Incorrect Lambda reading - Rovergauge - AFM

[bomb]

QBee - just tried to send you a PM, but PH isn't playing today !

[davep]

QBee said:

Hopefully Blitz will correct me if i am wrong. When I was installing my Clive Y my TVR guy got the firm impression that the white position on the resistor was a position that equated to no resistor, so its purpose was purely to allow you to switch between all the possibilities, including no resistor, without having to remove the resistor. He thus decided that, as he wanted me to run with lambdas, that i didn't need the tune resistor so didn't fit it. hence why mine has no resistor.

It's been said before, but to repeat:

If you do not have a NAS version of the 14CUX code (as in PRC 8747), where the code is locked to Map 5, or have not modified the Tune Resistor code to do the same, the car will need a tune resistor irrespective of whether lambda sensors are fitted or not. The value of the resistor will define whether the EMS will run in open loop or closed loop (when appropriate). When there is no tune resistor fitted to the 14CUX ECU or the resistance value is too high the code will cause Map 0 (Limp Home Map) to be selected.

Edited by davep on Wednesday 29th October 13:44

[bomb]

I have checked my car history file and my tune resistor is at the 'Green' setting which equates to European non-cat tune.

[blitzracing]

So that probably means you are running a 3.9 Land Rover or RangeRover map then. Not a good idea on a 5ltr TVR- as the airflow scaling and fuel requirements are very different.

[bomb]

I dont know which map it is, but when the standard cat was removed and replaced with the Y-piece, it come with the tune resistor. This was fitted and switched to the 'non cat' mode. The car has run well for several years like that. Only recently has the running become upset, and leads me to think the AFM is faulty as it has the very same symptoms as mentioned above.

Is there simple a way to determine which map it has ?? ( and this will steer me towards how it should be set up correctly).

[blitzracing]

You need to use the freeware RoverGauge on a laptop, and an interface cable that plugs into the OBD port in the loom. You can then either copy the map as a binary file and examine it with the freeware tuner pro, or more simply RoverGauge shows you the values in the fuel map in a table, and you then need to compare this to a stock Rangerover map. Ive done most of the hard work for you with these two fuel tables- one is the TVR 5ltr non cat map versus the non modified Landover map- in my books they look much the same. Otherwise simply spend £50 at your local rolling road and do a full power mixture check- Id expect the wrong map to run lean at full power.

5ltr non cat.



Landrover tune




The best man to speak to on this if you want a modified map is Joolz of http://www.kitsandclassics.co.uk/ as it wont cost you an arm and a leg to get a bespoke map.

Edited by blitzracing on Monday 13th October 18:37

[bomb]

Thanks for that. I'll give it a dose of looking at !

[tvr marvin]

Hi again and thanks for your help!
Especially Mark, who was a great help!
In fact it was only a broken AFM. I got a used one now that works.

So here is what we learned:
-if you long term lambda trim values are high or low (beyond 60%) your AFM is probably dead.
-The Chinese copies are crap.
-When you have a cat car, the trim value is ignored (as Mark stated before).

I ordered a new one from Rimmers and it is a Chinese copy and it goes all the way rich in idle.
I measured the trim value and it was 1.9V (instead of the 1.8V for cat cars).
I adjusted the trim value and nothing changed.

Now I want to have something I can monitor the lambda values. Can someone tell me where I can get the plugs for the harness? And what part number they have?

Thanks!

Marvin

[blitzracing]

There is a company called Dalroad Norslo who did them, but they are not really geared up to sell in small quantiles, and that was just the housings, and the connectors themselves are out of stock (for good methinks). You are better off splicing into the black and white wires and putting on your own connector. Is this what you are thinking of?:

[tvr marvin]

I fixed it!
I had the same problem with a new (used original TVR) AFM (too low Lambda trim).
Than I realized that the AFM works fine with the bonnet open. I first thought it is due to the heat.
But later I realized I had worn off the insulation of the cable to the connector. When the bonnet was closed there was a short circuit on the aluminum heat shield under the bonnet.
Insulating the cables with electric tape cured all my problems!

Guys, check your cables. I know the original harness insulation is brittle, so that may cause all sorts of problems.

Have fun!

Marvin

[EGB]

tvr marvin said:

I fixed it!
I had the same problem with a new (used original TVR) AFM (too low Lambda trim).
Than I realized that the AFM works fine with the bonnet open. I first thought it is due to the heat.
But later I realized I had worn off the insulation of the cable to the connector. When the bonnet was closed there was a short circuit on the aluminum heat shield under the bonnet.
Insulating the cables with electric tape cured all my problems!

Guys, check your cables. I know the original harness insulation is brittle, so that may cause all sorts of problems.

Have fun!

Marvin

Interesting informative post this on understanding the conflicts between lambdas, ECU, AFM, Rover Gauge. worn insulation etc.
Just checked the insulation on cable on my off side lambda. Strewth! nearly worn through to the wires, where it's rubbing on a chassis member under the wheel arch. Found it in time and cable tied it away from the rubbing area.
TVR factory did a better job on the nearside.

