Problems with new Rover 4.6 Engine

[heli_madken]

The injectors have clips that secure into the rail, you can lift the entire rail out and crank with injectors all pointing into jars.

Although whether that's easier than just swapping to different cylinders and re-trying...debatable. But sometimes a visual can help.

Thanks for the help.

Thats an interesting one as I have an aftermarket trumpet base supplied and fitted by RPI. The flange of the trumpet base stops me from taking the fuel rail off. They did two things to stop me removing the trumpet base, firstly glued the trumpets themselves in place with something I cannot release. Secondly before gluing them in lets put a huge amount of torque into the securing bolts so an open ended spanner has no hope of getting them off (trumpets stop anything else from being used).

Fortunately its quite tall so there is just about enough room to get the injectors out otherwise I would have had to destroy the trumpet base. Getting them back in is going to be fun.

Its a good idea though, I could buy a second hand fuel rail and extend the wiring and fuel line be an interesting experiment.

[heli_madken]

So two new injectors fitted, engine all back together but no difference.

Not sure what the two clean exhaust ports are trying to tell me but the car wasnt running on 6 cylinders. Infra red thermometer tells me all exhausts are roughly the same temperature perhaps Odd bank running slightly cooler.

Power wise the engine seems ok, no different to what it has always been which is nothing startling a bit better than a 3.5. What I have now are very white spark plugs and on restart with a hot engine idle surges to 2000 rpm (base idle has been set). In addition although the engine runs cool at speed temperature climbs very quickly at idle.

I think I need to sort the week mixture out first. This is only a recent thing so not sure what has happened. I noticed when I had the Plenum off that the RPI supplied idle mixture control stepper motor is just a £12 ebay special so this is the first thing I can change. As I am getting good fuel pressure the only other thing that I can think of are the O2 sensors but according to RoverGauge these seem to be operating and not showing any errors.

Dissapointed I thought I had found something but all the work over the past week has been for nothing.

[SuperV8]

Maybe unplugging and replunging in your injectors cured a doggy contact for those two injectors to explain the clean exhaust ports?
White plugs obviously suggest its too lean, which can also cause high temps.

[heli_madken]

Thanks for your help.

The clean exhaust ports are the reason for me changing the two injectors but this has made no difference to the engine. At idle the motor feels unsteady, not misfiring as such just not smooth. This might just be the tune level. Overall it doesnt seem to produce much power and sounds strained when accelerating.

Had a good look at the valve gear nothing seams wrong.

I measured the height of the valve springs when compressed and they are all pretty much the same so I am getting full valve movement even on the two suspect cylinders 4 and 6.

Did a leak down test and results are all pretty much the same staying within the green band on the gauge around 20% loss. Certainly no sounds of air escaping or bubbles in the coolant so I am pretty happy that I have no leaking valves, rings or gasket issues.

I have fitted the new stepper motor so need to do a base idle.

[heli_madken]

Still struggling with this, I replaced the stepper motor which seems to have sorted the idle surge issue, I also fitted new O2 sensors as a precaution.

I have a new problem now, at a steady speed on a flat road the engine is silky smooth then suddenly backfires several times then goes back to being smooth again. Never done this before.

Plugs are very white still, not sure if this is overheating or a lean mixture.

Any ideas?

[Pistonsquirter]

When the engine is running do those cylinders have sparking spark plugs and squirting injectors? I cannot see this in your diagnosis?

You have not wasted a week, you have spent a week identifying what ISNT the problem.

As usual it is probably something simple, and perhaps the fault only manifests dynamically (when the engine is running), so i suggest pull plugs and injectors on the two bad cylinders and visually confirm both spark and fuel on both...

HH

[heli_madken]

Thanks for your help.

Yes I can confirm that I have spark and fuel arriving at all cylinders. All plugs are identical in colour and all exhaust manifold temperatures are the same.

I did find a fault with the fuel pressure regulator in that on ignition switch off pressure dropped to 0 psi straight away. This has been replaced but the overall lean condition still persists, plugs are very white.

I replaced the fuel filter and blew some air through all the lines no sign of any blockages.

The only thing left now (that I can think of) is the MAF. This is an aftermarket Bosch unit and I am having trouble trying to physically get a good voltage reading from it to test due to its plug design.

Just as a recap, the engine has never pulled as well as it should but the lean condition only appeared two months ago. Previously plug colour was good. I have replaced - O2 sensors, two suspect injectors (but I think these where actually ok as are the rest), spark plugs, stepper motor, ECU water temperature sensor, distributor, coil (tried with and without power amplifier and either a 0.8ohm or 1.2ohm coil) and fuel filter.

