Thread: Developments in Dust Sensor tech

Originally Posted by MandJ

Thanks for that, glad I decided to skip the Jaycar model.
Just a thought, have you considered connecting 3 bright LED lights for AQI in Green, Yellow and Red to the small counters, really grabs your attention and might be more meaningful to the typical men's shed.

Yeah I thought of that but there are practicalities developing that need to be sorted regarding suitable detector location, especially in large sheds.

Some observations I have noted since using these detectors in sheds.

I currently have one detector logging data at the Mens shed where I am a member.

The detector is located in a relatively safe (for the physical well being of the detector) place on top of a horizontal RSJ about 2.2m high up against the shed wall, above a line of work benches. Members usually perform hand work with the odd power tool use at the work bench below the detector.

I was crook and only able to observe it for about 30 minutes on Monday morning before I had to leave. With 5 members on light activities in the shed and no large machines running ( I think one bloke ran a Band Saw for about 30s), PM10C never got over 32 ppb. The rest of the members were doing some brick paving outside. I will pull the SD card next week and look at the results.

There are belt and disc sanders about 2.5m away on one side and 2 bandsaws about 3m on the other side.

This mens shed is quite large, 14 x 10 x 4.8 m. This makes a difference as any dust can rapidly disperse into that large volume and it may take hours to build up to higher levels. Mid afternoon measurements on cold days when many members are present might show something.

Dust extraction use is sporadic. It usually always on but the right gate valves may or may not be open - this is why I want to implement the pressure gauge power lock out switch.

Ventilation is highly variable and has a major impact on the meter readings - there are two 3.5m high roller doors - if these and the double front personal entry doors are open, and the Evap AC fan running, the ventilation effect appears to be VERY significant especially near the middle of the shed.

Even with ventilation, an operator could be using a dusty machine in a corner or up against a wall and be over the limit, while where the meter is located it could still be well below the limit so any flashing light would not fire up.

I have already seen this effect in another mens shed I visited on Wednesday where there was a bloke using a wood lathe without ANY dust extraction and the detector could be within 1 m of him at shoulder height and PM10C not go over 50 ppb. I did not want to get into the operators face and distract him, so only when I held the detector above his head would PM10C go over 50 ppb.

At a large country shed I visited a couple of months back they had 4-5, 3HP or less, DCs that vented inside their shed so the place was covered in dust. They were in the process of deciding what to do about ducted extraction.They had open double width roller doors at each end or the shed and with no DCs/machines running and both doors open the PMC10 dust levels were less than 10 ppb, outside it was 4-7 ppb. I wish I had been there when their machines were running but I did not have any data logging detectors available at the time.

It's becoming increasingly apparent that a single detector in even a moderately ventilated larger shed may produce limited results. While this in itself is worth knowing - I knew this already to some extent and is why I have always promoted ventilation - but it does it make measurements and alarms difficult to implement.

About the best location for a single detector is still probably at operators mouth level in the middle of a shed, but I don't want to dangle a detector 1.5m above the middle shed floor. It looks like it will have to be at least 2.5m above the shed floor and even then, if someone waved a long piece of wood around and with the flying fox electrical leads at 3.2m high there could still be problems . 2.5m would be OK for PM2.5 but it will underestimate the PM10C. I am now looking at using only PM2.5 and putting a detector at 3.3m above the floor.

The bigger the shed is (especially the taller it is) the more difficult making any measurements will be. Whereas if you have a small shed and you like to keep the doors and windows closed, the results from one detector is going to be much more reliable. A with and without doors/windows open will be a very interesting comparison.

Ultimately the only way an individual operators exposure is going to be relibaly assessed is via a personal/wearable detector. Yeah another project I am already thing about.

The next thing to consider is, what are the natural levels of dust outside a shed doing? I am increasing getting around to thinking a differential counting system may be needed to really assess this.

I knew this was going to be more difficult than air flow measurements.

Originally Posted by DaveVman

I just found this thread and it's very interesting to me. I'd do one myself but I don't have the time. Plus I personally have plenty of basic things to do to improve my shed environment without needing a meter to tell me so.
So I'm just letting you know that there's others following your progress. Keep us informed.

Thanks Dave.

