Thread: Developments in Dust Sensor tech

Been reading a bit of background to Urban dust levels to try and get a better feel for the latest info on this.

One interesting source is the WHO Data base of Global Urban ambient air pollution.
WHO | WHO Global Urban Ambient Air Pollution Database (update 2016)

As I said in another post, "Urban dust" was recently raised from the 8th most common cause of death to the 5th most common cause and its not just lungs that are affected which why there is a renewed interest in general air pollution.
From the WHO website:

As urban air quality declines, the risk of stroke, heart disease, lung cancer, and chronic and acute respiratory diseases, including asthma, increases for the people who live in them.

A few interesting bits of info, all figures are the annual averages for 2016 in mg/m^3 of PM10 particles)
For reference, MDF and WRC have an 8 hour working day average of 0.5. Hardwood is 1 and softwood is 5,

There are 3 cities >0.5!
Of the 20 cities above 0.2 all are in the middle east, Indian subcontinent, with one in China and one in Nigeria.
Australian cities/towns are <0.02

An interesting fact that comes out is although PM2.5 is the most commonly reported value and PM10 data is not collected for many cities but is estimated by multiplying PM2.5 data x 1.84.

While it may not necessarily apply to wood dust sources , it suggests that with the increasing availability of PM2.5 detectors like
2017 Mini LED LCD Plantower PMS7003 G7 laser PM2.5 detector meter with battery | eBay that wood workers may be able to make use of cheaper PM2.5 detectors and just multiply the read out by two to get the PM10 data.

This is significant in that most low cost detectors don't work all that well above 1 mg/m^3 and constant long terms exposure of these detectors above those levels will reduce their operating life time.

Hopefully I will get some time for more testing today.

I've been trying to do some testing all morning but its been raining and that really messes up the measurements.

A humid atmosphere changes the particle size profile because smaller particles start to agglomerate or stick together to become larger ones.
This draws large number of particles of <0.3 microns into the >0.3 micron range and the bigger ones get even bigger and they then fall out of suspension quicker.
This makes it difficult to make sense of particle size distribution measurements.

One of the most important measurements with any system is the ZERO reading and to do that you need a zero dust environment. Fortunately I have a lab grade HEPA filter room air cleaner in the shed. It has a >99.99% efficient filter at >0.3 microns. which is partially clogged so it generates very clean air immediately underneath the filter. Before commencing any activities my shed air has a background of ~5000 particles/L so (100.00 - 99.99%) or 0.01% of that is 0.5 particles/L or in other words zero. Both the Arti and the PMS gave zero readings , although it did take a couple of repeat readings to get there. This is normal because measuring regular dust contaminates the counters and they need to be flushed with super clean regularly. This will represent a problem for wood workers who inadvertently contaminate their counter. There might be a way to generate this Zero using a small highly efficient 0.3 micron air filter and a compressed air supply. If you can do this then repeated measurement of outside air will be the next best thing.

I also made the following concentration measurements but bear in mind what I said above

Before I started I measured PM10 dust levels of <2.5 ug/m^3 which is similar to what I have seen before.

Then I did the zero readings and then went about normal activities and stopping whenever I did anything new to take some measurements

I tried to generate some dust by sweeping the floor of the shed and swept up about a dust pans worth of stuff but it was 99% leaves and stuff the dogs had walked into the shed with their wet feet. The rest was mainly metal dust and plastic swarf. The semi wet floor also kept the dust level down. The levels I measured at operator head height were <26 ug/m^3 which was much less than the 12000 ug/m^3 I saw back in 2012? but my DC system was not as complete as it is now so there was a lot more wood dust on the floor back then and of course it was not raining.

I did a bit of metal polishing/cleaning of 2, 20c size piece of SS and a Al bracket about 50 x 25 mm in size with a Scotchbrite wheel and the levels went up 140ug/m^3 . This thing sure is sensitive which is good. I just hope it is not too sensitive when it comes to wood dust.

One thing I have determined some time ago is that it is increasingly inconvenient to have to contaminate my shed with dust to perform tests at higher dust. I can measure dust inside a chamber/box but that I think that is too artificial. To test in the shed just takes too long to generate the amount of dust and the it's a right pain to then have to clean it up again. What I might have to do is look around for people doing real work and test dust levels while they are doing it. e.g. putting a heap of wood through a thicknesser or sanding down something large - with or without dust extraction.

One again Bob, great information for those of us who decide to put a monitor like this in the workshop, really makes you realise how important it is to have an understanding of the environment you are working in and the pitfalls of using a device like this.

I'm really looking at making one of these down the track. I wrote graphing software in assembler for a 32bit micro a few years ago, it's still running in the Car, it logs ATF cooler circuit pressure and temperature from 4 sensors to detect cooling efficiency and peak temperatures during different driving conditions, adjustable update down to every 100ms on all sensors, it's fast enough to show instantaneous Torque converter output ATF temps (sensors in the fluid) and every gear shift with < 100ms response time Pressure sensors. Should have no trouble with something like this (also uses floating point maths and 32 bit counters so overrun (counter wrap / overflow) is extremely unlikely. It can display a constant moving graph over an adjustable time frame and can be scrolled back through time. I'm thinking that having a moving graph might be meaningful for a quick glance at the screen to see the effect of doing some dust making activity on a particular machine.

