weird 3.9 idling

This site contains affiliate links for which LandyZone may be compensated if you make a purchase.
A

adri1957

Member
Posts
34
Hi all,
have been here for long but never posted a thread myself, this time I think I need your advice or I won't get out of this mud.

engine:
RRC 3.9 Efi 1990
engine rebuilt completely (starting from cyl liners all the way up, everything that could be and was worth changing was changed..)
formerly running on LPG as well, fitting kept but not implemented yet

Situation:
engine starts ok as soon as key is turned, but only at the second attempt, i.e.: every time I have to switch on ign once, then no matter how long I try, it wont start. Second time, immediately after, as soon as I turn the key it will start.
Once started, it idles at 2K rev for a few seconds (may be 15") and then gets down to 1300-1500 and there it remains.
If I engage drive (auto box) it will get down to 1000-ish, but if I put more load (eg.steering) than it stalls and dies. Not every time though, sometimes it will manage to keep running.
When I drive it, it drives very smoothly, fuel consumption is within limits (4.6 Km/l), engine has power and feels ok.

Already checked items list:
dizzy air gap, ignition module relocated, good quality HT leads, new coil, new coolant temp sensor, throttle pot is ok as far as voltmeter reading are concerned, MAF ok same way, stepper motor (IACV) cleaned and perfectly working at least as per rovergauge (RG) commands.

Tried to set the base idle and during that procedure, idle was as low as I could possibly get (300 RPM as indicated by RG). This should allow me to rule out extra unmetered air as it would affect the idle even excluding the stepper motor.

When it stalls it is normally running temp (not cold) and after that it would not re-start easily, unless I play with gas, and this doesn't work reproducibly either. I found that if anything else fails then I can disconnect the coolant sensor and it will start on the fly, but then "idle" at 2.5K 2.8K RPM. Reconnecting the CTS will take at least 30 seconds to bring the idle down to 1500 RPM.

Because the high idle cannot be to extra air, then there must be a condition by which the ECU "thinks" this is the correct idle and keeps it. RG reads 7% throttle pot and 0% stepper at idling 1300 RPM, so it seems it cannot go below that.....

Any help reeeeeeallllly welcome

THanks

Adriano
 
Best guess you have an inlet leak somewhere - Pull the stepper, apply RTV silicone then put it back on. Failing that I would think possibly inlet manifold not quite seated right or possibly plenum to ram housing not happy so take off the plenum and ram housing etc, pull the valley gasket and replace and ensure everything is properly clean in the right place and properly sealed. HTH
 
Hello Kev, many thanks for your input.
I was originally thinking on the same line, i.e. some air leak somewhere along the places you listed. However, performing the base idle setting I noticed that by clamping the stepper to plenum pipe to avoid the stepper operated air bypass I could reduce the idle down to 300-500 RPM. This made me rule out any external and unwanted air leak, unless it is along the stepper-to-plenum connection, or it is an "intentional" effect of the regulation by the step motor.
Am I correct in this conclusion? If there were any air leaks further downstream (plenum to ram, ram to inlet manifold, and valley gasket) how could they be mended by clamping the rubber pipe bypassing air through the stepper?

In fact, while writing I realize that if one can manage to completely close the stepper (say by RG commands) and then disconnect its plug to keep it closed irrespective of the ECU, there are two cases: 1) idle down to 300 as by clamping the rubber bypass --> stepper to plenum seal faulty as you suggested, or 2) the idle remains at 1500 --> rubber bypass faulty. I have carefully inspected the latter though, I wouldn't know how there could be a leak.

Thanks
Adriano
 
That's right, but ....how do I check that? It looks as if it is doing it, but I have no reference to compare with. Are there any stated sizes, angles, measures or the like on the workshop manual that one could refer to? Has the throttle to be square to the plenum inlet? or has it to be at an angle?
I remember reading somewhere that there must be an incomplete closing somewhere along the circumference but cannot remember more t han that. Help on this would be greatly appreciated please
Many thanks

Adriano
 
Cooltide could have a point. Yes the throttle disc can be checked using a vernier, the settings are in the workshop manual. Whilst you could check the throttle disc without removing the plenum I would suggest taking the plenum off and giving it all a thorough cleaning with carb cleaner and making sure everything is spotless then you can check/set the throttle disc, while you're about it you can also clean the ram housing and inlet trumpets and check for loose trumpets.

