Thread: Extracting/minimising metal dust in the workshop for better health

Slow speed wet grinders (they have to be slow speed otherwise there would be water everywhere... : -) overcome this metal dust problem by entrapping the metal swarf in the 'wash'. So, those who use them are sorted out!

But, dry grinding at full or half speed is known to broadcast the fine metal swarf that remains suspended in the air for up to an hour in which time it can be breathed in and ingested.

BobL has taken some readings on the levels of metal dust that remain airborne after grinding...

https://www.woodworkforums.com/f65/metal-dust-measurements1-183227

The simplest preventative measure is to wear an effective respiratory mask for at least an hour after grinding. Also taking the normal care when removing the mask and regularly replacing it or its filter.

The extent to which the metal dust that settles on the floor and other surfaces will become airborne again when disturbed by subsequent activity in the workshop, like walking around, may add to the exposure time.

BobL has designed and built a metal dust collector that feeds into his regular DE system after the metal has been removed from the air stream. It not only removes the metal dust from his workshop but also eliminates the risk of sparks getting through to the wood dust end of his DE system.

https://www.woodworkforums.com/f200/metal-dust-catcher-209392?highlight=metal+dust

Another approach is to place magnets close to the grinder wheel to capture the metal dust particles before they escape beyond the grinder. HSS does contain sub-2μm non-magnetic carbides, but the majority of the swarf particles will contain some magnetic Fe and be picked up by the magnets. However, the magnets will not manage to capture everything, so a respirator should still be used. This setup may not be as efficient as Bob’s metal dust pre-extractor, but it will reduce the amount of settled metal dust spread around to be disturbed and get airborne again. And, it is definitely better than only using a mask while grinding.

That is the approach that I have taken until I've implemented something like Bob's setup.

Note: See warnings about using magnets near the grinder in my next post in which I'll cover my current magnet setup/experimentation.


What other approaches are people using for metal dust control/extraction in their workshops?

Here is my magnet tray setup for my high speed grinder. As you can see I'm still using my cereal box prototype. The rare earth magnets are taped to the bottom and one to the back of the tray.

Magnet metal dust collector after 6 mths.jpg

Warning: If you have a steel body CBN wheel be very careful that these strong magnets don't pull onto the wheel and then let go when spinning at full revs!!!! Mine are magnetically attached to the metal slide of my sharpening jig that the tray sits on.

Grinder setup with magnet collector tray.jpg

This is the accumulated swarf (17g) that has been captured by the magnets over the last six months of sharpening my woodturning tools.

6 month's of dust captured.jpg

I wanted to see how much metal swarf was escaping the magnets in the tray so I hung some smaller magnets about 30cm above the grinder (the small white bag in the image above). The following was what was collected in the magnet tray from grinding just one gouge. It's a noticeable amount.

Amount of dust from one gouge grind.jpg

... and here is the dust picked up by the sentinel magnets above the grinder, which is almost imperceptible after grinding that one gouge.

dust from one gouge grind above grinder.jpg

Warning: There is a risk of the fine swarf attached to the magnets being ignited by the sparks off the grinding wheel and starting a fire fanned by the breeze off the wheel. The fine swarf is held by the magnets in such a way that it is prone to igniting in the same way that fine steel wool can ignite. I have had quite a pyrotechnic display in the swarf inside the tray on a number of occasions. I'm going to have to replace my prototype cardboard tray with something that is more fireproof! I’m using CBN/Diamond grinding wheels that generate minimal sparks, but expect this fire risk would be considerably greater with the heavy spark showers that come off Alox and Carborundum wheels.

I also wear a positive air pressure mask/respirator whenever I'm in the workshop, so I was interested to see if its filters had picked up any magnetic metal dust. I have only had the magnet arrangement on the grinder for six months, but was interested to see if the filters had collected any metal dust over the last 12 months. This won't tell me if there was less metal dust coming through in the last six months, but it will give me a baseline going forward. Here is what I picked up magnetically from the pre-filter...

metal dust on pre-filter.jpg

I didn't pick up anything on the HEPA filter itself.

