Making a particle filter that doesn't wear out for my plasma cutter
2 anni fa
2,074,924
Support these projects on patreon: patreon.com/stuffmadehere
Check out the subreddit: tinyurl.com/smhere
When I was designing the downdraft for my plasma cutter I was concerned that it might eject too much metal dust out of my shop and into my driveway. It turns out that I was right and as I've been cutting more I've noticed a real uptick in accumulated metal on the ground. To address this problem I built a filtering system designed to remove the bigger particles from the exhaust. Since I'm not concerned with smoke and really extremely fine vapor I was able to make a "solid state" filter that takes advantage of the inability of a heavy particle to follow a tight corner. It was inspired by cyclonic seperators on larger shop vacuum systems. From the testing that I've done it seems to work quite well and I'm pretty happy with the results.
These videos usually take a ton of work and a lot of money in tools and materials. I've made a patreon if you're interested in supporting the creation of these projects: www.patreon.com/stuffmadehere
Here's some of the tools that I use in this video:
Hypertherm powermax 45xp with machine torch: amzn.to/2zfoyAv
Hypertherm fine cut consumables (great for sheet metal) amzn.to/34SjMom
100mm linear stage for plasma Z axis - easier than building: amzn.to/3cAeEb3
Anemometer - cheap but seems to work well: amzn.to/2ROtkeL
Downdraft fans - these things chooch!: amzn.to/2VKFbM5
The best marker ever. Always in my pocket: amzn.to/3ewHGtL
Cyclonic dust collector: amzn.to/3ezyghf
Check out the subreddit: tinyurl.com/smhere
When I was designing the downdraft for my plasma cutter I was concerned that it might eject too much metal dust out of my shop and into my driveway. It turns out that I was right and as I've been cutting more I've noticed a real uptick in accumulated metal on the ground. To address this problem I built a filtering system designed to remove the bigger particles from the exhaust. Since I'm not concerned with smoke and really extremely fine vapor I was able to make a "solid state" filter that takes advantage of the inability of a heavy particle to follow a tight corner. It was inspired by cyclonic seperators on larger shop vacuum systems. From the testing that I've done it seems to work quite well and I'm pretty happy with the results.
These videos usually take a ton of work and a lot of money in tools and materials. I've made a patreon if you're interested in supporting the creation of these projects: www.patreon.com/stuffmadehere
Here's some of the tools that I use in this video:
Hypertherm powermax 45xp with machine torch: amzn.to/2zfoyAv
Hypertherm fine cut consumables (great for sheet metal) amzn.to/34SjMom
100mm linear stage for plasma Z axis - easier than building: amzn.to/3cAeEb3
Anemometer - cheap but seems to work well: amzn.to/2ROtkeL
Downdraft fans - these things chooch!: amzn.to/2VKFbM5
The best marker ever. Always in my pocket: amzn.to/3ewHGtL
Cyclonic dust collector: amzn.to/3ezyghf
Nice 'Prototype'! The most dangerous aspect of a prototype, when it works it becomes permanent. I have all kinds of examples of this phenomenon I've built! :-)
@Stuff Made Here How's the "particle filter that doesn't wear out for my plasma cutter" holding up? I'm mostly curious of your manual cleaning cycle and maintenance of the filter.
@Stuff Made Here Yeup, should've made it out of steel to begin with! Ah well, you probably have a new plasma cutter by this point!
My whole shop. Haha
@M LP Making all particles fall within the same radius of the magnet would severely limit air flow (Particles would have to fall within the same radius of the magnet to stop any non similar behaviour)
@Stuff Made Here You could put a magnet in the filter for attracting all the particles
"There's nothing more permanent than a temporary solution" -AVE
@Colm Reynolds ooga unga permanent uhhh uggh temprary guh -grog circa 89,000 BC
@BlueFire Animations that's not in the Bible, I'm just basing it on conjecture.
