Roger Graham on sun 30 may 04
In the Australian electronics magazine "Silicon Chip" for April 2004 there's
a construction project called "A Smart Mixture Display for Your Car".
Basically it's an add-on gadget for the motor enthusiast, to track the car's
fuel mixture, rich or lean, by monitoring the engine's EGO sensor. The
article includes information I've not seen before, explaining something
often noticed about changes in oxyprobe readings during the course of a kiln
firing.
One would expect that for a given degree of reduction (that is, for a given
air-to-fuel ratio) the millivolt reading from a sensor would remain
constant. Not so, it seems. The output decreases as the sensor heats up.
I've noticed the decrease in sensor output time after time as a firing
progresses, but assumed it was due only to changes in air-to-fuel ratio as I
adjusted burner pressure or damper.
I have a small test kiln fitted with an $800 oxyprobe down inside, and a $5
EGO sensor at the flue. In a typical firing, at 700 degrees C the oxyprobe
reads
780 millivolts. At 1200 degrees C, the reading is 600 mV, for what seems to
be the same amount of reduction.
The figures for the $5 EGO sensor, same firing, are 950 mV at the lower
temperature, falling to 800 mV at the higher temperature. The decrease is
much the same amount for either instrument, as the temperature rises.
The "Silicon Chip" article is accessible on-line at www.siliconchip.com.au,
though disappointingly only if you pay an amount for the download. It comes
down strongly in favour of a particular EGO sensor, described as "NTK
manufactured Ford E7TF 9F472 DA " , and includes a graph showing the
response curves of the sensor at three different temperatures. After a bit
of a search on the internet, I've found a very similar graph elsewhere (the
URL is http://www.innovatemotorsports.com/news3.php)
Doesn't make much difference in practice. Just interesting to now have an
explanation for the decrease in meter readings as the firing progresses.
Roger Graham, near Gerringong, Australia
http://members.optusnet.com.au/~rogergraham
Chris Morgan on tue 1 jun 04
Disclaimer:
My high school physics has gotten pretty rusty. Read the following with a
skeptic eye.
This is how I make sense of it:
Heating a metal increases its resistance, correct? So a hotter EGO sensor
would have a higher resistance overall than a cooler one, resulting in a
greater voltage drop across the the circuit, per Ohm's law. Make sense to
anyone else. One wonders what happens when a cool sensor is placed in the
flame. Can you observe the voltage drop as it heats?
Chris Morgan
On Sun, 30 May 2004 20:51:50 +1000, Roger Graham
wrote:
>I have a small test kiln fitted with an $800 oxyprobe down inside, and a
$5
>EGO sensor at the flue. In a typical firing, at 700 degrees C the oxyprobe
>reads
>780 millivolts. At 1200 degrees C, the reading is 600 mV, for what seems to
>be the same amount of reduction.
>
>Send postings to clayart@lsv.ceramics.org
>
>You may look at the archives for the list or change your subscription
>settings from http://www.ceramics.org/clayart/
>
>Moderator of the list is Mel Jacobson who may be reached at
melpots@pclink.com.
Roger Graham on wed 2 jun 04
Further to the apparent decrease in the signal from an EGO sensor, as the
temperature rises. One proposal put forward is that the effect is due to the
internal resistance of the probe rising as it gets hot. Yes, it probably
does (think "positive temperature coefficient of resistance") but it's
surely not a reason for the decrease.
The millivolt output from an oxyprobe isn't due to a change in resistance,
but to the presence of a greater number of negatively-charged oxygen ions on
one face of the zirconium oxide pellet (the side exposed to the atmosphere),
and a smaller number on the side exposed to the hot kiln gases. The output
doesn't depend at all on a flow of current through the device. . The
electronics people would describe an oxyprobe as a "high impedance source",
capable of giving a voltage output but not expected to deliver much current.
Its internal resistance is probably very high, but with a suitable meter
that doesn't matter. One aims to measure the voltage difference between the
two faces of the pellet, with a high-impedance meter which draws almost
zero current from the probe
Another proposal was that combustion is more efficient as the temperature
rises. "the given amount of gas is achieving better combustion as the heat
rises, thus taking the kiln out of reduction."
