[xj-s] Adaptive idle fueling trim ...?

XJ-S
1

Gentlemen:
The text below is taken verbatim from the JDHT for the 6.0L.
I have read and re-read multiple times, but have to admit –
I have no idea what they are trying to say.

Apologies for the ‘dumb’ question, but would someone be able
to provide the explanation in layman terms?—
Adaptive idle fueling trim
In order to ensure optimum performance throughout the life
of the vehicle, the fuel injection ECM software contains an
adaptive idle fueling function that automatically trims the
fuel injector idle pulse duration strategy. The total
available trim to the nominal injector pulse duration is +/-
20%. The function eliminates the manual adjustment of idle
trim. Adaptive fueling is performed by the ECM software only
when there are no diagnostic trouble codes (DTCs) present
and the following preconditions are met:

  1. Throttle closed
  2. Engine speed below 900 rpm
  3. Air injection disabled after engine start
  4. Closed loop fueling control enabled
    If there are no DTCs present and the preconditions are met,
    the ECM cancels purge flow and adapts the idle fueling.
    Between fueling adaptations, there is a delay of approx. 8
    minutes, during which the preconditions must be met. If the
    preconditions are interrupted, the delay will be lengthened.

This is probably another futile attempt on my part to
(over)analyze the manual in an attempt to understand what
the 36CU does in the first few minutes after start-up.

So far I’ve learned that A) it does have 1 minute timer,
after which it switches form open to closed loop (#4 above);
B) stops secondary air-injection when the CTS read 47�C (#3
above).

If my interpretation of the last sentence is correct, this
would mean that approx. a minute after start-up, the ECU
‘adapts the idle fueling’ = cuts of the injector pulse? (if
not DTCs and if idle speed is 900 rpm and the gas pedal is
not touched).

Coincidentally, or perhaps not, 1-3 minutes after start-up
is the timeframe of my idle stumble issues detailed here:

http://forums.jag-lovers.org/av.php?1620512o18

Thanks much in advance for all responses,
Steve

–
'95 XJS V12 6.0L, saphire/tan
–Posted using Jag-lovers JagFORUM [forums.jag-lovers.org]–

// please trim quoted text to context only

2

First I’ve seen this, but I believe I can explain it:

Adaptive idle fueling trim
In order to ensure optimum performance throughout the life
of the vehicle…

Referring to the significant issues developed with the early 3.6
engine. Without any actual fault, normal wear would result in the
engine idling so badly that it’d need an overhaul to run right again.
AJ6 Engineering developed a mod for adjusting the idle trim in those
cars.

the fuel injection ECM software contains an
adaptive idle fueling function that automatically trims the
fuel injector idle pulse duration strategy.

It automagically adjusts the idle mixture for you.

The function eliminates the manual adjustment of idle
trim.

That doofy little knob on the 6CU and 16CU ECU goes away.

Adaptive fueling is performed by the ECM software only
when there are no diagnostic trouble codes (DTCs) present…

Nothing’s broken.

…and the following preconditions are met:

  1. Throttle closed

It’s gotta be at idle.

  1. Engine speed below 900 rpm

It’s gotta be at idle, not coasting down a mountain.

  1. Air injection disabled after engine start

It’s gotta be fully warmed up, not running on the cold start map.

  1. Closed loop fueling control enabled

The oxygen sensors must be working.

If there are no DTCs present

Nothing’s broken.

and the preconditions are met,
the ECM cancels purge flow and adapts the idle fueling.

It’s gotta shut off the purge flow through the carbon canister
because that would richen the mixture a bit and confuse the
adjustment process. It’d be interesting to know HOW it shuts off the
purge flow! Is there a solenoid valve in the purge line?

Between fueling adaptations, there is a delay of approx. 8
minutes, during which the preconditions must be met. If the
preconditions are interrupted, the delay will be lengthened.