I am fairly sure this will be something simple and probably quite small meaning it won't be obvious so you might want to just bite the bullet and remove plenum etc, give it all a good clean then check and set everything up again from scratch, I know it is a nuisance but when all is said and done it only takes a couple of hours to pull the top off and put it back and costs shirt buttons which is better than having weeks of frustration and guessing and like so many others I've been there and done that more than once! . HTH
 
Hi, I will certainly follow this suggestion. As I said the engine has just been rebuilt, and in principle everything has been cleaned, but you never know and as you said it's worth going through it rather than just keep guessing + trial&error. I must admit that despite having rebuilt the whole thing, I assumed the throttle was OK and didn't check it more than by eye so I will carry on a deeper check while dismantling the top. Incidentally, does anyone know if there is a reason why the seals between inlet manifold, ram housing and plenum are not made of joints but left to the (awful in my opinion) silicone spreading strategy? Would it pay off taking some time in cutting those shapes out of appropriate cardboard (sorry for my lack of technical english here, I ignore the name of it)?

Many thanks
Ciao
Adriano
 
Last edited:
hello folks,
I have performed the suggested checkups. In fact cooltide did have a point....there was a difference of 8 mm in the throttle valve diameter distance to the plenum bore measured as per the workshop manual (it should have been 0,5 mm). Fortunately I had a spare plenum from another engine I use as source of bits'n'pieces so the problem was easily solved. The original valve was bent in both axes so it couldn't really provide an appropriate seal.
Having also checked the remaining layers (manifold, ram housing etc) and having found everything else was ok I put the whole thing back together, but.... with very much the same problem as initially stated. So I will try to give more detailed info for the ones who will very kindly try to help and solve it.

1) I measure parameters from rovergauge indicators, so temp, rpm, opening of IAVC, MAF metering etc are taken from there. In addition I have a good quality digital Vmeter I use to check tensions as per the manual for throttle pot and MAF terminals.

2) After the plenum exchange the situation is very similar to the one I had with the initial setup: high idle RPM (1200 instead of 1500RPM, stalling under load). I tried to perform the base idle setting again, given the change I made and the problem was very similar to what I had when trying to do it with old throttle: idle very stable at approx 300 RPM, but when I try to set it to 550 it starts to be erratic (cannot see how as the ISVC function is excluded) dropping down to 250 and raising up to 1300, then after a while doing so it will eventually drop too low and dye. Restarting is not easy again and it requires some play with the gas pedal or it won't start.
Despite all electricals give the correct reading upon Vmeter measurement, from time to time the fault codes for throttle pot and stepper motor will light up, especially if one plays with the gas, revving repeatedly the engine.
I checked the stepper, cleaned it and greased it and it seems to work perfectly. While idling at 1200 rovergauge tells me that stepper is at 0%, but if I issue a "close valve" command of 30steps form the PC it will close it further, meaning that 0% is some sort of arbitrary position the ECU thinks of but that is not real.
Throttle pot gives correct initial volts and increases smoothly upon valve opening, up to the expected maximum value.
I seem not to have the idle mode/target RPM info from rovergauge available at any time, as if ECU didn't know it is idling (or didn't want to...)
What could be the reason for an erratic base idle? The same reason seems to affect idling under operating conditions, such that when further load is added (drive or steering or both) and at the same time idle drops to a minimum it then dies.

Please, any V8 guru with any clues...?

Many thanks
Ciao
Adriano
 
Hi, if timing scatter is the "spontanous" random variation of timing (dizzy wear etc etc) I thought of that but wouldn't know how to check it. Timing obviously varies with idle oscillations, problem is whether it is the cause or the effect or oscillating idling.

For sure ignition has something weird as the timing strobo light doesn't work with every ignition lead, although I am absolutely certain all leads give sparks (tetsted them individually). Also the same lead will sometimes work and sometimes not. I never investigated the principle by which the strobo light works, but would assume the current getting through the leads generates a magnetic field in the clamp, and this acts as a switch on the strobo light. If the light doesn't work the switch is not effective, but I have no clues as to what this means and whether this could cause erratic idling.. Poor current? Too nice insulation around the lead?
If that is a potential cause then it must be downstream from dizzy cap as I could hardly see a mechanism by which there is a lack of current once every eight strokes and always at the same point....
 
adri1957 said:
Hi, if timing scatter is the "spontanous" random variation of timing (dizzy wear etc etc) I thought of that but wouldn't know how to check it. Timing obviously varies with idle oscillations, problem is whether it is the cause or the effect or oscillating idling.

