Idle issue with Rover 3.9 EFI : MG Engine Swaps Forum : MG Experience Forums : The MG Experience

Re: Idle issue with Rover 3.9 EFI #41
Rover827 Rich Truett Berkley, USA	Posted: April 22, 2012 11:17AM Registered: 1 year ago Posts: 8

Mark:

I have to give you a very public THANK YOU for your advice. I just finished checking the C02 levels at idle and found they were off in outer space somewhere. Worse, the AFM would not respond when I turned the allen wrench adjuster. No matter. I had another one and installed it. I checked that one and the settings were perfect for a non-cat vehicle.

So: the result is that by following your advice and setting the base idle speed, making sure my C02 levels were also properly set and by my own switching to the green tune resistor, my 1981 TR8 4.0-liter running the TVR Chimeara fuel map now has more power, idles perfectly, and runs extremely smoothly. I could not be happier.

The lesson here for all is: the Rover injection system is unforgiving of modifications to the PCV system, won't tolerate vacuum leaks, and must be set within factory parameters for it to behave properly. But when you do get all things dialed in, then you know why men spend crazy money and countless hours messing with the Rover V8 -- it is a sweet, sweet engine when all is in order.

Go here for a great tutorial on the Rover injection system: http://www.g33.co.uk/fuel_injection.htm

Sponsored Links. You can hide this ad & support this site by upgrading to a Gold Membership ~ click here for more info.

Reply Quote

Re: Idle issue with Rover 3.9 EFI #42
blitzracing Mark Thompson Hungerford, United Kingdom	Posted: April 22, 2012 12:32PM Registered: 1 year ago Posts: 23

Thanks- makes all that typing worth while :-)

Back to Chris's initial issue, as both lambdas are reading rich, you have some fairly limited options. First test get some air into the plenum by whatever means, loosening off the blanking bungs off is a good way, you may need to take it out and simply hold it against the hole to get enough air in, and see if the Lambda readings on the ECU mate drop and the probes start switching. This will prove the ECU and probes are working correctly if they do start switching.

If this works then you are most likely looking at AFM voltages or fuel pressure. You have swapped the AFM, and the voltage readings look OK on the surface, but you only have around 150mv to play with at idle before the ECU cant compensate for the wrong AFM reading- so bad earths, or poor connections can be enough to do this.

Fuel pressure- this is held at 3 bar below the pressure in the plenum and at the injector tip by the regulator. Its difficult to measure as there is no take off on the fuel rail on the earlier 3.9 engines, so you need to add a T piece and gauge where the fuel rail is fed from the pump- To get the reading you need to disconnect the rubber pipe from the plenum so the fuel pressure will be 3 bar as the regulator is now vented to atmosphere, and get the fuel pump to run. You can do this at idle, but it makes the engine run very rough / rich as the idle pressure is reduced normally by the high plenum vacuum with the throttle plate shut.

You may also want to check the fuel return line, if this gets kinked / blocked it will make the fuel pressure go up when when fuel demand is lowest- ie idle

Reply Quote

Re: Idle issue with Rover 3.9 EFI #43
rhinsb Rich Hagon Santa Barbara, California, USA 1971 MG MGB GT "Money Pit"	Posted: April 22, 2012 05:27PM Registered: 6 years ago Posts: 1,168

Chris, I think you need to fly Mark over for a beer/car diagnostic party! winking smiley

Sponsored Links. You can hide this ad & support this site by upgrading to a Gold Membership ~ click here for more info.

Reply Quote

Re: Idle issue with Rover 3.9 EFI #44
mgv8glen Glenn Towery Dover De., USA	Posted: April 22, 2012 10:08PM Registered: 5 years ago Posts: 2,330

I would like to thank Chris for this post & everyone that has given a reply!!!!! This has been one of the BEST! I will take what Mark posted & read over it till I turn the light off tonight. This week I am going to change the K&N vent back to what it was from the factory. Will ck. & make sure my dipstick is making a good seal. Thanks to all!!!!!!

