Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

air pollution particulate sensor package

155 views
Skip to first unread message

gyroma...@gmail.com

unread,
Jun 30, 2014, 3:05:42 PM6/30/14
to
Hi,

This message is related to my earlier request for advice in the topic 'finding electronics design and fabrication expertise for a project'.

Because of the kindness of the group members here, I am posting more details about the project and aims.

The overall aim is to develop and deploy inexpensive 'personal' air pollution monitors. For the environment in which I am interested, a major pollutant concern is particulates arising from biomass combustion, coal burning, and car exhaust.

Although these may be inconsistent aims, I am interested in having a device that...
- can measure particulate concentrations (and perhaps temperature and humidity levels)
- is low cost, so that many devices can be deployed within a constrained project budget
- has a good degree of concordance (at least qualitatively) with accurate stationary monitoring stations
- can be read with an app on a cellphone or similar device (bluetooth?)
- is convenient to the wearer of the device, e.g, is robust and has the ability to run on batteries for a few days at a time

I anticipate that a user (or environmental health scientist or health care professional guiding the users) might want to record reading at several points during the day, not continuously.

There are several projects I have found on the web focused on devices to quantify particulates in the air, but I don't think that they satisfy the aims above.

The closest instrument that I could find is detailed in a PhD dissertation by David Holstius, a chapter of which I have posted here:

https://www.dropbox.com/s/azs0dqzgpffppl1/Holstius2014dissertation_Ch3.pdf

Because of my lack of training in the field of electronics, I do not know if the detection method, components, and overall design chosen by this individual would be appropriate as a starting platform to achieve the aims above.

Any input or suggestions would be greatly appreciated.

Thank you again.

-gyro

John Larkin

unread,
Jun 30, 2014, 4:43:17 PM6/30/14
to
On Mon, 30 Jun 2014 12:05:42 -0700 (PDT), gyroma...@gmail.com
wrote:

>Hi,
>
>This message is related to my earlier request for advice in the topic 'finding electronics design and fabrication expertise for a project'.
>
>Because of the kindness of the group members here, I am posting more details about the project and aims.
>
>The overall aim is to develop and deploy inexpensive 'personal' air pollution monitors. For the environment in which I am interested, a major pollutant concern is particulates arising from biomass combustion, coal burning, and car exhaust.

Cars aren't bad. Diesels, in busses, trucks, and construction/farm
machinery, are.

I'm a bit skeptical about the evils of CO2, but particulates are nasty
for people and the planet.

Particulates melt snow, which is especially evil.



--

John Larkin Highland Technology, Inc

jlarkin att highlandtechnology dott com
http://www.highlandtechnology.com

jurb...@gmail.com

unread,
Jun 30, 2014, 4:52:57 PM6/30/14
to
You forgot to put that in your public directory so we cannot see it yet. I do have a better idea what you are talking about now.

You want to detect particulate matter in the air, and this is going to be optically because you just do not have many other choices.

Here is the problem, mathematically, from zero to 100 %, you must define 100 % as zero. Totaslly clean air is the setpoint, the zero, the reference. If the nibers do not start there, it will be hard to use them. It's like the audio example before, you don't say 99.7 % NOT distortion, you say 0.3 % distortion. Kinda flipped over. This is exactly the same in this case.

Take a calibrated light source. Then a detector. Whatever drop there is from the light source to the detector is the result. This is no all that easy. What's more, I doubt you will be able to do it on a smart phone. This requires hardware.

Now comes the $64 question, what kind of numbers do you want ? PPM of corbon particulates ? I would almost venture to say there is an off the shelf solution for that. In fact, what abput smoke detectors, but instead of feeding an alrm, feed it to meter or whatever. A buffer to get the levels you wanrt and then something to display the result.

If you need accuracy, then we are talking instrumentation. That is a bit harder than building a (n old style) TV set. I can do it a little bit but I am not great at it. I got a Wavetek 111 and I am not only impressed by its technology, but damn near enamoured. Fucker makes a sine wave out of a triangle at .05 % THD ! Frequency independent. Has a cool type of feedback circuit. It requies the input to be EXACLT the amplitude it expects. I just got into it like last year, but the damn thing was builtin 1 1970 !

Damn I am behind the times.

So, with your thing, I figure you should set the setpoint in a vacuum. Start with that. Whatever the output is at that point from the optical sensor then, consider that the setpoint.

Circuitry ? Well that is kinda like knowing when to use a PNP transistor instead of an NPN. It is flipped. Upside down or backwards, whatevr you want to call it.

jurb...@gmail.com

unread,
Jun 30, 2014, 4:58:53 PM6/30/14
to
Another thing is the wavelength of light you use in the detector. If you define the subjedct as particulate mattr that is VISIBLE, tha ti one thing. you oculd use UV or IR.

the main question here iss - what results do you want ? Exactly.

RobertMacy

unread,
Jun 30, 2014, 5:22:00 PM6/30/14
to
On Mon, 30 Jun 2014 13:43:17 -0700, John Larkin
<jla...@highlandtechnology.com> wrote:

>> ...snip...
>
> Cars aren't bad. Diesels, in busses, trucks, and construction/farm
> machinery, are.
>
> I'm a bit skeptical about the evils of CO2, but particulates are nasty
> for people and the planet.
>
> Particulates melt snow, which is especially evil.

Cars aren't bad. What about thousands of brake linings turning to powedery
dust at a Stop sign ??

RobertMacy

unread,
Jun 30, 2014, 5:58:25 PM6/30/14
to
On Mon, 30 Jun 2014 12:05:42 -0700, <gyroma...@gmail.com> wrote:

> ...snip....
>
> https://www.dropbox.com/s/azs0dqzgpffppl1/Holstius2014dissertation_Ch3.pdf
>
> ...snip...

Epithets deleted! that dropbox keeps squirreling around inside my system
and trying to do something, searching for things?

Twice that URL crashed my system [now THAT's malware control!, simply
crash if someone tries to muck about]

(I think dropbox got upset when I wouldn't sign up)

Anybody have a 'clean' URL for this?

Jim Thompson

unread,
Jun 30, 2014, 6:08:14 PM6/30/14
to
Depends on whether the dust falls to the ground or is fine enough to
become airborne.

...Jim Thompson
--
| James E.Thompson | mens |
| Analog Innovations | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| San Tan Valley, AZ 85142 Skype: skypeanalog | |
| Voice:(480)460-2350 Fax: Available upon request | Brass Rat |
| E-mail Icon at http://www.analog-innovations.com | 1962 |

I love to cook with wine. Sometimes I even put it in the food.

John Larkin

unread,
Jun 30, 2014, 6:13:21 PM6/30/14
to
On Mon, 30 Jun 2014 14:22:00 -0700, RobertMacy
<robert...@gmail.com> wrote:

Probably bigger particles. It's the really tiny stuff, like 100 nm,
that makes it deep into your lungs.

Phil Hobbs

unread,
Jun 30, 2014, 6:13:50 PM6/30/14
to
The dust is pretty large compared with diesel smoke particles, so it
doesn't penetrate very far into your lungs. They got rid of the
asbestos awhile back, which was a pity. Chrysotile asbestos is nasty,
serpentine basically isn't.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC
Optics, Electro-optics, Photonics, Analog Electronics

160 North State Road #203
Briarcliff Manor NY 10510

hobbs at electrooptical dot net
http://electrooptical.net

John Larkin

unread,
Jun 30, 2014, 6:16:31 PM6/30/14
to
On Mon, 30 Jun 2014 13:52:57 -0700 (PDT), jurb...@gmail.com wrote:

>You forgot to put that in your public directory so we cannot see it yet. I do have a better idea what you are talking about now.
>
>You want to detect particulate matter in the air, and this is going to be optically because you just do not have many other choices.
>
>Here is the problem, mathematically, from zero to 100 %, you must define 100 % as zero. Totaslly clean air is the setpoint, the zero, the reference. If the nibers do not start there, it will be hard to use them. It's like the audio example before, you don't say 99.7 % NOT distortion, you say 0.3 % distortion. Kinda flipped over. This is exactly the same in this case.
>
>Take a calibrated light source. Then a detector. Whatever drop there is from the light source to the detector is the result.

Unless you use reflection, like in an optical smoke detector. Then 0%
is the baseline. He could probably count individual particles.

RobertMacy

unread,
Jun 30, 2014, 6:27:52 PM6/30/14
to
On Mon, 30 Jun 2014 15:08:14 -0700, Jim Thompson
<To-Email-Use-Th...@on-my-web-site.com> wrote:

> On Mon, 30 Jun 2014 14:22:00 -0700, RobertMacy
> <robert...@gmail.com> wrote:
>
>> On Mon, 30 Jun 2014 13:43:17 -0700, John Larkin
>> <jla...@highlandtechnology.com> wrote:
>>
>>>> ...snip...
>>>
>>> Cars aren't bad. Diesels, in busses, trucks, and construction/farm
>>> machinery, are.
>>>
>>> I'm a bit skeptical about the evils of CO2, but particulates are nasty
>>> for people and the planet.
>>>
>>> Particulates melt snow, which is especially evil.
>>
>> Cars aren't bad. What about thousands of brake linings turning to
>> powedery
>> dust at a Stop sign ??
>
> Depends on whether the dust falls to the ground or is fine enough to
> become airborne.
>
> ...Jim Thompson

From the way that stuff migrates all over the wheels on my car, I'd say
it's pretty 'airborne'

Remember when the linings were made of asbestos? Which subsequently have
been removed from the market. Coincidentally? after reports finding high
levels in lungs of people who live near stop signs vs middle of the
blocks, more open areas.

Lasse Langwadt Christensen

unread,
Jun 30, 2014, 6:43:30 PM6/30/14
to
it the particle size that's important, if they are big enough it is is just
dirt it is when they are small enough to get deep into the lungs and stay there is gets dangerous, like asbestos and according to some exhaust from modern high pressure diesels and gasoline direct injection



-Lasse

Phil Hobbs

unread,
Jun 30, 2014, 7:12:07 PM6/30/14
to
On 6/30/2014 6:16 PM, John Larkin wrote:
> On Mon, 30 Jun 2014 13:52:57 -0700 (PDT), jurb...@gmail.com wrote:
>
>> You forgot to put that in your public directory so we cannot see it yet. I do have a better idea what you are talking about now.
>>
>> You want to detect particulate matter in the air, and this is going to be optically because you just do not have many other choices.
>>
>> Here is the problem, mathematically, from zero to 100 %, you must define 100 % as zero. Totaslly clean air is the setpoint, the zero, the reference. If the nibers do not start there, it will be hard to use them. It's like the audio example before, you don't say 99.7 % NOT distortion, you say 0.3 % distortion. Kinda flipped over. This is exactly the same in this case.
>>
>> Take a calibrated light source. Then a detector. Whatever drop there is from the light source to the detector is the result.
>
> Unless you use reflection, like in an optical smoke detector. Then 0%
> is the baseline. He could probably count individual particles.

Yup. Dark field light scattering is the easy way to do it, if you don't
need to go much below 0.5 microns. That's really simple--two air tubes,
a simple vacuum pump, a laser diode, a lens, a collecting mirror, and a
photodiode/TIA.

If the OP doesn't need size information, a condensation nucleus counter
is the way to go. Those draw the air sample into a supersaturated butyl
alcohol vapour, so any particles nucleate droplets.

Interestingly, the droplet radius grows linearly with time, so even a
really really small particle (10 nm or so) produces just about the same
size droplet as a 1-um particle.