The whole reason you don’t go into closed loop at idle is because
you’d get an unstable idle as the mixture cycles up and down. So,
what this system does is adjust the mixture only once every 8
minutes! When it happens, you’d probably get a slight shift in RPM,
but who would notice a slight shift every 8 minutes? And, as it
says, if any of those other things get tripped up, it gets LONGER
than 8 minutes!

– Kirbert

// please trim quoted text to context onlyOn 18 Sep 2013 at 12:11, sbobev wrote:

3

In reply to a message from Kirbert sent Wed 18 Sep 2013:

''The whole reason you don’t go into closed loop at idle is because

you’d get an unstable idle as the mixture cycles up and down.’’

But, isn’t it in closed loop by the precondition statement 4? But
as you say, the 8 minute time is like a big damper to keep it from
hunting with the minute mixture changes.

The idle trimmer pot is still in the 26 CU, but the method you
mention in ‘‘The Book’’ to adjust it cannot be used. There is no
feedback monitor socket, and the particular wires going to pins 4
and 10 of the ECU are no longer voltage values, but carry serial
data to the ‘‘JDS Diagnostic Plug’’. So unless one has a JDS
diagnostic gizmo or similar, there is no way of knowing what
turning the pot is doing unless you have good ears for idle quality.–
lockheed 92 XJS Cpe/97 LT1 Miami FL/ 96 XJS Cv 4.0 Austin TX
Austin, TX, United States
–Posted using Jag-lovers JagFORUM [forums.jag-lovers.org]–
–Support Jag-lovers - Donate at http://www.jag-lovers.org/donate04.php –

// please trim quoted text to context only

4

In reply to a message from Kirbert sent Wed 18 Sep 2013:

‘‘Is there a solenoid valve in the purge line?’’

There are two parallel solenoid purge valves (one for A Bank and
one for B Bank) - operated by the ECU - next to the canister in the
area forward of the left front wheel well.–
lockheed 92 XJS Cpe/97 LT1 Miami FL/ 96 XJS Cv 4.0 Austin TX
Austin, TX, United States
–Posted using Jag-lovers JagFORUM [forums.jag-lovers.org]–
–Support Jag-lovers - Donate at http://www.jag-lovers.org/donate04.php –

// please trim quoted text to context only

5

In reply to a message from Kirbert sent Wed 18 Sep 2013:

Thanks Kirby, I was almost sure you’ll respond.
I understand the preconditions, but what continues to confuse
me (and perhaps you too) is the closed/open loop operation. It
seems that you think the ECU is NOT in closed loop at idle,
but the fact is – the 36CU is always in closed loop 1 minute
after start-up.

The engine idles with O2 sensors feedback, which could be the
reason why almost all 6.0L are running relatively poorer (at
750 rpm) compared to the 5.3L.

My recent test confirms it – removing the O2 sensors forces
the ECU to use the pre-programmed maps and triggers a DTC.
The fueling is not trimmed/adapted and the engine idles and
runs very well. A bit richer and dirtier, but smoother all the
way from cold to normal operating temp.

Steve–
The original message included these comments:

First I’ve seen this, but I believe I can explain it:
The whole reason you don’t go into closed loop at idle is because
you’d get an unstable idle as the mixture cycles up and down. So,
what this system does is adjust the mixture only once every 8
minutes! When it happens, you’d probably get a slight shift in RPM,
but who would notice a slight shift every 8 minutes? And, as it
says, if any of those other things get tripped up, it gets LONGER
than 8 minutes!

–
'95 XJS V12 6.0L, saphire/tan
–Posted using Jag-lovers JagFORUM [forums.jag-lovers.org]–
–Support Jag-lovers - Donate at http://www.jag-lovers.org/donate04.php –

// please trim quoted text to context only

6

In reply to a message from lockheed sent Wed 18 Sep 2013:

Thanks Alan.
I believe this is the reason why the idle speed increases
slightly when there is a DTC flashed – the 36CU does not shut
those valves and the engine gets some more air through the
carbon canister vent.

I don’t know how many times I’ve read the manual, but every
time I go back to it, I discover something new or something
I’ve missed before.