You are dead right.
It shouldn't require a meter to tell you to open your doors and windows up when you make dust.
Or to include large doors or windows when designing a shed.

Thank Bob, yes I knew that going in, main reason why I wanted the sensor in a small box with an 8 Ah phone battery and Wireless transmitter. In a small workshop I can have this located at head height close to me as all our machines are close to the walls. I can also monitor multiple sensors with this unit. Thing is, I'm very sensitive to chemicals and wood dust from a sinus point of view, it's already found a source of pollutant before I smelt it and had to leave the room for a few minutes until it dissipated - AND - I had cross ventilation between me and her part of the WS. That pollutant was my Wife ironing some brand new cloth, the display alerted me to something and about 3 minutes later I felt the effect, asked SWMBO if she was doing any craft and, as it was last night, she was ironing again, only this time far worse because of the dyes and chemicals in the new unwashed cloth.

For anyone interested and to finish off, I'm posting two pictures, the first is the finished Dust Sensor head.

I have just finished this and have not tested in the WW workshop as yet, I'll likely start a separate thread for my results for anyone interested. It has already revealed problems with cross flow in our separate general hobby workshop, there are dead zones in a number of places so a change is in order there, especially with respect to my soldering station and SWMBO craft area.

I should also be able to verify dead zones in the wood workshop with respect to cross flow ventilation, allowing me to add or relocate roof extraction or cross flow inlets to the best locations. Obviously this will take some testing on all machines and type of work being done.

I plan to take my time as I'm in no hurry. As we use various machines for longer periods of time I will be able to tell if the extraction and cross flow at the operators head position is being compromised, and make changes accordingly instead of just guessing. This should, over time, give me an idea of how long I need to Ventilate the WS and run the DC after various WW jobs with respect to time spent at the machine and type of work being preformed. Yes I could just run it all day, but power bills are crippling and along with other utility's going up, I'm inclined to not needlessly waste money.

Finally, while some timbers give me an instant reaction and the current WS extraction and ventilation has virtually eliminated that, very small concentrations of these and others don't show up until hours later, by having this sensor at head height and testing all areas of the WS for levels over time, I hope to build up enough data and to either take extra precautions or improve DE even more, perhaps both.

I'l love to get a 5HP Clearview cyclone and run 10" mains with 6" drops, hell why not run 10" all the way, but unfortunately I can't afford ANY of that, and at my age I'm not likely to ever be able to.

Below: This unit is quite small and is fully independent of any power supply or leads, it has a 5 AH Liron battery and can be placed at head height or anywhere in the workshop, it's simple to move from machine to machine in a few seconds.




Below: This is the multifunction Micro and screen in it's case, It's used for everything, like a touch screen controlled accurate current and voltage reference, Function generator, general Logic logging, RF scanner and so on. As shown, it's running the Dust application mentioned in previous posts.




There is one final note in the PMS 7003 Sensor manual:

2. The air inlet and outlet are located in the plane close to the user's machine wall and the outside of the pores for the best installation side Type, such as can not be achieved, then the outlet around 2cm without shelter. Between the air inlet and the outlet. The structure isolates the airflow and prevents the airflow from flowing directly from the outlet to the air intake inside the user's machine.

I have not done that yet, simple to add and I'll do that before testing any further, but so far, the testing has given 100% repeatable results. The unit appears to be very sensitive and consistent in the results it returns for the various irritants in the craft workshop.

Cheers.

Nice work MandJ. Way way above my pay grade. You know if you can get those built to sell to the woodworking market you just might have enough for a CV. Would the receiver be able to take the information from multiple transmitters? Maybe to show how fast the dust migrates from a machine, the direction and which sizes of particles first move and follow. Not sure it's even feasible or even needed, just wondering. I for one would love to follow your findings as you play with it.

I spent most of today at the mens shed fixing dust problems.

Thurs morning is not an official shed open time but there were a couple of blokes doing special jobs and no one using machinery or power tools. During this time the dust counter was consistently below PM10C of <10 ug/m^3.

Around midday members started turning up for the arvo session and because I was fitting a new top and seal to the cyclone bin the DC was out of action. I was not supervision so I said nothing as I just wanted to see what would happen and several of the members started using machinery without DC..