Even though I do significantly more MW than WW, and I have sarcoidosis which is possibly through to be triggered by metal dust, I do worry less about metal dust than i do about wood dust.

The main reason is that metal dust is between about 7 and 15x faster at falling out of the air than wood dust so unless you are spraying metal dust (Sparks) directly into your face then its unlikely to be a problem.

Perhaps of greater significance than metal dust are the abrasives, and the binding agents used to bond the abrasives to belts or surfaces. These fall out of the air between about twice and 5 times faster than wood dust.

Unless you are into serious reshaping of large numbers of large tool bits the amount of fine metal/abrasive/binder dust generated by regulAR sharpening is likely to be minor compared to that generated by turning or sanding.

Metal workers do have other probably more pressing issues to deal with eg Metal heat treatment, welding and plasma cutting fumes, and the vapours from various lubes and coolants.

BTW if you want to see my metal dust catcher catching sparks coming off a thin kerf metal cutting wheel this may be of interest.
Thin Kerf cutting wheel attachment for bench grinder - Page 4

I'd begin with dust control and never feel safe with a mask on for just 60 minutes.
The P2.5 crap will still be hanging around.

Don't forget that some of that dust, a large fraction, will be smashed abrasive particles
and smashed/vaporized adhesive used to construct the disk.

I carve and shape steel, copper and various seashells for carving inlay, branding irons and so on.
My work station is the cut open bottom of a 2 liter milk jug connected to a ShopVac with a plaster fines bag in it.
Nothing gets away. 1mm chunks don't matter at all.

I wear
hearing protection
full face shield ( exploding cutoff wheels really sting)
gas particle cartridge mask, I can smell the cooked adhesive vapors.

In short, my face is right in it for the small work.

Use a dry grinder with a dust hood. I do a ton of grinding to make chisels. When I grind, I set a box fan near the spark stream and send the dust toward the open garage door. I can heavy grind for half an hour and have no black snot.

Just don't park your car right outside of the garage and do this.

Coarser wheels or belt will greatly decrease the fine dust. Cbn wheels make a very fine dust compared to a ceramic belt or coarse tool room wheel.

Originally Posted by D.W.

Use a dry grinder with a dust hood. I do a ton of grinding to make chisels. When I grind, I set a box fan near the spark stream and send the dust toward the open garage door. I can heavy grind for half an hour and have no black snot.

Just don't park your car right outside of the garage and do this.

Coarser wheels or belt will greatly decrease the fine dust. Cbn wheels make a very fine dust compared to a ceramic belt or coarse tool room wheel.

Ah yes, our snotometers are a very economical indicator of what we are breathing and your box fan arrangement seems to be doing a good job there according to yours, Dave.


I rarely do as much heavy grinding as you. Like most woodturners, I'd be back at the grinder frequently but briefly creating fine airborne metal dust particles approximately every 15-20mins, that is if I was using just one turning gouge and resharpened that whenever it lost its edge. Instead I tend to sharpen half a dozen gouges at a time, but that only takes a few minutes and with longer intervals in between. Nevertheless, I'm concerned about breathing in the fine metal dust that I'm generating and especially so with the exotic alloys that are now in most of our HSS turning tools.


I think you are right about the CBN wheels creating finer airborne metal dust, but I don't have a way of measuring that with any confidence. I can measure the fine PM dust levels in the workshop after grinding but have no way of knowing what fraction of that is metal or wood dust.