Cheers

Phil "former crud guy" Hobbs

Jeff Liebermann

unread,
Jun 30, 2014, 7:23:04 PM6/30/14
to
On Mon, 30 Jun 2014 13:52:57 -0700 (PDT), jurb...@gmail.com wrote:

>You want to detect particulate matter in the air, and this
>is going to be optically because you just do not have many other choices.

It's called a nephelmometer and uses light scattering. I have a
Nikken #1394 Air Quality Monitor:
<http://802.11junk.com/jeffl/crud/Nikken-AQM.jpg>
I don't think this model is sold any more. The photo is me using it
to find the source of the burning smell in the office. It turned out
to be an overheated resistor inside the UPS in the photo. The unit is
quite sensitive. I can stamp my feet on the rug and cause the graph
to almost go full scale from the dust. Same with a couch or chair.
You could probably clone the function quite easily.

More:
<http://www.nwcg.gov/pms/pubs/SMG/177-186.pdf>
<http://en.wikipedia.org/wiki/Nephelometer>
Overkill:
<http://www.esrl.noaa.gov/gmd/aero/instrumentation/neph_desc.html>


--
Jeff Liebermann je...@cruzio.com
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558

k...@attt.bizz

unread,
Jun 30, 2014, 7:24:36 PM6/30/14
to
Yes, particle size is key. Over 10um, the danger isn't so high. The
respiratory system tends to clean this stuff out before it gets the
lungs. At 2.5um the particles are a perfect size to get into the
lungs and stay there. At some point, though, particulates will pass
through the alveoli walls, into the blood stream (where they may be
toxic or simply cleared from the system).

When I look for dust management for woodworking, I make sure the
filters will filter 1um particles (about the best that can be found).
Wood isn't generally toxic (though some species certainly are),
though, so 1um is likely good enough.

Bill Sloman

unread,
Jun 30, 2014, 7:57:08 PM6/30/14
to
On Tuesday, 1 July 2014 06:58:53 UTC+10, jurb...@gmail.com wrote:
> Another thing is the wavelength of light you use in the detector. If you define the subject as particulate matter that is VISIBLE, that is one thing. you could use UV or IR.
>
> the main question here is - what results do you want ? Exactly.

http://en.wikipedia.org/wiki/Light_scattering

Rayleigh scattering is intensely wavelength dependent - to the sixth power of the particle diameter, and inversely proportional to the fourth power of wavelength.

It would be fun to use a bunch of light sources to measure scattering at a number of different wavelengths, and deconvolute the results to say something about the concentrations and sizes of the different particles doing the scattering.

You'd need Phil Hobbs to make it work ...

--
Bill Sloman, Sydney

Steve Wilson

unread,
Jun 30, 2014, 8:24:21 PM6/30/14
to
Jeff Liebermann <je...@cruzio.com> wrote:

> On Mon, 30 Jun 2014 13:52:57 -0700 (PDT), jurb...@gmail.com wrote:
>
>>You want to detect particulate matter in the air, and this is going
>>to be optically because you just do not have many other choices.
>
> It's called a nephelmometer and uses light scattering. I have a
> Nikken #1394 Air Quality Monitor:
> <http://802.11junk.com/jeffl/crud/Nikken-AQM.jpg>
> I don't think this model is sold any more. The photo is me using it
> to find the source of the burning smell in the office. It turned out
> to be an overheated resistor inside the UPS in the photo. The unit is
> quite sensitive. I can stamp my feet on the rug and cause the graph
> to almost go full scale from the dust. Same with a couch or chair.
> You could probably clone the function quite easily.

I could not find a user manual for your Nikken. If you have a pdf and can
spare the time, could you upload it to your web site?

This is a very interesting way to use a particle counter. I have some
questions:

How does the Nikken operate?

Is there a table showing the particle concentration vs size?

What type of resistor was it?

Why did the resistor give off particles?

Was it hot enough to give off smoke?

Could you detect the source by smell?

I use the Dylos DC1100 PRO, which is a laser particle counter. It shows
the particle count for two size ranges of particles: 0.5 micron to 2.5
microns, and greater than 2.5 microns. I believe it discriminates between
particle sizes by measuring the width of the pulse as a particle passes
through the light gap.

http://www.dylosproducts.com/ornodcproair.html

I have two units and I am always amazed at how well the readings agree
between them.

I wonder how the Dylos and Nikken would compare in ability to
discriminate between different size particles. Does the Nikken have more
size ranges?

Phil Hobbs

unread,
Jun 30, 2014, 9:00:03 PM6/30/14
to
AFAIK size measurement is always done by thresholding the pulse heights.
The scattered light power goes as the sixth power of the particle
radius. (I've recently reverse-engineered several such units for two
court cases, and they all worked that way.)

Cheers

Phil Hobbs

Phil Hobbs

unread,
Jun 30, 2014, 9:05:28 PM6/30/14
to
Back in my mis-spent youth (or maybe early middle age) I did a lot of
particle counting work. Not exactly that kind, but particle mapping in
6 dimensions (x, y, z, radial velocity, size, and time), and
collaborated a bit with some colleagues who were doing composition
sorting for particles in fluids, using the complex refractive index of
the particle to sort out bubbles, metals, nonmetals, and carbon.

At the moment I'm doing some higher speed stuff of the same sort, but
it's kind of slow going because I don't have the mechanical design and
fabrication facilities I really need for this.

Steve Wilson

unread,
Jun 30, 2014, 9:14:47 PM6/30/14
to
Phil Hobbs <ho...@electrooptical.net> wrote:

> AFAIK size measurement is always done by thresholding the pulse
> heights.
> The scattered light power goes as the sixth power of the particle
> radius. (I've recently reverse-engineered several such units for two
> court cases, and they all worked that way.)

> Cheers

> Phil Hobbs

Thanks. I was working from some description I found online, but that
makes a lot more sense. It is far easier to discriminate pulse amplitudes
than pulse widths.

I wonder if the scattered light power flattens out above some radius? For
example, if the particle size is above 2.5 microns, does the scattering
change? For example, Raleigh scattering is effective for particles much
smaller than the wavelength of the light. I believe it goes as the fourth
power of wavelength:

http://en.wikipedia.org/wiki/Rayleigh_scattering

Other approximations are used for larger particles:

http://en.wikipedia.org/wiki/Light_scattering_by_particles

It's hard to tell which approach would be best suited for typical air
quality measurements, such as mold spores and dust particles.

Phil Hobbs

unread,
Jun 30, 2014, 9:46:16 PM6/30/14
to
On 6/30/2014 9:14 PM, Steve Wilson wrote:
> Phil Hobbs <ho...@electrooptical.net> wrote:
>
>> AFAIK size measurement is always done by thresholding the pulse
>> heights. The scattered light power goes as the sixth power of the
>> particle radius. (I've recently reverse-engineered several such
>> units for two court cases, and they all worked that way.)
>
>> Cheers
>
>> Phil Hobbs
>
> Thanks. I was working from some description I found online, but that
> makes a lot more sense. It is far easier to discriminate pulse
> amplitudes than pulse widths.
>
> I wonder if the scattered light power flattens out above some radius?
> For example, if the particle size is above 2.5 microns, does the
> scattering change? For example, Raleigh scattering is effective for
> particles much smaller than the wavelength of the light. I believe it
> goes as the fourth power of wavelength:
>
> http://en.wikipedia.org/wiki/Rayleigh_scattering

Yes. When the particles are big enough to be really opaque, they start
looking like geometric obstacles, so that the scattered light goes like
the intercepted area.

A fine point is that as the particles get larger, the light scatters
into a smaller and smaller range of angles around the incident k vector.
>
> Other approximations are used for larger particles:
>
> http://en.wikipedia.org/wiki/Light_scattering_by_particles
>
> It's hard to tell which approach would be best suited for typical
> air quality measurements, such as mold spores and dust particles.

Dark field light scatter, where the laser beam, air flow, and receive
path are mutually perpendicular is good medicine. Being 90 degrees off
axis reduces the dynamic range, because larger particles, which scatter
a great deal more light, scatter most of it in the forward direction.

Interferometric systems, which are sensitive to amplitude rather than
intensity, don't suffer as badly from this. (Power goes as amplitude
squared.)

Steve Wilson

unread,
Jul 1, 2014, 12:07:44 AM7/1/14
to
Phil Hobbs <ho...@electrooptical.net> wrote:

> On 6/30/2014 9:14 PM, Steve Wilson wrote:
>> Phil Hobbs <ho...@electrooptical.net> wrote:

>> I wonder if the scattered light power flattens out above some radius?
>> For example, if the particle size is above 2.5 microns, does the
>> scattering change? For example, Raleigh scattering is effective for
>> particles much smaller than the wavelength of the light. I believe it
>> goes as the fourth power of wavelength:
>>
>> http://en.wikipedia.org/wiki/Rayleigh_scattering
>
> Yes. When the particles are big enough to be really opaque, they start
> looking like geometric obstacles, so that the scattered light goes like
> the intercepted area.

> A fine point is that as the particles get larger, the light scatters
> into a smaller and smaller range of angles around the incident k
vector.

>> It's hard to tell which approach would be best suited for typical air
>> quality measurements, such as mold spores and dust particles.

> Dark field light scatter, where the laser beam, air flow, and receive
> path are mutually perpendicular is good medicine. Being 90 degrees off
> axis reduces the dynamic range, because larger particles, which scatter
> a great deal more light, scatter most of it in the forward direction.

> Interferometric systems, which are sensitive to amplitude rather than
> intensity, don't suffer as badly from this. (Power goes as amplitude
> squared.)

> Cheers

> Phil Hobbs

Boy, this sure got complicated very quick! So many issues on scattering
angles, changes with particle size, orientation, intercepted area, etc.

It's difficult to figure out what the system is actually measuring. Of
course, I understand this is one of your specialties, and you really did
some marvellous work at IBM on measuring particles in the presence of
huge interference.

And thanks for the note on interferometric measurements. That's another
area I need to look at.

Very interesting thread. Thanks. I will now be a lot more sceptical when
I see some marketing hype on a new particle instrument.

Jeff Liebermann

unread,
Jul 1, 2014, 12:13:17 AM7/1/14
to
On Tue, 01 Jul 2014 00:24:21 GMT, Steve Wilson <no...@nospam.com>
wrote:

>> It's called a nephelmometer and uses light scattering. I have a
>> Nikken #1394 Air Quality Monitor:
>> <http://802.11junk.com/jeffl/crud/Nikken-AQM.jpg>

>I could not find a user manual for your Nikken. If you have a pdf and can
>spare the time, could you upload it to your web site?

Sure. I'll scan it later tonite.

>This is a very interesting way to use a particle counter. I have some
>questions:
>
>How does the Nikken operate?

My guess(tm) is by optical scattering. The light show is inside the
black box with the air hose intake. See photos below. I didn't take
apart the box, but with 4 wires going into the box, it's a likely
guess.

>Is there a table showing the particle concentration vs size?

No. The manual has a table by concentration:
Level Color Particles/Liter
1 blue 500
2 yellow 1000
3 yellow 2000
4 yellow 3000
5 red 4000
6 red >5000
I presume that it's measuring dust particles, not aerosols.

>What type of resistor was it?

It had a ceramic case, so it could have been either metal film, wire
wound, some kind of bulk resistivity component. Most of the smoke
seemed to be coming from the scorched PCB.

>Why did the resistor give off particles?

It didn't. The PCB is what was producing the smoke.

>Was it hot enough to give off smoke?

It was hot enough to scorch the PCB, so I didn't think it would be
useful to also burn my fingers. Also, it's not very safe to shove my
fingers into a live AC powered device. I had an IR thermometer handy,
but neglected to use it.

>Could you detect the source by smell?