Edited by EGB on Sunday 10th May 14:04

[tvr marvin]

Hi,
I have to revise my last post. It is not fixed. I figured that is the heat. I wrapped up the AFM in heat reflecting foil and it stayed cooler. That made it read more air and let in more fuel.
However, there are some problems:
A) I cannot really insulate the AFM, after some time it will always getting hot and act up. I don't know how to keep it cool.
B) I am not sure if it is just the case of some old AFM or if they are always like that? Is it possible that they become heat sensitive when aged (happened to my Grandma)?

Finally: Does anyone ever opened one and maybe I can tune it to compensate for that effect? Because hot environment is normal and being a bit rich when cold would be ok.

Marvin

[blitzracing]

You cant work with electronics thats heat sensitive in any way- the AFM is designed not to be temperature sensitive, so if the output is shifting as it gets hot the electronics is on its way out, end of story. How do you know the mixture is shifting?-This would only be visible on green tune if you had an AFR gauge fitted. On white tune the ECU should just go and compensate, unless its so far out you run out of fuel trim.

[tvr marvin]

Hi Mark,
Honestly I am not sure if it is the AFM or not. What I see is that the ECU has a long-term fuel trim of 20% to -35% as long as things stay cool. Whenever I go a bit faster (fast mountain roads etc) the AFM gets really hot and ECU long term trim goes to -100%.
Car runs so lean that it misfires. The misfires I can feel as part-load hesitation.
I suspect the AFM, because whenever I let her idle with open bonnet, the AFM stays cooler and everything works fine. If I drive her after idling for a while, no part-load hesitation.

Also, If I reset the ECU (by switching power to the ECU off and on again) and fuel maps are default, she runs fine and I have no part load hesitation.

I tested that with a second AFM, which seemed to be ok. But after some hot runs the other AFM does the same - Now I don`t know if the AFM is shot also or if my error is somewhere else.
I also tried a Chinese copy of the AFM and it makes the ECU go over rich from the start ( +100% fuel trim).

Could it be something else?
Does anyone have a known to be good AFM somewhere?
What else can I test (I tested the voltages at the AFM, but both tested ok).

What can I do?

Thanks!
Marvin

[blitzracing]

The AFM should be easy enough to diagnose- just take a Rovergauge reading in direct mode at idle before the engine gets too hot, and the same again on a hot engine, and see if there is a significant difference. Its important the RPM is the same for both tests.

[PH430]

I have got the same issue.
On not totaly hot engine the direct reading is 32% on totaly hot it comes down to 29/30%.
The heat affects the output of the AFM. On hot engine somtimes the engine would look to stall, but the ecu keep it alive.

I have got a 20am AFM on stock with a new MA chip. Mark; is this AFM better on output then the stock 5am?

[tvr marvin]

Hi Mark,
Thanks for your help.
Here is what happened. Had 25°C outside today and started her up. After initial warm-up and O2 sensors started switching the ECU immediately went to -100% fuel trim (long term). Even the short term was around -70% to -90% on top of that.
MAF reading was 39% to 40% at idle (850 to 950 RPM)


After running a bit and getting stuck in a traffic jam, reading dropped down to 36% to 37%.
Long term trim recovered to -47% to -64%.


Car ran nice all the time!

Since I have a 123Tune ignition, I reduced the vacuum advance by 2° to 8°@14"Hg. I had almost no part load hesitation.

Is it possible that I am chasing a phantom problem? Maybe everything is ok? I am just worried that if some adjustment values are maxed out, something is not quite right?


Marvin

P.S. Mark, I read your comment on adapting the 20AM to the 14CUX and your problem with the electronics at 130°C. If they worked at 130°C but could not cope with the temp swings between cold start-up and said 130°, there is one solution: Heat them up to 130°C.
A friend of mine develops electronics for a living and he told me that. If you have temperature sensitive electronics, heat them to a value (controlled electric heat circuitry) that ambient temps do not reach. Than they will be stable all the time.

[blitzracing]

PH430 said:

I have got the same issue.
On not totaly hot engine the direct reading is 32% on totaly hot it comes down to 29/30%.
The heat affects the output of the AFM. On hot engine somtimes the engine would look to stall, but the ecu keep it alive.

I have got a 20am AFM on stock with a new MA chip. Mark; is this AFM better on output then the stock 5am?

Depends on your engine size- the 20 AM let more air in as its bigger, that can liberate a few ponies given the correct chip if you are over about 250 BHP, but makes little difference on engines below this. I think its worth about 10 BHP on a stock 5ltr.

[blitzracing]

Marvin- are you definitely running the correct map for the engine? If all the sensor values are good and stable, it would be the next place to look.

[tvr marvin]

Hi,
I didn't change anything and it says map 5. I thought that is correct for a 5-litre?
Car is completely stock except for the pre-cats being removed and the 123ignition.
But I run the normal Rover ignition curve (except for 2° less vacuum advance).





Marvin