Unknown percentage is metal dust.jpg

I'm mostly using a #360 CBN 8" wheel running at 2,800rpm. There is an entry in the Science of Sharp blog that has some SEM images of the swarf off a #320 hand stone which illustrate what the finest particles of metal can look like under high magnification. Note the 2μm scale.

sp320_swarf_06.jpg
source: SWARF! – scienceofsharp

If you didn't know the scale you were looking at there you could easily just see that as the unmagnified swarf that comes from metal machining. The darker piece in the bottom right is a piece of abrasive that has broken away from the stone. Some of the metal swarf is as fine across as a few 0.1μm and with those fibre-like shapes the particles would be quite aerodynamic and tend to stay airborne for longer.

I haven't been able to find an image of the swarf coming off a similar grit CBN wheel, but running at high speed, however, expect the particles sizes will be even smaller and more fractured.

Originally Posted by NeilS

I think you are right about the CBN wheels creating finer airborne metal dust, but I don't have a way of measuring that with any confidence. I can measure the fine PM dust levels in the workshop after grinding but have no way of knowing what fraction of that is metal or wood dust.

It's pretty easy to work this out.
Metal dust falls out of the air quickly so when you first go into the shed any dust present in teh shed will be environmental or wood dust.
Note those values as background values
Then grind some metal and note the dust values on your meter - the difference between that and the background will be due metal and abrasive dust.
You will have to grind a lot of metal to see a clear difference.
If you do this post the counts so we can all see what is going on and I'll perform an uncertainty analysis for you.

BTW - those dust meters like the one you are showing are calibrated for environmental dust with a nominal density of about 1g/cc so you have to multiply by 8 for metal and about 3 for abrasives. With CBN there will be virtually no abrasive dust involved.

Originally Posted by BobL

when you first go into the shed any dust present in the shed will be environmental or wood dust.
Note those values as background values
Then grind some metal and note the dust values on your meter - the difference between that and the background will be due metal and abrasive dust.

That is what I did...

Pre-grind reading

Background PM count.jpg

Post-grinding reading

Unknown percentage is metal dust.jpg

BUT, then the next day (having had a little think overnight) I arrived at the shed when everything would have been settled and I turned on the grinder without doing any grinding.... and got a similar reading to post-grinding???

I wasn't sure if it was relaunched wood or metal dust from the breeze off the grinder wheels.

I'll try to take some more readings under more controlled conditions (like a thorough cleanup before taking any readings) and post any different results here.

Originally Posted by NeilS

BUT, then the next day (having had a little think overnight) I arrived at the shed when everything would have been settled and I turned on the grinder without doing any grinding.... and got a similar reading to post-grinding???

I wasn't sure if it was relaunched wood or metal dust from the breeze off the grinder wheels.

I'll try to take some more readings under more controlled conditions (like a thorough cleanup before taking any readings) and post any different results here.

Try using a reading outside the shed as a baseline but don't be surprised if that's greater than what's in the shed.

Originally Posted by NeilS

Ah yes, our snotometers are a very economical indicator of what we are breathing and your box fan arrangement seems to be doing a good job there according to yours, Dave.


I rarely do as much heavy grinding as you. Like most woodturners, I'd be back at the grinder frequently but briefly creating fine airborne metal dust particles approximately every 15-20mins, that is if I was using just one turning gouge and resharpened that whenever it lost its edge. Instead I tend to sharpen half a dozen gouges at a time, but that only takes a few minutes and with longer intervals in between. Nevertheless, I'm concerned about breathing in the fine metal dust that I'm generating and especially so with the exotic alloys that are now in most of our HSS turning tools.


I think you are right about the CBN wheels creating finer airborne metal dust, but I don't have a way of measuring that with any confidence. I can measure the fine PM dust levels in the workshop after grinding but have no way of knowing what fraction of that is metal or wood dust.

Unknown percentage is metal dust.jpg

I'm mostly using a #360 CBN 8" wheel running at 2,800rpm. There is an entry in the Science of Sharp blog that has some SEM images of the swarf off a #320 hand stone which illustrate what the finest particles of metal can look like under high magnification. Note the 2μm scale.

sp320_swarf_06.jpg
source: SWARF! – scienceofsharp

If you didn't know the scale you were looking at there you could easily just see that as the unmagnified swarf that comes from metal machining. The darker piece in the bottom right is a piece of abrasive that has broken away from the stone. Some of the metal swarf is as fine across as a few 0.1μm and with those fibre-like shapes the particles would be quite aerodynamic and tend to stay airborne for longer.