In terms of testing - smoke tester shows no leaks on inlet system or on exhaust, RoverGauge shows no errors, O2 trimming is normal, stepper acts normal, fuel pressure is now 40 psi (tried 42 and 38), ignition timing is strobed at 10' (with the 123ignition distributor I have tried various base settings and curves without any effect currently gone back to old distributor). Compression test showed #4 and #6 to be weaker but within 10%, leak down test showed all ok. I am not loosing any coolant and inspection camera does not show any washed pistons. Mechanically engine does not make any nasty noises. Tick over is a little rough but steady think this is just the state of tune. Also checked that the crank pully timing marks coincide with TDC.

RPI supplied an ECU with a Tornado chip marked as being for the 4.2 engine most likely in a standard form which I feel is responsible for the poor performance, I have written to Tornado for advice but not had a response. I dont think this is responsible for the sudden lean running though.

After replacing the fuel pressure regulator I have not had the misfire in my last post but I am not confident this has been solved yet, on the previous journey it was a good 5 miles before the engine started missing.

I think as stated it is something simple. Everything I do makes no difference to performance or the lean condition which is frustrating even when I wildly alter the timing curve on the distributor. I have tried to see if the cam timing is out with the camera but cant quite make it out, I cant see RPI making this mistake though.

I am reluctant to replace yet another relatively new part but I may have go for a replacement MAF just to rule it out.

[Ian Anderson]

Is it perhaps something simple as a bad earth/ground connection

Perhaps try a couple of jump leads to back up existing battery to chassis and chassis to engine.
Not sure how many time things end up being a bad earth

Ian

[heli_madken]

Thanks Ian thats an interesting one.

I currently have two earth leads to the engine block at the front, one goes directly to a bolt sharing the battery earth on the chassis and another on the other side of the engine goes to the coil mounting on the inner wing and to the chassis.

Looking around on the net I see references to earth straps on the back of the cylinder heads? I think at least one of them was for radio interference. I dont recall having these on the original engine. I think it is worth my while making a couple up wont do any harm.

[stevieturbo]

get a wideband on it to see what actual mixtures are doing ( assuming you've already checked fuel pressure too )

Is there any adjustment screws on the MAF, what design is the MAF on those ? Is it an old flapper or something more modern ?

[heli_madken]

Yep checked fuel pressure, assuming gauge is accurate its currently at 40 psi. I only have the standard narrowband O2 sensors fitted is there a piece of kit that you can read wideband sensors with? One thing I havent done yet is to create a logfile using RoverGauge.

Its a 14CUX/Hotwire system, other than a remote CO trimmer there is no adjustment on the MAF

[stevieturbo]

There are so many these days, a few cheap examples

https://diy-efi.co.uk/product-category/ ... ambda-kits

https://ldperformance.co.uk/product/wid ... tJOOqwA8_w

https://www.tegiwa.com/search?type=prod ... CONTROLLER

[GDCobra]

Its a 14CUX/Hotwire system, other than a remote CO trimmer there is no adjustment on the MAF

I'm a bit (well more that a bit actually) confused by that. From investigations I've done with the 14CUX the CO trimmer is effectively a separate item from the MAF although physically integrated in the standard MAF. It also only has an affect on non-cat, open loop systems.
In the code the reading from the CO trim adds a trim (which can be positive or negative) to the fuelling calculation, it does this by using the data location used for short term lambda trim hence CO trim and lambda feedback are mutually exclusive. Can't understand why RPI would have this device on a lambda equipped engine. Well I guess that fact that it was RPI helps me understand.

Another thing I'm confused about is that you say that the engine is running lean but how can that be the case on a closed loop system? If the fuel pressure was low or for some other reason the calculated fuelling was too short the feedback from the lambdas would trim up the fuel delivery to correct it. I realise that there are some conditions where the system will revert to open loop but during cruise this won't be the case.
Makes me wonder if the lambdas are not giving an accurate reading for some reason, most obvious would be an air leak in the exhaust but I'd expect this to give a leaner than actual reading and cause the ECU to add fuel.
It would be interesting to see what Rovergauge reports regarding your fuel trims, I think I'd be focussing on getting some logs before firing the parts cannon any more.

I agree with the advice to put a wideband Lambda on it but unless you use a probe up the exhaust this will mean getting a bung welded to one or both sides of the exhaust.

