calibration test gas mixtures [was Re: NTK UEGO calibration]

[Jason Haines]

As promised, here are some sample gas mixtures for different air fuel
ratios. These gas mixtures were provided to us by Horiba for use with our
wide band meters at work but they should apply to any wide band meter. The
bubbler is a device (looks like a bong) that is designed to add water vapor
to the mixture to more closely simulate the gas mixture that would be
encountered by an O2 sensor (with water being a natural byproduct of
gasoline combustion). You can do this without the bubbler, you just loose a
little bit of accuracy.

With bubbler:

The gas mixture you want for 0.85 lambda (12.5:1 AFR) is:
 Carbon monoxide 3.4%
 Hydrogen  2.0%
 Carbon dioxide 10.0%
 Nitrogen  balance

The gas mixture you want for .8 lambda (11.76:1 AFR) is:
 Carbon monoxide 6.64%
 Hydrogen  2.57%
 Carbon dioxide 10.0%
 Nitrogen  balance

A gas mixture for 1.0 lambda (14.7:1 AFR) is:
 Carbon dioxide 13.0%
 Nitrogen  balance

A gas mixture of 13.0 AFR:
 Carbon monoxide 4.1%
 Hydrogen  0.75%
 Carbon dioxide 10.0%
 Nitrogen  balance


Without bubbler:

A gas mixture of 13.0 AFR:
 Carbon monoxide 4.0%
 Hydrogen  0.733%
 Carbon dioxide 10.0%
 Nitrogen  balance

I don't have a lean (>1.0 lambda) gas mixture specification yet. If you need
one you can use atmospheric as roughly 20.6 - 20.9% 02. I assume you could
also use a known 02 % concentration gas (with remainder being something
semi-inert like nitrogen).

You will want to flow the gas past the sensor(s) at roughly 2-3 liters per
minute. You do not want to have any back pressure on the system when
calibrating the sensors (although it would be interested to see how the
results changed).

These mixtures assume normal gasoline. I don't know how much error not
having the moisture content or the correct fuel specifications will cause.
If you look at the difference between the mixtures for the 13:1 AFR with and
without the bubbler, you can see the level of accuracy that is being
targeted here. The difference between 4.1% and 4.0% for the CO and 0.75% and
0.733% for the Hydrogen. =

You will want to get a +/- 1% certified blend but you shouldn't have to pay
for an EPA certified blend (you don't need that level of accuracy or
paperwork). One source for the gas mixtures is Scott Specialty Gases
(215-766-8861) - they have offices all over the country. Our tank came from
Praxair (440-237-8770). I think they also have offices all around the US.
You can get large tanks like those used for welding gases or you can get
disposable cylinders but you pay a lot for the disposable cylinders because
they normally make the larger cylinder in the custom blend and then fill the
disposables from that so you pay for the extra gas anyway.

I hope this if of use to some people.


Jason




>
> > Heater in the WB sensor *will* ignite a propane or butane mix w/ air -
> > detonate is more like it.  Did that.
> >
> > I'm just a minor in chem from 40 years ago, so its a bit misty....really
> > dunno how these things work.
> >
> > I used argon/CO2 MIG mix to get stoich; ran a couple of WBs, compared to
> > the switching of a NB sensor.  Used a 1 qt Mason jar w/ holes in the lid
> > for the sensors & small hole to admit gasses.  Convection from the
> > heater(s) mixes things well, adding a very small muffin fan didn't
change
> > anything.  Starting w/ air in the jar (offscale lean), running argon in,
> > AFRs fall to 14.7:1.  Adding propane or butane at this point is safe,
but
> > doesn't take much to *really* swing the AFR rich.  Far as I can tell,
*no*
> > O2 and *no* hydrocarbon = stoich - add either O2 or hydrocarbon gas to
get
> > the full range of AFRs.
> >
> > Short of a bomb deal, I couldn't come up w/ a reproducable means of
mixing
> > known amounts of gasses & come up w/ a 'cal' gas.  These are available
> > commercially, mostly for EPA stuff & come w/ all sorts of assays &
> > certificates w/ a price tag to match - try BOP gasses.
> >
> > Barry
> >
> > At 04:39 PM 8/21/01 -0400, you wrote:
> > >For those of us that are not chemists, is there a cheap way that is
> > >also safe (not that I don't like to blow things up as long as I am not
> > >one of those things).
> > >
> > >What about Propane?  Or Butane?  Something readily available, and if
> > >so how would one go about testing?  Will the heater in the O2 sensor
> > >start an explosion with Propane or other?
> > >
> > >Where would one get nitrogen or argon?  Is there an easy way to make
> these
> > >gases (that is also safe)?
> > >
> > >Thanks,
> > >
> > >Steve F
> > >
> > >-----Original Message-----
> >
>
> --------------------------------------------------------------------------
> --
> > To unsubscribe from diy_efi, send "unsubscribe diy_efi" (without the
> quotes)
> > in the body of a message (not the subject) to
majordomo at lists.diy-efi.org
> >
> >
>
> --------------------------------------------------------------------------
--
> To unsubscribe from diy_efi, send "unsubscribe diy_efi" (without the
quotes)
> in the body of a message (not the subject) to majordomo at lists.diy-efi.org
>

----------------------------------------------------------------------------
To unsubscribe from diy_efi, send "unsubscribe diy_efi" (without the quotes)
in the body of a message (not the subject) to majordomo at lists.diy-efi.org