No. The entire office was full of burning resistor stench. There was
no visible smoke. I've had some experience chasing down overheating
computers and power supplies inside server farms. The concentration
of potential smoke sources and the high volume dispersible nature of
the air flow in an ISP's server farm makes this a difficult problem.
Lacking an IR imager, I just shoved a rubber hose into my nose, and
sniffed for maximum smoke stench among the multitude of exhaust ports.
About ever 15 minutes, I had to go outside to clear my nose. The
olfactory sensitivity comes as a side effect from being a werewolf:
<http://members.cruzio.com/~jeffl/nooze/werewolf.txt>

Sniffing for smoke in my small office created a different problem. The
flow of HVAC air would disperse smoke over the entire office making it
seem like everything was belching smoke. I immediately turned off the
ceiling fans. That helped but required that I purge the air about
every 10 minutes to prevent a build up of too much smoke.

I also made the mistake of thrashing around the office, disturbing
books and boxes laden with dust. Just touching a computah would cause
it to unload an almost invisible cloud of dust into the air via the
power supply or case fan. It was difficult to find the source without
being able to touch or move anything.

The overly sensitive Nikken AQM also created a problem. It would go
tilt fairly quickly even at a fair distance from the source of smoke.
I needed a smoke attenuator which was provided by restricting the air
flow with a somewhat lint free rag and some duct tape.

Once the procedure was established, it was very easy to find the
source of the smoke.

>I use the Dylos DC1100 PRO, which is a laser particle counter. It shows
>the particle count for two size ranges of particles: 0.5 micron to 2.5
>microns, and greater than 2.5 microns. I believe it discriminates between
>particle sizes by measuring the width of the pulse as a particle passes
>through the light gap.
>
>http://www.dylosproducts.com/ornodcproair.html
>
>I have two units and I am always amazed at how well the readings agree
>between them.
>
>I wonder how the Dylos and Nikken would compare in ability to
>discriminate between different size particles. Does the Nikken have more
>size ranges?

No. The Nikken was designed for people with allergies or health
issues and may be associated with aroma-therapy which is hinted in the
terse manual. A look at their web pile will show their intended
audience. It is NOT a precision instrument.
<http://www.nikken.com>
My guess(tm) is that it is simply looking for large particles, which
comprise most of the lung irritants.

Photos:
<http://www.learnbydestroying.com/jeffl/Nikken%201394%20Air%20Quality%20Monitor/>

The hole in the intake nozzle is for the translucent tube that feeds
the intake of the light box.
<http://www.learnbydestroying.com/jeffl/Nikken%201394%20Air%20Quality%20Monitor/Nikken-AQM-01.jpg>

CAM-003 Light box:
<http://www.learnbydestroying.com/jeffl/Nikken%201394%20Air%20Quality%20Monitor/Nikken-AQM-02.jpg>

PCB. Note the two push buttons and about 5 pots:
<http://www.learnbydestroying.com/jeffl/Nikken%201394%20Air%20Quality%20Monitor/Nikken-AQM-04.jpg>
<http://www.learnbydestroying.com/jeffl/Nikken%201394%20Air%20Quality%20Monitor/Nikken-AQM-03.jpg>

Exhaust fan:
<http://www.learnbydestroying.com/jeffl/Nikken%201394%20Air%20Quality%20Monitor/Nikken-AQM-05.jpg>

View from the exhaust end:
<http://www.learnbydestroying.com/jeffl/Nikken%201394%20Air%20Quality%20Monitor/Nikken-AQM-06.jpg>
The metal plate in the middle is floating and not connected to the
other sheet metal. My guess(tm) is that it's a static brush, intended
to strip off any charge from the dust particles that might cause them
to stick to the metal duct walls.

Serial number sticker:
<http://www.learnbydestroying.com/jeffl/Nikken%201394%20Air%20Quality%20Monitor/Nikken-AQM-07.jpg>

Jeff Liebermann

unread,
Jul 1, 2014, 12:42:58 AM7/1/14
to
On Mon, 30 Jun 2014 21:13:17 -0700, Jeff Liebermann <je...@cruzio.com>
wrote:

>Photos:
><http://www.learnbydestroying.com/jeffl/Nikken%201394%20Air%20Quality%20Monitor/>

I dumped 3 PDF files into the above directory which are scans of the
English language parts of the user manual.

jurb...@gmail.com

unread,
Jul 1, 2014, 12:53:51 AM7/1/14
to
>"(I think dropbox got upset when I wouldn't sign up) "

The OP fucked up. he did not put it in his "public" directory. If you don't put it in "public", nobody can get to it but you, or anyone on YOUR account.

jurb...@gmail.com

unread,
Jul 1, 2014, 12:58:58 AM7/1/14
to
>"Unless you use reflection, like in an optical smoke detector. Then 0%
is the baseline. He could probably count individual particles"

That may solve a problem but it creates another.

Then the mirror has to be cleaned.

Now if this thing is allowed to have moving parts we can deal with that. In fact even a lens would have to be cleaned sooner or later but a mirror would of course be worse. Emitting surface, detecting surface and then reflecting surface. Bad enough without that reflector.

But then, that is all up to the designer. Maybe it would be better. We are still a bit scant on details until - well - whatever.

DTJ

unread,
Jul 1, 2014, 1:01:56 AM7/1/14
to
On 01-Jul-14 3:05 AM, gyroma...@gmail.com wrote:
> Hi,
>
> This message is related to my earlier request for advice in the topic 'finding electronics design and fabrication expertise for a project'.
>
> Because of the kindness of the group members here, I am posting more details about the project and aims.
>
> The overall aim is to develop and deploy inexpensive 'personal' air pollution monitors. For the environment in which I am interested, a major pollutant concern is particulates arising from biomass combustion, coal burning, and car exhaust.
>
> Although these may be inconsistent aims, I am interested in having a device that...
> - can measure particulate concentrations (and perhaps temperature and humidity levels)
> - is low cost, so that many devices can be deployed within a constrained project budget
> - has a good degree of concordance (at least qualitatively) with accurate stationary monitoring stations
> - can be read with an app on a cellphone or similar device (bluetooth?)
> - is convenient to the wearer of the device, e.g, is robust and has the ability to run on batteries for a few days at a time
>
> I anticipate that a user (or environmental health scientist or health care professional guiding the users) might want to record reading at several points during the day, not continuously.
>
> There are several projects I have found on the web focused on devices to quantify particulates in the air, but I don't think that they satisfy the aims above.
>
> The closest instrument that I could find is detailed in a PhD dissertation by David Holstius, a chapter of which I have posted here:
>
> https://www.dropbox.com/s/azs0dqzgpffppl1/Holstius2014dissertation_Ch3.pdf
>
> Because of my lack of training in the field of electronics, I do not know if the detection method, components, and overall design chosen by this individual would be appropriate as a starting platform to achieve the aims above.
>
> Any input or suggestions would be greatly appreciated.
>
> Thank you again.
>
> -gyro
>


Maybe this could be a starting point. They are cheap enough.

<https://www.google.com.au/search?q=sharp+dust+sensor>

Adrian Jansen

unread,
Jul 1, 2014, 1:19:43 AM7/1/14
to
On 1/7/2014 8:13 AM, Phil Hobbs wrote:

> The dust is pretty large compared with diesel smoke particles, so it
> doesn't penetrate very far into your lungs. They got rid of the
> asbestos awhile back, which was a pity. Chrysotile asbestos is nasty,
> serpentine basically isn't.
>
> Cheers
>
> Phil Hobbs
>
According to my reference, its serpentine ( blue, mostly ) asbestos
which is the bad one, chrysotile ( white ) asbestos is relatively safe.
Can you quote a source for the data ?

--
Regards,

Adrian Jansen adrianjansen at internode dot on dot net
Note reply address is invalid, convert address above to machine form.

daku...@gmail.com

unread,
Jul 1, 2014, 1:43:01 AM7/1/14
to
Hello Sir,
Thanks for the detailed description, it
appears that this is a large project,
and would need considerable time, effort
and money. There are two aspects --
detection and measurement. The first is
straightforward -- a simple phototransistoe,
would work as a detector - like in a
smoke detector. Measurement is tough,
since then your device has to be calibrated
against a test bench. What is the test-bench ?

Steve Wilson

unread,
Jul 1, 2014, 4:34:44 AM7/1/14
to
Jeff Liebermann <je...@cruzio.com> wrote:

> On Tue, 01 Jul 2014 00:24:21 GMT, Steve Wilson <no...@nospam.com>
> wrote:
>
>>> It's called a nephelmometer and uses light scattering. I have a
>>> Nikken #1394 Air Quality Monitor:
>>> <http://802.11junk.com/jeffl/crud/Nikken-AQM.jpg>
>
>>I could not find a user manual for your Nikken. If you have a pdf and
>>can spare the time, could you upload it to your web site?
>
> Sure. I'll scan it later tonite.

Please don't spend the time. I found a used one on eBay and I'm sure it
will come with a manual.

>>This is a very interesting way to use a particle counter. I have some
>>questions:
>>
>>How does the Nikken operate?
>
> My guess(tm) is by optical scattering. The light show is inside the
> black box with the air hose intake. See photos below. I didn't take
> apart the box, but with 4 wires going into the box, it's a likely
> guess.

I'll see if I can take it apart and get the actual circuit.

>>Is there a table showing the particle concentration vs size?
>
> No. The manual has a table by concentration:
> Level Color Particles/Liter
> 1 blue 500
> 2 yellow 1000
> 3 yellow 2000
> 4 yellow 3000
> 5 red 4000
> 6 red >5000
> I presume that it's measuring dust particles, not aerosols.

I'm interested to find the size range. I'll compare it with the Dylos and
see if that helps.
I know the stench of burning pcb. It goes everywhere. Unfortunately, I
also know the source. It is usually something I did.

> I also made the mistake of thrashing around the office, disturbing
> books and boxes laden with dust. Just touching a computah would cause
> it to unload an almost invisible cloud of dust into the air via the
> power supply or case fan. It was difficult to find the source without
> being able to touch or move anything.
>
> The overly sensitive Nikken AQM also created a problem. It would go
> tilt fairly quickly even at a fair distance from the source of smoke.
> I needed a smoke attenuator which was provided by restricting the air
> flow with a somewhat lint free rag and some duct tape.
>
> Once the procedure was established, it was very easy to find the
> source of the smoke.

>>I wonder how the Dylos and Nikken would compare in ability to
>>discriminate between different size particles. Does the Nikken have
>>more size ranges?
>
> No. The Nikken was designed for people with allergies or health
> issues and may be associated with aroma-therapy which is hinted in the
> terse manual. A look at their web pile will show their intended
> audience. It is NOT a precision instrument.
> <http://www.nikken.com>
> My guess(tm) is that it is simply looking for large particles, which
> comprise most of the lung irritants.

It is interesting that it responds to the particles from an overheated
pcb. I don't know what the size range would be, but that may be something
I could try when the unit arrives and I can compare it with the Dylos.

> Photos:
> <http://www.learnbydestroying.com/jeffl/Nikken%201394%20Air%20Quality%2
Thanks for the time and effort you put in to answer my questions.


Steve Wilson

unread,
Jul 1, 2014, 4:44:42 AM7/1/14
to
Jeff Liebermann <je...@cruzio.com> wrote:

> On Mon, 30 Jun 2014 21:13:17 -0700, Jeff Liebermann <je...@cruzio.com>
> wrote:

>>Photos:
>><http://www.learnbydestroying.com/jeffl/Nikken%201394%20Air%20Quality%2
>>0Monitor/>

> I dumped 3 PDF files into the above directory which are scans of the
> English language parts of the user manual.