The first half of this proposal sounds sensible. It may well be true. But
surely if the gas is combining more efficiently with oxygen, there must then
be LESS oxygen ions available in the kiln atmosphere, and this would give a
HIGHER voltage difference between the two surfaces of the probe.
The graphs put out by the EGO manufacturers show millivolt output on the
vertical axis, versus "fuel-to-air ratio" on the horizontal axis. Presumably
they are able to control the fuel-to-air ratio to keep it constant during
their testing. And they claim that the sensor output truly does decrease as
the probe gets hotter. Keeping the gas-to-air ratio constant isn't so easy
for a potter, since both gas pressure and damper opening are altered as the
firing progresses.
What happens in practice here is that the burner pressure is increased every
half hour or so to maintain a more-or-less constant rate of temperature
rise, say 200 Celsius degrees per hour. At each increase in pressure, when
the burner flame gets fiercer as more gas comes through the jet, there is a
need for more air.This is achieved by sliding the damper along to increase
the size of the flue opening.
The damper is opened until the readings of the oxyprobe and the exhaust
sniffer both simultaneously drop rapidly (oxidising). Then if the damper is
closed just a few millimetres, the readings both rise at once. The response
curve of this kind of sensor shows a very abrupt rise as the fuel-to-air
ratio is adjusted, a sudden jump from a low reading (say 80 mV on an EGO
sensor) to a high one (say 500 mV). The potter makes some kind of judgment,
based on the size of the flame from the flue, or the murky atmosphere, or
log notes from previous firing , and sets the damper to give what seems to
be an appropriate amount of reduction. And then notes the readings on the
meters.
Constant amount of reduction? Who can tell? But interestingly, when the
kiln is adjusted by intuition like this, and meter readings are recorded,
the values are lower as the kiln gets hotter. So after a few firings it's
sufficient to set the damper by meter readings alone, and be sure that
reduction is indeed happening even if you can't see a flame at the flue.
Roger Graham, near Gerringong, Australia
http://members.optusnet.com.au/~rogergraham
Lee Love on wed 2 jun 04
Does anybody know anyone that fires an oxyprobe with a woodkiln? Euan
said he tried it and it is pretty much useless. In a wood kiln, you
cycle through oxidation, reduction and neutral atmospheres between each
stoke.
In my little wood kiln, when I start stoking above the grates, I stoke
2Kgs of wood on each stoke, stoking each side alternately, allowing one
to burn clean while the other is reducing. At 700*C, I half close
the damper. By 900*C, still stoking 2Kgs off wood per stoke, the kiln
puts itself into reduction. It is combusting less efficiently at this
temp. At 1100*C, I have to reduce the amount of wood I stoke to
1.5Kgs, or else the kiln chokes out and the heat rise stops. It
become more efficient and requires less wood to continue temperature
rise. Toward the end, I close the damper a little more.
--
Lee in Mashiko, Japan http://mashiko.org
http://journals.fotki.com/togeika/Mashiko/ Commentary On Pottery
Hank Murrow on thu 3 jun 04
Dear Roger;
Thanks so much for your clarity on the workings of the OxyProbe (or
Kiln Finger as some of us have taken to calling it). I have mused for
some time on the possibility of driving the damper on my Doorless
Fiberkiln with a linear actuator and the gas pressure with an electric
valve, using output from the OxyProbe. This way, the damper could be
adjusted as the gas pressure changes to keep the atmosphere constant.
Some difficult (for me) work remains in developing the software for
such a control. However, I am really enjoying making pots, so work on
this has taken a back seat.
Again, Thanks,
Hank, still on the workshop tour and readying for the Centered in CT
workshop.
On Jun 2, 2004, at 4:36 AM, Roger Graham wrote:
> Further to the apparent decrease in the signal from an EGO sensor, as
> the
> temperature rises. One proposal put forward is that the effect is due
> to the
> internal resistance of the probe rising as it gets hot. Yes, it
> probably
> does (think "positive temperature coefficient of resistance") but it's
> surely not a reason for the decrease.
>
> The millivolt output from an oxyprobe isn't due to a change in
> resistance,
> but to the presence of a greater number of negatively-charged oxygen
> ions on
> one face of the zirconium oxide pellet (the side exposed to the
> atmosphere),
> and a smaller number on the side exposed to the hot kiln gases. The
> output
> doesn't depend at all on a flow of current through the device. . The
> electronics people would describe an oxyprobe as a "high impedance
> source",
> capable of giving a voltage output but not expected to deliver much
> current.
