[Diy_efi] DIY-WB and methanol

[steve ravet]

steve ravet wrote:
> > My question is fairly simple:  Will the DIY-WB provide accurate A/F r=
atio
> > information when a mixture of gasoline and methanol is present, oxidi=
zed by
> > nitrous oxide?

Here's what I found in my Trash folder (there's more good info in that
thread):

Subject: Re:    [Efi332] Nitrous and WBO sensing
Date:    Wed, 24 Jul 2002 16:45:19 -0700
From:    Garfield Willis <garwillis at msn.com>

The only good way to deal with questions like this is to understand the
fundamentals about how the sensor "senses" AFR. Especially in the case
of mere oxygenates (like N=B2O-NOS, which is just oxygen in a compact
form, tied to some nitrogen; the nitrogen isn't combustible), it DOES
seem like it should be intuitive that nothing is altered in the AFR
sensing. After all, we're just adding more Os, and attempting to supply
the correct amount of fuel corresponding to those added Os, right? So
that's probly everyone's (including mine) intuition at first glance;
it's not a different FUEL per se, we're still just running straight
petrol, just that we're getting more "air" into the engine, and then
adding more fuel, so the AFRs should still be "petrol" AFRs, right?

Right, they *should*. And if our NTK sensors we all use to read AFR were
a type that counted molecules of O2 and molecules of CxHy, and gave us a
ratio, that would be that. Unfortunately, life isn't that simple, and
the sensors we're stuck with at present don't work that way.

The problem arises because our sensors only pick up "partial pressures"
of the gases they're sensitive to, in the exhaust stream. That's why
even mere diluents like water injection, have to be accounted and
corrected for in use. Because they end up in the exhaust stream, and
dilute the partial pressures of the "active ingredients" our sensors
actually measure. Soooo ANYTHING that alters the assumed background
constituents of the exhaust gas, upon which the sensor's calibration
curve is based, will affect the AFR readings and produce errors.

Some of you will already be saying, Geezus what a mess, I'm outta here.
Too complicated. Yeah, I sympathize. Especially when none of the other
AFR vendors have bothered to attempt to raise the level of consciousness
on this key issue (remind you at all of those similar EBP issues? :)

But for those intrepid enough to hang in there, here are the salients:
in any combustion process, there are 3 components from our proformance
standpoint. (1) Fuel, combustibles, hydrocarbons that burn, (2)
oxygenates, molecules or partial molecules that contain oxygen that can
feed the fire along with the O2 injested via inducted air, and (3)
diluents, anything in the inducted or injested mix that is neither 1 or
2. Nitrogen in the air that's inducted is a diluent, atomized water
inducted into the chambers is a diluent, and the nitrogen in NOS is also
a diluent, because it ends up as elemental . Diluents are, well,
diluents (or dilutents, not really a word, but you get the idea),
because they're inert *to the chemistry of the combustion process*, but
they may be either unavoidable, in the case of the 73%N=B2 in inducted
atmospheric air, OR they may be advantageous directly, like water, as a
coolant, or piggybacked onto an oxygenate, and thereby unavoidable, as
in N=B2O/NOS. Actually, since NOS is in liquid form and cools on
expansion, it's ALSO an advantageous coolant.

OK, so got the taxonomy? Fuels, Oxygenates, Diluents. EVEN with straight
petrol combustion, normally aspirated, all THREE are in play. The petrol
is a straight fuel (usually :), the air is a mixture of oxygenates and
diluents. They all get weighed going in, and coming out, and presto,
AFR!  NOT...heh, they all go in, turn into gases and vapors, some of
which our sensor is sensitive to, others not, and out they come in the
exhaust tract.

The calibration curve many are familiar with, for AFR vrs. Ip or Vout
for these NTK sensors, is based on an ASSUMED ratio of all these
byproducts ONLY for straight petrol combustion with ordinary earth
*air*. IOW, that calibration curve is based on there being a certain
percentage of Nitrogen in air (a mere diluent), and ALSO a certain
percentage of h2o vapor present in the exhaust as a result of the
combustion (so-called 'wet' exhaust gases), again a mere diluent.

The reason these percentages of mere diluents is important isn't because
the sensor is sensitive to them (it isn't), but because what the sensor
IS going to be picking up is the partial pressures of the gases (H=B2,
HC/VOC, CO, O2) that it is sensitive to, BUT these partial pressures
will be 'diluted' or relative to the amount of inert gas components
they're mixed with.

To put this in perspective, let's say you plan to go one step further
than water injection, and outdo all your buddies at the track; you're
going to inject liquid nitrogen!! You've found nirvana and a way to
store LN=B2, cheaper, easier, and lighter than water. You're out there
tuning away with your AFR meter, wondering if the readings you'd expect
with straight petrol will apply with your inert coolant injection. They
WON'T! Why not, it's inert!? you say. And true it is, BUT you've just
added an unexpected diluent into the exhaust stream, and your AFR sensor
is going to see that as a decrease in the partial pressure of whatever
gases it's sensing. You haven't changed the number of O2 molecules
combining with your hydrocarbon molecules, and infact, as far as the
*air* in AFR is concerned, you haven't even changed that! BUT, if you're
running on the lean side of stoich, it's going to indicate richer than
you really are, and if you're running on the rich side of stoich, your
uncorrected AFR instrument is going to indicate you're leaner than you
are.

--=20
Steve Ravet
steve.ravet at arm.com
ARM,Inc.
www.arm.com

_______________________________________________
Diy_efi mailing list
Diy_efi at diy-efi.org
http://www.diy-efi.org/mailman/listinfo/diy_efi