3.9 efi problem

[Grumpy2]

Throttle pot within limits

Perhaps a further clarification might help

The engine cuts out in two ways

1. When coming to a standstill I can see the revs drop as I dip the clutch to around 1100rpm then to around 900 and once at a standstill 2-3 seconds later down to around 700rpm. It’s at this point that the engine may or may not cut out.

2. When idling perfectly well for some time, say at traffic lights, in neutral with the clutch disengaged (foot off the clutch pedal) the engine will go lumpy for a second or two and then cut out.

In both cases the engine will restart with no problem (so far) and run fine until it decides to cut out again in one of the two circumstances outlined above.

The only other feature that may or may not be present is a slight weakness pulling away from standstill. But that could be me as my ‘normal’ car is an auto

The engine runs well, smooth and powerful and with decent economy too. Plugs are correct colour although getting the Exhaust gasses through an MOT is always a bit hit and miss but we get there.

[ChrisJC]

I am wondering if there is something simpler going on here, like an air leak giving a variable AFR.

Chris.

[Ian Anderson]

What is your timing doing?

Could the vacuum advance be picking up the wrong side of the butterfly? Possibly manifold incorrectly drilled.

Are you running brake vacuum boosters are they the source of spurious air?

Ian

[Grumpy2]

Don’t think I’ve mentioned it’s a Morgan Plus 8 1994. Totally standard (for a Morgan) except for fancy exhaust manifold and side pipes. It was fine for the first three years of my ownership the started stalling at idle about a year ago. Was ok during the cooler months but it didn’t get used much during the cooler months. Other than servicing nothing has been done to the engine that may have created the problem.

I strongly suspect something has failed intermittently, component, pipe, connector. But what?

It has twin cats with lambda sensors, brake servo, and uses the discovery/Range Rover EFi loom and sensors. Dissy is standard. It doesn’t overheat and runs at around 85 degrees.

Oh, timing. I’ve checked TDC with a dial gauge and it’s actually at 4 degrees after DC. Showing as 4 degrees advanced so 8 degrees advance in total.

[GDCobra]

Bench tested and generating a signal fine. All nice a clean inside with no defective wiring. Where did you get a replacement from?

I need to look out the packaging in my garage and let you know, weirdly after ordering it I found it was actually despatched by a company not far from where I live.

One other thing worth looking at is the signal from your ignition coil, I initially thought this was not too important, only needed so the system could detect the engine was running to keep the fuel pump running but in fact it is the primary trigger for the system to perform it's fuelling calculation and inject fuel. If you don't have a good signal there you can run into problems. The connection from the coil does not go directly to the 14CUX ECU, there is a resistor in series with it (6K8 IIRC) although this is quite a large package it may be 'hidden' inside the loom.

[garrycol]

Coming late to this - do you have an auto or manual ECU? The manual ECU has speed input and when the ECU detects no speed it goes to the idle program - without speed input the ECU will still idle the engine but it can be a bit rough. The auto ECU does not have the speed input so goes to a nice idle - it is recommended that if the the system is going into a car with electronic speed input to always go the auto ECU.

Just another thought - no actual experience just advice from those who have used the 14CUX system in another car.

Garry

[Grumpy2]

As above I understand the ecu is told if it’s auto or manual via pin 34 and either 12v, 0v or in the car if a manual via a 510ohm resistor to earth. Mine is the latter.

Cheers

Ps I’m considering an ignition module from bearmach but it lists two for a 94 EFi 3.9! Typical

[DaveEFI]

As above I understand the ecu is told if it’s auto or manual via pin 34 and either 12v, 0v or in the car if a manual via a 510ohm resistor to earth. Mine is the latter.

Cheers

Ps I’m considering an ignition module from bearmach but it lists two for a 94 EFi 3.9! Typical

Think the difference is two or three pin. But the third pin only for a screened cable, and can be used without. When Lucas was still around, they stopped supplying the two pin and included a plug so you could change to three pin.

[GDCobra]

As above I understand the ecu is told if it’s auto or manual via pin 34 and either 12v, 0v or in the car if a manual via a 510ohm resistor to earth. Mine is the latter.