Again thanks for all the work you are putting on this - looking forward to more data and thoughts on the device.

Mike.

Originally Posted by MandJ

One again Bob, great information for those of us who decide to put a monitor like this in the workshop, really makes you realise how important it is to have an understanding of the environment you are working in and the pitfalls of using a device like this. .

Yep - it would be all to easy to accept its output as gospel but given it's not designed to take into account the particular requirements for WW activities it would be careless to accept its output as gospel and go with it.

Graph output would be good - have just been talking to son about a larger screen implementation.. At first I thought it would not be sensitise but now I think it might be too sensitive. As soon as the weather dries up a bit we'll know.

Again thanks for all the work you are putting on this - looking forward to more data and thoughts on the device.

No worries Mike, fortunately I'm interested in all this sort of stuff so it's not that big of a chore.

More testing this morning.

I'm still not convinced the absolute numbers the PMS is putting our for the PM10 are correct but I have increasing confidence in reporting relative numbers i.e. this is X times higher that Y.

This morning I started by testing the outside air which had a PM10 concentration of between 5 and 6.5 ppb (Perth's annual average is 17 ppb).

What I mainly wanted to test was how much dust was coming out of the DC enclosure when nothing was being collected inside the shed.
The reason for this is that as increasing emphasis is placed on air pollution by various authorities I wonder how long it will be before woodies will be called to task on how they manage their wood dust especially in inner suburban situations.

The reason for measuring while no dust is being collected is because that is usually the longest period of time a DC is running i.e. cut. . . . do other stuff . . . . cut . . . . . do other stuff, so the DC is usually running but collecting nothing more often than it is collecting. If your DC leaks even when collecting nothing it will add to the atmospheric dust load and may become a cause for concern to authorities.

Testing a DC for leaks is very time consuming because a number of measurements are needed over outside of the DC. When the DC is in an enclosure where the air escapes in a preferential direction - then it's easier to tell if its leaking and a single measurement may suffice.

So with the DC running and collecting nothing the PM10 dust coming out of the enclosure was <2.5 ppb so less than half that of the surrounding atmosphere.
This means if the DC is leaking it's still collecting more dust than it is emitting so free running it is actually cleaning up the atmosphere!

If you get one of these counters and set it up and do this test you need to let the DC run for ~10 minutes or so before testing the ongoing air. The reason for this is when you first turn on the DC the levels coming out of the DC might be much higher than the atmosphere. In my case it was 5x higher .

BUT this is probably just residual dust generated by slowly decomposing materials inside the enclosure that has built up since you last ran the DC. All materials (even epoxy paint) does that to some extent or other. Running the DC for some time will flush this out and provide a more realistic reading.

Unfortunately if it is leaking this measurement doesn't tell you where it is leaking. If it is a large leak then those are usually visible e.g. under the band clamp or a hole in the bags may show up as a red-brown stain. If it is a small leak then you have little choice but to change the plastic collection bag(s) - chips and shards of wood that swirl around in plastic collection bags can puncture or abrade small holes in these bags - and/or change the filters.

The next test I did was place the detector on top of the TS guard and recorded dust levels while ripping 1m long steps of 16 mm MDF, and jarrah pieces about 25mm thick and 450 mm log.
When the DC is on I could not detect any differences in the dust levels compared to outside shed air.
This is not something I would be game to do wth the $5000 particle counter from work but its a different thing with the PMS - if I stuff it up it only costs $34 to replace.
Anyway the guard is clearly working.

I could not resist doing a couple of the MDF cuts without the DC on (guard still in place). This produced 55 ppb just above the guard and 131ppm behind the guard. Most of the visible dust seems to come out of the holes in the TS cabinet and fall to the floor. When I put the PMS on the chair next to the TS cabinet the reading went up to 370 ppb. As I said I wouldn't do this sort of testing with the expensive sensor from work so that is one useful thing about the PMS. However I am loathe to do more of these high level dust tests in my shed as it just makes a mess.

Bob if you don't stop posting these great test ideas and results, I'm going to weaken and get one sooner that I had planned

Hah! my plan is working - I have just ordered 3 more - I plan to eventually deploy a couple as stand alone loggers in mens sheds but the first tests will be to see how consisted they are with each other - sort of a repeat of the test posted in post #31 above.

BTW I should have added that after testing the cutting of the MDF without the DC running and now my nose is running - gawd I hate that stuff.

This arvo I just left the PMS running in the middle of the shed at head height and as I completed things I wrote down the readings.
Shed background was 2.7 -6.3 ppb

With DC on :
Belt sanding: a large piece of rusty steel - 26 ppb, small pieces of Al - 23 ppb.

With DC off
Cutting a 20 mm steel rod - 457 ppb, ScotchBrite Wheel polishing small SS parts - 350 ppb

Most interesting. Shouldn't take me too long to work out what needs DC and what doesn't.

Took the particle counter to my sisters place yesterday arvo to show the weekly family gathering.