I've been away for a while and only skimmed through the later posts. Did you ever verify that the timing marks are in the correct place?
Seems you did fit a 123 distributor but then swapped it back out. Why did you revert? Was there a problem with it?

[heli_madken]

Thanks for your help and taking so much time with your response it is appreciated.

The MAF is a bosch unit and comes with the remote CO trimmer replacing the standard in-housing trimmer. I have this set currently at the 1.5v recommended for a non-cat car but have tried other settings with no obvious difference. As I understand it the lambda's would kind of overide this anyway which I think is what you are saying. I think the trimmer is fitted only in case I ellected not to use Lambdas.

RoverGauge doesnt show any massive amounts of trimming going on at idle, I could not previously take any logs on the move as my laptop was in effect broken but I have a new one now so I can get some logs on the move. I have replaced Lambda's to rule out any issues with faults.

Yes timing marks checked with a dial gauge.

Originally the car didnt run with any great power but did not show any signs of running lean. My problems seemed to start after fitting the 123Ignition distributor.

One of the problems with diagnosing this engine is I seem to have a constant stream of failed components (all supplied by RPI) I was experiencing issues with damp finding its way into the RPI Power Amplifier resulting in coil failure. I wrote to 123Ignition for advice on which power amplifier to use and they stated that with thier 1.2Ohm coil an amplifier isnt needed but I do wonder now. I put the old distibutor back in which itself does have a problem in that it does not advance enough to see if the hot running issue went away but in turn now the new missfire problem has occured so the car has not run enough to get any reading from the plugs.

I am assuming that the mixture is lean only from the white plugs (all plugs look identical) but of course this could also mean very high cylinder temperatures. Not sure what could be causing this other than perhaps an ignition related issue. One more investigation I want to do is the RPI supplied engine wiring loom. This was a mess with loose connectors, cut wires, corroded plugs and generally untidy. It had come from an LPG equipped car and had the supply wires to the fuel injectors cut through and simply then wrapped back into the loom. Took me some time to find the issue as first start attempts where a failure. I intend taking it off, unwrapping it and checking evereything. One puzzling thing for me is the presence of two 6.9ohm resistors on the feed to the ECU from the coil where there should only be one as far as I know, need to trace what this is doing.

The latest failure I discovered was the fuel regulator, again supplied by RPI. I thought I would try a higher pressure (42Psi) again but noticed on switching off the pump the pressure dropped immediately. So this was replaced and now everything works as it should. The last run didnt show the missfire problem but I am not convinced this was causing the issue but will take advice. I also replaced the fuel filter and blew through all lines to make sure there are no restrictions.

Funny you should mention exhaust leaks, I had previously checked for leaks using a smoke tester after the problem started but after the last drive I did hear the give away ticking noise and sure enough I have a leak from the manifold. This has happened in the past, the flanges just dont line up well so I have bought a one piece flange and will cut off and reweld the manifold to this. Either the weather or being disheartened with the whole thing has stopped me from doing this so the car hasnt run since.

I cant help but think there is something fundamentaly wrong somewhere, maybe mechanical or electrical. The funny thing is that no matter what I do or replace performance never really alters, changing timing makes no difference. I get the same no real power at low RPM better over 3500 rpm but not startling. I do find myself looking at carburettors increasingly.

[GDCobra]

OK, I'll break this down a bit.

The MAF is a bosch unit and comes with the remote CO trimmer replacing the standard in-housing trimmer. I have this set currently at the 1.5v recommended for a non-cat car but have tried other settings with no obvious difference.
As I understand it the lambda's would kind of overide this anyway which I think is what you are saying. I think the trimmer is fitted only in case I ellected not to use Lambdas.

The lambdas don't kind of override the CO trimmer one or the other is used depending on open or closed loop tune.
I guess having the trimmer fitted in case you swap to an open loop tune is not a bad idea but it won't be doing anything for you while you are running closed loop.

RoverGauge doesnt show any massive amounts of trimming going on at idle, I could not previously take any logs on the move as my laptop was in effect broken but I have a new one now so I can get some logs on the move. I have replaced Lambda's to rule out any issues with faults.

If no trimming is going on at idle that would indicate that the mixture must be fairly good hence nothing to do or the signal from the lambdas is not correct or something has been tweaked in the 14CUX which would be odd but then we are talking about RPI!
Hopefully you'll get some logging done on the move and this may reveal something.

Yes timing marks checked with a dial gauge.