Steve–
The original message included these comments:

‘‘Is there a solenoid valve in the purge line?’’
There are two parallel solenoid purge valves (one for A Bank and
one for B Bank) - operated by the ECU - next to the canister in the
area forward of the left front wheel well.

–
'95 XJS V12 6.0L, saphire/tan
–Posted using Jag-lovers JagFORUM [forums.jag-lovers.org]–
–Support Jag-lovers - Donate at http://www.jag-lovers.org/donate04.php –

// please trim quoted text to context only

7

But, isn’t it in closed loop by the precondition statement 4?

No. Closed loop is enabled, but the system only actually operates in
closed loop above idle. At least, that’s what I’d gather from
this. It makes sense because closed loop didn’t operate at idle in
earlier cars, and if you forced it into closed loop (unplugging a
loop connector in the trunk) you could get an idle that customers
would take the car back to the dealer and gripe about. The advent of
heated oxygen sensors helped matters, but apparently not enough to
enable full-time closed loop at idle.

The idle trimmer pot is still in the 26 CU, but the method you
mention in ‘‘The Book’’ to adjust it cannot be used.

Great! Since the description said the automatic adjustment is
limited to +/- 20%, perhaps you can adjust this thing when it’s more
than 20% off. Move it back to a point where the automatic adjuster
stays within range, so to speak. Just guessing.

There is no
feedback monitor socket, and the particular wires going to pins 4 and
10 of the ECU are no longer voltage values, but carry serial data to
the ‘‘JDS Diagnostic Plug’’.

Yeah, but you could figure it out. Heck, you could forget that plug
and read the voltages from the oxygen sensors directly; just adjust
the mixture so that one is flickering and the other is indicating a
bit rich.

– Kirbert

// please trim quoted text to context onlyOn 18 Sep 2013 at 17:10, lockheed wrote:

8

There are two parallel solenoid purge valves (one for A Bank and one
for B Bank) - operated by the ECU - next to the canister in the area
forward of the left front wheel well.

Well, there ya go. Ironically, there were purge valves on the
earlier cars too, installed by a recall – but they weren’t operated
by the ECU. Rather, they were vacuum-operated, controlled by
tappings on the LH throttle housing.

– Kirbert

// please trim quoted text to context onlyOn 18 Sep 2013 at 17:37, lockheed wrote:

9

I believe this is the reason why the idle speed increases
slightly when there is a DTC flashed – the 36CU does not shut
those valves and the engine gets some more air through the
carbon canister vent.

Well, yeah! Good point! But are we saying that the purge valves
stay closed for the entire 8 minutes unless one of the criteria falls
out? Or do the purge valves only remain shut briefly while the
system adjusts the idle mixture?

It’d be kinda interesting to check this feature. Somebody hook up a
light bulb across one of those purge valves and watch how they go on
and off, just during normal operation and when you unplug an O2
sensor.

– Kirbert

// please trim quoted text to context onlyOn 18 Sep 2013 at 18:27, sbobev wrote:

10

– the 36CU is always in closed loop 1 minute
after start-up.

Well, if you think about it, this idle mixture adjustment is a kind
of closed-loop operation. It is clearly looking at the O2 sensors,
which means it makes sense that it’d flag a DTC when they get
disconnected. As compared to the running-speed closed loop, though,
this “adaptive” thing apparently A) has a MUCH slower refresh rate, 8
minutes vs. what, several times a second? and B) probably has a
memory that keeps the last setting even after the car is shut off.

That does make it seem odd, though, that upon unplugging the O2
sensors the car would IMMEDIATELY go into a richer running mode. I
would have expected it to stay at the setting it had last. But, now
that I think about it, that might not be good from the designers’
standpoint, as that O2 sensor might remain unplugged (or defective) a
loooong time, and the last adjustment might not be appropriate over
the long haul. So perhaps they decided to immediately move to a
failsafe operation, a little bit rich so nobody gets hurt.

