19th May, 2002
Well another month passes and another month closer to summer! The continuous travel with work is making it very hard to get any real time in on the Rush. Still, plenty of time in hotel rooms and 'planes to think about my next move in the build.
I've spent much of the last week or so researching how best to identify where my fuel problems lie. What I really needed was a tame Land Rover mechanic complete with electronic test book to analyse my problem/symptoms. During one of my recent attempts to sort things out I noticed that the ECU warning lamp output was 0 volts meaning that it was unhappy (the other side of the lamp on the dash is wired to 12v). Normal running should see this output at 12v. This means that the ECU is running a get you home map which is very rich, and since I have too much fuel going in anyway (3.9 Efi system on a 3.5 remember) this was surely not going to help matters. It seemed obvious that the ECU disliked being disconnected from some of its ancillaries which provide inputs to the ECU to control the stepper motor and the fuel rate. These inputs are:-
In order to determine how best to trick the ECU into thinking it has the above still connected, I began to search out any useful info on the 'web and found (seemingly everyone else knew) a very useful zip file in the Range Rover Knowledge Book section of www.rangie.com. Once downloaded and unzipped, it was a complete scan of the Range Rover workshop manual test procedure for a hot-wire Efi system. Armed with this new information, I worked out that by adding three strategically placed 10k resistors into the ECU plug/loom, I should be able to convince the ECU all was fine and dandy. I've updated my Efi drawing to show these changes, click here to take a gander.
So yesterday I went into battle. Man versus machine, take no prisoners. Firstly I completely 'buzzed' the Efi loom from each pin of the 40 pin plug to where the wire actually connected to the engine/car loom. All was 100% correct. Next I added the 10k resistors (sounds like a cooking recipe!) into the loom. Whilst leaving that on one side to marinade, I wired up one of the dash warning lamps to the ECU warning lamp output.
With the above being done, I switched the ignition on and hey ho, the Efi warning lamp did not illuminate! A further turn of the switch saw the engine run. Although the idle was still lumpy, I took the positive view that now I knew that the ECU was happy, my poor running problems must entirely be due to the excess of fuel entering the engine. That being said, the engine no longer seemed to 'hunt' and I felt that the day's work had made an improvement to the engine's running.
Just to ensure that everything Efi was as it should be, with the help of my father I set about following the LR test procedure. Everything passed with the exception of one item. With the ECU plug connected and the ignition on, pin 39 should be 9.5v, I measured 0v. This input to the ECU is the engine speed sense from the -ve side of the coil. With the ECU plug unplugged and ignition on, pin 39 reads 12v as you would expect. It seems as though the ECU is pulling pin 39 down to 0v. This might well be due to me shorting the -ve side of the coil down to earth as a result of a minor wiring mishap under the dash. However, the engine does run, if not a little erratically. It would seem that the next step is have the ECU tested.
Provided the ECU is OK, all I can do now to solve the fuelling issue is to have the ECU re-chipped. I shall e-mail RPI and John Eales (now that they seem to have a website) to enquire as to the precise cost. However, there might be another way...reduce the fuel pressure in the fuel rails.
Currently the fuel rail pressure is always 2.5 bar above the inlet manifold pressure. This is accomplished by the fuel pressure regulator valve and tank return. It MIGHT be possible to change this for a variable 1:1 pressure regulator valve from Fuel System Enterprises. This all depends if they make a valve that will adjust down to 2 bar (an unusual request that, to actively wish to restrict fuel flow!). A call tomorrow should find out the answer.
Well the above mentioned Fuel System Enterprises were able to supply a variable base pressure regulator and this was duly fitted to the Rush. Unfortunately once fitted and with the pressure set to 2 bar, it was quickly proven that it really made little difference to the idle.
This last item caused me much grief!! In accordance with my RR manual, I bled the slave cylinder such that no air came out of the bleed nipple. However, the clutch would not disengage. It appeared at first sight that the slave cylinder push rod needed to be pushed too far into the bellhousing, beyond the range of the slave cylinder piston in order to disengage the clutch. Being slightly unsure of the provenance of my bellhousing and clutch fork/pivot, and mindful that there are at least three different lengths of V8 clutch pivot, I concluded that I had the wrong length pivot fitted. No choice but to have the engine and 'box out again...ARGH!!!
Having researched the Rimmer catalogues, I concluded that I needed a UKC13 pivot from an SD1. I ordered it and it duly arrived....the same as the one that I already had. Deeply confused, I contacted RPI and spoke to a guy called Holly (I have no idea if that is his real name, or a nickname!). He was as confused as I was and said there was little he could do but if I e-mailed over some digital photos they might be able to shed some light on the matter. However something he said about slave cylinders and plungers made me go back to the workshop last Monday and have a closer look at what was going on.
I wanted to see just how far the slave cylinder plunger would move up its barrel. So I pulled it and it came right out of the barrel. It then became immediately clear what the problem was.....2-3 cu.inches of air inside the cylinder.....DOH!. Quite simply, I had a very bad case of the wind. It took 10 seconds to expel the air with the help of my father. With the cylinder back in the bellhousing, it then became clear that the plunger can push the fork far enough to disengage the clutch and I had the correct pivot all the time. Right now the gearbox and bellhousing are sitting on a workmate with the slave cylinder connected to the clutch pipe and pedal. The system is now fully primed such that the clutch pedal will not move since the slave plunger has pushed the fork against the far end of its groove in the bellhousing and is a solid as a rock. I will need to drain off a small amount of clutch fluid in order to allow the plunger to depress back into the cylinder prior to refitting the bellhousing and gearbox to the engine.
Oh and one final problem here, one of the slave cylinder threads in the bellhousing has 'pulled'. I might well have to drill through the blind hole and bolt straight through the bellhousing (IYSWIM) to stop the slave cylinder pulling straight out of the housing the first time I try the clutch for real.
Back to the engine and there's not much left to test/change to try and improve the running. After much advice I decided to have the ECU tested and as you can see from the Lucas report, there was no fault found.
So once the engine is in again, I need to consider what the next step is. It seems that I really have only two choices, either re-chip the ECU (£300+VAT) or fit a Weber and Eidelbrock manifold (£500+VAT). The latter is more attractive since if in the future I up the capacity, a set of jets is considerably less expensive than another chip. And anyway, I'm not convinced that a 3.9 hot-wire system can be effectively re-chipped down to 3.5 requirements.
In the meantime, I will continue with the other jobs to get the car complete before the end of the summer. I reckon, engine aside, I need only another five full (10 hr) days on the car to get it finished to SVA standard.
I reckon that there's nowt really wrong and that a combination of factors is the root cause of the problem. And indeed the consensus at work is that I'm chasing my tail and that I should take the thing for a good run to really see if the overfuelling is an issue or whether continually running it at idle has caused all of the carbon fouling. It might simply come down to 'round hole' or 'funny shaped hole' in the bonnet.
For now its push on with other parts of the build whilst I ruminate on the next move.