Searching \ for '[OT] Dust measurement (with a PIC attached)' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: www.piclist.com/techref/microchip/devices.htm?key=pic
Search entire site for: 'Dust measurement (with a PIC attached)'.

Exact match. Not showing close matches.
PICList Thread
'[OT] Dust measurement (with a PIC attached)'
1999\10\12@015159 by Rob Symmans

flavicon
face
This may be a little off topic but I figure it may be of interest to
others:

I'm working on a small data collection / logging system to measure RH &
temperature. It would be nice to also measure the amount of dust in the
area under test.I figured with the diversity of the "list" some one may
have done this at some stage. I have found some limited information on
systems which use an optical method employing a laser diode but little
other information seems available. The systems I have found output a
mg/unit volume figure dependent on the amount of dust in the air sample.
Any clues as to how this is done?? Linear CCD?

rob

1999\10\12@031030 by John De Villiers

flavicon
face
The mars rover does this to measure the amount of dust that settles on the
unit for future rover missions.

As you have prolly seen from the pictures, this thing has lots of solar
panels ontop of it. The glass layer ontop of this has a cutout in one corner
. The piece of glass there is hinged and actuated by a piece of memory wire.

So what they do is every now and then they lift the glass and measure the
voltage difference between the exposed cell and one of the other cells.

The difference is hopefully due to dust ontop of the panels. Could also be a
martian kid following his newfound toy around.  ;-)


John
> {Original Message removed}

1999\10\12@042118 by Peter Keller

flavicon
face
There is a vacuum cleaner on the market (philips) which makes use of a
dust-detector. as much as i know, there is an optical sensor which
recognizes the dust particels and counts them. if you find more information,
please let me know.
peter

Rob Symmans schrieb:

{Quote hidden}

1999\10\12@092331 by paulb

flavicon
face
Rob Symmans wrote:

> Any clues as to how this is done?? Linear CCD?

 Nothing so complex.  I'd think in terms of a black-painted box with
light-traps (zig-zag paths) through which the dust-laden air can travel
and in the dark, a laser firing in one end, a "dummy load" at the other
and a photocell (any size) along the side of the cavity.

 (Slow) pulses will be generated as the dust travels across the beam.

 Dummy load looks like this (all painted matt black):
    __ __ __ __ __
   |  |  |  |  |  |
   |\               <-- beam enters.
   |__|__|__|__|__|
 Annular baffles prevent reflected light escape, oblique deflector at
end prevents direct reflection.  Laser should be similarly baffled.

 You can however increase gain by reflecting the beam back along a
*separate* path.

 Warning - dust detectors need regular maintenance!
--
 Cheers,
       Paul B.

1999\10\12@094613 by Art Allen, KY1K
flavicon
face
I've been looking for the same information for quite a few years, and have
gotten bits and pieces, but nothing much beyond the 'this should work'
block diagram::>

I work in a related communications industry, so I do not have a problem
with laser diode drivers, or detectors/related hardware of any sort.

I've yet to see information that was solid enough for an experienced person
to construct anything that might work, so my guess is that non-professional
optics people would have even bigger problems.

If anyone gets any real information on this topic, please email me--offlist
preferred. I am happy to provide technical assistance regarding the
photonics aspect of this project to anyone who can come up with complete
and feasable particle counter method.

I would strongly advise anyone working on this project to work towards
building a particle counter first-direct measurement of the number of
particles per some unit volume is a very tall order::>

Regards,

Art

At 01:39 PM 10/12/1999 +0800, you wrote:
{Quote hidden}

1999\10\12@125714 by Anne Ogborn

flavicon
face
Since it's not been made clear, the laser ones work by firing a laser
beam through the air column. The beam is invisible except where
it encounters dust (you can easily see this phenomenon with a students
lab laser). The detector is a photocell that picks up the scattered light.

>  Nothing so complex.  I'd think in terms of a black-painted box with
>light-traps (zig-zag paths) through which the dust-laden air can travel
>and in the dark, a laser firing in one end, a "dummy load" at the other
>and a photocell (any size) along the side of the cavity.


You also need to plan your air path carefully.  Paul's light trap is
also a wonderful dust trap.

Another approach, one more suitable for a 'dosimeter' type
measurement (and much more sensitive as a result) than
a 'current value' measurement, is to pass air through a paper
filter of small area by using a pump.

Put an LED on one side and a photocell on the other - the dust
will slowly obscure the LED and the photocell R will increase.

Another 'dosimeter' style measurement -
wind a fine wire into a tiny coil (a 110V lamp filiment might be good).
and place in airstream. Occasionally interrupt the airstream (so it
doesn't cool the thing) and put a regulated current through it and a
resistor.
measure the voltage drop.  The wire will have a temp. coefficient,
so the voltage drop should increase with the temp. The wire will
reach a temp determined by when the electrical power converted
to heat is = to the heat dissapated to the still air.
The dissapated heat will depend on the coating of dust 'insulation'
on the wire.

You shouldn't try to get the wire too hot, or you'll fry the dust coat.

Another dosimeter method -
supply test air with a pump to a T junction. One side should lead
to a small area HEPA filter. the other should lead to a pressure sensor.
Back pressure will rise as the filter fills with dust.

1999\10\12@125923 by Anne Ogborn

flavicon
face
come to think of it,
dust measurement is discussed in "Rocket Propellants", by Sarner.
I once interviewed for a job with Archer Daniels Midland interfacing
various sensors & actuators in a flour mill to a computer.
I know that included particle counters - there's apparently a whole
body of literature on particle counting and sizing.

