Instructions to change fuel maps on 14CUX Griffith, Chimaera

[FlipFlopGriff]

robertf03 said:

Was there ever a 4.2 rover offered in the UK?

Wedge 420SE and SEAC.
FFG

[stevesprint]

FlipFlopGriff said:

Wedge 420SE and SEAC.
FFG

Thanks Paul,
Although most Wedge's run the old flapper system I believe a small number near the end of production have the 14CUX system. It would be interesting to see a 420SE Prom as TVR quoted 300bhp.

[stevesprint]

davep said:

Steve, early programming convention usually required data definitions (DB - data byte and DW - data word) to be declared at the beginning of the code, and the fuel maps are simply data definitions. As to why 14CUX's individual non-map related data definitions suddenly appear immediately after Map 0's ADC table and not Map 5, we'll probably never know, but there would have been a logical reason for doing so (put all the important stuff at the beginning maybe)

Thanks Dave,
I expected to see the data before the code, but I’m just surprised to see the non map related data slap bang in the middle of map 0 between the main fuel table and the throttle pot direction table. Also Map 0's ADC table is in a different place compared to the other maps.
It would have been logical if map 0 had the same data structure as the other maps with the non-map related data grouped together after map 0 or 5. I guess the data layout was dictated as the software evolved.
The main thing is we now know the layout and are very grateful for Colin & Dan's years of hard work.

Edited by stevesprint on Saturday 1st March 16:14

[cmb]

I just released RoverGauge 0.6.0, which has a couple fixes and a couple new features. Here are the main points:

• Added display of MAF CO trim (only when open loop)
• Added display for tune ident byte, as well as checksum fixer byte
• Switched to "odd" / "even" terminology when referring to cylinder banks as this is unambiguous
• Fixed minor bug with fuel pump run request
• Fixed fault code descriptions for bank-specific faults

Because Google changed their file storage policies, the downloads have been moved to a Google Drive folder:
https://drive.google.com/folderview?id=0B83FOZ5t1n...

Enjoy. Let me know if you find any problems.

[davep]

cmb said:

I just released RoverGauge 0.6.0, which has a couple fixes and a couple new features. Here are the main points:

• Added display of MAF CO trim (only when open loop)
• Added display for tune ident byte, as well as checksum fixer byte
• Switched to "odd" / "even" terminology when referring to cylinder banks as this is unambiguous
• Fixed minor bug with fuel pump run request
• Fixed fault code descriptions for bank-specific faults

Because Google changed their file storage policies, the downloads have been moved to a Google Drive folder:
https://drive.google.com/folderview?id=0B83FOZ5t1n...

Enjoy. Let me know if you find any problems.

Yet more useful features, especially the open loop function! The RG screen interface is becoming really comprehensive - thanks Colin.

[blitzracing]

cmb said:

I just released RoverGauge 0.6.0, which has a couple fixes and a couple new features. Here are the main points:

• Added display of MAF CO trim (only when open loop)
• Added display for tune ident byte, as well as checksum fixer byte
• Switched to "odd" / "even" terminology when referring to cylinder banks as this is unambiguous
• Fixed minor bug with fuel pump run request
• Fixed fault code descriptions for bank-specific faults

Because Google changed their file storage policies, the downloads have been moved to a Google Drive folder:
https://drive.google.com/folderview?id=0B83FOZ5t1n...

Enjoy. Let me know if you find any problems.

Does the MAF CO trim only appear when its on a non catalyst map and running? I dont see it on the front screen, but Im not plugged into the ECU at the moment....

Also something Ive yet to try as its not something I cant read on the road at speed- What happens to the short term trim values when you hit open loop on a closed loop map- do they just zero ?

[stevesprint]

cmb said:

I just released RoverGauge 0.6.0, which has a couple fixes and a couple new features. Here are the main points:

• Added display of MAF CO trim (only when open loop)
• Added display for tune ident byte, as well as checksum fixer byte
• Switched to "odd" / "even" terminology when referring to cylinder banks as this is unambiguous
• Fixed minor bug with fuel pump run request
• Fixed fault code descriptions for bank-specific faults

Colin,
Thanks for another round of updates and new features, especially from all us open loopers. I’m sure it will help us understand the mystery behind the CO trim. I personally will also find the addition of the ident bytes and checksum fixer very useful.
Thanks again
Steve

[danbourassa]

blitzracing said:

Does the MAF CO trim only appear when its on a non catalyst map and running? I dont see it on the front screen, but Im not plugged into the ECU at the moment....

Also something Ive yet to try as its not something I cant read on the road at speed- What happens to the short term trim values when you hit open loop on a closed loop map- do they just zero ?

Mark, the screen defaults to closed loop but, if an open loop map is detected, the trim bars will disappear, the short/long term radio buttons will remain but be greyed out and the CO trim voltage will appear.

When a closed loop map goes open, $8000 is written to the short term trim values which is the zero point.

[robertf03]

I did some datalogging on the 4.6 with the modified fuel map on Saturday. The vacuum gauge and the load row were surprisingly close. I wish I had performed this check prior to the engine swap.

I mentioned before I had a problem with the 3.9 running rough and rich to the point where I had to pull the fuel pump fuse and crank the engine over to clear it out while on Imogene Pass. The problem seems to have followed the 4.6. With the wideband at the tailpipe (removed catalysts) I was seeing AFRs in the 12.5 to 14.7 range after the inital startup enrichment, then rovergage showed the lambdas climb and eventaully max out to the right for fuel correction. I eventually got an o2 sensor fault while driving around.