Very interesting. I'm sorry I didn't catch you in time to save you the
effort. But now I am really interested in taking the thing apart to find
out how it works. Thanks to Phil's work, we have a pretty good idea what to
expect in the circuits. If those are present, it's a good lead to what's
inside the box.

I'll take some pictures if I can get in without destroying it.

Steve Wilson

unread,
Jul 1, 2014, 5:29:32 AM7/1/14
to
Jeff Liebermann <je...@cruzio.com> wrote:

> I dumped 3 PDF files into the above directory which are scans of the
> English language parts of the user manual.

Thanks again. Just for fun, I uploaded 3 files showing the operation of the
Dylos.

This shows a normal count in my apartment. The two counters agree to about
1%, which I found phenominal.

http://www.sed.comuf.com/particle/dsc09392.jpg

Here shows a higher count in a stairwell. Still very good agreement.

http://www.sed.comuf.com/particle/dsc09396.jpg

Here is a prototype particle filter I am working on. About 40dB reduction
in particle count. That's pretty good for 0.5 to 2.5 micron particles. They
are very hard to filter without a HEPA, which is very expensive for good
air flow.

http://www.sed.comuf.com/particle/3b9515e7.jpg


Steve Wilson

unread,
Jul 1, 2014, 5:56:20 AM7/1/14
to
Steve Wilson <no...@nospam.com> wrote:

PS, I'm mainly interested in the left column, which shows particle count
for sizes between 0.5 and 2.5 micron.

The right column shows particle sizes above 2.5 micron. I'm not so much
interested in these because they don't have the same tendency to lodge in
the lungs. They also tend to run about one-tenth the number of the
smaller particles, so if you know one, you automatically know the other.
Within statistical error, of course:)

I have no idea what the particle count would be for sizes below 0.5
micron.

There seems to be little information on the health effects, or what size
ranges are filtered by ordinary PM 100 air filters. The 3M description
covers a very broad range:

http://www.3m.com/product/information/P100-Particulate-Filter-
Respirator.html

The health effects seem to be ill-defined. Here is an excerpt from a
student lesson:

"The article �Particles in Practice: How Ultrafines Disseminate in the
Body� raises questions about how ultrafine particles smaller
than 100 nanometers (or 0.1 �m) are able to be absorbed into the body and
distributed in the cells. They do not appear to be
encapsulated by macrophages and seem to be entering cells and transported
throughout the body by other, yet unknown,
mechanisms. This has significant implications for the potential health
effects caused by particles in this size range."

https://www.niehs.nih.gov/health/assets/docs_a_e/ehp_student_edition_less
on_particles_size_makes_all_the_difference.pdf

These issues are very significant for the design of particle filters to
improve indoor air quality.

Phil Hobbs

unread,
Jul 1, 2014, 7:02:20 AM7/1/14
to
On 7/1/2014 1:19 AM, Adrian Jansen wrote:
> On 1/7/2014 8:13 AM, Phil Hobbs wrote:
>
>> The dust is pretty large compared with diesel smoke particles, so it
>> doesn't penetrate very far into your lungs. They got rid of the
>> asbestos awhile back, which was a pity. Chrysotile asbestos is nasty,
>> serpentine basically isn't.
>>
>> Cheers
>>
>> Phil Hobbs
>>
> According to my reference, its serpentine ( blue, mostly ) asbestos
> which is the bad one, chrysotile ( white ) asbestos is relatively safe.
> Can you quote a source for the data ?

Thanks for prodding me--looks like I was completely off track.

The Wikipedia page talks about chrysotile and serpentine being the same
thing, and amosite and crocidolite (which I'd never heard of) as the
nasty carcinogenic ones.

Of course even completely inert mineral needles will cause fibrosis if
you inhale enough of them.

Whether that's enough to justify forcing millions of drivers to use
inferior brake pads is a policy question. (Nowadays brake pads have to
warm up to achieve their maximum effect, whereas asbestos ones didn't.)

Phil Hobbs

unread,
Jul 1, 2014, 7:49:16 AM7/1/14
to
On 7/1/2014 12:58 AM, jurb...@gmail.com wrote:
>> "Unless you use reflection, like in an optical smoke detector. Then
>> 0%
> is the baseline. He could probably count individual particles"
>
> That may solve a problem but it creates another.
>
> Then the mirror has to be cleaned.

That's a perennial problem with optical sensors. I once talked with a
guy who made optical sensors for cracks in railway rails. Trains are
_dirty_, and the only way they were able to keep their optics clean was
to continuously blow compressed air out of the optical port.
>
> Now if this thing is allowed to have moving parts we can deal with
> that. In fact even a lens would have to be cleaned sooner or later
> but a mirror would of course be worse. Emitting surface, detecting
> surface and then reflecting surface. Bad enough without that
> reflector.
>
> But then, that is all up to the designer. Maybe it would be better.
> We are still a bit scant on details until - well - whatever.
>

John Larkin

unread,
Jul 1, 2014, 10:21:33 AM7/1/14
to
On Tue, 01 Jul 2014 15:19:43 +1000, Adrian Jansen <adr...@qq.vv.net> wrote:

>On 1/7/2014 8:13 AM, Phil Hobbs wrote:
>
>> The dust is pretty large compared with diesel smoke particles, so it
>> doesn't penetrate very far into your lungs. They got rid of the
>> asbestos awhile back, which was a pity. Chrysotile asbestos is nasty,
>> serpentine basically isn't.
>>
>> Cheers
>>
>> Phil Hobbs
>>
>According to my reference, its serpentine ( blue, mostly ) asbestos
>which is the bad one, chrysotile ( white ) asbestos is relatively safe.
> Can you quote a source for the data ?

Serpentine is California's State Rock. It's pretty, green with white streaks.




--

John Larkin Highland Technology Inc
www.highlandtechnology.com jlarkin at highlandtechnology dot com

Precision electronic instrumentation

John Larkin

unread,
Jul 1, 2014, 11:56:33 AM7/1/14
to
On Mon, 30 Jun 2014 21:58:58 -0700 (PDT), jurb...@gmail.com wrote:

>>"Unless you use reflection, like in an optical smoke detector. Then 0%
>is the baseline. He could probably count individual particles"
>
>That may solve a problem but it creates another.
>
>Then the mirror has to be cleaned.

Mirror? A laser or LED shoots out light and a photodiode catches light scattered
by particles. Shine a laser pointer across the room and see the dust or smoke
particles... no mirror involved.

Photoelectric smoke detectors work like that.

George Herold

unread,
Jul 1, 2014, 12:31:37 PM7/1/14
to
On Monday, June 30, 2014 7:23:04 PM UTC-4, Jeff Liebermann wrote:
> On Mon, 30 Jun 2014 13:52:57 -0700 (PDT), jurb...@gmail.com wrote:
>
>
>
> >You want to detect particulate matter in the air, and this
>
> >is going to be optically because you just do not have many other choices.
>
>
>
> It's called a nephelmometer and uses light scattering. I have a
>
> Nikken #1394 Air Quality Monitor:
>
> <http://802.11junk.com/jeffl/crud/Nikken-AQM.jpg>
>
> I don't think this model is sold any more. The photo is me using it
>
> to find the source of the burning smell in the office. It turned out
>
> to be an overheated resistor inside the UPS in the photo. The unit is
>
> quite sensitive. I can stamp my feet on the rug and cause the graph
>
> to almost go full scale from the dust. Same with a couch or chair.
>
> You could probably clone the function quite easily.
>
>
>
> More:
>
> <http://www.nwcg.gov/pms/pubs/SMG/177-186.pdf>

Oh good, I was going to say that besides optical you could
also pass the air though a filter and then see/weight what the filter collects
it's not that fast though.

George H.

Mikko OH2HVJ

unread,
Jul 1, 2014, 2:56:23 PM7/1/14
to
Jeff Liebermann <je...@cruzio.com> writes:

> On Mon, 30 Jun 2014 13:52:57 -0700 (PDT), jurb...@gmail.com wrote:
>
>>You want to detect particulate matter in the air, and this
>>is going to be optically because you just do not have many other choices.
>
> It's called a nephelmometer and uses light scattering. I have a
> Nikken #1394 Air Quality Monitor:
> <http://802.11junk.com/jeffl/crud/Nikken-AQM.jpg>

Nice pictures and especially I like the domain name! Could the
metal tunnel be an electrometer circuit ? The center plate might be
connected to the LMC6001, which is quite common in aerosol
electrometers. What's inside the black box marked with 'CAM-003' ?

--
Mikko OH2HVJ

Jeff Liebermann

unread,
Jul 1, 2014, 3:43:22 PM7/1/14
to
On Tue, 01 Jul 2014 21:56:23 +0300, Mikko OH2HVJ
<mikko.sy...@nospam.fi> wrote:

>Jeff Liebermann <je...@cruzio.com> writes:
>
>> On Mon, 30 Jun 2014 13:52:57 -0700 (PDT), jurb...@gmail.com wrote:
>>
>>>You want to detect particulate matter in the air, and this
>>>is going to be optically because you just do not have many other choices.
>>
>> It's called a nephelmometer and uses light scattering. I have a
>> Nikken #1394 Air Quality Monitor:
>> <http://802.11junk.com/jeffl/crud/Nikken-AQM.jpg>

>Nice pictures and especially I like the domain name! Could the
>metal tunnel be an electrometer circuit ?

Yes, it could but I have no clue how one would measure dust
accumulation using an electrometer. Besides, the forward mounts for
the insulate plate uses nylon insulators, which are hygroscopic, and
would leak badly if uses as an electrometer. Yet, there's a very low
input current op amp on the board. If some spare time finds me, I'll
investigate the patents and maybe try to reverse engineer the PCB.

>The center plate might be
>connected to the LMC6001, which is quite common in aerosol
>electrometers.

<http://www.ti.com.cn/general/cn/docs/lit/getliterature.tsp?genericPartNumber=lmc6001&fileType=pdf>

>What's inside the black box marked with 'CAM-003' ?

I don't know. I have to go back inside anyway. I just noticed on one
of my photos that I forgot to reinsert all the connectors. Argh.

Jon Elson

unread,
Jul 1, 2014, 5:18:41 PM7/1/14
to
RobertMacy wrote:

> On Mon, 30 Jun 2014 13:43:17 -0700, John Larkin
> <jla...@highlandtechnology.com> wrote:
>
>>> ...snip...
>>
>> Cars aren't bad. Diesels, in busses, trucks, and construction/farm
>> machinery, are.
>>
>> I'm a bit skeptical about the evils of CO2, but particulates are nasty
>> for people and the planet.
>>
>> Particulates melt snow, which is especially evil.
>
> Cars aren't bad. What about thousands of brake linings turning to powedery
> dust at a Stop sign ??
Get a hybrid. Generally, the brakes last over 100K miles before needing
replacement. They are only used to stop the car from below 9 MPH or
so.

Jon

Jon Elson

unread,
Jul 1, 2014, 5:25:35 PM7/1/14
to
jurb...@gmail.com wrote:


> Take a calibrated light source. Then a detector. Whatever drop there is
> from the light source to the detector is the result. This is no all that
> easy. What's more, I doubt you will be able to do it on a smart phone.
> This requires hardware.
So, a far better way is to shoot a narrow beam of light down a
dark tube, and have a sensor slightly off to the side of the beam,
looking toward the source. The forward scattering of light is vastly
stronger than the backwards reflection. But, it is very dependent on
the particle size and species. So, you need a calibration factor
for the type. This is how photoelectric smoke detectors work, and with
a collimated laser that is pulsed, you can get a lot of light to
work with.