Reply Quote

Re: Idle issue with Rover 3.9 EFI #45
Duncan Duncan Cowen Vancouver, Canada 1956 MG MGA 1500	Posted: April 23, 2012 01:57AM Registered: 5 years ago Posts: 1,360

In reply to # 2047629 by mgv8glen Will ck. & make sure my dipstick is making a good seal. Thanks to all!!!!!!

That dipstick thing is tricky, most of us forget it. I remember it making a difference to the idle on my old '73 B if the dipstick was in or not, and it was clearly stated in the manual that the engine was supposed to be airtight. I replaced the o-ring on that dipstick and moved on to other leaks!

Making the PCV system of the car as close to factory is also a good thing! I remember that that old '73 MGB barely made it home during a drive not long after I got it, because my sister took it for a run on the highway and the emissions hoses had been plugged with ball bearings etc. The blow-by built up enough pressure in the engine that the tappet chamber gasket blew out the side of the engine, and the resulting vacuum leak was so big the car wouldn't run without the choke on. That's what got me putting the emissions back to factory on that car!

On my Mini, same careful application of ball bearings in emissions hoses (why did people do this??). The poor Mini didn't blow any gaskets though, just blew all the oil out of every orifice until it ran down the inside surface of the bonnet, and then sprayed out of a hole at the front onto the windscreen. Yummy!

Get it as close to factory as you can. A great place to start, at least!

Sponsored Links. You can hide this ad & support this site by upgrading to a Gold Membership ~ click here for more info.

Reply Quote

Re: Idle issue with Rover 3.9 EFI #46
cgill Chris Gill Salmon Arm, British Columbia, Canada 1971 MG MGB GT 1979 MG MGB "BOPR V8 Conversion"	Posted: April 25, 2012 12:28AM Registered: 6 years ago Posts: 2,832

Hi guys.

Just a quick update. I checked to make sure the dipstick was well sealed. It looks okay but I'll probably get another o-ring to be safe. I sealed up the flame trap so it is okay.

I pulled all the spark plugs and they are light gray so it's running lean. The spark plugs were all the same colour suggesting that it isn't one cylinder causing the slight miss. I adjusted the gap on each of them to .035. Is this the correct gap? Based on what I read it should be bewteen .033 and .038.

I then took a closer look at my oxygen sensor wiring. I have the two white/orange wires from the oxygen sensors wired to the slate coloured wire on the Advance Autowire harness. This is originally for a fuel pump warning buzzer.. I'll try to dig up a diagram tomorrow so you can see what is going on.

I have two other wires coming from the ECU that are connected to the AAW harness: A blue with pink (or maybe purple) stripe going to the Black with yellow stripe AAW wire (originally for he inertia cutout switch), and a White with gray stripe from the ECU going to the white with a green stripe on the AAW harness (goes from the power relay to the ignition switch).

I wired it this way based on input from my mechanic friend but I have no idea if it is correct. The one other thing I did tonight was disconnect both oxygen sensors. The car still had the slight miss and erratic idle.

All ground wires from the EFI system and ECU are good (paint removed from grounding point).

Could this be an ignition and fuel pressure issue? I have my fuel pressure set at 38psi at idle. Unfortunately I don't have the gauge with me right now so I can't see if it is increasing under load. It is an in tank fuel pump built by Todd Budde.

I am really hoping this will be some 'eureka' moment when I finally figure it out....

Thanks again for all your advice and patience as I work through this!
Chris

My other obsession...photography!

http://www.flickr.com/photos/77646803@N00/

Reply Quote

Re: Idle issue with Rover 3.9 EFI #47
blitzracing Mark Thompson Hungerford, United Kingdom	Posted: April 25, 2012 06:29AM Registered: 1 year ago Posts: 23