Cheers

Ps I’m considering an ignition module from bearmach but it lists two for a 94 EFi 3.9! Typical

I agree with you Grumpy2. Although there are some different ECUs around this was due to detail changes during its lifetime and possibly location (USA cars had the tune resistor locked out for instance), for a given era of ECU I don't think there is any difference between ECUs, the system 'knows' it is hooked up to a manual or auto from pin 34 which will flip between open and ground on an auto and a 510ohm resistance on a manual.

All versions I've known have the road speed input, this is used for a number of things:
Speed limit - Fuelling cutoff above a certain speed (112MPH rings a bell), some manufactuers disabled this.
Moving idle - Idle speed is held at around 12-1400RPM until vehicle speed drops below around 3MPH, apparently to make gear changes easier
Indicating the engined is being propelled by the vehicle motion to prevent the system trying to hunt for an idle speed during over-run and hence closing the IAC valve.

[garrycol]

As above I understand the ecu is told if it’s auto or manual via pin 34 and either 12v, 0v or in the car if a manual via a 510ohm resistor to earth. Mine is the latter.

Cheers

Thats not my understanding - I was told that the ECUs are different and have different part numbers - but as I said I have no personal experience.

[Grumpy2]

Finally got a chance to play today

before today the gear selected was drive/park and now its categorically manual, the 510 ohm resistor has had some impact

base idle was off at 600 as previous set, now at 525 with bypass airway blocked. Idles nicely at 750rpm connected.

lambda readings have changed markedly too

So I've made a change but if the stalling has stopped only time will tell. Need to get a few miles under the belt now I think. Next week......

(

previous screen dump

current screen after fiddling

don't know why the pics are sideways)

[GDCobra]

Finally got a chance to play today

before today the gear selected was drive/park and now its categorically manual, the 510 ohm resistor has had some impact

base idle was off at 600 as previous set, now at 525 with bypass airway blocked. Idles nicely at 750rpm connected.

lambda readings have changed markedly too

So I've made a change but if the stalling has stopped only time will tell. Need to get a few miles under the belt now I think. Next week......

(ED478C70-027B-4126-BFB3-3EEAFC7C1B2D.jpegC7893C27-4724-4C27-95E2-5EBD774A02EC.jpeg

don't know why the pics are sideways)

Hi Grumpy
Not sure if I understand what has been changed here but there are some significant differences in your RG readings.

From you post it seems that you have added teh 510ohm resistor and maybe you tweeked the base idle. Was anything else done?

510Ohm resistor
You are now getting an indication of "Manual" for the selected gear which is consistent with having the resistor connected, my understanding is that the only diffence this should make is to set your idle speed to include the increment (typically 100 RPM) which is the same as you would get with an auto' setup when in park/neutral. You can see that the target idle is 700 in both cases which is consistent with this, if you were to leave the wire open circuit or ground it (simulating the gearbox switch operating) you would see a difference in the target idle speed. I believe this will drop with the condition for "in gear" but can't remember if that is open or ground.
I don't think this has contributed to the changes you see in the readings though. Although I've been wrong before

Base idle
If I understand correctly you have dialled the BI down from 600 to 525 however this does not seem to be consistent with what RG is showing. In your "before" graphic you have a target idle of 700 and an IAC valve reading of 22% (I can't see what the actuat RPM reading is as not in your graphic but I'm assuming is was nominally 700).
You now have an IAC reading of zero (nothing even showing on the scale) and you indicate that the engine speed was 750RPM, even higher than the target idle. That would indicate that the idle screw adjacent to the throttle plate is controlling the idle to the extent that even with the IAC wound fully in the system cannot achive the 700RPM target value. In other words the Base Idle is actually 750RPM.
I'm surprised the system isn't showing an error for this, I thought it got upset if the reading was below a few %, a figure of around 10% is in my head but as I've never had this issue I can't say for sure.
Either way this is a good indication that your BI is too high for the system to operate as intended but this is a "trick" used by some to get around issues with the road speed sensor and other IAC related problems.

Lambda trim
This is a bit odd and may indicate an issue although I'm not sure I'm seeing the full story.
Your "before" graphic indicates teh short term trims were maxxed out in the +ve direction, the system was doing all it could to increase the fuel being sent to the engine by increasing the injector open time on both banks. It could be that this maximum amount just got the correct mixture or even at max' fuel trim the mixture may still have been too weak.
I'm surprised the system was not showing a warning here too.
This would suggest low fuel flow/pressure getting to the injectors, I'd count out O2 sensor issues as this would most likely only affect one bank.