1999\10\12@131223 by Wagner Lipnharski

picon face
I think one good idea is just using two LEDs and Two photosensors. One
pair will have a clean path (using a pipe) with the same distance of the
pair inserted into the dusted area.  Comparing both photosensors output
you will have a delta voltage representing somehow a "dust percentage".
Just filter both photosensor outputs to reflect at least the last 5
seconds of measurements, and create a table for it. Nice and easy.

I already used something like that for a liquid measurement, at a
softdrink company (they need to measure the darkness of the caramel into
the liquid).

Wagner

1999\10\12@152610 by Thomas McGahee

flavicon
face
I have built smoke detectors that actually respond to the
particulate matter in the air. Since dust is also particulate matter,
you might find this method useful.

The smoke/dust detector consists of a long container that contains
internal light baffles that allow air to move relatively freely through
the structure, but does not allow light in past a certain point.
The entire inside including light baffles is painted with
FLAT-black paint to reduce light reflections.

I have made two variations on this theme. The first uses a simple
bright LED as the light source. The LED is mounted such that it
faces a light sensor that is several inches away. A fan is mounted
at the top of the structure and blows air INTO the structure.
MOST of the smoke/dust will travel through, but SOME will stay
inside, so periodic cleaning is essential.

If there is dust in the air, then the light reaching the sensor
is diminished. The more dust there is, the less light will reach
the sensor.

The second variation uses two mirrors to increase the light path
considerably while keeping the physical size of the structure
reasonably small. The light source is aimed at one of two mirrors
that face one another and which are parallel to each other. The
light beam is adjusted so that it hits the mirror at an angle
other than 90 degrees. This causes the beam to bounce back and
forth between the mirrors, and to also move DOWN the mirrors until
it hits the light sensor. Because the light path is so much
longer, this method makes the device that much more sensitive
to small changes in dust concentration. You can use a laser for
the light source, but I have had good results using just a
super-bright red LED. You can use optics to narrow the beam,
or do like *I* do and just live with a wide, diffuse beam.

I will often use another super-bright red LED as my light sensor.
There are times when I want a digital readout of smoke/dust
level. Often a simple op-amp circuit is the answer, and I use
a really cheap digital voltmeter as my readout. You can always
use a PIC, of course.

Oh, yes, one more point of possible interest: I have found it
very useful to sometimes use an op-amp circuit in a closed-loop
feedback mode to keep the light falling on the sensor constant.
This means that as smoke/dust increases, the voltage and current
delivered to the LED/resistor pair will likewise increase,
as the light is made brighter to compensate for the smoke/dust.

You do have to open the thing up and clean it every so often.

If you use a laser and you restrict the beam at the *sensor* end
so that the size of the beam hitting the sensor is small,
you can view the output on an oscilloscope and see fluctuations
caused by the dust particles. The larger the area of light
that falls on the sensor, the more of an "average" you will
get. One implication of this is that it is actually possible
to count the dust particles per unit of time if they are big
enough. The measured volume in this case is the circular
area of the *received* beam multiplied by the length of the
beam. (Assuming a collimated beam, of course!)  Convert
fluctuations to pulses, and count the pulses. As the distance
between source and sensor increases, the system becomes more
non-linear, since the probability of TWO or more particles
or aggregate of particles obscuring the beam at the same time
will increase. This non-linearity is less problematical with
the diffuse-sensor approach, since there you are dealing with
averages to begin with.

Fr. Tom McGahee


{Original Message removed}

1999\10\12@161514 by hris Fanning

flavicon
face
> I think one good idea is just using two LEDs and Two photosensors. One
> pair will have a clean path (using a pipe) with the same distance of the
> pair inserted into the dusted area.  Comparing both photosensors output
> you will have a delta voltage representing somehow a "dust percentage".
> Just filter both photosensor outputs to reflect at least the last 5
> seconds of measurements, and create a table for it. Nice and easy.
[snip]

Aren't there optical smoke detectors that work the same way?  This might
be an off the shelf smoke detector part...?

Chris

1999\10\12@163406 by w. v. ooijen / f. hanneman

picon face
> There is a vacuum cleaner on the market (philips) which makes use of a
> dust-detector. as much as i know, there is an optical sensor which
> recognizes the dust particels and counts them. if you find more
information,
> please let me know.

I have one of those, but I never used the dust detector.
As far as I remember it is just a LED and detector on both sides
of the shaft.

Wouter

1999\10\13@043353 by Steinar Olsen

flavicon
face
Hi rob,

You may measure dust with a optical laser system.
this can be done 2 ways,

1. Use direct light transmission (loss of light due to dust in path from
100% light setting) This is unaccurate for low concentration, but suitable
for heavy dust.

2. use scatterd light measurement (light scattered when it hits small
particles
in the air) This is accurate for small conc. of dust. Collect scattered ligt
with a lens, and cover the center of the lens to prevent direct laser light
from hitting the sensor.

good luck




{Quote hidden}

Steinar Olsen
Electronics Engineer
_____________________________________________________________
Norsk Elektro Optikk A/S        Work E-Mail:  spam_OUTSteinarTakeThisOuTspamneo.no
Solheimsveien 62A               Home E-Mail:  .....StolsenKILLspamspam@spam@c2i.net
P.O. Box 384                    Direct:       +47 67 91 11 58
N-1471 SKAARER                  Phone:        +47 67 97 47 00
NORWAY                          Telefax:      +47 67 97 49 00
=============================================================

More... (looser matching)
- Last day of these posts
- In 1999 , 2000 only
- Today
- New search...