That's interesting, I'd be interested to know how you got on with that and what equipment was used. I tried to do it by making a spark plug adaptor to a plunger DTI but due to the angle the spark plug enters the cylinder it did not work, wasn't pushing directly into the DTI.

Originally the car didnt run with any great power but did not show any signs of running lean. My problems seemed to start after fitting the 123Ignition distributor.

I'm a bit confused why the distributor would make the car run lean, I could accept that if the spark output was weak the mixture may not burn but this would result in a rich mixture (fuel not completely burned by the time the exhaust stroke starts).
Personally I'd persist with the 123 distributor as your old style distributor and power amplifier seem to also have issues.

One of the problems with diagnosing this engine is I seem to have a constant stream of failed components (all supplied by RPI) I was experiencing issues with damp finding its way into the RPI Power Amplifier resulting in coil failure. I wrote to 123Ignition for advice on which power amplifier to use and they stated that with thier 1.2Ohm coil an amplifier isnt needed but I do wonder now. I put the old distibutor back in which itself does have a problem in that it does not advance enough to see if the hot running issue went away but in turn now the new missfire problem has occured so the car has not run enough to get any reading from the plugs.

Not a great situation, been there myself so I feel you pain.
This is why I'd stick with the 123, takes a dodgy distributor and associated gizzmo's out the the equation.
I'm assuming your 23 is new.

I am assuming that the mixture is lean only from the white plugs (all plugs look identical) but of course this could also mean very high cylinder temperatures. Not sure what could be causing this other than perhaps an ignition related issue. One more investigation I want to do is the RPI supplied engine wiring loom. This was a mess with loose connectors, cut wires, corroded plugs and generally untidy. It had come from an LPG equipped car and had the supply wires to the fuel injectors cut through and simply then wrapped back into the loom. Took me some time to find the issue as first start attempts where a failure. I intend taking it off, unwrapping it and checking evereything. One puzzling thing for me is the presence of two 6.9ohm resistors on the feed to the ECU from the coil where there should only be one as far as I know, need to trace what this is doing.

Got to say I really hate the A word, assumption and presumption are the mother and father in my view!
You really need to find out for sure where you mixture is. If the lambdas are working properly this should be a better indication of where your mixture is.
I think debugging the loom is a good idea, not sure it would help with your issues but can't do any harm.
There were 2 resistors in my loom, these are to limit the current from the ignition flyback pulse. IIRC these were both connected to the coil negative, one went to the ECU and the other to the rev' counter (unconnected in my case and probably yours)

The latest failure I discovered was the fuel regulator, again supplied by RPI. I thought I would try a higher pressure (42Psi) again but noticed on switching off the pump the pressure dropped immediately. So this was replaced and now everything works as it should. The last run didnt show the missfire problem but I am not convinced this was causing the issue but will take advice. I also replaced the fuel filter and blew through all lines to make sure there are no restrictions.

This is not necessarily the regulator, I have similar behaviour on mine and believe it is due to the fuel pump. There is a check valve on the pump outlet (or rather on my original pump there wasn't) so when the pump stops the downstream rail retains its pressure. When I replaced the pump earlier this year I fitted the valve. This was never a problem in reality, it may help start-up particularly if under-bonnet temperatures are high (vapour lock) but once running it should have no influence.

Running a higher fuel pressure may result in a richer mixture but the closed loop processing should dial this back, it may give you a richer mixture when the system goes into open loop but my understanding is that the trim values calculated when in closed loop will still be applied.

One issue could be fuel pressure when running. Fuel pressure is all well and good but if the flow is restricted this will drop as the fuel is consumed.
This could be due to restriction anywhere between the tank and the regulator and is very difficult to measure as the line pressure will vary with plenum pressure (vacuum). I made up a rig to check mine by electronically monitoring fuel pressure and manifold pressure then combining the values to give a differential value. Unfortunately this is a bit of a Heath Robinson (prototype) system and needs to two sensors piping in so I can't easily lend it to you.

I cant help but think there is something fundamentaly wrong somewhere, maybe mechanical or electrical. The funny thing is that no matter what I do or replace performance never really alters, changing timing makes no difference. I get the same no real power at low RPM better over 3500 rpm but not startling. I do find myself looking at carburettors increasingly.

I'd really steer clear of reverting to carburettors, you could put a lot of work into that and if the issue isn't in the fuelling area. There must be a problem which can be solved on this.

Have you thought of running the ECU in open loop mode? Only needs a change to the tune resistor.