Looking at the "After" graphic it can be seen that the sort term trim is now within range on both banks, odd being on the rich side, even lean, but is being corrected by the system.

Looking at your graphic from a previous post I can see that the trim was being adjusted there within the limits but that time both banks were being enriched, odd bank more than even, which is actually as similar offset to your "after" graphic but with more enrichment.

Obviously this is a limited amount of information (would be interesting to see the full screen image and also the Long Term trim values) but I think I'd be looking at fuel delivery (checking fuel pump, filter and lines and monitoring fuel pressure when running) based on the variation in trim value you are getting.

It may also be a good idea to completely power down the ECU (remove the permanent live) which will remove the memorised trim values and allow the system to re-calculate them. And keep an eye on both the short and long term values thereafter.

Don't know how you are getting your screen images, looks like you are taking photo's. May be bette to use "Windows+Shift+S" to get a screen grab of the whole RG screen.

[Grumpy2]

So you think I’ve made it worse? I don’t profess any expertise.

Base idle with the airway plugged was 525 rpm. When reconnected the idle was a very steady 750rpm. Bypass varying between 0 and 15% As you notice the bypass showed 0 I. The pic. Before adjusting the Allen screw the base idle was 600rpm rising to 700-750 when reconnected. Idle was unsteady, bypass fluctuating between 4-25%.

What else did I do. In order to add the resistor I disconnected the battery, unplugged the ecu and connected pin 34 to earth via a 510 ohm resistor. Before reconnecting the ecu I gave both sets of contacts a spray with contact cleaner.

I don’t think I have a method of checking the fuel pressure either stationary or running.

Thanks for you continued detailed replies.

Oh, images are photos as I’m using a cranky old laptop for RG but iPad to post here

[GDCobra]

So you think I’ve made it worse? I don’t profess any expertise.

Base idle with the airway plugged was 525 rpm. When reconnected the idle was a very steady 750rpm. Bypass varying between 0 and 15% As you notice the bypass showed 0 I. The pic. Before adjusting the Allen screw the base idle was 600rpm rising to 700-750 when reconnected. Idle was unsteady, bypass fluctuating between 4-25%.

What else did I do. In order to add the resistor I disconnected the battery, unplugged the ecu and connected pin 34 to earth via a 510 ohm resistor. Before reconnecting the ecu I gave both sets of contacts a spray with contact cleaner.

I don’t think I have a method of checking the fuel pressure either stationary or running.

Thanks for you continued detailed replies.

Oh, images are photos as I’m using a cranky old laptop for RG but iPad to post here

Strange that you have the base idle set at 525 but then it raises when you up block the bypass even though the valve position appears to be zero. Makes me wonder if the valve is not fully closing even when wound fully in, that's the only way I can see the idle rising when un-blocking the bypasss, also the only reason why you should have a value greater than the target setting. If the valve is already at zero postion the system cannot reduce it further.

When you disconnected the battery you will have cleared all the stored data from the 14CUX, this would have included the lambda trim values and may be related to why these have changed (for the better I'd say) this is not a problem, the system will re-establish them as soon as the engine is warmed up and certainly after it has been driven. If some garbage data was stored this would be a good thing.
I'm not too well up on this area as my car does not run cat's or closed loop.

I believe you can get a fuel pressure gauge which will screw into the schrader valve on the fuel rail. Don't forget pressure will vary with manifold vacuum.
I'm assuming the vacuum connection between plenum and fuel pressure regulator has been checked?

[Grumpy2]

thanks again - you clarity and understanding are most welcome. I could replace the idle valve and see what happens as I have 3 after being told the the first one had failed and the second one failed because it was a cheap copy. But then it stalled again with the third (bearmach) valve in exactly the same manner. I'll pop it out over the weekend and see what's what.

Vacuum pipe intact and creating a reading when running - I have a vacuum gauge.

I'll have a look for the fuel rail valve and see if I can buy a suitable gauge. (anything in particular you'd recommend?)

thanks again