For sure ignition has something weird as the timing strobo light doesn't work with every ignition lead, although I am absolutely certain all leads give sparks (tetsted them individually). Also the same lead will sometimes work and sometimes not. I never investigated the principle by which the strobo light works, but would assume the current getting through the leads generates a magnetic field in the clamp, and this acts as a switch on the strobo light. If the light doesn't work the switch is not effective, but I have no clues as to what this means and whether this could cause erratic idling.. Poor current? Too nice insulation around the lead?
If that is a potential cause then it must be downstream from dizzy cap as I could hardly see a mechanism by which there is a lack of current once every eight strokes and always at the same point....
Click to expand...
A basic timing gun or even a fancy one set at 0 btdc connected to no.1 lead will show it up, and yes it will usually be distributor internals but ignition amp, ignition amp wiring may be suspect. I personally can't think of any other cause of erratic behaviour whilst setting base idle with stepper motor blanked, no air leaks and correct fuel pressure. But ready to be proven wrong:p
 
cooltide said:
A basic timing gun or even a fancy one set at 0 btdc connected to no.1 lead will show it up
Click to expand...
Sorry to keep on this but....how would it? I only know how to use the gun for checking the timing but wouldn't know what to expect in case of timing scatter. My problem is that the mechanical advance is always active and it is not possible to disconnect it as one can do with vacuum advance. Hence if the timing affects idle and is itself erratic (scattering) then idle would be affected, but if idle is affected by other factors then it will definitely result in timing change via the mechanical advance mechanism. Can they be uncoupled? Is there another way to spot primary scatter as opposed to scatter secondary to varying RPM at idle?
 
Well you do have a valid point. When I have experienced scatter with a defunct distributor, it was all over the place, quicker than the changes in idle speed.
iirc I had a spare distributor with new piggy back amplifier which cured it. Long time ago.
Unfortunately I never got around to dismantling the old dizzy. Think it's in a box somewhere with US tippexed on it
 
If the ecu isn't switching on idle stabilisation the options are a bit limited. I would have thought Rovergauge would show when idle stabilisation is active, certainly the Ecumate does.

You will need to look at throttle potentiometer voltage. Ignore % value it is the voltage that is important, you also need to ensure the voltage chanegs smoothly with no spikes or dead spots. At idle, throttle closed the throttle pot voltage needs to be something like 0.27v (do a search for the right voltage), also, the speed sensor input needs to be showing the right voltage although I forget what that should be. The other things to double check are ignition timing and the integrity of the vac advance unit.

If all the inputs check out then I might think about a substitute ecu. Have you thought about the O2 sensors - I assume it is a catalyst car. If it is a catalyst car the ecu will look at O2 inputs once it is up to temperature to modify the engine idle mixture so this could be something to look at.

One other thought, remove the air filter hose so you can get to the front of the air flow meter. Start the engine and let it get to an idle condition then SLOWLY place your hand in front of the air flow meter covering it one finger at a time until you completely cover the air flow meter which should cause the engine to stall. If the engine doesn't quite stall you have an inlet air leak. Good luck.
 
Hello folks, many thanks for all these attempts and thinking time devoted to my problem.
It is really a puzzling situation as most of the things have already passed the test, yet the car is not behaving.
Rovergauge has an indicator of idle mode, it's just my car that doesn't get it lighted up...
Throttle pot gives the correct voltage to start and also a smooth shift, or at least as smooth as my Vmeter+my eyes can tell.
Yesterday it developed another symptom: when I accelerate (at a modest rate) the engine will hesitate and miss, and not only under load. If I push the foot down then there is no hesitation, but if I drive "normally" it will reproduce it quite easily, while it is more difficult to reproduce it while running the engine with the car parked. You can tell it is not electrical but related to petrol supply as it is not a complete cut, what I cannot tell if it is too lean or too rich, the hesitation would suggest lean but the smell does suggest rich... Anyway, I wondered if the pot is faulty intermittently near the idle/light throttle position and this causes an abnormal petrol injection. I was thinking to bring a cable from the pot socket into the car and keep the Vmeter connected while driving to see if there is a position whereby it gives unreproducible readings and if that coincides with the hesitation, or variable reading can be observed while hesitating. If I keep the foot in the same position the symptom can last tens of seconds.
Another condition I know could give this intermittent symptom is if the recirculating petrol going back to the tank will not get directly back into the pump, so that the pump will not be filled up and will be pumping air bubbles. This could be particularly evident when the car moves.
All these factors (pot + pumping bubbles) might as well affect the idling randomly as I reported initially.
However, I really don't know how to check the pump factor, or better, I can't think of taking the tank down (my RR hasn't got the inspection panel) to visually inspect the pump assembly.... :-((
As for the last suggested test yes, it will stall and idle if the air filter hose is closed by hand, so no air leaks, which was also suggested by the fact that I can take the idle down to 300 RPM excluding the stepper during the base idle setting.
So the candidates now are throttle pot and fuel line back to the tank.
How do you guys feel about that?