You have the tool you need the verify the lamda wiring, the ECUmate. Unplug the lambda probes and the reading on the ECU mate should show lean, on both sides, as you have removed the voltage source. I would run the engine until you get the rich reading from the probe, then disconnect a probe and see what the reading does with the engine still running. You need to do this to get the ECU to read the lambda's in the first place, I doubt the ECU mate can read the voltages directly, has has to be processed by the CPU in the ECU first, so the lambda circuit should be active. If it still reads rich, then something is very wrong with the wiring. The probes need a 12 volt feed to the heater, not just to warm them up, but also to provide a voltage source for the resistive element in the probe, and in once case I saw the 12 volt supply had shorted to the lambda output and shoved 12 volts into the ECU. It did not bat an eyelid, but just tried to lean off the mixture as it thought a 12 volt signal from the probe was rich! I dont think however this is your issue, as both sides would have to be shorted that seems unlikely. In terms of plug colours- they are only off use if you run the engine for a while in the failing region and then kill the engine, (called a plug chop) as the colour will change say from full load to idle if the conditions are different, so its not a very good test. It is possible that if the lambda outputs are wrong, that the ECU is leaning the mixture right off to try an compensate, and making it run rough. This is why its worth getting extra air into the plenum to see if the probes pick this up :-) With the probes unplugged it should also generate a fault code after a period as well as the ECU should pick up the fact that the mixture is now "lean" for a period.

If you fuel pressure was 38 psi at idle, its not so far out to cause an issue, unless the fuel return has become blocked for some reason since the test was made.

have a read of this:

http://www.g33.co.uk/images/PDFS/14cux%20faulot%20finding.pdf

Mark.

Reply Quote

Re: Idle issue with Rover 3.9 EFI #48
Duncan Duncan Cowen Vancouver, Canada 1956 MG MGA 1500	Posted: April 25, 2012 10:50AM Registered: 5 years ago Posts: 1,360

Chris, I didn't know you'd had to splice different looms together to use the AAW setup. As I mentioned in the second post of this thread, a wiring mistake on my part caused the engine to run so rich it would die below about 1500-2000rpms. That wiring mistake was because I hooked a computer from one car into a loom from another, which is essentially the same thing you've done. It's easy to make a mistake doing that! And an easy mistake to fix, too.

When I connected my ECM to my loom, I made an Excel sheet showing what wire went where, like ribs off a spine. The "vertebrae" were the components (eg O2 voltage, AC request, IGN, etc), and the "ribs" were the wire colours. On one side, the wires from the ECM, the port on the ECM they went to, and their colour. On the other side, the colour of the wire in the loom, and the connector they went into.

How'd you plan the connections, and how many colours are there in the AAW harness? Any chance you mistook two similar coloured wires??

Reply Quote

Re: Idle issue with Rover 3.9 EFI #49
cgill Chris Gill Salmon Arm, British Columbia, Canada 1971 MG MGB GT 1979 MG MGB "BOPR V8 Conversion"	Posted: April 26, 2012 10:14AM Registered: 6 years ago Posts: 2,832

Mark and Duncan, once again thank you for the advice and encouragement. I really appreciate it.

Mark, that article on troubleshooting oxygen sensor fault codes was great. I am going to test them on Sunday (I'm on my way to California at the moment - damn work is getting in the way of my fun!). I also picked up an electronic ignition to give that a shot because it might be multiple issues going on and I thought it wouldn't hurt to try.

I will post an update once I have had a chance to do the testing on Sunday.

Cheers,
Chris

My other obsession...photography!

http://www.flickr.com/photos/77646803@N00/

Reply Quote

Re: Idle issue with Rover 3.9 EFI #50
Duncan Duncan Cowen Vancouver, Canada 1956 MG MGA 1500	Posted: April 26, 2012 01:28PM Registered: 5 years ago Posts: 1,360

Electronic ignition is an excellent upgrade!! I'm surprised you went with EFI and yet still had points. Have a safe trip, and when you get back, I look forward to hearing about any changes the electronic ignition make!

Reply Quote

Re: Idle issue with Rover 3.9 EFI #51
cgill Chris Gill Salmon Arm, British Columbia, Canada 1971 MG MGB GT 1979 MG MGB "BOPR V8 Conversion"	Posted: April 30, 2012 12:45PM Registered: 6 years ago Posts: 2,832

A quick update...