Jon

Jon Elson

unread,
Jul 1, 2014, 5:29:42 PM7/1/14
to
John Larkin wrote:


> Unless you use reflection, like in an optical smoke detector. Then 0%
> is the baseline. He could probably count individual particles.
>
>
If you want to count particles, you need a VERY small beam spot, so
you focus a beam down to a needle point, and also focus the detector
on the same spot, usually at right angles. Then, you can put this
in a chamber with a mild vacuum pump, and have the air sample introduced
through micro-sized tubing, like big hypodermic tube. This can focus
a jet of sample air into the focal region of the source and detector.
You will get microsecond pulses as the particles zip by.

Jon

Jon Elson

unread,
Jul 1, 2014, 6:01:09 PM7/1/14
to
Jeff Liebermann wrote:


> Yes, it could but I have no clue how one would measure dust
> accumulation using an electrometer. Besides, the forward mounts for
> the insulate plate uses nylon insulators, which are hygroscopic, and
> would leak badly if uses as an electrometer. Yet, there's a very low
> input current op amp on the board. If some spare time finds me, I'll
> investigate the patents and maybe try to reverse engineer the PCB.
You might look for "aerosol charger" related to some work we did
about 40 years ago. You need a radioactive source, although the ones in
home smoke detectors are probably usable. The beta particles (excited
electrons) ionize the air sample, which charges the particles. These are
then trapped in a filter, and the charge bleeds off to an electrometer.
We smoked the filters with a soot source to make the filter more conductive,
then had to insulate the filter from the vacuum system so the only path
was through the electrometer. This was a bit tricky, we used a thin mylar
film, IIRC. You have to use a specific electrometer op amp and one
of those Victoreen glass-encased gigiOhm resistors.

Jon

Jon Elson

unread,
Jul 1, 2014, 6:06:08 PM7/1/14
to
daku...@gmail.com wrote:


> Thanks for the detailed description, it
> appears that this is a large project,
> and would need considerable time, effort
> and money. There are two aspects --
> detection and measurement. The first is
> straightforward -- a simple phototransistoe,
> would work as a detector - like in a
> smoke detector. Measurement is tough,
> since then your device has to be calibrated
> against a test bench. What is the test-bench ?
You can create uniform 1 um particles with di-octyl
pthalate, kind of a relative of the stuff that makes
the hot butter smell at the movie theater popcorn
machine. Then, you need a particle counter, as we
have discussed. if you know particle size and the count
in some measured air sample, then you can compute the
particle density in the ambient air, by count, mass/volume
and other measures.

There are other schemes for generating known particle density.

All this should be readily searchable in the air pollution
literature from 40 years ago.

Jon

k...@attt.bizz

unread,
Jul 1, 2014, 6:22:12 PM7/1/14
to
On Tue, 01 Jul 2014 16:18:41 -0500, Jon Elson <jme...@wustl.edu>
wrote:
No thanks. Brakes are cheap.

Phil Hobbs

unread,
Jul 1, 2014, 7:36:44 PM7/1/14
to
The best method that I know of is to use a very dilute solution of
polystyrene latex spheres. They're grown from solution, and are very
uniform in size. You dilute them in very pure water (18.2 megohm-cm
deionized, and filtered with an 0.014 um ultrafilter), with a very small
amount of very clean surfactant (Triton from Fisher Scientific is the
best), then atomize the solution with a medical aspirator (as used e.g.
for administering Ventolin to pulmonary patients).

The water droplets form two populations: those drops containing one
sphere, and those containing none. (You adjust the concentration so
that doublets are very rare.) Because the water is so clean, the
empties dry down very very small, leaving basically just the individual
spheres floating round.

Adrian Jansen

unread,
Jul 1, 2014, 8:04:48 PM7/1/14
to
On 1/7/2014 9:02 PM, Phil Hobbs wrote:
> On 7/1/2014 1:19 AM, Adrian Jansen wrote:
>> On 1/7/2014 8:13 AM, Phil Hobbs wrote:
>>
>>> The dust is pretty large compared with diesel smoke particles, so it
>>> doesn't penetrate very far into your lungs. They got rid of the
>>> asbestos awhile back, which was a pity. Chrysotile asbestos is nasty,
>>> serpentine basically isn't.
>>>
>>> Cheers
>>>
>>> Phil Hobbs
>>>
>> According to my reference, its serpentine ( blue, mostly ) asbestos
>> which is the bad one, chrysotile ( white ) asbestos is relatively safe.
>> Can you quote a source for the data ?
>
> Thanks for prodding me--looks like I was completely off track.
>
> The Wikipedia page talks about chrysotile and serpentine being the same
> thing, and amosite and crocidolite (which I'd never heard of) as the
> nasty carcinogenic ones.
>
> Of course even completely inert mineral needles will cause fibrosis if
> you inhale enough of them.
>
> Whether that's enough to justify forcing millions of drivers to use
> inferior brake pads is a policy question. (Nowadays brake pads have to
> warm up to achieve their maximum effect, whereas asbestos ones didn't.)
>
> Cheers
>
> Phil Hobbs
>
The Wiki is wrong there. Serpentine and chrysotile are two entirely
different minerals, with different chemical composition, and biological
reactions. Just why they are both classed as "asbestos", and treated as
the same evil thing is just one of those mysteries which seems to stick
around forever.

rickman

unread,
Jul 1, 2014, 9:34:43 PM7/1/14
to
I worked for the Consumer Product Safety Commission at the time this
regulation was being developed. At that time they were not sure just
want caused lung disease. I remember my boss was attending meetings
with EPA working on this. They seemed to feel the problem was fibers
rather than the material itself, but they didn't know just what
constituted a fiber. He said one prominent researcher defined a fiber
as anything 3 times as long as it is wide. Another researcher responded
that his car was a fiber under that definition. lol

People in this group love to criticize anyone at any time. We know a
lot more now and have the huge advantage of not being under the gun to
save lives which is exactly what they were mostly concerned with at that
time. They couldn't delay while they waited for more studies, so they
did the best they could with what they had.

If anyone feels the laws need to be reworked, then get involved and help
shape some new legislation.

I find especially ridiculous the idea that today's brake pads don't stop
the vehicles well enough. lol. I remember the junk being made in the
US in the early 70's. Today's cars are the safest we have ever driven.

--

Rick

Phil Hobbs

unread,
Jul 1, 2014, 9:41:33 PM7/1/14
to
Metal brake pads have to warm up to work properly, whereas asbestos ones
didn't. Of course, anyone who ever uses "lol" for anything might not
understand actual data.

John Larkin

unread,
Jul 2, 2014, 12:57:25 AM7/2/14
to
Yeah, beats hauling a ton of batteries up and down hills.

Jan Panteltje

unread,
Jul 2, 2014, 1:36:08 AM7/2/14
to
rickman:
>Today's cars are the safest we have ever driven.

Defective car keys, exploding airbags, engine controllers sensitive to RF.
batteries on fire.
Poisonous cooling substances.
The list is endless.

You are wrong.

John Devereux

unread,
Jul 2, 2014, 3:33:21 AM7/2/14
to
That is one of your dafter posts Jan...


--

John Devereux

gyroma...@gmail.com

unread,
Jul 2, 2014, 8:22:09 AM7/2/14
to
On Monday, June 30, 2014 2:52:57 PM UTC-6, jurb...@gmail.com wrote:
> You forgot to put that in your public directory so we cannot see it yet. I do have a better idea what you are talking about now.
>
>
>
> You want to detect particulate matter in the air, and this is going to be optically because you just do not have many other choices.
>
>
>
> Here is the problem, mathematically, from zero to 100 %, you must define 100 % as zero. Totaslly clean air is the setpoint, the zero, the reference. If the nibers do not start there, it will be hard to use them. It's like the audio example before, you don't say 99.7 % NOT distortion, you say 0.3 % distortion. Kinda flipped over. This is exactly the same in this case.
>
>
>
> Take a calibrated light source. Then a detector. Whatever drop there is from the light source to the detector is the result. This is no all that easy. What's more, I doubt you will be able to do it on a smart phone. This requires hardware.
>
>
>
> Now comes the $64 question, what kind of numbers do you want ? PPM of corbon particulates ? I would almost venture to say there is an off the shelf solution for that. In fact, what abput smoke detectors, but instead of feeding an alrm, feed it to meter or whatever. A buffer to get the levels you wanrt and then something to display the result.
>
>
>
> If you need accuracy, then we are talking instrumentation. That is a bit harder than building a (n old style) TV set. I can do it a little bit but I am not great at it. I got a Wavetek 111 and I am not only impressed by its technology, but damn near enamoured. Fucker makes a sine wave out of a triangle at .05 % THD ! Frequency independent. Has a cool type of feedback circuit. It requies the input to be EXACLT the amplitude it expects. I just got into it like last year, but the damn thing was builtin 1 1970 !
>
>
>
> Damn I am behind the times.
>
>
>
> So, with your thing, I figure you should set the setpoint in a vacuum. Start with that. Whatever the output is at that point from the optical sensor then, consider that the setpoint.
>
>
>
> Circuitry ? Well that is kinda like knowing when to use a PNP transistor instead of an NPN. It is flipped. Upside down or backwards, whatevr you want to call it.



I am sorry for the mistake. Please let me know if you can access the document here:

http://goo.gl/IBhTZW

( https://dl.dropboxusercontent.com/u/557195/air_quality_sensors_papers/Holstius2014dissertation_Ch3.pdf )

Thank you all for your very interesting comments in this thread.

-gyro

Massoud

unread,
Jul 2, 2014, 12:52:11 PM7/2/14
to
gyroma...@gmail.com wrote in
news:727ea200-b117-4942...@googlegroups.com:

...
>
> Because of my lack of training in the field of electronics, I do not
> know if the detection method, components, and overall design chosen by
> this individual would be appropriate as a starting platform to achieve
> the aims above.
>
> Any input or suggestions would be greatly appreciated.
>
> Thank you again.
>
> -gyro


Check "Dylos dc1100 pro." It is cheap and reliable.


Mass.









--- news://freenews.netfront.net/ - complaints: ne...@netfront.net ---

haitic...@gmail.com

unread,
Jul 2, 2014, 3:27:53 PM7/2/14
to
On Monday, June 30, 2014 9:05:28 PM UTC-4, Phil Hobbs wrote:
> On 6/30/2014 7:57 PM, Bill Sloman wrote:
>
> > On Tuesday, 1 July 2014 06:58:53 UTC+10, jurb...@gmail.com wrote:
>
> >> Another thing is the wavelength of light you use in the detector.
>
> >> If you define the subject as particulate matter that is VISIBLE,
>
> >> that is one thing. you could use UV or IR.
>
> >>
>
> >> the main question here is - what results do you want ? Exactly.
>
> >
>
> > http://en.wikipedia.org/wiki/Light_scattering
>
> >
>
> > Rayleigh scattering is intensely wavelength dependent - to the sixth
>
> > power of the particle diameter, and inversely proportional to the
>
> > fourth power of wavelength.
>
> >
>
> > It would be fun to use a bunch of light sources to measure scattering
>
> > at a number of different wavelengths, and deconvolute the results to
>
> > say something about the concentrations and sizes of the different
>
> > particles doing the scattering.
>
> >
>
> > You'd need Phil Hobbs to make it work ...
>
> >
>
>
>
> Back in my mis-spent youth (or maybe early middle age) I did a lot of
>
> particle counting work. Not exactly that kind, but particle mapping in
>
> 6 dimensions (x, y, z, radial velocity, size, and time), and
>
> collaborated a bit with some colleagues who were doing composition
>
> sorting for particles in fluids, using the complex refractive index of
>
> the particle to sort out bubbles, metals, nonmetals, and carbon.
>
>
>
> At the moment I'm doing some higher speed stuff of the same sort, but
>
> it's kind of slow going because I don't have the mechanical design and
>
> fabrication facilities I really need for this.
>
>
>
> Cheers
>
>
>
> Phil Hobbs
>
>
>
>
>
> --
>
> Dr Philip C D Hobbs
>
> Principal Consultant
>
> ElectroOptical Innovations LLC
>
> Optics, Electro-optics, Photonics, Analog Electronics
>
>
>
> 160 North State Road #203
>
> Briarcliff Manor NY 10510
>
>
>
> hobbs at electrooptical dot net
>
> http://electrooptical.net

Phil
What kind of mechanical design facilities are you looking for?
jb

gyroma...@gmail.com

unread,
Jul 2, 2014, 4:20:29 PM7/2/14
to
On Wednesday, July 2, 2014 10:52:11 AM UTC-6, Massoud wrote:
> ...
>
> >
>
> > Because of my lack of training in the field of electronics, I do not know if the detection method, components, and overall design chosen by this individual would be appropriate as a starting platform to achieve the aims above.
>
> >
>
> > Any input or suggestions would be greatly appreciated.
>
> >
>
> > Thank you again.
>
> >
>
> > -gyro

>
> Check "Dylos dc1100 pro." It is cheap and reliable.>
>
>
> Mass.
>


Thank you for the suggestion.

