Thread: Dust Concentration Measurements

Here is some preliminary dust sensor data for a Perth mens shed.
I won't identify them but they have allowed me to use the data for discussion purposes.

The shed is a barn like structure - maybe 15 x 15 m with a 5m peaked roof.
Large double roller door on one side always kept open at all times when shed is operational.
Line of windows down two sides - not always open.

Sensor located at 2m above shed floor (on top of a BandS) ~1/3rd of the distance in from shed wall.

Open 5 half days a week. Mon/Tues/Thurs/Sat/Sun - closed on Fri and Weds - this is reflected in the lack of dust peaks for those days on the graph
Lots of members seemed to be using hand tools on small jobs

Clearvue Max located outside shed with 9" mains but only 4" ducting to machines (against my advice BTW - but there you go)
All powered sanding is done outside.

Even though they only have 4" ducting It looks like it's working for them - this shows once again the benefit of a big shed and lots of ventilation and sanding all done outside.

Each data point represents 30 s worth of data, so only once for 30s does the PM10 concentration go over the OHS limit of 500 ppb.
Even the periods for which the shed goes >100 ppm is not that bad.

I am going back to measure another weeks worth of data and will try to get a more central location
Have recommended they beef up ventilation further.

MS2.jpg

This afternoon I had to sand and undercoat 4 preschool kids size chairs and 2 small tables.
These had been made at the mens shed and they had a pine frame and MDF tops/backs.

It was a nice day so I decided to do this outside and wear my "Wearable" dust sensor to see what it showed.
I was using a mouse sander but I ran out of sanding pads so finished the job by hand sanding.

Below shows the sequence of events.
For about 1000 seconds (~15 minutes) I sanded the tables with the mouse (Mouse Sanding 1).
Then I had a break and went and grabbed a coffee and hung out a load of washing
Then back to mouse sanding for about another 15 minutes (Mouse Sanding 2)
By then I had no sanding pads left so I switched to hand sanding for ~30 minutes.
Then undercoat

Outside-sanding.jpg

It's interesting to see the concentration can approach the recommended 500 ppb limit for MDF even when sanding outside but the difference between mouse sanding 1 and 2 was interesting.
It was actually quite breezy and looking back on the BOM website I see the average windspeed during "Mouse sanding 1" was 17 kph with gusts up to 30 kph.
During "Mouse sanding 2" the average wind speed was 13 km/h and wind gusts of up to 28 km/h.
This suggest that you don't want to be sanding MDF outside when the wind speed is less than ~10 km/h.
Well, . . . not without wearing a mask anyway.

Bob could you tell us what the mouse sander was?

Originally Posted by Bohdan

Bob could you tell us what the mouse sander was?

Ryobi. it has a dickie little dust collection bag on the back - but it collects very little dust.

Have you collected data for the particle distributions produced by different types of tools? It is reassuring to see that the major part of particles are at the big end of your range.

Bob would it be possible for you to get your hands on a festool sander and vac to test to see if their dustfree claims are worth the money?

Perhaps take your dust sensor to a dealer that has a demo setup for you to try as for the first time we would have some quantitative data on their claims.

Originally Posted by rob streeper

Have you collected data for the particle distributions produced by different types of tools?

None of the particle size sensors (either mine or the ones I have borrowed from work) that I have access to are able to measure any dust larger than 10 microns, or able to cope with the volume of the stream of sawdust produced by a tool. Locating these sensors too close to the path of the sawdust stream coming from the tool will damage them beyond repair.

Measuring dust distributions output from a tool requires using the tool in an enclosed chamber with fans that constantly stir up the dust. Then special sensors that can cope with the volume of dust and measure particles up to 100 micron are used.

I have the particle distribution data for today's measurements, but don't forget my wearable sensor is on my shoulder which means it will not be the same as particle distribution production by the tool because the bigger particles may not even get up to shoulder height.

It is reassuring to see that the major part of particles are at the big end of your range.

Once needs to be careful inferring a particle distribution from different size particle concentrations.

Remember a single 10 micron particle = 1000 1 micron particles so the chances are that the are many many more smaller particles.

The dust distribution would also be modified by the filter bag attached to the sander, the wind and distance of the sander to the sensor.

I have given up on assessing tool dust output - what concerns me more is what the operator is likely to breathe in - hence the wearable.