More or less at the same time one member started cutting up sections of 250 x 50 mm Jarrah using a SCMS for bread boards. He was about 10m from the counter, another was using a DP also about 10 m from the counter, (He was the only one wearing a dust mask) and a third was using a router about 4m from the counter. There were about 5 others doing things ranging from hand tools to metal work to hand sanding.

I wasn't keeping a super close eye on the counter but the highest particle counter reading I saw towards the end of about a 5 minute burst of routing was 900 ug/m^3 and when the routing stopped it dropped quickly to 700 and then slowly to around 200 and kept falling to less than 50 ug/m^3 after a few minutes.

I told the member using the router to call me before he was going to do more. This time I turned on the evap AC fan and with the router doing another 5 minute burst the PM10C didn't go above 70 ug/m^3. It would have been more than this near the operator but it does shows how important ventilation is in the overall scheme of things.

Within minutes of using the AC fan the concentration was dow around 4ug/m^3 which was probably the concentration of external air.

Originally Posted by QC Inspector

Nice work MandJ. Way way above my pay grade. You know if you can get those built to sell to the woodworking market you just might have enough for a CV. Would the receiver be able to take the information from multiple transmitters? Maybe to show how fast the dust migrates from a machine, the direction and which sizes of particles first move and follow. Not sure it's even feasible or even needed, just wondering. I for one would love to follow your findings as you play with it.

Thanks, yes I can monitor multiple transmitters. One of the things I have to do is to add a temperature and humidity sensor like Bob did, however what I normally do is to quickly program a tiny 8 or 14 pin micro to insert the T&H values into the data stream from the Sensor, so no extra overhead to read this in the main unit, and if the Sensor is outside, then T&H readings are taken right at the sensor.

I have a second sensor on the way, that can be used to compare outside dust levels with the lowest levels reached in the workshop, and as you said, also see just where the dust migrates to in the workshop, that may be important in stopping dust build up in dead zones, especially in a workshop with lots of wall cabinets and high shelves, last thing I want is to breath in a face full of fine timber dust that has built up over time.

BTW I have three selectable graph to choose from and can zoom in or out as needed, there are 3 other screens that show everything that comes from the Sensor. Graphing shows the two most important being PM2.5 and PM10, there is also an AQI screen just showing those two big gauges for PM10 and PM25, colours and calibration match the AQI index in the rest of the program. All data is logged automatically to a 32GB MicroSD card and log times can be from 1 .5 hours to 14 hours, which is all I wanted for the WS. However it would have no trouble logging over a week or month if required, AQI would be averaged over that time frame or as desired.

Some brilliant information here gents - it has certainly convinced me to order a PMS7003 or 3 in the coming weeks.
Looking forward to more updates and outcomes of your designs.

Finally got to do some measurements, my wife spent about 1.5 hours making a bowel on the lathe from a dry Camphor laurel blank. I started with the sensor mounted at head height directly above the lathe.

To recap the DC set up: External housed 3HP DC, the pick up for the lathe is a short run of 150mm ducting around 2m long. The pickup is a BobL made 150mm bell mouth. The workshop is 6M x 3M and has a huge intake (with equally huge fan built in) in one wall of the WS, the opposite end to the lathe, shed vents out via a roof exhaust, obviously the DC intake when running and a vent above the door.

I tried placing the sensor on the lathe downwind from the cross flow air, I tried it just above and slightly behind the bell mouth and every place I could think of to get the maximum readings.

I have a second sensor which it the type 7003M version with improved intake and flow path (manufactures description) both show similar results.

Air quality is pretty good out here, normal background yesterday and today: PM2.5 = 0 , PM10 = 0, PM1= 0.

FYI - RAW total particle in 0.1 litre of air > X um: 0.3um = 108, 0.5um = 28, 1um = 4, 2.5um = 0, 5um = 0, 10um = 0

The highest readings I got over 1.5 hours (including some sanding) were PM2.5 of 7 ug/m3 and PM10 went to 8 for a few seconds.

Typically, there is a lot of wood curl shavings on the floor and over the lathe, I had to remove the build up shavings trapped in the intake grill about a dozen times.

The DC is shut down as soon as the work had finished, however the cross flow is left running and the huge shed door is opened.