I believe that for my application, the price point of the Dylos is too high for widespread deployment. Also, as mentioned in my other post, it is desirable that the unit be able to be read (via bluetooth?) by mobile devices, such as cellphones.

Phil Hobbs

unread,
Jul 2, 2014, 4:23:04 PM7/2/14
to
> Phil
> What kind of mechanical design facilities are you looking for?
> jb
>

It's more the fabrication that's the issue. Aligning stuff to a few arc
seconds with ball-screw tilt stages is a mug's game.

josephkk

unread,
Jul 2, 2014, 11:19:16 PM7/2/14
to
On Mon, 30 Jun 2014 18:13:50 -0400, Phil Hobbs <ho...@electrooptical.net>
wrote:

>On 6/30/2014 5:22 PM, RobertMacy wrote:
>> On Mon, 30 Jun 2014 13:43:17 -0700, John Larkin
>> <jla...@highlandtechnology.com> wrote:
>>
>>>> ...snip...
>>>
>>> Cars aren't bad. Diesels, in busses, trucks, and construction/farm
>>> machinery, are.
>>>
>>> I'm a bit skeptical about the evils of CO2, but particulates are nasty
>>> for people and the planet.
>>>
>>> Particulates melt snow, which is especially evil.
>>
>> Cars aren't bad. What about thousands of brake linings turning to
>> powedery dust at a Stop sign ??
>
>The dust is pretty large compared with diesel smoke particles, so it
>doesn't penetrate very far into your lungs. They got rid of the
>asbestos awhile back, which was a pity. Chrysotile asbestos is nasty,
>serpentine basically isn't.
>
>Cheers
>
>Phil Hobbs

Be glad if you don't live in an area with naturally occurring asbestos. I
live near lots of them.

?-)

josephkk

unread,
Jul 2, 2014, 11:31:10 PM7/2/14
to
On Tue, 01 Jul 2014 21:41:33 -0400, Phil Hobbs <ho...@electrooptical.net>
wrote:

>
>> If anyone feels the laws need to be reworked, then get involved and help
>> shape some new legislation.

Actually the current legislation is nearly correct. It is set up so that
regulation does the adapting, by leveraging the best repeatable science
available.
>>
>> I find especially ridiculous the idea that today's brake pads don't stop
>> the vehicles well enough. lol. I remember the junk being made in the
>> US in the early 70's. Today's cars are the safest we have ever driven.
>>
>
>Metal brake pads have to warm up to work properly, whereas asbestos ones
>didn't. Of course, anyone who ever uses "lol" for anything might not
>understand actual data.

Metal in brake pads came in well before asbestos went out. Its
superiority in thermal fading resistance to non-metal brake pads made it a
market winner.
>
>Cheers
>
>Phil Hobbs

Phil Hobbs

unread,
Jul 3, 2014, 9:25:20 AM7/3/14
to
Brake fade isn't caused primarily by hot pads, it's caused by boiling
brake fluid. And earlier metal pads also contained asbestos.

Cheers

Phil

haitic...@gmail.com

unread,
Jul 3, 2014, 9:44:56 AM7/3/14
to
On Monday, June 30, 2014 3:05:42 PM UTC-4, gyroma...@gmail.com wrote:
> Hi,
>
>
>
> This message is related to my earlier request for advice in the topic 'finding electronics design and fabrication expertise for a project'.
>
>
>
> Because of the kindness of the group members here, I am posting more details about the project and aims.
>
>
>
> The overall aim is to develop and deploy inexpensive 'personal' air pollution monitors. For the environment in which I am interested, a major pollutant concern is particulates arising from biomass combustion, coal burning, and car exhaust.
>
>
>
> Although these may be inconsistent aims, I am interested in having a device that...
>
> - can measure particulate concentrations (and perhaps temperature and humidity levels)
>
> - is low cost, so that many devices can be deployed within a constrained project budget
>
> - has a good degree of concordance (at least qualitatively) with accurate stationary monitoring stations
>
> - can be read with an app on a cellphone or similar device (bluetooth?)
>
> - is convenient to the wearer of the device, e.g, is robust and has the ability to run on batteries for a few days at a time
>
>
>
> I anticipate that a user (or environmental health scientist or health care professional guiding the users) might want to record reading at several points during the day, not continuously.
>
>
>
> There are several projects I have found on the web focused on devices to quantify particulates in the air, but I don't think that they satisfy the aims above.
>
>
>
> The closest instrument that I could find is detailed in a PhD dissertation by David Holstius, a chapter of which I have posted here:
>
>
>
> https://www.dropbox.com/s/azs0dqzgpffppl1/Holstius2014dissertation_Ch3.pdf
>
>
>
> Because of my lack of training in the field of electronics, I do not know if the detection method, components, and overall design chosen by this individual would be appropriate as a starting platform to achieve the aims above.
>
>
>
> Any input or suggestions would be greatly appreciated.
>
>
>
> Thank you again.
>
>
>
> -gyro

I encourage you to talk with the docs on the phone at the NIH SBIR National
Heart Lung Institute, as I remember. You said you were not skilled at
electronic design - these docs know even less. The NCI also a possibility.
It's going to take the resources to develop what you want - The docs are quite
encouraging, often.

And, whether or not you know it, you are a legitimate developer of low cost
particulate detection systems. If you talk with the docs on the phone, you
would be primarily asking them questions, such as, "Is this a legitimate
topic to apply for?" and "Do I need to partner with a researcher in this field?"
(probably both yes.)


k...@attt.bizz

unread,
Jul 3, 2014, 9:58:49 AM7/3/14
to
Natural asbestos isn't a big problem. It's the results of the
machining of the stuff that makes the nasties. Rocks don't often get
into the lungs.

George Herold

unread,
Jul 3, 2014, 12:45:24 PM7/3/14
to
On Wednesday, July 2, 2014 4:23:04 PM UTC-4, Phil Hobbs wrote:
> On 07/02/2014 03:27 PM, haitic...@gmail.com wrote:
>
<snip>
> > Phil
> > What kind of mechanical design facilities are you looking for?
> > jb
>
> It's more the fabrication that's the issue. Aligning stuff to a few arc
> seconds with ball-screw tilt stages is a mug's game.

I'm reluctant to post 'cause you are most likely three thoughts ahead of me.. but anyway,

There are some wickedly fine pitched threads now.

longer lever arm

Double micrometer..(is that the right name?)
(It's got tick marks of 1 micron (or something like that.. I can look it up.))

Use a piezo stack and turn a pot. The quick and dirty way to do this
is just to stick the piezo underneath the tip of the ball-tipped adjustment screw.
AE0203D04F is what we use.
Here?
http://store.bravoelectro.com/ceramicpiezoactuators-ae0203d04fbrpiezoactuatormaxdisplacement46m2x3x5mm-p-1502.html

George H.

Jeff Liebermann

unread,
Jul 3, 2014, 8:22:07 PM7/3/14
to

Steve Wilson

unread,
Jul 3, 2014, 10:44:38 PM7/3/14
to
Jeff Liebermann <je...@cruzio.com> wrote:

> On Mon, 30 Jun 2014 21:13:17 -0700, Jeff Liebermann <je...@cruzio.com>
> wrote:
>
> I added 2 more photos of the inside of the light box to the web pile:
> <http://www.learnbydestroying.com/jeffl/Nikken%201394%20Air%20Quality%2
> 0Monitor/Nikken-AQM-08.jpg>
> <http://www.learnbydestroying.com/jeffl/Nikken%201394%20Air%20Quality%2
> 0Monitor/Nikken-AQM-09.jpg
> <http://www.learnbydestroying.com/jeffl/Nikken%201394%20Air%20Quality%2
> 0Monitor/> Looks like it's an optical scattering system.

Thanks. That is very interesting. In Nikken-AQM-09.jpg, the BA10324A
looks like a slow op amp, similar to the LM358

http://datasheet.octopart.com/BA10324A-Rohm-datasheet-8321777.pdf

I'm surprised the particles really move that slowly through the focal
point that the BA10324A can follow them.

I see the symbol for Q2, but I can't find Q1. Did they miscount, or is it
somewhere on the other side of the board?

D1 and D2 look like dual diodes. Is this some kind of log converter? Your
photos are excellent, but it's kind of hard to trace the circuit looking
at black traces on a black soldermask. I'll wait until mine arrives so I
can trace the circuit.

Maybe an optical illusion, but the board seems very thick looking at the
two screw holes.

I'm glad you were able to get it apart without destroying it. That means
I can try monitoring the signals.

Mine should arrive next week and I'll compare the readings with the
Dylos.

It would be nice if the Nikken can get down to 0.5 micron. It is battery
powered and would be a lot more convenient than lugging the Dylos around
with the power cord. It's a bit awkward trying to find an AC outlet,
especially if it's far from where I want to monitor.

Very interesting. Thanks, Jeff

Jeff Liebermann

unread,
Jul 4, 2014, 12:19:21 AM7/4/14
to
On Fri, 04 Jul 2014 02:44:38 GMT, Steve Wilson <no...@nospam.com>
wrote:

>Jeff Liebermann <je...@cruzio.com> wrote:
>
>> On Mon, 30 Jun 2014 21:13:17 -0700, Jeff Liebermann <je...@cruzio.com>
>> wrote:
>>
>> I added 2 more photos of the inside of the light box to the web pile:
>> <http://www.learnbydestroying.com/jeffl/Nikken%201394%20Air%20Quality%2
>> 0Monitor/Nikken-AQM-08.jpg>
>> <http://www.learnbydestroying.com/jeffl/Nikken%201394%20Air%20Quality%2
>> 0Monitor/Nikken-AQM-09.jpg
>> <http://www.learnbydestroying.com/jeffl/Nikken%201394%20Air%20Quality%2
>> 0Monitor/> Looks like it's an optical scattering system.

>Thanks. That is very interesting. In Nikken-AQM-09.jpg, the BA10324A
>looks like a slow op amp, similar to the LM358
>http://datasheet.octopart.com/BA10324A-Rohm-datasheet-8321777.pdf

LM358 is a dual op amp. The BA10324A is a quad op amp. Probably more
similar to an LM324.
<http://www.ti.com/lit/ds/symlink/lm324ka.pdf>

>I'm surprised the particles really move that slowly through the focal
>point that the BA10324A can follow them.