Originally Posted by Bohdan

Bob would it be possible for you to get your hands on a festool sander and vac to test to see if their dustfree claims are worth the money?
Perhaps take your dust sensor to a dealer that has a demo setup for you to try as for the first time we would have some quantitative data on their claims.

See comments above first.

I don't want to get involved with any dealers - even if they would let you "borrow" a tool (I doubt they would) borrowing tools means being beholding to them.
Besides new products are not really the ones we should be testing - I'd rather test stuff that has been "out and about" for a year or two to see what they are like as this is what most operator sees for most of a tools working life.

BTW I have a Festool sensor but I don't have their dust extractor.

A few years ago i tried to specifically test a few high quality tools and their VCs using the dust sensor from work. I tried careful waving the sensor near the working tool - it's near impossible to effectively operate a tool and simultaneously wave the sensor around near (but not too close to) the dust because I don't want to damage the sensor AND more importantly the novelty effect of doing this lasts about 30 seconds and then it gets dead boring.

In the end I ended up locating the sensor about 2m away from the tool but I don't believe this reflects what the operator breathes in.
Recently both M&J and I have determined that locating a sensor over on a bench away from the dust source simply does not capture what the operator experiences - hence the wearable sensor.I

By way of a general comment, when higher quality tools and VCs are brand new they seemed to perform much as described by the manufacturer but older (even just a few months) VCs didn't work as well. Unless they are scrupulously maintained, by the time they are a couple of years old most VCs are pretty dirty and leak. The tool manufacturers will then simply tell you to service the VC or to get a new one. It also depends how they are used. If the workpiece blocks tool air holes too much they don't work as well.

As I keep harping on about more important than testing the tool in an artificial short term experiment is to test what the operator breathes in long terms during the course of normal/real operations. So every time I use a machine or tool for a real project I will (If I have time and I remember) strap on the wearable - this means I can get on with real work and allow the wearable to record the data at the same time. Hence I can collect many hours worth of data without taking up too much time. The data should also be much more "real world" than most artificial testing.

If someone wants to give me a new festool sander and vac that I can use over the next couple of years I would be happy to test it.

This is not my data but here is the particle size distribution for particles <10 microns output by a sander using 80 grit paper.
This is done in a special enclosed chamber that seals in all the dust and uses a fan to fluff it all up so the sensor can measure all the dust.
There are lots more particles above 10 microns but I'm not showing these here as I am not worried about them.

Screen Shot 2017-11-17 at 11.22.00 pm.png


Below is the particle size distribution the wearable dust sensor detected for the highest measure value of dust concentration (380 ppb) on the graph in post #18 above.

Notice how different it is from the graph above.
The following will explain the difference
- I was using 120 and 180 grit paper
- the back ground in the garden has not been subtracted
- the filter on the back of the sander will tend to filter out the big particles and let the finer ones though
- More finer particles will also tend to evade capture to start with
- the big particles will fall away from the sander to the ground - may not even get up to shoulder height
- the finer particles will fog the area (our back yard) and build up and stay suspended in the air for much longer that the bigger particles - now given I was outside and there was a variable breeze this should blow all the dust away but it clearly did not.

This shows why the wearable is so important in this sort of testing.
Screen Shot 2017-11-17 at 11.34.42 pm.png

The background has not been subtracted - in terns of the 0.3 - 0.5 micron particles the count was ~4000 particles/L
It will not change the shape of the curve much

Another week of dust data from the Mens Shed where I am a member.

This time the sensor has been moved so its just above 3 large machines (big belt sander, drum sander and table saw) - its also directly under the HD Evap AC outlet which is used (sometimes) as a forced ventilator.

The gap in the middle of the graph is once again the weekend.

MMSNov2017.jpg

Fairly similar to last time, even with a large pulse on Friday afternoon from the weekly clean up.
Notice that even before the shed opened on the friday the background was around 10 ppb.
This time it was a really big clean up for the annual fair which was held on the Saturday.
I was told that a couple of blokes got busy with compressor hoses.
The pulse shows that for ~8 minutes the dust was over 500 ppb , reaching a maximum of 808 ppb before coming down below 100 ppb in about 8 minutes.
I was not sure they had the evap AC unit fan running.