– Kirbert

// please trim quoted text to context onlyOn 18 Sep 2013 at 18:13, sbobev wrote:

11

In reply to a message from Kirbert sent Wed 18 Sep 2013:

The ECU works a bit differently, I think. It flags the DTC for
disconnected CTS immediately, but the DTC for O2 sensor is set
after a minute, when it tries to switch to closed loop.

Once this happens, the ECU is in ‘limp-mode’. No feedback.

But, let’s assume all #1 through #4 are met. All is normal.
Does the 8 minute delay mean the car has to stay in the
driveway in Park for the adaptive fueling trim to happen?

Or the 8 minute delay means that the car starts ‘from memory’
and can be driven, and after 8 minutes, if I happen to be
standing at a red light, #1 through #4 are met again and the
ECU reprograms itself?

I am sorry to be asking so many questions, but not
understanding the operation per the manual, while trying to
address idle-related issues bothers me quite a lot.

Steve–
The original message included these comments:

– the 36CU is always in closed loop 1 minute
after start-up.
Well, if you think about it, this idle mixture adjustment is a kind
of closed-loop operation. It is clearly looking at the O2 sensors,
which means it makes sense that it’d flag a DTC when they get
disconnected. As compared to the running-speed closed loop, though,
this ‘‘adaptive’’ thing apparently A) has a MUCH slower refresh rate, 8
minutes vs. what, several times a second? and B) probably has a
memory that keeps the last setting even after the car is shut off.

–
'95 XJS V12 6.0L, saphire/tan
–Posted using Jag-lovers JagFORUM [forums.jag-lovers.org]–
–Support Jag-lovers - Donate at http://www.jag-lovers.org/donate04.php –

// please trim quoted text to context only

12

In reply to a message from sbobev sent Thu 19 Sep 2013:

You’re now way past my knowledge. When I get home I’ll
start studying the Manual…–
John. '95 XJS 6.0L convertible. Southlake, TX
–Posted using Jag-lovers JagFORUM [forums.jag-lovers.org]–
–Support Jag-lovers - Donate at http://www.jag-lovers.org/donate04.php –

// please trim quoted text to context only

13

In reply to a message from sbobev sent Thu 19 Sep 2013:

I think that the label ‘‘Adaptive idle fuel trim’’ is a bit
misleading. In my view these adaptations are not made to
control idle speed, that is done by the ECU idle strategy. I
think the AIFT is just designed to trim long term fluctuations
due to wear and to adjust between cars for small differences
in things like fuel injector flow etc. I think all it does is
change the baseline for the ECU idling strategy, a bit like a
gain function. It does not alter the strategy per se.

Once the car has warmed up, adaptive fueling will take place
providing the preconditions are met. If you were to wait at
idle for a further 8 minutes the system will go through the
routine again and you will hear the loud click from the purge
valves opening and closing.

If you just get in the car, crank the engine, slip into drive
and take off as one would normally do, no adaptations would
take place.

Long and short, I don’t think that the AIFT is a likely
candidate for your idle woes.

Also, I do not believe that the 6.0 has a 45 sec timer for the
hot start as per the 5.3 The 6.0 used the ECU to control hot
starts instead.

Cheers,

Allan–
1994 XJS 4.0 2+2 Convertible 5 Speed Manual
–Posted using Jag-lovers JagFORUM [forums.jag-lovers.org]–
–Support Jag-lovers - Donate at http://www.jag-lovers.org/donate04.php –

// please trim quoted text to context only

14

But, let’s assume all #1 through #4 are met. All is normal.
Does the 8 minute delay mean the car has to stay in the
driveway in Park for the adaptive fueling trim to happen?

Yeah, I wondered about that, too. I certainly hope not! It would
make more sense if you sat in the driveway for one minute, and then
at a stoplight for one minute, and then several more such stops, and
when the total time at idle added up to 8 minutes, it’d go ahead and
trim the idle fuelling. If it had to wait until you’re stuck in a
traffic jam, I’m not sure it’d EVER trim!

