EGR Bypass for CRD

[glend]

I have decided to go ahead and bypass or shutdown my EGR, as I feel that this has been causing my rather lumpy idle of late, which is very annoying at crawl speed.

It is well known that pre-2005 models can disable EGR with the blocking of the small vacuum line as covered in other posts here (or though CRDJON's small can approach). For years 2005 onward, the EGR is electroncially controlled via sensors, and the MAF plays a key role.

The easiest approach is to simply unplug the MAF on the top of the airbox, but this generates a CEL and code (usually a P0102).

The Liberty Owners in the US have developed the messy SEGR kit to splice into the wiring loom and spoof the EGR signals to keep it closed and to eliminate CELs and codes.

Green Diesel Engineering (GDE) has developed an ECU flash that minimises (or mostly eliminates) EGR activity (and of course they must tell you that this should not be used on the road in the USA - which their buyers generally ignore).

I have been involved in email discussions with GDE about software differences in USA and Export models. They have confirmed that Export software is different, and there are things that they will change in software flash eventually delivered to Australia or European markets. GDE ECU flash availability in Australia is probably at least six months off, if/when they find a local distributor to handle ECU swaps and burns.

So last night I went ahead and unplugged the MAF and did some testing. First time it was unplugged (prior to starting up) it did not generate a code or CEL, and I must admit I got excited about finding that out. Subsequent starts did generate a P0102 and a CEL. These are easily cleared with my scanner but annoying none the less. I am assuming that some sensor state was not cleared through the first unplugged start but was altered later. The MAF can supposedly be unplugged after the start up without generating a CEL or code (but that will get real old very quickly if you have to pop the bonnet each time you want to start the vehicle).

Performance wise my previous lumpy idle and shudder crawl has disappeared with EGR no longer opening (and idle is where it was heavily used). Getting moving off the line is much crisper and the engine seems to respond better at small throttle adjustments. Idle RPM has dropped slightly but it is very smooth now. I haven't done Cmohr's "headlight smoke test" yet but it is obvious in performance that it is not dumping hot gases and soot into my intake tract. I can still generate smoke out the back (through overfueling by standing on the go peddle).

I'll carry my scanner in the car from now on, but probably only clear it before trips (where you might want to know if a new CEL comes on, (and why)).

I'd appreciate SAKJ getting back to me on how he managed to disable the CEL and code on his late model CRD.

I've duct taped up the MAF plug and socket, but will reinstall it for Jeep service (especially because I am still under warranty).

Here is a question: If my EGR was causing my idle and crawl performance problems should I go ahead and plug it back in to try and get it to fail - so that I can get it replaced under warranty, or just leave it disabled? I have about six months of warranty left.

BTW shutdown is not noticably different, although I don't seem to hear the FCV closing now.

Anticipated benefits (in addition to driveability and performance ) are increase fuel economy and generally cooler running temperatures, elimination of further soot buildup in the intake body, cleaner Boost Control Sensor, cleaner oil, etc.

I'll come back to this thread when I get some reasonable fuel economy data.

[Lancer]

That's interesting. I live in England, and got very excited when I first read about the GDE flash (I know, I need to get out more). When I first emailed GDE about any differences between US and UK/export software, they told me that there weren't any! In fairness, this was several months ago, and the differences may have become apparent since then. It's a bit moot right now, as I'm unemployed and can't afford it, but it's certainly a mod I'm keen to do once I'm back in work - especially given the economy and driveability improvements reported by users.

[Auberon]

Please understand what you are dealing with before you fiddle if that is possible. Following is an exceptionally brief outline of what you are toying with:

The Standard Embedded Bosch Control strategy for the EGR and TVA systems uses proportional plus Integral (PI) control structure with feed-forward terms. The MAF is measured upstream of the compressor and is used to close the loops on the EGR and TVA systems. The 2 outputs are switched by the controller by suitable logic and divided into the EGR and TVA systems. The gains of the 2 PI controllers depend on suitable maps resulting from the engine fuelling and speed.

The setpoints of these two coordinated controllers are derived from extrinsic mapping driving actuators to maintain combustion conditions as closely as possible to the stoichiometric ratio.

In the open-loop mode it cannot guarantee accurate tracking due to engine variability, aging and driving environment. The BOSCH strategy employs emperically determined parameters (by VM's own calibration engineers) but there are LIMITS to this automatic strategy as this tends to assume linearity.

There is also a relationship to which I have pointed before that relies on the sensor creating a feedback loop to the standard ECU through a series of COMPLEX steps based on first principles.

Predictive controllers have the advantage that actuator positioning is no longer needed. This is probably what MAY allow you to block off ports....at a pinch on certain models but your gains would be questionable. Generally, setpoints should be simultaneously reached and should cover the relevant operating parameters (pre-determined from known data sets).

None of this can be done without consideration of engine operating conditions including cetane number, fuel freshness, air density, RH, TP, load, oil type etc.

So my advice is to tread carefully.
Cheers
Auberon

[glend]

I don't understand any of that, nor do I necessarily care about the state of the logic in the system designed to bugger up my engine in the first place. I do understand that unplugging the MAF has solved my lumpy surging idle and improved low speed drivability. LOST users who have taken this step have shown that there appears to be no long term negative affect on the engine.

[Pipeliner]

I understood stoichiometric ratio - the rest struck me as the sort of stuff designed to totally confuse anyone who isn't a contol engineer. Probably very wise and full of important stuff but totally unintelligible like most PhD Theses.

[glend]

I'm going to take it in to Jeep in the next couple of weeks and see if I can get the EGR replaced under warranty. After reading through the Section of the Service Manual that deals with the EGR I found the following quote:

"A malfunctioning EGR system can cause engine stumble, sags, or hesitation, rough idle, engine stalling and poor driveability." That's pretty much all of my symptoms bar the stalling.

To me, the diagnostics that I have conducted point to the EGR failing. Both powering down the vehicle and unplugging the MAF stopped the problem, which indicates to me that either it is the ECM or the EGR valve. After powering down the problem returned shortly after (three or so trips) so that may indicate its not the ECM.

Despite raising this with Jeep at my last service, they say they could find no 'codes' to deal with nor did their alledged drive test notice it. They have a new Service Manager at Peter Warren and after the botched job they did on my EGR pipe replacement they owe me some slack on this one.

[cmohr]

Glen, if they say that again, that its not presenting during the test, get straight into the car with the service manager and go for a drive, show them the problem.