I haven't been able to find an image of the swarf coming off a similar grit CBN wheel, but running at high speed, however, expect the particles sizes will be even smaller and more fractured.

Yes on the fan, and yes on the heavy grinding being not as representative of grinding tools from time to time, but I think keeping the fines in an airstream somewhere is a good idea. The trouble for my CBN wheel is that with the baldor grinder, the vacuum hood guard doesn't fit (the wheel is too wide).

I can tell you what I see with the various belts, though, and I have a couple of pictures of the heavier stuff that settles below.
1) with CBN or trizact, I see a cloud of fine dust coming off of the wheel. With CBN, sometimes, you have to get the right view across the cut into an open space to see it, but the fines are prolific. I only have a picture of most of what settles - I'd bet on average the fines go farther. On a 5000 foot a minute belt grinder with trizact, the cloud is far more pronounced. The fan is at my feet (where the contact wheel is directed - to make sure that stuff goes out - I'll say why later along with the nose part, there are other issues).
2) with ceramic belts in the 36-60 range (they are aggressive and coarse cutting), what comes off is pigtails. These are partially unhardened and partially hardened steel (The hardened steel is the minority of this - I do that grinding last, but it's fairly easy to set a vacuum hose from a cyclone at the bottom of the separate belt sander that I use for that ( a typical belt sander but with a good flat platen, which my belt grinder doesn't have). CBN isn't really suited well for unhardened metals, so it's out here. Ceramic cuts so well, though, that if you lean into it with carbon steels, it will actually create a combustion point where the spark is and that makes smoke (I want nothing to do with that, either) and not just dust. It will set off the smoke detector in another room
3)..I thought I had a three, but I can't remember it now.

I think there may be a case for using ceramic belts instead of grinder wheels to lessen the load, but that's a different setup, the belts aren't cheap, and they're generally coarse.

Here's the pictures of the settled swarf at about 75x optical - first the CBN wheel (again, I see the dust - it floats away - this is what settles under the grinder and is coarser than the overall swarf profile)
S20210827_0001.jpg

This is off of a 60 grit ceramic belt (there is a table under this sander, so this metal actually fuses together in a mat - it's 3 dimensional and hard to get a good picture of with a very cheap handheld scope).

S20210827_0002.jpg

I shook this a little bit to get the fines out of it - but at magnification, it's clear to see that the fines are still little longer pigtails

S20210827_0003.jpg

There is much less dust in the air from the belt sander (it's far slower than the contact wheel grinder - 2200 feet per second). It's not a great comparison, and I don't have dust laying around right now that would be as I don't have a hardened set of chisels to grind right now, but I will be doing that in the near future and get better comparables.

At slower speed, the belt sander does seem to make less fines, though, and the high speed belt grinder makes large pigtails like this, but the speed complicates this by making combustion in some cases near the point of grind (the combustion is like a solid flame of burning metal about the size of a golf ball in terms of volume - there's no way it's good to breathe it).

As far as the fan goes - the other part of this for grinding for me, I got in the bad habit of not masking up when I wanted to try something for a minute or two, not masking up, and not opening the door. This was early in the process - now any grinding is door open, and fan - no matter how hot or cold it is outside. What I noticed without the fan, even with the door open is that not only would I have dusty snot, but my eyes would hurt the next day. The dust gets in our eyes - we can absorb it. I'm fairly sure that some of it rusts in our eyes (or something of the sort) and irritates them.

With high alloy steels like turning tools, there's definitely less spark, but who knows if trading heavy metals as dust vs. spark in simpler steels is a good trade - neither is probably good.