OTOH, it would be a neat trick. You bring your Jag into the dealer
and BMC about a lousy idle. Fill out the paperwork, run off and do
something else for four hours. The mech starts the car and lets it
idle in the parking lot for 10 minutes. You come back, he charges
you $350 for an idle trim adjustment, and you drive away happy
because your car idles great again!

– Kirbert

// please trim quoted text to context onlyOn 19 Sep 2013 at 19:20, sbobev wrote:

15

In reply to a message from Kirbert sent Fri 20 Sep 2013:

Correct, the adaptation routine interval is accumulative. I
expect that the system uses a simple counter that is
start/stopped as preconditions warrant. When the counter gets
to a certain point and pre-conditions exist, adaptation takes
place. After which the counter is reset and the process
repeated.

Again the purpose here is not to control idle speed, merely to
provide offset to the idle strategy baseline with the
expectation that in normal circumstances changes would be few
and far between (i.e. over the life of the vehicle)

Also all that has been mentioned are the conditions under
which the adaptation routine runs, NOT what it looks for to
make a change. I expect that certain parameters would have to
be out of range in order for the routine to actually make a
change in baseline offset. For instance for a certain engine
speed or load, perhaps TPS voltage is above x volts or below y
volts. It is probably a reasonably wide range as the idle is
actually controlled by the ECU varying injector pulse
duration. Obviously there is a practical range that the ECU
can accommodate, hence the need for an occasional offset.

Cheers,

Allan–
The original message included these comments:

Yeah, I wondered about that, too. I certainly hope not! It would
make more sense if you sat in the driveway for one minute, and then
at a stoplight for one minute, and then several more such stops, and
when the total time at idle added up to 8 minutes, it’d go ahead and
trim the idle fuelling. If it had to wait until you’re stuck in a
traffic jam, I’m not sure it’d EVER trim!

–
1994 XJS 4.0 2+2 Convertible 5 Speed Manual
–Posted using Jag-lovers JagFORUM [forums.jag-lovers.org]–
–Support Jag-lovers - Donate at http://www.jag-lovers.org/donate04.php –

// please trim quoted text to context only

16

In reply to a message from AllanG sent Fri 20 Sep 2013:

Hi Allan:

No one is saying the adaptive fueling trim controls idle
speed. My idle speed is close to 750 rpm at normal operating
conditions. I do see small ‘oscillations’, which to my hear
appear periodic with a repeat period of about 2-3 seconds.
Idle drops/increases by 25-50 rpm.

Mr. Bywater advised that this is normal and exactly what the
car should do (O2 sensor feedback takes some time at idle).
I’ve (reluctantly) accepted it for what it is, but I am
still trying to figure out what the ECU does 1-3 minutes
after start-up. During that period, I have some complaints.
One of the hypothesis I’ve got is that the fueling is
incorrect for some reason.

With DTC FF44, adaptive idle fueling trim does not happen,
engine warms up fine in open-loop. The closed-loop
operation, while the engine has not reached normal operating
temperature, is clearly the problem as removing O2 sensors
feedback appears to ‘cure’ the stumble.

The question is – 3 sets of O2-sensors cannot be bad (the
chances are astronomical) – what else could be messing the
fueling only 1-3 minutes after start-up. CTS, CPS, TPS,
idle switch are all within specs. Vacuum is good, fuel
pressure is fine.

I brought the ‘Adaptive idle fueling trim’ topic because it
occurred to me that since I’ve had the car running in the
garage for some time (certainly longer than 8 minutes at
times), while conducting various tests, it might have
‘reset’ the baseline.

But, if indeed it works as you describe it, likely nothing
has happened.

BTW, will removing the battery clear it from the memory,
like the DTCs are done with?

Thanks much for chiming in.