Just started the process of shoehorning most of the contents of about 1/3 the surface area of the house (Dining, kitchen family) into the other 2/3'rds so we can get the Jarrah floor boards resurfaced as the board in these rooms were last done 23 years ago and are well overdue. The boards in the family room are cupped so a fair bit of wood will need to be removed to get them flat. There are also some large gaps but we will just have to live with those.

The moving process would not be that bad except the whole house is cluttered with 38 years of accumulated stuff. Fortunately the contents of the whole kitchen do not need to be over with the stuff in the cupboards being left in place. The least cluttered room is an upstairs bedroom so about 40 boxes of stuff have been carted up the stairs. SWMBO has a bung left hand wrist so muggins here gets to do that. About a dozen boxes of stuff have gone going into a shower cubicle in the laundry.

After doing that for 2 days plus moving the furniture i'm pretty well stuffed. Just the main entertainment unit to unpack and a sofa and 2 arm chairs.

The furniture is going into the lounge (which is really SWMBO craft room), downstairs bedroom and the back veranda which is fortunately just outside the family room.

Why am I talking about this in this thread?
Well, this gives me the opportunity to run a recording dust detector during the sanding process to see what sort of dust levels they generate.

Is anything known about the particle distributions produced when sanding different types of timber?

Originally Posted by rob streeper

Is anything known about the particle distributions produced when sanding different types of timber?

In general the harder the wood the finer the dust for the same process.

Some data here
Annals of Occupational Hygiene Vol 44.6 pp 455-466, 2000
"A study on dust emissions, particle size distribution and formaldehyde concentrations during machining of MDF"
Authors are Chung et al.

The particle size distribution produced by using a tool is a PIT to measure and is not the same as what makes it inside a woodworker.

This is the distribution arising when pine is sanded with 80 grit paper.





This show what particles typically get inside a wood worker.
DustDeposition.jpg
A wood worker has to have their mount and nose directly in the stream of sawdust to get particles bigger than 10 microns inside them.
This does happen especially for wood turners
For most other operations only sizes less than about 10 microns get inside you.

in the past it was thought that particles less than about 0.3 microns were of minimal concern because
1) not many of these are produced during wood working and
2) whatever was breather in was almost 100% breather out.

These days the finer ones (PM2.5) are implicated in non-pulmonary diseases like heart and strokes so more attention is being paid to them.

The floors are sanded and have had the sealer and first coat of finish applied.

I had a dust sensor running since monday but because I did not use a battery back up, and the floor finisher's big sander tripper the breaker for the circuit the sensor was running I lost the data for all of monday and tuesday up till about 10.30.

So all I have is the data from 10:30 am on Tuesday to when the sanding was completed wednesday lunchtime.
Unfortunately this missed most of the time from when the big belt sander was in operation for most of Monday.

The data shows a 1000 ppb peak starting at about the 5000 second mark which is when the large belt floor sander was being run with belts as coarse as 30 grit because the boards in the family room that were badly cupped. During the time the data shows the dust levels averaged around 500 ppb for about two hours but you can see the levels dropped pretty quickly after that was completed.

The large belt sander only had a 25L needlefelt bag by which dust was collected.

After the large belt sander, large and small ROS were in use. The large ROS uses a 400 mm diameter velcro pad on which 4 standard 150 mm discs are place. Grits up to 240 were used with this sander - dust extraction from this unit was via a 50 mm hose and a Industrial Strength Nilfisk vacuum cleaner and a cyclone. It uses a washable PET filter and paper cartridge filter.

The smaller ROS was a ruggedised unit with the pad poking out the front so it could get under things like stairs etc and was mainly used to sand the edges. This was attached to a smaller Nilfisk vacuum cleaner. I was really impressed with these vacuum cleaners although I guess given the $2500 price tag for the bigger unit it would want to do the business.

Around the 90000 second mark is where the smaller vacuum cleaner plus a Nilfisk backpack VC were used to

I was also impressed at how little dust made it into other rooms in the house. I had a dust sensor running in the bedroom where it never got above ~6ppb which is normal a bedroom given all the fabrics etc. In the study 2 rooms further away the levels didn't reach more ~2ppb which is about average for that room..


Screen Shot 2017-12-06 at 5.10.41 pm.jpg

The graphic for the above post seems to have got lost so I'll post it again.
Screen Shot 2017-12-06 at 5.10.41 pm.jpg