> Its internal resistance is probably very high, but with a suitable
> meter
> that doesn't matter. One aims to measure the voltage difference
> between the
> two faces of the pellet, with a high-impedance meter which draws
> almost
> zero current from the probe
>
> Another proposal was that combustion is more efficient as the
> temperature
> rises. "the given amount of gas is achieving better combustion as the
> heat
> rises, thus taking the kiln out of reduction."
>
> The first half of this proposal sounds sensible. It may well be true.
> But
> surely if the gas is combining more efficiently with oxygen, there
> must then
> be LESS oxygen ions available in the kiln atmosphere, and this would
> give a
> HIGHER voltage difference between the two surfaces of the probe.
>
> The graphs put out by the EGO manufacturers show millivolt output on
> the
> vertical axis, versus "fuel-to-air ratio" on the horizontal axis.
> Presumably
> they are able to control the fuel-to-air ratio to keep it constant
> during
> their testing. And they claim that the sensor output truly does
> decrease as
> the probe gets hotter. Keeping the gas-to-air ratio constant isn't so
> easy
> for a potter, since both gas pressure and damper opening are altered
> as the
> firing progresses.
>
> What happens in practice here is that the burner pressure is increased
> every
> half hour or so to maintain a more-or-less constant rate of temperature
> rise, say 200 Celsius degrees per hour. At each increase in pressure,
> when
> the burner flame gets fiercer as more gas comes through the jet, there
> is a
> need for more air.This is achieved by sliding the damper along to
> increase
> the size of the flue opening.
>
> The damper is opened until the readings of the oxyprobe and the exhaust
> sniffer both simultaneously drop rapidly (oxidising). Then if the
> damper is
> closed just a few millimetres, the readings both rise at once. The
> response
> curve of this kind of sensor shows a very abrupt rise as the
> fuel-to-air
> ratio is adjusted, a sudden jump from a low reading (say 80 mV on an
> EGO
> sensor) to a high one (say 500 mV). The potter makes some kind of
> judgment,
> based on the size of the flame from the flue, or the murky atmosphere,
> or
> log notes from previous firing , and sets the damper to give what
> seems to
> be an appropriate amount of reduction. And then notes the readings on
> the
> meters.
>
> Constant amount of reduction? Who can tell? But interestingly, when
> the
> kiln is adjusted by intuition like this, and meter readings are
> recorded,
> the values are lower as the kiln gets hotter. So after a few firings
> it's
> sufficient to set the damper by meter readings alone, and be sure that
> reduction is indeed happening even if you can't see a flame at the
> flue.
>
> Roger Graham, near Gerringong, Australia
>
> http://members.optusnet.com.au/~rogergraham
>
> _______________________________________________________________________
> _______
> Send postings to clayart@lsv.ceramics.org
>
> You may look at the archives for the list or change your subscription
> settings from http://www.ceramics.org/clayart/
>
> Moderator of the list is Mel Jacobson who may be reached at
> melpots@pclink.com.
>
>
murrow.biz/hank
Ilene Mahler on thu 3 jun 04
----- Original Message -----
From: "Hank Murrow"
To:
Sent: Thursday, June 03, 2004 9:08 AM
Subject: Re: Kiln Exhaust Sniffer Revisited.
> Dear Roger;
>
> Thanks so much for your clarity on the workings of the OxyProbe (or
> Kiln Finger as some of us have taken to calling it). I have mused for
> some time on the possibility of driving the damper on my Doorless
> Fiberkiln with a linear actuator and the gas pressure with an electric
> valve, using output from the OxyProbe. This way, the damper could be
> adjusted as the gas pressure changes to keep the atmosphere constant.
> Some difficult (for me) work remains in developing the software for
> such a control. However, I am really enjoying making pots, so work on
> this has taken a back seat.
>
> Again, Thanks,
>
> Hank, still on the workshop tour and readying for the Centered in CT
> workshop.
>
>
> On Jun 2, 2004, at 4:36 AM, Roger Graham wrote:
>
> > Further to the apparent decrease in the signal from an EGO sensor, as
> > the
> > temperature rises. One proposal put forward is that the effect is due
> > to the
> > internal resistance of the probe rising as it gets hot. Yes, it
> > probably
> > does (think "positive temperature coefficient of resistance") but it's
> > surely not a reason for the decrease.
