Mixture adjustment on a Rover 3.9 V8 without Lambda Sensors.
Discussion
Sorted.
I've got pretty good results with a remap taking ~10% off the injected fuel. I've now got hot idle ~13.5:1, mild acceleration around 14.5:1, lean cruise ~15.5:1 and hard acceleration around 13:1.
If anyone else is interested in doing the same, here is the guide: http://www.stevesprint.com/remap-14cux and the thread here which gets longer by the week: http://www.pistonheads.com/gassing/topic.asp?h=0&a...
You need Roverguage to see which part of the fuel-map you are currently running in.
You need a wide-band oxygen sensor and gauge to know how rich/lean you are currently running.
You need TunerPro (free tuning software) and the TunerPro 14CUX definition (XDF) file to easily identify maps and edit them.
You need "14CUX Toolkit" to correct the checksumms and duplicate your changes to the rest of the chop.
You need a chip-burner an 28C256 EEPROM chips (28 pin) to write and re-write with your changes.
One bit which wasted a bit of my time, Rovergauge said i was on "Map 1". But in the Tunerpro XDF that was referred to as "Map 0".
Making changes to Map 1 acheived nothing, but Map 0 responded as it should.
Steve covers this in the fine-print on the page linked above. Just took me a while to find it.
I'm now looking forward to burning less than 20 litres/100km for the first time since I've owned this thirsty beast.
I've got pretty good results with a remap taking ~10% off the injected fuel. I've now got hot idle ~13.5:1, mild acceleration around 14.5:1, lean cruise ~15.5:1 and hard acceleration around 13:1.
If anyone else is interested in doing the same, here is the guide: http://www.stevesprint.com/remap-14cux and the thread here which gets longer by the week: http://www.pistonheads.com/gassing/topic.asp?h=0&a...
You need Roverguage to see which part of the fuel-map you are currently running in.
You need a wide-band oxygen sensor and gauge to know how rich/lean you are currently running.
You need TunerPro (free tuning software) and the TunerPro 14CUX definition (XDF) file to easily identify maps and edit them.
You need "14CUX Toolkit" to correct the checksumms and duplicate your changes to the rest of the chop.
You need a chip-burner an 28C256 EEPROM chips (28 pin) to write and re-write with your changes.
One bit which wasted a bit of my time, Rovergauge said i was on "Map 1". But in the Tunerpro XDF that was referred to as "Map 0".
Making changes to Map 1 acheived nothing, but Map 0 responded as it should.
Steve covers this in the fine-print on the page linked above. Just took me a while to find it.
I'm now looking forward to burning less than 20 litres/100km for the first time since I've owned this thirsty beast.
Tune is ~95% sorted and I'm very happy with it. Fuel consumption is still unknown (will take several tanks to get an average) but it's already doing ~300km to the fuel light (dented tank, reduced volume) where previously it had done 136km to the fuel light.
After the previous rough tune (leaning out whole chunks of the map to get it within range) it was time for some fine-tuning. I setup roverguage on a laptop in the passengers seat to display the fuel table cell currently being used. My lovely (but reluctant) assistant had the job of logging as I drove. Juggling a laptop and a clip-board with a printed out a spreadsheet of the current fuel map with space under each cell to write the current A/F reading.
Now automatics are a PITA for this. Try as you might to get a specific load/rpm point and you can't. It'll change down and you're off into a whole nother part of the map.
But I decided that doesn't actually matter. If I can't purposefully get into those parts of the map, then I shouldn't accidentally be in them either and whatever is there can do just fine.
The operating range is a big swoop from idle with 25-38% load going to a bit over 3000rpm. More throttle (higher load) is required to get more rpm so that whole upper right part of the map isn't accessible.
On the bottom left I found it similarly impossible to get a full load reading at 2000rpm. I could get into full load only at 2,700-3,100 and of course to the rev-limit.
I found big lean spots (~15.5:1) at about 3000rpm at full load, higher was around 13:1. Interestingly at this 3000rpm point I'm now instructing the injectors to be fully open (255 on a scale of 0-255) and this is working well.
I'm working on target A/F of ~13.5:1 at low load (idle), 16:1 (lean cruise) for low load at 1500-3000rpm. Stoich by about 75% load and tapering to ~12.5:1 at full load.