The floor and lathe are cleaned via an externally mounted Vacuum cleaner fed via a mini cyclone dust separator, obviously there was no change in readings during clean up.

My take on this first test is that the DC set up for the Lathe with this type of timber is effectively removing fine particles below 10um. The next test is to do this without cross flow, then without the DC, and finally with both off.

I was surprised at the relatively low readings taken about 1.5m from the DC housing exhaust opening, air blowing past the sensor, highest I got was PM2.5 of 17 ug/m3 and PM10 around the same figure, however in fairness, I had cleaned and washed the 4 year old filter bags about 2 months ago with only some light sanding and band saw work been done in that time.

Not WW dust topic - but:

I had the sensor sitting on the dining room table the other nigh, cooked a bit of steak in the fry pan and had readings hitting 630 ug/m3 for PM2.5 and 750 for PM10, I really must fix the exhaust fan above the stove - was a cold night and had the doors closed, readings were taking so long to come down that I relented and opened the doors.

I just haven't had a chance to do any real readings since finishing the unit, so hopefully some more meaningful results will come as circumstances permit.

Originally Posted by MandJ

I was surprised at the relatively low readings taken about 1.5m from the DC housing exhaust opening, air blowing past the sensor, highest I got was PM2.5 of 17 ug/m3 and PM10 around the same figure, however in fairness, I had cleaned and washed the 4 year old filter bags about 2 months ago with only some light sanding and band saw work been done in that time.

.

Light sanding and BS work will have conditioned your filters nicely.

The sort of numbers you are getting are pretty similar to what I see on my system.

You can approximately measure the efficiency of the filters by measuring the dust concentration of the air being collected in inside the shed with the dust concentrations going out of the DC. There is no need to make any extra dust while doing this.


What this shows is
a) how well sealed your DC is,
b) how efficient the filters are especially at the PM2.5 - PM10 dust removal.
c) that you are moving large amounts of air during extraction which dilute the dust concentrations.

You could in effect vent that DC inside the shed but without continuously monitoring with a particle counter you would never know where you stood.

Thanks for the FB. To be honest, I'd have been disappointed if the readings were anything other that what I got at the Lathe, with that bell mouth and minimal loss between the pick up and 3hp DC it should have been as good as I could get, but nice to have it confirmed. The sanding station and Band saw cutting tall veneers and logs are where I'm really looking forward to doing some testing. The router table DC design is so good that I expect nothing to show up there, but I won't get a chance to test that until the weekend.

BTW, yes, I used silicon sealer on every join and every seam on the DC housing, but over time, and as more saw dust builds in the bags, I can't stop small leaks around the bag mounting collars, but as you have posted before, that is always going to be a problem with bags, and of course the occasional seam leak- I also have such open access to the DC housing (one complete long side is a door) that I can wipe the inside of the housing completely clean (very little clean up needed though) each time the bags are emptied and cleaned.

The new PMSA003 particle counter arrived today.
It supposedly has the same specs as the PMS7003 that M&J and I have used in our other counters.

Here are a few pics - the main difference is the A003 is about 25% smaller.

This seems to show the fan on the A003 is larger but the impeller is the same size but the opening to the fan on the A003 is bigger - this doesn't make for more airflow because that is limited by the sizeof the outlet which is the same on both counters.
PMSA0031.jpg

PMSA0032.jpg

This shows another difference in that the fan (black box) on the A003 can be unscrewed and rotated by 90º which take the inlet further from the outlet reducing the likelihood of the fan rebreathing the same particles.
Reminds me of moving the outlet vent of a DC away from the shed inlet.

PMSA0033.jpg

Was about to order my dust sensors from Aliexpress and stumbled across these ones from Element14 - HPMA115S0-XXX - HONEYWELL - Gas Detection Sensor, PM2.5, Laser Scattering, Air, 15 %, HPM Series | element14 Australia

Being Honeywell branded and for ~$35 delivered, I think I will give them a go.
Datasheet link - http://www.farnell.com/datasheets/23...hoCkqQQAvD_BwE

Anything in the specs that you can see gents that would make these units unfit for WW usage?

I'd hold off until Bob answers as it looks a bit like some the lower cost units. A lot less data out of this thing by the quick look I had. I know Bob looked at quite a few before going with the unit we have.