> >
> > The millivolt output from an oxyprobe isn't due to a change in
> > resistance,
> > but to the presence of a greater number of negatively-charged oxygen
> > ions on
> > one face of the zirconium oxide pellet (the side exposed to the
> > atmosphere),
> > and a smaller number on the side exposed to the hot kiln gases. The
> > output
> > doesn't depend at all on a flow of current through the device. . The
> > electronics people would describe an oxyprobe as a "high impedance
> > source",
> > capable of giving a voltage output but not expected to deliver much
> > current.
> > Its internal resistance is probably very high, but with a suitable
> > meter
> > that doesn't matter. One aims to measure the voltage difference
> > between the
> > two faces of the pellet, with a high-impedance meter which draws
> > almost
> > zero current from the probe
> >
> > Another proposal was that combustion is more efficient as the
> > temperature
> > rises. "the given amount of gas is achieving better combustion as the
> > heat
> > rises, thus taking the kiln out of reduction."
> >
> > The first half of this proposal sounds sensible. It may well be true.
> > But
> > surely if the gas is combining more efficiently with oxygen, there
> > must then
> > be LESS oxygen ions available in the kiln atmosphere, and this would
> > give a
> > HIGHER voltage difference between the two surfaces of the probe.
> >
> > The graphs put out by the EGO manufacturers show millivolt output on
> > the
> > vertical axis, versus "fuel-to-air ratio" on the horizontal axis.
> > Presumably
> > they are able to control the fuel-to-air ratio to keep it constant
> > during
> > their testing. And they claim that the sensor output truly does
> > decrease as
> > the probe gets hotter. Keeping the gas-to-air ratio constant isn't so
> > easy
> > for a potter, since both gas pressure and damper opening are altered
> > as the
> > firing progresses.
> >
> > What happens in practice here is that the burner pressure is increased
> > every
> > half hour or so to maintain a more-or-less constant rate of temperature
> > rise, say 200 Celsius degrees per hour. At each increase in pressure,
> > when
> > the burner flame gets fiercer as more gas comes through the jet, there
> > is a
> > need for more air.This is achieved by sliding the damper along to
> > increase
> > the size of the flue opening.
> >
> > The damper is opened until the readings of the oxyprobe and the exhaust
> > sniffer both simultaneously drop rapidly (oxidising). Then if the
> > damper is
> > closed just a few millimetres, the readings both rise at once. The
> > response
> > curve of this kind of sensor shows a very abrupt rise as the
> > fuel-to-air
> > ratio is adjusted, a sudden jump from a low reading (say 80 mV on an
> > EGO
> > sensor) to a high one (say 500 mV). The potter makes some kind of
> > judgment,
> > based on the size of the flame from the flue, or the murky atmosphere,
> > or
> > log notes from previous firing , and sets the damper to give what
> > seems to
> > be an appropriate amount of reduction. And then notes the readings on
> > the
> > meters.
> >
> > Constant amount of reduction? Who can tell? But interestingly, when
> > the
> > kiln is adjusted by intuition like this, and meter readings are
> > recorded,
> > the values are lower as the kiln gets hotter. So after a few firings
> > it's
> > sufficient to set the damper by meter readings alone, and be sure that
> > reduction is indeed happening even if you can't see a flame at the
> > flue.
> >
> > Roger Graham, near Gerringong, Australia
> >
> > http://members.optusnet.com.au/~rogergraham
> >
> > _______________________________________________________________________
> > _______
> > Send postings to clayart@lsv.ceramics.org
> >
> > You may look at the archives for the list or change your subscription
> > settings from http://www.ceramics.org/clayart/
> >
> > Moderator of the list is Mel Jacobson who may be reached at
> > melpots@pclink.com.
> >
> >
> murrow.biz/hank
>
>
____________________________________________________________________________
__
> Send postings to clayart@lsv.ceramics.org
>
> You may look at the archives for the list or change your subscription
> settings from http://www.ceramics.org/clayart/
>
> Moderator of the list is Mel Jacobson who may be reached at
melpots@pclink.com.
| |
|