Thanks for your help

Ciao
Adriano
 
Hi all,
some new symptoms accumulated in the meanwhile and I thought I might update the thread.
Idling is now down to1k, but for no apparent reasons. Nothing mechanical done after the plenum chamber replacement. May be it learns from experience....
Idling actually still fluctuates, now even with cold engine (it used to so it only when warm), and yet the engine will die if I engage drive or steer while idling.
During a steep downhill road with snow I had to use LR and 2nd gear and I noticed that every 60-90" something happened that reduced the speed (as if I clamped the petrol line) 4-5 times in the following 20 seconds, and then back to normal for another 60-90". I think that is the equivalent of the hesitation I have when accelerating, as if the petrol was not there. I put a length of clear plastic tubing between fuel pressure regulator and return pipe and observed the fuel flowing. There is plenty of air bubbles that possibly accumulate in weird places of the fuel rail, according to gravity, slope etc etc and that I suspect might as well be injected instead of petrol and cause the random missing and idling. There must be something damaged in the pump assembly, such as an air bubble filter or the like. I haven't been able to measure fuel pressure, however the symptom that doesn't fit with all this is the smell running too rich that comes immediately after it misses or even stalls while idling. On one side one would think the pump malfunction is causing the engine to run lean from time to time, on another side the smell suggests it is running rich.....
Could it be that when it runs too lean due to the air bubbles the ECU will try to compensate injecting more petrol and if bubbles have gone that more will be too rich?
Anyway a new pump is on its way and hopefully that should improve things. Presently, with less than a full tank hesitation is really heavy.

In the meanwhile something weird to the electricals also happened. The coolant level light went on while there was sufficient coolant in the tank. The visible electrical connections look ok, so I looked at the diagram to understand what was in common on that line and discovered that the coolant level switch is connected to fuse A5, and via the flasher unit also to fuse B3. When I went back to check the fusebox I found both fuses blown, but new fuses didn't cure the coolant warning light and neither blew again. The flasher unit and warning lights are not woking at the moment, even with new fuses. I checked all the remaining fuses and - surprise surprise....- when I removed the rear screenwiper fuse (C3) the engine stopped. I repeated the test several times and the two events are definitely connected, thought I have absolutely no clues at what could the connection be....and if it is normal or a symptom of something else going wrong.

I wonder what's next...... o_o

Ciao

Adriano
 
adri1957 said:
Hi all,
some new symptoms accumulated in the meanwhile and I thought I might update the thread.
Idling is now down to1k, but for no apparent reasons. Nothing mechanical done after the plenum chamber replacement. May be it learns from experience....
Idling actually still fluctuates, now even with cold engine (it used to so it only when warm), and yet the engine will die if I engage drive or steer while idling.
During a steep downhill road with snow I had to use LR and 2nd gear and I noticed that every 60-90" something happened that reduced the speed (as if I clamped the petrol line) 4-5 times in the following 20 sedamaged in the pump assembly, such as an air bubble filter or the like. I haven't been able to measure fuel pressure, however the symptom that doesn't fit with all this is the smell running too rich that comes immediately after it misses or even stalls while idling. On one side one would think the pump malfunction is causing the engine to run lean from time to time, on another side the smell suggests it is running rich.....
Could it be that when it runs too lean due to the air bubbles the ECU will try to compensate injecting more petrol and if bubbles have gone that more will be too rich?
Anyway a new pump is on its way and hopefully that should improve things. Presently, with less than a full tank hesitation is really heavy.

In the meanwhile something weird to the electricals also happened. The coolant level light went on while there was sufficient coolant in the tank. The visible electrical connections look ok, so I looked at the diagram to understand what was in common on that line and discovered that the coolant level switch is connected to fuse A5, and via the flasher unit also to fuse B3. When I went back to check the fusebox I found both fuses blown, but new fuses didn't cure the coolant warning light and neither blew again. The flasher unit and warning lights are not woking at the moment, even with new fuses. I checked all the remaining fuses and - surprise surprise....- when I removed the rear screenwiper fuse (C3) the engine stopped. I repeated the test several times and the two events are definitely connected, thought I have absolutely no clues at what could the connection be....and if it is normal or a symptom of something else going wrong.

I wonder what's next...... o_o

Ciao

Adriano
Click to expand...
 
You must log in or register to reply here.

Similar threads

H
Finding source for vibration/lump at idle
Replies
2
Views
512
HACKCHILD
H
C
3.9 efi no fuel, wont start, tested injectors right bank 8 ohms.........
Replies
6
Views
645
cljcars
C
J
Struggle Idling when cold 1985 Land Rover 90
Replies
2
Views
330
Knappster
Knappster
D
3.9 V8 auto 14CUX limp mode/no power
Replies
3
Views
771
AllanWorms
AllanWorms
M
Desperate. Don't know anymore.
Replies
17
Views
1K
boguing
boguing
Back
Top