[DML] Re: Volvo F Series CO Values

["endotex23" <endotex23@xxxxxxxxx>]

Hello Martin. Sorry for the delay and the length of this reply. I
realize you're probably familiar with much of this but others may be
interested in the details.

You wrote: 

>FYI under warm-up conditions, lambda sensor disconnected, and,
>>crucially, *when the full throttle microswitch is pressed* all have
>>exactly the same effect, electrically speaking. This assumes your
>full throttle switch is working.< 

I wrote I didn't follow you, so you clarified:

>My point is a simple one. If you want to play with the lambda system
in open-loop, it's much easier to just press the full throttle switch
than to disconnect the lambda sensor.<

Technically that is not correct, it's NOT electrically the same. As I
said before, the duty cycle with a disconnected O2 sensor and the one
with the WOT switch closed are different. They are not far apart, yes,
but different enough (it can be as much as 20%) to matter if you want
to set things up right. Measure it for yourself and see. It's also
plainly stated in the service manual. You can get away with doing it
your way but it's sloppy.

Then you wrote:

> Attempting to adjust this makes the lambda system "fight" you, and
it only changes once you reach 100% or 0% duty cycle on the frequency
valve's PWM.<

Your point is valid but but does it mean it's OK to set mixture in
closed loop, ie; with the sensor connected? No. Rather than bore the
readers with a long treatise I'll simply say again that doing so will
cause you to end up shifting the loop's output range to a point where
it's ability to make full scale corrections at other times like during
normal driving conditions, fuel component drift with age, ect, is
inhibited. It's basic control theory and doesn't apply only to
automotive applications. I'll be glad to expand on this if asked.

You CAN adjust mixture with the loop closed as long as you monitor
dwell. But (as you correctly stated) the CO will stay the same (you
erroneously claim 1% but we'll get to that) the FV duty cycle will
move around and you'll end up with the loop shifted if you also don't
pay attention to the dwell meter. Put your analyzer on the car,
measure the dwell (or scope the FV signal), adjust mixture, and you'll
see what I mean. But don't take my word for it, the service manual is
pretty clear. Which brings us to...

Martin wrote:

> The DeLorean manual does not mention the exhaust taps, nor any
instructions for using them, but they are there.<

The Delorean manual does indeed mention them, under Section B1 "Lambda
System Operating Test". (It's on page 188 in my manual. (Btw, how did
you originally learn about the plugs?). And it brings us back to why
the CO should NOT be 1% with the loop closed. If I can't convince you
based on the known design properties of O2 sensor based mixture trim
systems then at least take the word of the service manual. 

Under that procedure refer to Step 2:  "Remove both exhaust pipe
plugs. Plugs are located in left and right pipes at exhaust
manifold". 

Step 5: "Disconnect oxygen sensor".

Step 7: "CO should be 1% @ +/- .3% curb idle speed". 

Note that "should" really means "adjust it to that if it isn't". The
procedure says "should" because this is, after all, a test procedure
and not an adjustment procedure.

After the other steps (in which you check the controller's response to
a zero and full scale sensor signal using a battery and further test
the WOT switch input), take particular note of Step 12: "Reconnect the
Lambda sensor" and Step 13: "CO should drop below 1%". The fact alone
that the dwell changes between open loop and closed loop tells us the
CO level MUST change. It can't remain at 1% because the duty cycle
will change when you connect the sensor. (This assumes you adjusted it
to 1% in open loop).

Also note the final Step 16: "Remove exhaust gas probes and install
exhaust pipe plugs".

Read the procedure again. Since it clearly states to measure the
exhaust stream *at the pipe plugs* (which are upstream of the catalyst
if one is installed) can you explain how this is any different than a
car without catalyst? Such as as the ones you work with? 

Even ignoring the manual's confirmation, fuel trim systems for
emission control are designed to maintain a stohciometric setpoint
because the O2 sensor is built to output around 450 millivolts at a
Lambda of 1 and the rest of the system is calibrated around that
point. Since CO and O2 are inversely proportional it is mathematically
impossible to have a Lambda of 1 with a CO level of 1%. 

And since the O2 sensor and Lambda loop is calibrated to control at a
setpoint of Lambda 1 the output of the sensor will be shifted rich by
a 1% CO mixture and a properly functioning loop wll act to correct
that deviation. Hence a closed emission loop (with or without catalyst
using any standard or wide band O2 sensor) that remains near 1% CO is
either: A) Itself not operating correctly or B) Is trying to
compensate for an external problem in the fuel system outside of it's
operating range.

If CO stays at 1% closed loop you either have an engine problem or
your analyzer may simply need to be calibrated. (I don't know what you
have but if it's not NDIR based and if you're not using cal gas on a
regular basis it's readings are little more than a guess). As per the
service manual, a leak in the exhaust system or unmetered air entering
the engine are two possible causes of CO levels that don't drop under
closed loop. Or it may simply be as you said, the mixture is
misadjusted to the point the duty cycle can no longer correct for it.
The bottom line is CO should NOT remain at 1% when the loop is closed,
catalyst or not. Both the manual and anyone with a knowledge of Lambda
systems will tell you this. Nor should you adjust (but it's ok to
check) the mixture with the O2 sensor connected for the reasons stated
earlier. 

I know the Deloreans I've work on (when set up properly) don't stay at
1% CO in closed loop when measured before the catalyst. Nor do other
vehicles so equipped. Less than one half of one percent is a typical
value for K Jet/Lambda equipped engines and it's only that high
because the system is such a lousy one compared to injection duration
based EFI. A well operating system will control to much less. Fwiw,
measuring "engine out" emissions (rather than tailpipe emissions) is a
common diagnostic technique in the smog business because a catalyst
can mask engine problems. It's why test pipes are made. Did you really
think the primary reason for availabilty of test pipes is so people
can bypass their catalyst for other reasons? ;)

Note that I'm assuming thoughout all this that other factors on a
Delorean effecting mixture after warm up (primary regulated fuel
pressure and control pressure for example) are working correctly.
Otherwise they too would cause a shift in the loop's output.

By the way, 1% CO may be "healthy" (more power) for the engine but
it's not healthy in any other way. Inefficient combustion is not only
more polluting in a direct sense but by reducing fuel economy it
increases refinery and a myriad of other emissions generated by
increased oil exploration and harvesting. Not to mention the costs
involved. Over millions of vehicles it all adds up. I'll be the first
to admit the Delorean is somewhat gutless in it's stock configuration
but enriching (with the associated increase in CO) *alone* doesn't get
you very much. 

Greg 








To address comments privately to the moderating team, please address:
moderators@xxxxxxxxxxx

For more info on the list, tech articles, cars for sale see www.dmcnews.com

To search the archives or view files, log in at http://groups.yahoo.com/group/dmcnews 
Yahoo! Groups Links

<*> To visit your group on the web, go to:
    http://groups.yahoo.com/group/dmcnews/

<*> To unsubscribe from this group, send an email to:
    dmcnews-unsubscribe@xxxxxxxxxxxxxxx

<*> Your use of Yahoo! Groups is subject to:
    http://docs.yahoo.com/info/terms/