Fuel pressure checks out and the injectors checked out, though I'm wondering if its an intermittent pump that only functions correctly when observed.

I've heard of testing zirconia o2 sensors with a torch and voltmeter, but I don't think that would apply to a titanium o2 sensor. These are new sensors in the past 6 months. Any ideas on how to bench test these?

Other than the faults, the modified 3360 chip without any MAF scalar modifications seems to be running very close to acceptable. The only major problems other than overfueling and the o2 faults are it will bog down at idle and stall, just as the 3.9 would at altitude. CO trim is at 1.848 volts per the NAS spec. I'm stumped. I've done everything I can to eliminate the possibility of exhaust leaks. copper coating on fresh gaskets, ground and rewelded the flanges.

We had a freeze and its been too cold to work on it, but I'll try again this weekend.


logs are at http://www.flemcodesign.com/logs.zip if you are interested

Edited by robertf03 on Wednesday 5th March 05:39

[blitzracing]

The probes are easy to test- first these are resitive probes, so need a 12 volt feed to get an output. This is taken from the heater supply on the red wire to ground- so check you have 12 volts with the engine running as this is a switched feed (like the pump). You can simply measure the output of the probes with a test meter and the engine running (warm engine). Test between the black (+ve ) and white (ground). A good probe will cycle between 0 and about 1.2 volts if everything is working. If its stuck at zero volts, clamp the fuel return pipe from the regulator with plyers to increase the fuel pressure and richen the mixture- the probe should jump to 1.2 volts or so. You have a dead probe if this does not happen. Likewise if its stuck at 1.2 or more volts, then let some air into the plenum by removing one of the bungs or servo pipe, to weaken the mixture, and it should drop to 0 volts. If the ECU has registered a fault, the lambda voltages must have stopped switching for some period (as in minutes), as it does not go into fault that easily. A good read here as well:

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

In terms of rich running, a couple of culprets could be an overvoltage AFM,(Check the RG readings- should be around 32-35 % direct reading at idle) or fuel pressure not being controlled by plenum vaccuum- it will ready 37 PSI with no vacuum, but should be a lot lower at idle with the engine running.

Edited by blitzracing on Wednesday 5th March 13:01

[danbourassa]

robertf03 said:

logs are at http://www.flemcodesign.com/logs.zip if you are interested

What do you think happened here? (see MainVoltageGlitch.png)

https://drive.google.com/?tab=wo&authuser=0&am...

Edited by danbourassa on Thursday 6th March 00:00

[robertf03]

You know, the more I think about it the more I don't remember reinstalling the large bundle of ground wires that attaches to the left rear cylinder head. Maybe one of my helpers did. I'll have to look at that when I get home.

Hopefully this is all related and an easy fix

[spitfire4v8]

Dan .. how do you import and view a data log session made by someone else in rovergauge?

[danbourassa]

Robert,
I graphed the Lambda trims from one of the log files and noticed that both banks have a severe positive bias. They can be found in the same place as before:

https://drive.google.com/?tab=wo&authuser=0#fo...

This would show the RG bar graphs predominantly to the right. Is this what you saw? It would also increase fuel.

There is another graph called HO2S_and_Reference. This is actual data taken during a road test. It shows the small amount of voltage swing from the sensors. A high level is only about 55 counts (where 255 counts equals 5 Volts) which equates to about 1.1 Volts. The graph also shows a reference voltage which is around 23 counts or about 0.45 Volts. Software compares the Lambda reading with this reference to determine if it's high or low. The point of all this is to show that we are talking about only tenths of a Volt here. Land Rover went to the trouble of even shielding the wires that run to the sensors.

I'm not sure about this but I suspect that the reference voltage is actually a remote ground (heater return) sense on the far end of the wiring harness at the Lambda connectors. It may depend solely on the heavy current return to slightly raise the voltage from zero. If this is the case, it seems that it would be easy to upset this voltage and put a bias on the readings. Unfortunately, RG doesn't display this reference (another upgrade, Colin?). Make sure the heater coils are not burned out and that they are getting power. The fuel pump relay supplies this. There are two terminals marked 87. White/purple goes to the fuel pump. White/orange (2 wires) go to the Lambda sensors. Using a yellow relay here could cause problems.

[danbourassa]

spitfire4v8 said:

Dan .. how do you import and view a data log session made by someone else in rovergauge?

I use Excel. I open Excel first and then File/Open. That way it asks about the format. You want delimited data (not fixed width). Then select 'comma'.

[spitfire4v8]

^^ thanks Dan .. not done any data logging yet so will do as you do

[robertf03]

danbourassa said:

yellow relay here could cause problems.

Ha, guess what color my fuel pump relay is!

[danbourassa]

robertf03 said:

Ha, guess what color my fuel pump relay is!

Chartreuse?

[robertf03]

no, I thought it was yellow but I confused it with another one. It is the correct silver color

I did verify the ground wires for the o2 heaters and shield were not hooked up. Oops. I'll try again this weekend.

[robertf03]

Much better. O2 is switching nicely.

A little fine tuning and I think this one will be good enough.

http://www.flemcodesign.com/logs1.zip