change in the pressure based on each piston stroke from the pump.
Back to aftertreatment! The purging phase is required to draw back the unused DEF into the reservoir where the ECU can monitor the temperature to ensure it doesn’t freeze and block the system. There are some slight differences between the two systems and how it affects the waveform. Looking at the Denox 2.2 first we come back to the one way check valve.
As the system is shutting down and the purging cycle begins both systems switch the 4/2 flow control valve so the pump is no longer pulling DEF from the tank but instead is pulling it back from the supply lines. 4/2 means that the valve has 4 ports and 2 positions. With the Denox 2.2 there is a one way check valve in the return line which will allow the DEF to pass back to the tank when the system is primed and dosing. As soon as the flow control valve switches the DEF is now being pulled back through the supply line and quickly as well due to the pump control being back up at 90%. This causes the one way check valve to shut and starts to pull a vacuum, or pressure below atmospheric. In order to prevent the pipework from collapsing the injector is pulsed at a high frequency to allow air in the system and so controls the amount of vacuum generated. This means the system can draw the fluid back quickly back to the tank once the vehicle has been switched off or the activate test stopped. This section is important when looking for blockages as per Komatsu case study where the injector was blocked due to over heating - https://www.picoauto.com/library/case-s ... blue-fault.
The Cummins doesn’t have this check valve though and as such the rate of change during the purging stage is much slower than the Denox 2.2. We also see that the pump speed is much slower as shown in the duty cycle of the control and feedback frequency. I unfortunately didn’t capture as much time as I should have to see how long it would take before the pressure in the system returned back to the atmosphere. One thing to note though is the DEF lines on the Cummins system are heated which isn’t something I’ve seen on the Denox 2.2. This could explain why it doesn’t have such an urgency on getting the DEF back to the tank as should the temperature drop too much I would assume it would heat the lines as well.
Hopefully this shows that despite the two systems being slightly different, there are similarities. By understanding the differences we can use the same signals between the two, and with the tools in Pico, to help diagnose issues with these aftertreatment systems. I hope this helps.
All of PACCAR (NA and EU) swapped to the Cummins system with the 2021 emissions updates, I haven't seen anyone dive this deep into it yet!
One note - in NA, PACCAR has been using heated DEF lines since the DEF lines were introduced (2010). I'm not sure if EU used them before though.
Think the heated pipes are market dependent but I would have assumed the pipes here in Europe would have been although I didn't see anything on them to suggest they were.
Lee Sharp, aka @Sharpy, had the opportunity to take one apart. This module had an issue with incorrect temperature and pressure readings.
Strange idea to use tracks rather than direct connections but I guess it is more serviceable this way.
Thanks for the note regarding NA and the feedback. Hoping to bring some of this content to guided tests in the future to help HD techs.
Great post keep up the good work Ben