Note the two lenses on the IR emitter and detector. It looks like
they focus the light to a point in the middle of the circular area
where the tube delivers air flow. It's probably difficult to detect
with a diffused light source, but one that's concentrated almost to a
point should be fairly simple (famous last assumptions).

>I see the symbol for Q2, but I can't find Q1. Did they miscount, or is it
>somewhere on the other side of the board?

Probably other side. I couldn't get to the back side of the PCB as
the leads for the IR emitter and detector were both soldered from the
bottom of the board. There may be a mess of components on the circuit
side but I could tell without unsoldering.

>D1 and D2 look like dual diodes. Is this some kind of log converter?

No clue and I've already reassembled the device (this time plugging in
the connector I forgot last time).

>Your
>photos are excellent, but it's kind of hard to trace the circuit looking
>at black traces on a black soldermask. I'll wait until mine arrives so I
>can trace the circuit.

I'm still trying to figure out what the insulated plate in the middle
of the air duct is doing. I forgot to see if there's any high
voltage when I had it apart.

>Maybe an optical illusion, but the board seems very thick looking at the
>two screw holes.

I didn't measure it but it looked like ordinary 0.032 inch G10/FR4.

>I'm glad you were able to get it apart without destroying it. That means
>I can try monitoring the signals.

Yep. It was something to do while Windoze 8.1 was slowly installing.

>Mine should arrive next week and I'll compare the readings with the
>Dylos.

Cheap on eBay for about $30.

>It would be nice if the Nikken can get down to 0.5 micron.

I could probably blow some dust into it and see. However, my
guess(tm) that is works best with large particles that scatter the
light better.

>It is battery
>powered and would be a lot more convenient than lugging the Dylos around
>with the power cord. It's a bit awkward trying to find an AC outlet,
>especially if it's far from where I want to monitor.

Yep. 3.7v Li-Ion cell phone battery which is easily replaceable.

>Very interesting. Thanks, Jeff

Steve Wilson

unread,
Jul 4, 2014, 1:14:45 AM7/4/14
to
Jeff Liebermann <je...@cruzio.com> wrote:

> On Fri, 04 Jul 2014 02:44:38 GMT, Steve Wilson <no...@nospam.com>
> wrote:

>>Thanks. That is very interesting. In Nikken-AQM-09.jpg, the BA10324A
>>looks like a slow op amp, similar to the LM358
>>http://datasheet.octopart.com/BA10324A-Rohm-datasheet-8321777.pdf
>
> LM358 is a dual op amp. The BA10324A is a quad op amp. Probably more
> similar to an LM324.
> <http://www.ti.com/lit/ds/symlink/lm324ka.pdf>

Picky. I said similar. The LM324 is a quad 358. I was talking about the
basic input structure and bandwidth. Differential PNP so the common mode
extends to zero volts, and low bandwidth.

>>I'm surprised the particles really move that slowly through the focal
>>point that the BA10324A can follow them.

> Note the two lenses on the IR emitter and detector. It looks like
> they focus the light to a point in the middle of the circular area
> where the tube delivers air flow. It's probably difficult to detect
> with a diffused light source, but one that's concentrated almost to a
> point should be fairly simple (famous last assumptions).

That's the issue. How fine is the focal point, and how fast do the
particles move through?

>>I see the symbol for Q2, but I can't find Q1. Did they miscount, or is
>>it somewhere on the other side of the board?

> Probably other side. I couldn't get to the back side of the PCB as
> the leads for the IR emitter and detector were both soldered from the
> bottom of the board. There may be a mess of components on the circuit
> side but I could tell without unsoldering.

Maybe wise. The IR emitter may be sensitive to ESD. But I'd expect some
vias to connect to the other side.

>>D1 and D2 look like dual diodes. Is this some kind of log converter?

> No clue and I've already reassembled the device (this time plugging in
> the connector I forgot last time).

I'll trace the circuit and see what it does.

> I'm still trying to figure out what the insulated plate in the middle
> of the air duct is doing. I forgot to see if there's any high
> voltage when I had it apart.

Yes, I thought that was strange. Wouldn't high voltage need a corona
generator upstream? Maybe it's a light baffle.

>>I'm glad you were able to get it apart without destroying it. That
>>means I can try monitoring the signals.
>
> Yep. It was something to do while Windoze 8.1 was slowly installing.

Tee hee. I'm running XP in a vm. Takes about 45 seconds to recover from a
crash that crosslinks clusters. Happened again today. No problem.

>>Mine should arrive next week and I'll compare the readings with the
>>Dylos.

> Cheap on eBay for about $30.

There were a half dozen or so when I looked. Some went as high as $400.
They are all gone now. Mine was $17.99 plus $16.34 shipping.

>>It would be nice if the Nikken can get down to 0.5 micron.

> I could probably blow some dust into it and see. However, my
> guess(tm) that is works best with large particles that scatter the
> light better.

The large particles don't do as much damage as the small ones, which
lodge in your lungs. According to the Dylos measurements, the small ones
are about ten times more numerous than the large ones. But I found a
different environment yesterday where the ratio was closer to 16%. That
should be telling me something, but I don't know what yet.

I'm really interested to see some waveforms. As far as I can tell, the
Dylos and Nikken may operate the same way, so I'm expecting similar
results. I'm hoping to be able to tweak the Nikken to get down to 0.3
micron or lower. Maybe add a comparator to exclude the large ones, and a
pot for variable thresholds.

Jeff Liebermann

unread,
Jul 4, 2014, 1:42:58 PM7/4/14
to
On Fri, 04 Jul 2014 05:14:45 GMT, Steve Wilson <no...@nospam.com>
wrote:

>That's the issue. How fine is the focal point, and how fast do the
>particles move through?

Well, the spot diameter could be measured. The emitter doesn't look
like anything special. Just an ordinary LED operating in near IR
region. You can see near IR with a digital camera (try it on your TV
remote control). A little cancerous cigarette smog, and the light
beam should be visible.

>Maybe wise. The IR emitter may be sensitive to ESD. But I'd expect some
>vias to connect to the other side.

I'm sure I could have removed it without much difficulty. I just
didn't think it was worth the risk. I'm more worried about trashing
the calibration than breaking anything.

>Yes, I thought that was strange. Wouldn't high voltage need a corona
>generator upstream? Maybe it's a light baffle.

Please note that it was a guess(tm) as to the purpose of the insulated
plate. I still don't understand its function as the plate is NOT in
the air flow going to the optical dust sensor. I'm fairly sure it's
not a high voltage system because of the nylon insulators and the lack
of any high voltage components on the PCB. More Googling, I guess.

How the IR dust sensor works:
Sharp:
<http://www.digikey.com/articles/techzone/2014/may/sniffing-the-air-sensors-for-monitoring-air-quality-and-safety>
<http://www.digikey.com/us/en/techzone/sensors/supplier/Sharp_Microelectronics__425.html>
<http://www.sharpsma.com/webfm_send/1488>
<https://www.sparkfun.com/products/9689>
<http://www.howmuchsnow.com/arduino/airquality/>

Shinyei:
<http://www.sca-shinyei.com/particlesensor>
<http://www.sca-shinyei.com/pdf/PPD42NS.pdf>
Taking it apart (with schematic):
<http://takingspace.org/wp-content/uploads/ShinyeiPPD42NS_Deconstruction_TracyAllen.pdf>
Outdoor air monitoring in Beijing:
<http://aqicn.org/sensor/shinyei/>
More of the same:
<https://www.google.com/#q=shinyei+PPD42ns>

>There were a half dozen or so when I looked. Some went as high as $400.
>They are all gone now. Mine was $17.99 plus $16.34 shipping.

That's what happens when people read this newsgroup.
There are still 3 for sale/auction:
http://www.ebay.com/itm/151346164608
http://www.ebay.com/itm/191220658083
http://www.ebay.com/itm/331247326123

>The large particles don't do as much damage as the small ones, which
>lodge in your lungs. According to the Dylos measurements, the small ones
>are about ten times more numerous than the large ones. But I found a
>different environment yesterday where the ratio was closer to 16%. That
>should be telling me something, but I don't know what yet.
>I'm really interested to see some waveforms.

According to my random reading, one can distinguish between various
types of dust sources by the waveform.

>As far as I can tell, the
>Dylos and Nikken may operate the same way, so I'm expecting similar
>results. I'm hoping to be able to tweak the Nikken to get down to 0.3
>micron or lower. Maybe add a comparator to exclude the large ones, and a
>pot for variable thresholds.

That's down to the size of mold and possibly bacteria.

You can differentiate particle size by pulse width. The Shinyei
detector claims 1 micron minimum on the data sheet. My guess(tm) is
that something in either the amplifiers that follow or the air speed
limit the upper frequency response. The faster the particle goes by,
the narrower the pulse. You might be able to slow down the air flow
in order to "stretch" the pulse width.

Another possible way might be to artificially increase the particle
size by making small particles clump together as in an electrostatic
precipitator. Equal air flow through two different ducts. One duct
would give the dust particles a positive charge, the other negative.
When recombined, they would clump together forming roughly spherical
ummm... clumps, which should produce a much larger pulse width.

Good luck.

Steve Wilson

unread,
Jul 6, 2014, 6:20:02 AM7/6/14
to
> On Fri, 04 Jul 2014 05:14:45 GMT, Steve Wilson <no...@nospam.com>
> wrote:

>> That's the issue. How fine is the focal point, and how fast do
>> the particles move through?

> Well, the spot diameter could be measured. The emitter doesn't
> look like anything special. Just an ordinary LED operating in near
> IR region. You can see near IR with a digital camera (try it on
> your TV remote control). A little cancerous cigarette smog, and
> the light beam should be visible.

I wonder why they want to use near ir. I would think blue, as in
Blu-Ray, or a uv led would be better for catching the small
particles that do the real damage.

OTOH, it might be possible to use an RGB led and get more
information about the size distribution by changing the color.

>> Maybe wise. The IR emitter may be sensitive to ESD. But I'd
>> expect some vias to connect to the other side.

> I'm sure I could have removed it without much difficulty. I just
> didn't think it was worth the risk. I'm more worried about
> trashing the calibration than breaking anything.

I will calibrate the unit against the Dylos. Then I can do whatever
I want and check it against the Dylos to verify the calibration.

> How the IR dust sensor works:

> Sharp:

> <http://www.digikey.com/articles/techzone/2014/may/sniffing-the-air-
sensors-for-monitoring-air-quality-and-safety>
> <http://www.digikey.com/us/en/techzone/sensors/supplier/Sharp_Microelec
tronics__425.html>
> <http://www.sharpsma.com/webfm_send/1488>
> <https://www.sparkfun.com/products/9689>
> <http://www.howmuchsnow.com/arduino/airquality/>

> Shinyei:

> <http://www.sca-shinyei.com/particlesensor>
> <http://www.sca-shinyei.com/pdf/PPD42NS.pdf> Taking it apart (with
> schematic):

> <http://takingspace.org/wp-
content/uploads/ShinyeiPPD42NS_Deconstruction_TracyAllen.pdf>
> Outdoor air monitoring in Beijing:

> <http://aqicn.org/sensor/shinyei/> More of the same:

These are extremely valuable links. Thanks for taking the time to
search and post them.

I'm always amazed at how you can find so many useful links in such a
short time. You should start classes and show us how you do it.

> According to my random reading, one can distinguish between
> various types of dust sources by the waveform.

It would be very interesting if you had any links. I'd like to find
out what this stuff is and where it comes from.

>> As far as I can tell, the Dylos and Nikken may operate the same
>> way, so I'm expecting similar results. I'm hoping to be able to
>> tweak the Nikken to get down to 0.3 micron or lower. Maybe add a
>> comparator to exclude the large ones, and a pot for variable
>> thresholds.