I will never have the patience for systems that don't make any dust, though, not even just refreshing bevels. What I've found with two other wheel grinders that I have is that when the vacuum is hooked to the dust port on them, very little dust escapes, especially of the fine types.

Some interesting observations there DW.

I got rid of all my oxide grinding wheels, not just because of the grinding dust aspect but because wheel dressing made large amounts of dust.

Grinding is thus performed with
a) belts on 50 mm (limited dust extraction) and 150 mm belt sanders (good dust extraction)
b) 300mm disc sander (limited dust extraction)
c) flap sander discs inside a fume hood - the large swarf goes everywhere but the fine dust is taken care of.
d) two CBN and two diamond wheels , for tooling tips. Each time these are used is usually <30s.

The other things that make large amounts of dust are Scotchbrite polishing mops. The mops themselves also wear way and generate large volumes of dust.

Stick Welding also generate large volumes of dust although I have increasingly moved towards TIG which doesn't use flux coated filler rods and at times no rods at all.
Welding is performed in a fume hood or outside.

I found a few metal dust posts in the MWF that may be of interest
Metal dust measurements1
Metal Dust Issues
Metal dust catcher

Yessir on the welding smoke!! Almost nothing that's combusted is good for us to breathe. I've heard many discuss the life shortening effects of large amounts of welding and improper (lack of) ventilation.

A hood is a good idea. If I didn't have a door to open and a fan to send the dust into the world, I would come up with a hood of some sort. The burning eyes the next day from oxidizing whatever it was in them....not a good thing.

I didn't think about this too much when I was just woodworking, but the amount of it that's produced making metal tools just makes it unavoidable.

Dust hood on a common wheel grinder (any type that has a hose hookup) is such a wonderful thing to catch all of those fines, too.

Installing a fume hood was one of the best things I did in my shed extension.
Fumehood8.jpg
The above picture shows the smaller squirrel cage fan that I installed when I first built it. The vice is on a 75 x75 x 5 mm SHS post and can be removed and the beige steel wings can wrap fully around the bench.

IMG_3880.jpg
This pic shows upgraded fan, and shows I have been using it for spray paining.
The wooden "I" shaped structure has a lazy susan on the top and is used for turning objects being painted.

IMG_3618p.jpg
More recent picture.

As well as spray painting I also weld, grind, and even undertake chemistry experiments under the hood.

Originally Posted by BobL

Try using a reading outside the shed as a baseline but don't be surprised if that's greater than what's in the shed.

Saw this post from you Bob when I got back from my workshop yesterday evening. As it turns out I had taken an outside reading, but unfortunately it isn't going to serve the purpose you intend.

I gave the workshop a good clean to get the background dust inside down to a baseline that wouldn't confound any of the test readings. It took me about an hour with the cyclone running to remove as much of the fine airborne particles stirred up by the cleanup. I also vacuumed the area around my grinders with the drum collector and exhaust positioned outside.

At the end of the cleanup and with the cyclone still running here is the reading I got...

Clean WS c DE running.jpg

As I would be taking the test readings without the cyclone running the plan was to take a series of readings after turning off the cyclone to get a baseline for inside the workshop without any activity going on in there. I know that the readings are up when I arrive at the workshop the next day and before I turn on the cyclone, but not the rate at which it rises without the cyclone going. As Bob points out, the particle count outside the workshop is often higher than inside the workshop and that infiltrates into the workshop over time.

Anyway, while monitoring the PM counter readings inside the workshop they suddenly went chaotic. Had this happen one other time and suspected that the neighbours had arrived home and lit their wood fire. This is the reading that I got outside the workshop...

Smoke outside WS.jpg

And, yes, distinct smell of smoke in the air. Not unexpected during winter up here in the Adelaide Hills. But, no point in trying to do any test readings inside the workshop with that intermittently infiltrating into the workshop.

I'll see what is happening there today. The forecast wind directions are in my favour...