On the way I had it running on the front seat of the car with the external air vent fully open and the fan running - I was amazed at how low the particle count was even on a main road (PM10.
I pulled over to the side of the road to make some measurements - it was breezy and had last rained about 90 minutes ago but the RH was only 54%.
The PM10 readings I measured was were between 2 and 5 ppb!
When I got home I checked the nearest AQI measurement site which was at South Lake about 5km away and at the same time I was measuring they measured a PM10 AQI of 6. That's 6% of the 50ppb recommended maximum, or 3 ppb - so there was pretty good agreement.
At my sisters place she had one of those fragrant oil mister things running and they generate around 25 ppb of PM10 admittedly is mostly water.

Interesting paper just out:
Co-benefits of climate mitigation: Counting statistical lives or life-years?, M.S. Andersen, Ecological Indicators (79), 11-18, August 2017.

http://www.sciencedirect.com/science/article/pii/S1470160X17301693

"For the affected victims of air pollution the reduction in average life expectancy is 10.7 years when considering all four latency scenarios, while the simulations suggest that 9 years should be a lower bound estimate. Within the span of a normal human life this is not an insignificant loss in life expectancy." (Per. Δ10 ugPM_2.5/m³⁾

This paper discusses urban dust pollution, not wood or metal shop dust. Nonetheless the effect is dramatic.

Thanks for the link Rob. Was speaking to a former colleague yesterday who is an Atmospheric researcher (mainly satellite data) but he also supervises a team that does ground based studies to calibrate satellite data. Anyway he was interested in these low cost sensors as they are about to study dust plumes from mines and mineral processing sites in our North West when they pass over townsites. They basically collect data simultaneously by Satellite, ground base vehicles and for vertical access they use planes and even balloons and increasingly for small scale events they use drones. With drones like planes it's all about weight V distant V height so they try to use lightweight low power payloads. I sent him the details and he said he will get back to me.

Originally Posted by BobL

Took the particle counter to my sisters place yesterday arvo to show the weekly family gathering.

On the way I had it running on the front seat of the car with the external air vent fully open and the fan running - I was amazed at how low the particle count was even on a main road (PM10.
54%.

Depending on what car you've got/age/whether you use genuine products as many manufacturers these days are deploying static/charcoal filters which are very effective at removing common dust/pollen etc.

Originally Posted by tonzeyd

Depending on what car you've got/age/whether you use genuine products as many manufacturers these days are deploying static/charcoal filters which are very effective at removing common dust/pollen etc.

Charcoal filters are more aimed at removing organics and odours from air than particles - they work best at lowish air speeds such as is found in AC air systems. To remove remove 0.3 micron particles from large volumes of air requires a many pleated HEPA filter.

Still waiting for some dry weather to continue the particle counter testing.

Meanwhile I see Plantower have released a new detector called the PMSA003.

There's not much info available about it but it seems to contain the same insides as the 7003 but in a smaller form factor.

The 7003 is 48x37x12 mm compared to the A003 which is 38 x 35 x 12 mm and slightly lower weight..

The A003 is targeted at what is called the "personal wearable" devices that are worn on the shoulder and powered by a battery pack on a belt. Also attached to the belt is a digital recording system that logs the wearers exposure.

The other useful feature is the intake fan direction can be oriented either at 0º or 90º to the air outlet. This reduces air/dust recycling so improved the response time.

AT $40 (versus $23 for the 7003) is still reasonably priced but bear in mind you have to add a special micro connection to access these devices.
https://www.aliexpress.com/item/PLAN...6-f7159e4ec289

Remember these are just the sensors and they have to be connected to something to get the data out and see it.

If you want something that immediately give you an output number then there is this

2017 Mini LED LCD Plantower PMS7003 G7 laser PM2.5 detector meter with battery | eBay

The weather has improved so I took the particle counter to the mens shed today. I was there to install a dishwasher but though I would just run the counter to see what I could see.

There were about a dozen members present but only 5 were wood working with the rest working on the outside of the shed, metalworking, chatting or doing admin. The sort of WW being done was all small to very small work, toys, boxes using mainly hand tools with the very occasional brief use of small Band Saw, lathe and small belt/disc sander. No one used any large machinery during the time I was there

Dust levels outside ether shed were <10 ug/m^3 for PM2.5 and < 15 for PM10 - Perth's annual average PM10 is 17 ug/m^3.

As usual they were NOT initially using the dust extractor but I said nothing and measured dust levels both near the workers and at head height in the middle of the shed.
The highest levels I recorded were PM2.5 of 12.8 ug/m^3 and PM10 of 35 ug/m^3 which was near the belt sander.
All measurements were much less than the lowest recommended dust levels for any wood (500 ug/m^3)

The situation was helped by the fact that the shed has a large volume so that the small amount of dust generated had a lot of room to expand into.
The second factor was it was quite windy and with 3 of the 4 shed main doors open the shed was quite well ventilated.

Mid morning someone finally turned on the dust extraction and the levels fells and stayed close to that outside the shed as described above.
I will go back on a Friday when a lot more active woodworkers usually attend.