The electronic ignition did wonders for the missing and general sputtering. It now pulls hard to 5K without hesitation at the higher rpms although there is the occasional brief miss when the RPMs are low and the engine is under load.

We put a voltmeter on the oxygen sensors (which are new NGKs) and couldn't see them cycling. We pulled a vacuum line with no change in the sensors. We also sprayed brake cleaner into the plenum with no change in the sensors. We are going to try using a higher quality voltmeter to see if that makes a difference but the ECU mate also didn't show any change in the Oxygen sensors with the readout. I checked the wiring and I have 12 volts to the heater wires and the grounds are good. So the only culprit could be the blue wire from the oxygen sensor to the ECU but I didn't touch this part of the rover harness so I can't see how that might be affected.

My other obsession...photography!

http://www.flickr.com/photos/77646803@N00/

Reply Quote

Re: Idle issue with Rover 3.9 EFI #52
blitzracing Mark Thompson Hungerford, United Kingdom	Posted: April 30, 2012 01:53PM Registered: 1 year ago Posts: 23

Can you disconnect the sensor output from the blue wire? If you can, just measure the voltage on the wire leading to the ECU (now disconnected). It should read 0 volts, and if it does not you have a loom issue. Then check the probe output open circuit- It should NEVER read more than about 1.4 volts or so, however rich the mixture is. If its above this, again you have a wiring issue. Dont forget to try and get some extra air into the plenum by introducing an air leak if you can remove one of the plenum bungs bit by bit and see if the probe output drops. It does not need to be connected to the ECU to work.

Reply Quote

Re: Idle issue with Rover 3.9 EFI #53
cgill Chris Gill Salmon Arm, British Columbia, Canada 1971 MG MGB GT 1979 MG MGB "BOPR V8 Conversion"	Posted: April 30, 2012 02:13PM Registered: 6 years ago Posts: 2,832

Hi Mark,

I'm not sure what you mean by disconnecting the sensor output from the blue wire. Can you clarify? I can get a probe on the blue wire from the oxygen sensor to the ECU if that is what you mean. I'll check out voltages on it when I get a chance.

I artificially enriched the mixture using brake cleaner sprayed into the plenum and also pulled a vacuum line to get extra air in with no difference in oxygen sensor operation.

Cheers,
Chris

My other obsession...photography!

http://www.flickr.com/photos/77646803@N00/

Reply Quote

Re: Idle issue with Rover 3.9 EFI #54
blitzracing Mark Thompson Hungerford, United Kingdom	Posted: April 30, 2012 02:25PM Registered: 1 year ago Posts: 23

This is the wire colours on the the 14CUX looms I have worked on- I cant guarantee your are the same, but basically if you disconnect the black wire from the probe that joins the screened blue(?) wire and test each wire independently. This will trace if something is wrong at the probe end, or if its a fault with the ECU wiring. The probes are quite happy with the signal wire disconnected, you can just read the voltage straight off the probe with a meter. You normally have a big cylindrical 3 pin connector that joins the probe to the loom(?) They are not that easy to pull to bits once they are together but you can try and get a small jewelers screwdriver down the side of the pin you want to remove to push the locking tab back and pull the pin out. If they are new probes are they crimped into the loom? They are difficult to solder if you cut the probe wires as they are steel, not copper. It can be done but need a good flux.

Loom = White/orange (Heater) Probe=Red
Loom = Screened blue (Probe signal) Probe = Black
Loom = Black ( Ground ) Probe = White

NOTE measure your voltages to the black loom earth, not chassis as may not be a quite the same level and the ECU earth- you will notice all the sensors have their own earth- nothing is grounded through the chassis.

Edited 3 time(s). Last edit at 04/30/2012 02:38PM by blitzracing.

Reply Quote

Re: Idle issue with Rover 3.9 EFI #55
blitzracing Mark Thompson Hungerford, United Kingdom	Posted: April 30, 2012 02:55PM Registered: 1 year ago Posts: 23

Just a thought- you have got the correct NTK resistive lambda probes? The more common voltage generating lambda probes have a reverse voltage curve- ie lean gives voltage, rich does not. I hate to think what the ECU would do if you fitted these.....