As well as correcting the values I have, I'm doing a lot of smoothing in between. There are some very strange dips and peaks which I'm hoping were related more to Pommy emissions tests than engine needs.
It's working very well. I have stable idle both hot and cold, I have 16:1 A/F at cruising from 80-100km/h and likely beyond. At application of full pedal it holds 12.5:1 until you let off.
I think I'll richen up the full load a bit to 12:1 to ward of what might be some light pinging (detonation) there. Otherwise it's very good. It feels like it has at least the 180 horses it originally had and even sounds purposeful.
I still haven't touched timing and I don't think I will. If I can get rid of all traces of detonation at 12:1 full load it'll be timed about perfect for the 91 octane I'm using. I'm happy it's advanced as much as it safely can be already and I'm not feeding this thing high-octane if I can tune around it for operating conditions I virtually never use.
If you run a 3.9V8, this is highly recommended.
After the previous rough tune (leaning out whole chunks of the map to get it within range) it was time for some fine-tuning. I setup roverguage on a laptop in the passengers seat to display the fuel table cell currently being used. My lovely (but reluctant) assistant had the job of logging as I drove. Juggling a laptop and a clip-board with a printed out a spreadsheet of the current fuel map with space under each cell to write the current A/F reading.
Now automatics are a PITA for this. Try as you might to get a specific load/rpm point and you can't. It'll change down and you're off into a whole nother part of the map.
But I decided that doesn't actually matter. If I can't purposefully get into those parts of the map, then I shouldn't accidentally be in them either and whatever is there can do just fine.
The operating range is a big swoop from idle with 25-38% load going to a bit over 3000rpm. More throttle (higher load) is required to get more rpm so that whole upper right part of the map isn't accessible.
On the bottom left I found it similarly impossible to get a full load reading at 2000rpm. I could get into full load only at 2,700-3,100 and of course to the rev-limit.
I found big lean spots (~15.5:1) at about 3000rpm at full load, higher was around 13:1. Interestingly at this 3000rpm point I'm now instructing the injectors to be fully open (255 on a scale of 0-255) and this is working well.
I'm working on target A/F of ~13.5:1 at low load (idle), 16:1 (lean cruise) for low load at 1500-3000rpm. Stoich by about 75% load and tapering to ~12.5:1 at full load.
As well as correcting the values I have, I'm doing a lot of smoothing in between. There are some very strange dips and peaks which I'm hoping were related more to Pommy emissions tests than engine needs.
It's working very well. I have stable idle both hot and cold, I have 16:1 A/F at cruising from 80-100km/h and likely beyond. At application of full pedal it holds 12.5:1 until you let off.
I think I'll richen up the full load a bit to 12:1 to ward of what might be some light pinging (detonation) there. Otherwise it's very good. It feels like it has at least the 180 horses it originally had and even sounds purposeful.
I still haven't touched timing and I don't think I will. If I can get rid of all traces of detonation at 12:1 full load it'll be timed about perfect for the 91 octane I'm using. I'm happy it's advanced as much as it safely can be already and I'm not feeding this thing high-octane if I can tune around it for operating conditions I virtually never use.
If you run a 3.9V8, this is highly recommended.
Kiwibacon said:
Interestingly at this 3000rpm point I'm now instructing the injectors to be fully open (255 on a scale of 0-255) and this is working well.
If this is the case, it is not ideal to run the injectors at 1005 duty.So are injector duties actually at 100%, or is there just another scalar or table somewhere that means the numbers on your main table are the restriction for more fuel ?
ie with older Link ecu's, Motec etc etc, there are other multipliers that basically give you more or less resolution on the main table. You could see say 255 on the main table, but it's simply the other multipliers are the restriction, not that you're actually out of ability to add more fuel.
stevieturbo said:
If this is the case, it is not ideal to run the injectors at 1005 duty.
So are injector duties actually at 100%, or is there just another scalar or table somewhere that means the numbers on your main table are the restriction for more fuel ?
ie with older Link ecu's, Motec etc etc, there are other multipliers that basically give you more or less resolution on the main table. You could see say 255 on the main table, but it's simply the other multipliers are the restriction, not that you're actually out of ability to add more fuel.
Short story is I don't know about the scaling. Longer story is I don't think I'll be at that specific operating point for long enough to cause any problems.So are injector duties actually at 100%, or is there just another scalar or table somewhere that means the numbers on your main table are the restriction for more fuel ?
ie with older Link ecu's, Motec etc etc, there are other multipliers that basically give you more or less resolution on the main table. You could see say 255 on the main table, but it's simply the other multipliers are the restriction, not that you're actually out of ability to add more fuel.
Only one load cell is at 255 (2700-3000rpm, full load). The others are significantly less yet provide the same A/F ratio. Which does suggest other multipliers interacting.
Kiwibacon said:
Short story is I don't know about the scaling. Longer story is I don't think I'll be at that specific operating point for long enough to cause any problems.
Only one load cell is at 255 (2700-3000rpm, full load). The others are significantly less yet provide the same A/F ratio. Which does suggest other multipliers interacting.
You could identify the actual injector duty cycle by inspection - use a multimeter with a duty cycle scale across one of the injector banks and monitor this against what the ECU requests for that load point.Only one load cell is at 255 (2700-3000rpm, full load). The others are significantly less yet provide the same A/F ratio. Which does suggest other multipliers interacting.
100SRV said:
You could identify the actual injector duty cycle by inspection - use a multimeter with a duty cycle scale across one of the injector banks and monitor this against what the ECU requests for that load point.
I could. But these vehicles have far more pressing matters to attend to.So a bit of an update.
It's been running brilliantly ever since. Fuel economy around 14-16 litres/100km which, to be honest, ain't bad for how I've been using it.
The one problem I did have, was erratic hot idle. Sometimes it would be excellent, other times the rpm would keep dropping until the oil light would flicker.
The main problem was I tuned in the idle in July (coldest month, middle of winter) and temperature is having a bigger effect than I thought.
So today (32C) I strapped the wideband gauge back on with the intent of finding out what that little "Carbon Monoxide" screw on the side of the MAF did.
Cold idle: 11.5:1. All good.
MAF voltage (blue-blue) checked at the prescribed 1.0V.
Take it for a run to get it hot and the idle has leaned out to stoich! Problem found then.
Adjust MAF screw so the volts at key on but engine off are at 0.5V.
Hot idle is now fluctuating between 14.5:1 and 15:1. Clearly the wrong direction.
Adjust MAF screw to 1.5V.
Hot idle is now ~13.5:1 and fluctuates between 13.2-13.6:1.
I do think I've cracked it. So if anyone needs to know, lower volts on the MAF is a leaner idle. Higher volts on the MAF is a richer idle.
A/F ratios at open road cruise are unaffected. I haven't had time to do an exhaustive check over the whole map.
May this help someone else in the future.
It's been running brilliantly ever since. Fuel economy around 14-16 litres/100km which, to be honest, ain't bad for how I've been using it.
The one problem I did have, was erratic hot idle. Sometimes it would be excellent, other times the rpm would keep dropping until the oil light would flicker.
The main problem was I tuned in the idle in July (coldest month, middle of winter) and temperature is having a bigger effect than I thought.
So today (32C) I strapped the wideband gauge back on with the intent of finding out what that little "Carbon Monoxide" screw on the side of the MAF did.
Cold idle: 11.5:1. All good.
MAF voltage (blue-blue) checked at the prescribed 1.0V.
Take it for a run to get it hot and the idle has leaned out to stoich! Problem found then.
Adjust MAF screw so the volts at key on but engine off are at 0.5V.
Hot idle is now fluctuating between 14.5:1 and 15:1. Clearly the wrong direction.
Adjust MAF screw to 1.5V.
Hot idle is now ~13.5:1 and fluctuates between 13.2-13.6:1.
I do think I've cracked it. So if anyone needs to know, lower volts on the MAF is a leaner idle. Higher volts on the MAF is a richer idle.
A/F ratios at open road cruise are unaffected. I haven't had time to do an exhaustive check over the whole map.
May this help someone else in the future.
blitzracing said:
The setting affects the mixure up to around 2500 rpm, but the effect reduces as the RPM goes up. The setting is much the same as the long term trim value that closed loop uses.
Thanks Mark.Do you know anything about hot/cold mixture adjustment? Can they be tweaked individually or is there a fixed offset?
Hi, I have a Landrover 3.5 V8 with a 3.9 Lucas ECU injection, this might help....
The throttle position sensor, MAF and ECU have to be setup with each other so they read each other correctly.
-Ensure there is no air leaks
-Throttle butterfly at closed top / bottom should be a 0.5 to 1mm gap between it and the plenum & dosent close over centre
-Throttle position sensor from the red cable to earth should be set at 0.35v and rise to 4.5 / 5v at full throttle
-Fuel presser should prime to 40psi on most engines
-Set ignition timing to 6 BTDC
-Next warm up engine & adjust the MAF meter to (1.5v non cat, 1.8v with a cat) Take reading between Blue/Red & Red/Black wire)
-Turn off engine & clamp air hose to stepper motor
-There is a idle adjustment screw on top of the plenum near to the throttle mechanism. Turn until you get 550 to 650 rpm
-Turn engine off remove clamp from air hose
-Now drive vehicle advancing the timing until it pinks under load. (Drive up a hill). Then retard a couple degrees so pinking doesn’t accur...
The throttle position sensor, MAF and ECU have to be setup with each other so they read each other correctly.
-Ensure there is no air leaks
-Throttle butterfly at closed top / bottom should be a 0.5 to 1mm gap between it and the plenum & dosent close over centre
-Throttle position sensor from the red cable to earth should be set at 0.35v and rise to 4.5 / 5v at full throttle
-Fuel presser should prime to 40psi on most engines
-Set ignition timing to 6 BTDC
-Next warm up engine & adjust the MAF meter to (1.5v non cat, 1.8v with a cat) Take reading between Blue/Red & Red/Black wire)
-Turn off engine & clamp air hose to stepper motor
-There is a idle adjustment screw on top of the plenum near to the throttle mechanism. Turn until you get 550 to 650 rpm
-Turn engine off remove clamp from air hose
-Now drive vehicle advancing the timing until it pinks under load. (Drive up a hill). Then retard a couple degrees so pinking doesn’t accur...
^^^^^
Don't seem to be having much luck with yours.
"Hi, Iv finally given up on my 3.9 injection system that was fitted to my 3.5 V8 Defender. When it goes wrong you can’t find anyone to work on it and the guys who know what they are doing cost a fortune, Parts cost a fortune even 2nd hand...!!! The internet is full of forums complaining about poor running issues with the early injection system so why are we converting from good old carbs to something that is over 25years old and has been pulled out of an old scraper....? Who out there is running carbs and can give us a little advice on what set up would work best...?"
www.landyzone.co.uk/land-rover/3-9-injection-syste...
Don't seem to be having much luck with yours.
"Hi, Iv finally given up on my 3.9 injection system that was fitted to my 3.5 V8 Defender. When it goes wrong you can’t find anyone to work on it and the guys who know what they are doing cost a fortune, Parts cost a fortune even 2nd hand...!!! The internet is full of forums complaining about poor running issues with the early injection system so why are we converting from good old carbs to something that is over 25years old and has been pulled out of an old scraper....? Who out there is running carbs and can give us a little advice on what set up would work best...?"
www.landyzone.co.uk/land-rover/3-9-injection-syste...
Edited by paintman on Tuesday 5th March 22:41
Hey! I've been looking for a thread like this online for ages! I always knew there was a way to get better consumption from this engine and to make it run more reliable and less rich. I have a few more questions for you, kiwibacon, if you don't mind? Is there a way I could get hold of you?
Also, would it be possible to upload your .bin file for us to use as a starting point?
Thanks!
Also, would it be possible to upload your .bin file for us to use as a starting point?
Thanks!
SALandy said:
Hey! I've been looking for a thread like this online for ages! I always knew there was a way to get better consumption from this engine and to make it run more reliable and less rich. I have a few more questions for you, kiwibacon, if you don't mind? Is there a way I could get hold of you?
Also, would it be possible to upload your .bin file for us to use as a starting point?
Thanks!
I suspect there's enough variation in individual engine behaviour at this age that my BIN file wouldn't help you. For some reason this car I've got was running far too rich but everything normally measurable was fine.Also, would it be possible to upload your .bin file for us to use as a starting point?
Thanks!
That makes sense. I was just wondering in terms of a starting point. Did you only adjust the fueling levels at those load points? You didn't change anything else besides that and the CO pot?