Steve–
The original message included these comments:

Correct, the adaptation routine interval is accumulative. I
expect that the system uses a simple counter that is
start/stopped as preconditions warrant. When the counter gets
to a certain point and pre-conditions exist, adaptation takes
place. After which the counter is reset and the process
repeated.
Again the purpose here is not to control idle speed, merely to
provide offset to the idle strategy baseline with the
expectation that in normal circumstances changes would be few
and far between (i.e. over the life of the vehicle)
Also all that has been mentioned are the conditions under

–
'95 XJS V12 6.0L, saphire/tan
–Posted using Jag-lovers JagFORUM [forums.jag-lovers.org]–
–Support Jag-lovers - Donate at http://www.jag-lovers.org/donate04.php –

// please trim quoted text to context only

17

In reply to a message from Kirbert sent Fri 20 Sep 2013:

Could Jaguar be so evil?

I hope the 36CU system was not intended to increase the
frequency of trips to the dealership… :-))

Last night I read something else in the JDHT manual, which
you may know, but I saw/paid attention to/ it for the first
time.

The 36CU watches closely the input from the CTS after start-
up from cold. Not only because it needs it to compute the
fueling but also because if the CTS signal doesn’t reach
values corresponding to 60�C (or so) within ca. 18000 engine
revolutions, it flags a code.

The explanation was provided (surprisingly) just below:

  1. To avoid wrong fueling in case of mulfunctioning CTS
  2. To give an early warning that the thermostat is likely
    gone and needs replacement

How cool is that?

Steve–
The original message included these comments:

OTOH, it would be a neat trick. You bring your Jag into the dealer
and BMC about a lousy idle. Fill out the paperwork, run off and do
something else for four hours. The mech starts the car and lets it
idle in the parking lot for 10 minutes. You come back, he charges
you $350 for an idle trim adjustment, and you drive away happy
because your car idles great again!

–
'95 XJS V12 6.0L, saphire/tan
–Posted using Jag-lovers JagFORUM [forums.jag-lovers.org]–
–Support Jag-lovers - Donate at http://www.jag-lovers.org/donate04.php –

// please trim quoted text to context only

18

The 36CU watches closely the input from the CTS after start-
up from cold. Not only because it needs it to compute the
fueling but also because if the CTS signal doesn’t reach
values corresponding to 60°C (or so) within ca. 18000 engine
revolutions, it flags a code.

Interesting! I wonder what happens when you start the car in, like,
sub-zero weather?

– Kirbert

// please trim quoted text to context onlyOn 20 Sep 2013 at 7:13, sbobev wrote:

19

In reply to a message from Kirbert sent Fri 20 Sep 2013:

They thought about that too – the 6.0L is fitted with a
coolant heater in the lower radiator hose. Standard cold
weather package. Pictures in my album.
Mine is a convertible. Go figure… :-))

BTW, 18000 revolutions is 15-20 min idling in Park or 7-10
min driving (the way I’d do it from cold, trying to not go
over 2000 rpm). Not unrealistic target, I think.

When I have the time, I will also post some interesting
reading about battery voltage and how the ECU handles it
during start-up, also found in the JDHT ROM. They provide
scanned images, not text, so copy and paste doesn’t work –
I have to re-type everything.

Steve–
The original message included these comments:

The 36CU watches closely the input from the CTS after start-
up from cold. Not only because it needs it to compute the
fueling but also because if the CTS signal doesn’t reach
values corresponding to 60�C (or so) within ca. 18000 engine
revolutions, it flags a code.
Interesting! I wonder what happens when you start the car in, like,
sub-zero weather?

–
'95 XJS V12 6.0L, saphire/tan
–Posted using Jag-lovers JagFORUM [forums.jag-lovers.org]–
–Support Jag-lovers - Donate at http://www.jag-lovers.org/donate04.php –

// please trim quoted text to context only

20

BTW, 18000 revolutions is 15-20 min idling in Park or 7-10
min driving…

Hmmmm. Just long enough to make owners wonder why the hell the car
runs fine for a few minutes and then starts flagging faults?

– Kirbert

// please trim quoted text to context onlyOn 20 Sep 2013 at 12:11, sbobev wrote: