Hey guys - I know there's a ton of info online about this, but it all seems to fall into either (1) "i'm going to do this" then no follow-up after it's completed or (2) "my friend did this" and I have very few details - so I'm having trouble getting an answer I'm confident in. Hoping to see if anyone has direct experience with my exact configuration and can answer a couple of questions.

I have a 1999 D1 which was originally fitted with a 4.0L GEMS. Prior owner swapped in a 4.6L GEMS from a P38 Range Rover, and left the original 4.0 ECU. The vehicle runs exceptionally well and is getting 15mpg (with big tires, roof rack, steel bumper), but I'm dealing with very high NOx levels that are causing me to fail emissions. (Note: new carb-compliant cats, O2 sensors, fuel injectors, spark plugs, and MAF sensor is clean / appears to be working properly). I'm leaning toward the theory that the engine is running lean due to the computer underfueling it.

My question: has anyone done this swap, and if so: did you need to reprogram / swap the ECU? Did you have do any other mechanical mods to support the slight change to the needed A/F ratio? I've seen a number of people say that the GEMS ecu was built with settings for both the 4.0 and 4.6, but that it needs to be reprogrammed to enable the 4.6 settings (otherwise it "works" but not at its best). Any experience with this?

The other factor is that the 4.6 had a different MAF, but I've seen zero stories of people successfully retrofitting a different MAF. I'm hopeful an ECU reprogram or swap is all that's needed.

Thanks, as always, for any input - y'all are the best.

 

The A/F is controlled by the O2s, so if the O2s are modulating properly on the live data scanner trends, the A/F is correct. How do the trends look?

Ignition timing for a larger volume probably needs to be advanced since a larger volume takes longer time to burn. Advancing timing at idle may produce more NOX, not less, not sure about this. The ignition timing is controlled by a map programmed into the ECU, based on engine load (based on TPS, MAF, RPM, ECT, CKP and maybe other sensors). So, there will be differences in the ECUs. I personally don't have experience swapping or reprogramming the ECU, and I don't need to do emissions checks. I do know that without the correct timing map, your engine will not run at peak efficiency or power.

 

I am in the process of doing this right now with my '98 - D1. Not everyone has done the conversion properly - I can name a few I have talked to personally that have not done the complete steps. One of the things that needs to be swapped is the cam shaft. The lobes are different on the D2 4.6L & Range Rover that most pirate from. If you have an original D1 camshaft, you may not be getting enough input and/or burn - I do not know the complete science or specs on this, but I have already set up my rebuilt heads & camshafts along with rocker arms & pushrods.

The other thing that I have found through all of the research I am working on is that, yes, you need to have this thing reprogrammed. A Rover dealership service bay is supposed to be able to do it. I have an old school friend of mine that does Land Rovers and he has talked about the need for reprogramming. He is someone who had done these kinds of things previously and it is a must. However, I do not know if there is a way to check if yours would have been reprogrammed.

 

I recently swapped my 98 disco 4.0 for a 4.6 p38. I ordered the engine from RPI engineering in England and changed the cam to a piper 270. Also had the heads ported etc since they had the engine apart. Also got the ECU flashed and reprogrammed to match the new engine. Changed to a HD flex plate while reinstalling the engine. Also, new shorter carbon fiber intake trumpets to allow for increased air flow in because I have a free flowing exhaust that lets more air out. Runs great and their website is full of great info for this adventure. They are helpful with questions along the way as well. FWIW

 

Thanks, guys - all really helpful responses... I'm starting to get more optimistic that a reprogram is all that's needed.

O2s look right - oscillating between ~.1 and ~.9 on both banks. The emissions tech who ran diagnostics also tested it on 3 different machines and found that lambda values look essentially perfect. On the other hand, I swapped the spark plugs for the same champion copper ones but 2 steps colder this weekend. Took a look at the old ones and they do seem to indicate that I'm running lean (see pic - old one is on the right).



Glad to hear another data point on this. I'm in touch with the prior owner who did the swap, and he confirmed that no changes were done to the ECU. The engine appears to be running really well in both his and my experience (and he's in a non-emissions state), so it was a "if it ain't broke" situation. But since it's apparently slightly "broke", I'm going to see about having a dealer reprogram it.

Thanks, really good resource. I may reach out to them to ask what other mods they'd recommend. The big question mark for me feels like the MAF sensor housing being a different size. Someone on a Facebook disco page suggested finding a way to choke down the diameter of the MAF housing so that air passes over the sensor more quickly and "tricks" the ECU into injecting more fuel. Interesting idea - going to start with a reprogram first, though.

 

I swapped a 4.6 from a 2002 Range Rover into my 98 Discovery about 10 years ago. I mainly followed this site Frankenrover - Swapping a Bosch D2 4.6 engine into a GEMS D1 Land Rover Discovery which helped a ton. From what I recall the only tough part was welding on the new crankshaft sensor. Everything else was pretty straight forward. I had the 4.6 ecu but never ended up using it and just kept the 4.0 ECU. Ran it this way for years and even passed NJ inspections. Radiator ended up cracking and blew the head gaskets. With rust, head gaskets and other life needs, I ended up selling it about 4 years ago but the swap was well worth it in the end.