Reply Quote

Re: Idle issue with Rover 3.9 EFI #56
Simon Austin Surrey, BC, Canada 1977 MG MGB V8 Conversion 1994 MG RV8	Posted: April 30, 2012 05:51PM Registered: 8 years ago Posts: 8,221

Chris,

What electronic ignition did you install?

"Speed costs........how fast you want to spend?"

Reply Quote

Re: Idle issue with Rover 3.9 EFI #57
cgill Chris Gill Salmon Arm, British Columbia, Canada 1971 MG MGB GT 1979 MG MGB "BOPR V8 Conversion"	Posted: April 30, 2012 06:33PM Registered: 6 years ago Posts: 2,832

Pertronix 1181 kit. About $100 at Lordco.

Mark, I bought them from a local parts store using the part number for my EFI system. I assume they are right. They look identical to my old ones. I might try to re-install the originals to see if there is any difference just in case....

My other obsession...photography!

http://www.flickr.com/photos/77646803@N00/

Reply Quote

Re: Idle issue with Rover 3.9 EFI #58
Simon Austin Surrey, BC, Canada 1977 MG MGB V8 Conversion 1994 MG RV8	Posted: April 30, 2012 09:00PM Registered: 8 years ago Posts: 8,221

Probably the same Pertronix I have. Buick distributor, correct?

"Speed costs........how fast you want to spend?"

Reply Quote

Re: Idle issue with Rover 3.9 EFI #59
cgill Chris Gill Salmon Arm, British Columbia, Canada 1971 MG MGB GT 1979 MG MGB "BOPR V8 Conversion"	Posted: April 30, 2012 10:50PM Registered: 6 years ago Posts: 2,832

Yep!

My other obsession...photography!

http://www.flickr.com/photos/77646803@N00/

Reply Quote

Re: Idle issue with Rover 3.9 EFI #60
cgill Chris Gill Salmon Arm, British Columbia, Canada 1971 MG MGB GT 1979 MG MGB "BOPR V8 Conversion"	Posted: May 09, 2012 12:52AM Registered: 6 years ago Posts: 2,832

I'm still waiting for a friend to come over with a fresh set of eyes to look at my setup becaue I am out of ideas. I want to test voltages at various points between the oxygen sensors and the ecu but would like an electrical guru with me to ensure I am doing things right.

Tonight I ruled out an issue with the actual oxygen sensors. I installed some original ones and they also went open loop. It has to be wiring but aside from having the ground wires for the oxygen sensors sharing a common ground with a couple of other components, that is the only questionable thing I have found.

One thing to note is that the oxygen sensor wires on the passenger side pass very close to the starter motor wires. Should I insulate the oxygen sensor wires here to avoid stray voltages?

On to the second issue - I can completely close the throttle plate by screwing down the base idle setting with the stepper unplugged but the engine will still run. I pulled the upper plenum and checked it for possible vacuum leaks, sealed up anything that looked remotely suspicious and then compared the throttle plate seal to another plenum I have. It seems that air can get around the throttle plate easier on my plenum compared to the second one I have if I blow compressed air at the plate. I have no idea why that would happen because there is no physical difference when you look at the throttle plates. I put mine back on and started the car. It ran the same. I didn't have time tonight to try tightening down the base idle screw again. I'll try that tomorrow night. If it still runs I am going to try the second plenum to see if that makes a difference.

My other obsession...photography!

http://www.flickr.com/photos/77646803@N00/

Reply Quote

The MG Experience

www.mgexp.com

MG Engine Swaps Forum

Idle issue with Rover 3.9 EFI

Add your reply here, or post your own questions!Members Sign In if you've already registered, orRegister a New AccountRegistration is free and takes less than a minute

Join Our Club

MGExp Menu

—

Adjust Text Size

More to Explore

Add your reply here, or post your own questions!

Members Sign In if you've already registered, or
Register a New Account
Registration is free and takes less than a minute