I have the same issue with my disco 1. I've also checked all the measureable values and all checks out. The only issue I have is that I believe the connection between my MAF and the ECU is a bit dodgy. The car runs great when it runs, however, after a long run, it will start to stutter and fail when I'm not accelerating. I believe this is due to the MAF "disconnecting" from the ECU and thus the ECU struggles to control the engine correctly. This inevitable causes the car to die and it's difficult to start as the car has blackened the spark plugs from it running so rich. I will run a new cable through between the two once I get a moment and the plug I've ordered comes. I'm also going to check the fuel system and the ignition system as I believe some of the cabling needs to be replaced.
After all that I would like to mess around with the ECU a bit...
In terms of the O2 levels you used in adjusting your mixture, did you install a permanent sensor in your exhaust? One that you could see from your dash all the time? Which one did you end up using - I've been looking for one online but the list is incredibly long...
Much appreciated!
I have the same issue with my disco 1. I've also checked all the measureable values and all checks out. The only issue I have is that I believe the connection between my MAF and the ECU is a bit dodgy. The car runs great when it runs, however, after a long run, it will start to stutter and fail when I'm not accelerating. I believe this is due to the MAF "disconnecting" from the ECU and thus the ECU struggles to control the engine correctly. This inevitable causes the car to die and it's difficult to start as the car has blackened the spark plugs from it running so rich. I will run a new cable through between the two once I get a moment and the plug I've ordered comes. I'm also going to check the fuel system and the ignition system as I believe some of the cabling needs to be replaced.
After all that I would like to mess around with the ECU a bit...
In terms of the O2 levels you used in adjusting your mixture, did you install a permanent sensor in your exhaust? One that you could see from your dash all the time? Which one did you end up using - I've been looking for one online but the list is incredibly long...
Much appreciated!
SALandy said:
That makes sense. I was just wondering in terms of a starting point. Did you only adjust the fueling levels at those load points? You didn't change anything else besides that and the CO pot?
I have the same issue with my disco 1. I've also checked all the measureable values and all checks out. The only issue I have is that I believe the connection between my MAF and the ECU is a bit dodgy. The car runs great when it runs, however, after a long run, it will start to stutter and fail when I'm not accelerating. I believe this is due to the MAF "disconnecting" from the ECU and thus the ECU struggles to control the engine correctly. This inevitable causes the car to die and it's difficult to start as the car has blackened the spark plugs from it running so rich. I will run a new cable through between the two once I get a moment and the plug I've ordered comes. I'm also going to check the fuel system and the ignition system as I believe some of the cabling needs to be replaced.
After all that I would like to mess around with the ECU a bit...
In terms of the O2 levels you used in adjusting your mixture, did you install a permanent sensor in your exhaust? One that you could see from your dash all the time? Which one did you end up using - I've been looking for one online but the list is incredibly long...
Much appreciated!
This vehicle also kept falling over at warm idle. The culprits were the idle air stepper motor and battery voltage. Changing the stepper motor helped but ultimately putting a brand new battery in was the cure. Looks like volts were falling enough that everything got too slow to react. Especially to temp changes (from driving to stopping etc).I have the same issue with my disco 1. I've also checked all the measureable values and all checks out. The only issue I have is that I believe the connection between my MAF and the ECU is a bit dodgy. The car runs great when it runs, however, after a long run, it will start to stutter and fail when I'm not accelerating. I believe this is due to the MAF "disconnecting" from the ECU and thus the ECU struggles to control the engine correctly. This inevitable causes the car to die and it's difficult to start as the car has blackened the spark plugs from it running so rich. I will run a new cable through between the two once I get a moment and the plug I've ordered comes. I'm also going to check the fuel system and the ignition system as I believe some of the cabling needs to be replaced.
After all that I would like to mess around with the ECU a bit...
In terms of the O2 levels you used in adjusting your mixture, did you install a permanent sensor in your exhaust? One that you could see from your dash all the time? Which one did you end up using - I've been looking for one online but the list is incredibly long...
Much appreciated!
I changed the map values wholesale until I got within range and then had an assistant with a laptop and clipboard watch which fuelling cell I was in while driving. My O2 sensor is an Innovate Motorsports wide-band unit and gauge. I have it rigged up as portable and clamp it into the end of the exhaust.
I have no way to adjust fuelling per cylinder or bank, but decided it doesn't matter that much.
Gassing Station | Engines & Drivetrain | Top of Page | What's New | My Stuff