> That's down to the size of mold and possibly bacteria.

I found you can go even lower, down to 3 nm and below. Here's one
example using condensation:

"Measuring d<3 nm particles with a Condensation Particle Counter"

ftp://ftp.nilu.no/ACCENT/files/sgro1.pdf

Mold spores, bacteria and viruses are serious health hazards. The
problem is how to distinguish between the pathogens and ordinary
debris.

OTOH, it may be better to concentrate on making a particle filter
that simply strips everything from the room air. Then it doesn't
matter what they are.

But I need a particle counter that goes below the 0.5 um of the
Dylos to verify the performance of the filter. The Nikken looks like
an ideal platform for experimentation. It has a light chamber that
is easy to get to, dual lens for better light handling, battery
power for portable operation, plenty of room for additional
circuits, probably easy to add USB for data transfer, and so on.

I'm really pleased and grateful that you posted the original info on
the burning pcb. It has led to much greater understanding of the
problem, and I will soon have a Nikken in my hands for further work.

> You can differentiate particle size by pulse width. The Shinyei
> detector claims 1 micron minimum on the data sheet. My guess(tm)
> is that something in either the amplifiers that follow or the air
> speed limit the upper frequency response. The faster the particle
> goes by, the narrower the pulse. You might be able to slow down
> the air flow in order to "stretch" the pulse width.

It would be very hard to maintain calibration using pulse width. The
calibration would depend on air velocity, which would depend on the
fan speed, how dirty the blades were, bearing friction, turbulence,
and so on.

As Phil mentioned earlier, pulse amplitude is pretty universal in
the high end counters, and would be much easier to give stable
results.

> Another possible way might be to artificially increase the
> particle size by making small particles clump together as in an
> electrostatic precipitator. Equal air flow through two different
> ducts. One duct would give the dust particles a positive charge,
> the other negative.

That is how condensation particle counters work. Here's an excerpt:

"In general, laminar-flow CPCs operate by drawing an aerosol sample
continuously through a heated saturator, in which alcohol is
vaporized and diffuses into the sample stream. Together, the aerosol
sample and alcohol vapor pass into a cooled condenser where the
alcohol vapor becomes supersaturated and ready to condense."

"Particles present in the sample stream serve as condensation sites
for the alcohol vapor. Once condensation begins, particles grow
quickly into larger alcohol droplets and pass through an optical
detector where they are counted easily."

http://www.ferret.com.au/ODIN/PDF/Showcases/100428.pdf

> When recombined, they would clump together forming roughly
> spherical ummm... clumps, which should produce a much larger pulse
> width.

And higher amplitude!

> Good luck.

Thanks!

haitic...@gmail.com

unread,
Jul 6, 2014, 9:00:36 AM7/6/14
to
On Thursday, July 3, 2014 12:45:24 PM UTC-4, George Herold wrote:
> On Wednesday, July 2, 2014 4:23:04 PM UTC-4, Phil Hobbs wrote:
>
interesting info - what is max displacement of th piezo?
voltage?

Jeff Liebermann

unread,
Jul 6, 2014, 11:22:34 AM7/6/14
to
On Sun, 06 Jul 2014 10:20:02 GMT, Steve Wilson <no...@nospam.com>
wrote:

>I wonder why they want to use near ir. I would think blue, as in
>Blu-Ray, or a uv led would be better for catching the small
>particles that do the real damage.

My guess(tm) is because the device is looking only for large particles
(>1 micron) such as cigarette smog and not small particles such as
mold and bacteria. The device has been around for a long time,
possibly before cheap UV emitters and before GaN, InGaN and AlGaN
detectors were commonly available. However, it might also be
something as simple as the age of the Nikken device, which probably
preceded cheap UV components.

>I'm always amazed at how you can find so many useful links in such a
>short time. You should start classes and show us how you do it.

No need for a class and there are only a few tricks:
1. Use Google images. If you're looking for a schematic of
something, search for schematic images, not for text.
2. Know thy buzzwords. The proper selection of search terms is
paramount. I usually dive into the Wikipedia article first to extract
the necessary buzzwords, and then search for them either individually
or as a group.
3. Don't rely completely on Google search. I like to use Blekko when
I'm stumped, but which does have a learning curve:
<http://blekko.com>
4. If you want research papers and/or patents, use Google Scholar
search:
<http://scholar.google.com>
<https://www.google.com/?tbm=pts>
5. When posting a link, try to reduce it to only the necessary parts,
removing all the tracking info, and irrelevant detail. That's not
easy on some searches and does require learning how the URL's work. I
guess that might be a worthy topic for a web page (but not a class).

The rest is just reading. I must admit that I'm more into speed than
accuracy, which results in my missing some obvious points, but overall
seems to work well.

I also like to read pages backwards, starting at the bottom. Many web
authors have the common affliction of describing problems and
solutions in reverse order. They write like a mystery story, where
the details are supplied first, and the important conclusions at the
end.

I hope this helps.

>> According to my random reading, one can distinguish between
>> various types of dust sources by the waveform.
>
>It would be very interesting if you had any links. I'd like to find
>out what this stuff is and where it comes from.

I would have posted a link if I had found something relevant. I'll
try again, but not for a few days. Busy with other projects.

>> You can differentiate particle size by pulse width. The Shinyei
>> detector claims 1 micron minimum on the data sheet. My guess(tm)
>> is that something in either the amplifiers that follow or the air
>> speed limit the upper frequency response. The faster the particle
>> goes by, the narrower the pulse. You might be able to slow down
>> the air flow in order to "stretch" the pulse width.

>It would be very hard to maintain calibration using pulse width. The
>calibration would depend on air velocity, which would depend on the
>fan speed, how dirty the blades were, bearing friction, turbulence,
>and so on.

I can't do anything to compensate for filth in the ducting, but
measuring air speed is fairly trivial with a hot wire anemometer. I've
built several for weather stations that are quite useful even at low
wind speeds. If fan bearing (more likely bushing) wear is deemed a
problem, then a feedback loop with the hot wire anemometer controlling
the fan speed should compensate for any variations in speed. Of
course, altitude and temperature enter into the equations, so this
isn't really trivial, but methinks can be accomplished.

The problem is that you seem to be designing an "instrument" rather
than something for home use. The requirements for each are quite
different in areas of accuracy, calibration, repeatability, etc. It's
much easier to take a precision instrument, and downgrade it for home
use, than to convert something crude into a precision instrument. For
example, your question about identifying particle composition might be
better addressed with gas chromatography or possibly a mass
spectrometer.


Marginally related anecdote: You don't really need all the technology
to count dust particles. I once had to deal with a serious airborne
pollution problem, twice. I'll leave out the details (unless someone
wants them). The problem was to measure how much dust was being
transported by the HVAC ducting. I didn't have time to rent air
quality instrumentation and just needed something fast. I purchased
some sticky back window shelf paper[1]. I placed some 100 cm squares
in the air stream, and gave it about 30 minutes to collect some dust.
With a microscope, reticule, and my miserable math, I estimated the
particle density from several small sample area. The microscope also
gave me a clue as to what I was counting. I made an order of
magnitude math error, but the dust problem was sufficiently bad that
it didn't matter. In one case, the dust was coming from a nearby
industrial plant with a malfunctional electrostatic precipitator. In
the other, the 1989 earthquake had broken open the HVAC ceiling
ducting joints which allowed the introduction of nearby blown
insulation into the air ducts.

[1] Use paper, not vinyl as the vinyl will build up a static charge
from the air flow.


--

Steve Wilson

unread,
Jul 6, 2014, 10:35:45 PM7/6/14
to
I just wanted to thank you for all the time and effort you have put in this
thread.

Your information has been very helpful and I will put it to good use.

George Herold

unread,
Jul 7, 2014, 8:59:36 AM7/7/14
to
On Sunday, July 6, 2014 9:00:36 AM UTC-4, haitic...@gmail.com wrote:
> On Thursday, July 3, 2014 12:45:24 PM UTC-4, George Herold wrote:
>
> > On Wednesday, July 2, 2014 4:23:04 PM UTC-4, Phil Hobbs wrote:
>

> > http://store.bravoelectro.com/ceramicpiezoactuators-ae0203d04fbrpiezoactuatormaxdisplacement46m2x3x5mm-p-1502.html
>
>
> interesting info - what is max displacement of th piezo?
> voltage?

I think it's 4 microns at 100V.. but that may be 150V.

George H.

haitic...@gmail.com

unread,
Jul 7, 2014, 11:10:53 AM7/7/14
to
>I'm always amazed at how you can find so many useful links in such a
>short time. You should start classes and show us how you do it.

Here is another "killer" way to get technical info:

Google "Google patents advanced"

The link will put you on Google patents.

Google will even link out from patents to relevant non-patent articles.

It's a law that a patent HAS to tell you not only the invention idea, but how
to build it.


Another: If you are looking for a technology area, go to worldcat.org. You can
then get most any book through Interlibrary Loan in ANY library. This includes
theses and downloadable articles.

haitic...@gmail.com

unread,
Jul 7, 2014, 11:18:21 AM7/7/14
to
On Monday, June 30, 2014 3:05:42 PM UTC-4, gyroma...@gmail.com wrote:
> Hi,
>
>
>
> This message is related to my earlier request for advice in the topic 'finding electronics design and fabrication expertise for a project'.
>
>
>
> Because of the kindness of the group members here, I am posting more details about the project and aims.
>
>
>
> The overall aim is to develop and deploy inexpensive 'personal' air pollution monitors. For the environment in which I am interested, a major pollutant concern is particulates arising from biomass combustion, coal burning, and car exhaust.
>
>
>
> Although these may be inconsistent aims, I am interested in having a device that...
>
> - can measure particulate concentrations (and perhaps temperature and humidity levels)
>
> - is low cost, so that many devices can be deployed within a constrained project budget
>
> - has a good degree of concordance (at least qualitatively) with accurate stationary monitoring stations
>
> - can be read with an app on a cellphone or similar device (bluetooth?)
>
> - is convenient to the wearer of the device, e.g, is robust and has the ability to run on batteries for a few days at a time
>
>
>
> I anticipate that a user (or environmental health scientist or health care professional guiding the users) might want to record reading at several points during the day, not continuously.
>
>
>
> There are several projects I have found on the web focused on devices to quantify particulates in the air, but I don't think that they satisfy the aims above.
>
>
>
> The closest instrument that I could find is detailed in a PhD dissertation by David Holstius, a chapter of which I have posted here:
>
>
>
> https://www.dropbox.com/s/azs0dqzgpffppl1/Holstius2014dissertation_Ch3.pdf
>
>
>
> Because of my lack of training in the field of electronics, I do not know if the detection method, components, and overall design chosen by this individual would be appropriate as a starting platform to achieve the aims above.
>
>
>
> Any input or suggestions would be greatly appreciated.
>
>
>
> Thank you again.
>
>
>
> -gyro

I grubbed around in funding programs, and EPA just completed a funding round in
March, 14.

They have funded 5 "PM" centers - Particulate Matter. One is at Harvard,
another 2 at UC universities, and I can't remember the other 2.

It would be interesting to ask them what their main limitation and frustration
is, instrument-wise. People love to talk about their problems.

Another way to approach this is NIH - They tend to focus on the biology, but
smart phone medical apps are hot.

jb

haitic...@gmail.com

unread,
Jul 9, 2014, 7:11:34 AM7/9/14
to
I just googled some more "particulate smartphone." There doesn't seem to be a
genuine optical detector that is low cost in existence.

This is kind of thing that NIH funds, of dubious commercial appeal.





0 new messages