@BlueFire Animations and the Lord said "everything in this world is temporary, and that is the one thing about it that will never change"
@Jivvi Is that as far back as it goes, or does it go farther?
@Ryan Dunn You're right. The Milton Friedman quote is "Nothing is so permanent as a temporary government program." He was paraphrasing an old Russian proverb "Nothing is more permanent than a temporary solution," which itself came from the even older Greek proverb "Nothing is more permanent than the temporary."
‘I know how to settle a score with engineers - data’ How dare you use our weapon against us
@underourrock but that would remove all the fun of frustration
@Graham Potter advertisers:
Engineer gaming
But he used the figures for the low setting for in filtered and high for filtered only proving that there was a reduction in flow
@Philip Cooksey it was beautiful! So beautiful I can barely look at it! God bless
Clever solution. Reminds me of a Tesla valve. It would be interesting to see if optimizing the shape and positioning of the blades could provide even better flow and filtering.
@Probably Not a Big Toe of our s ,:8,,. ;
Haha I was thinking he same thing!
I feel like the scale is a bit off
This is effectively a “chevron separator”. These are common on liquid vapor systems (like a cooling tower) where you want the vapor (air) to exit out the top of the unit without a significant degree of liquid (water) particles exiting along with. The liquid particles hit the walls to the separator, coalesce, and drip back down if designed property.
With that kind of airflow I'd be tempted to try putting a second layer in, made with smaller diameter tubing to try to catch the smaller particles.
The air is restricted he used two different fan settings to measure the filtered and non filtered
2:41
@Ominous Penguin Maybe use an electro magnet? Very powerful, and could be switched on with the machine running. Then switched off afterwards, leaving the metal to drop down to the bottom.
@Ominous Penguin Doesn't have to be magnetic material, just conductive. Any conductive metal moving through a magnetic field will experience a rotating electric current within itself, which generates a magnetic field that opposes the static magnetic field it's traveling through. This is how motors and generators work, but you can get a simpler more directly applicable example just dropping a magnet through an aluminum or copper tube. Look up eddy current braking.
So, three layers! Bigger -> Existing -> Smaller
I love “Fortunately we live in the 21st century, and they invented alchemy”
haha, yeah, that was a great line. Though, technically speaking, we have indeed invented alchemy, but it usually comes in more of a toroid than a roll. And uses very large amounts of energy relative to the mass converted. Particle accelerators are cool. Speaking of which... Hey Shane, next project idea for your shop: build a particle accelerator. :)
@Brandon quite literally the conservation of mass/energy They weren’t wrong
AND IT COMES ON A ROLL
That caught me of guard
When you want to quantify the exhausted particles make a picture of the sticky area, convert it to a threshold image, that marks the particles in black and the rest in white and count the black pixles. That would give your "probably about 5 % of the particles" estimation more credibility. :)
He sounds like William Osman but makes stuff as well as Adam Savage, it's the best of both worlds.
No he's Michael reeves and Adam savage combined
No
Way more technical. This guy is next level on youtube.
i like your tot style intro
I've always hated Adam Savage so......
I just found your videos and pretty much binge watched them all over two days. I'm an engineer who also likes to invent and make stuff. This video is the last one I viewed and based on what I saw in the others, I was mildly disappointed that you didn't write some code to measure the particle density for each of the tests to do more of a quantitative comparison. ;-) Speaking of code, you frequently mention that it would be boring to spend any time explaining it and perhaps in in the minority but I'd actually enjoy that. Oh, and keep including the failures that lead up to your success. That's the best part of you videos. Lastly, I have to compliment your wife. Her power to put up with you is off the charts and I love the eye rolls that go along with that.
Cool stuff! I considered doing a downdraft and did some experimenting with a makeshift downdraft with a ~30" fan as well as a water table on a rolling cart. I ended up doing a water table with a 55-gal drum that I pressurized with shop air to raise and lower the water. It captures just about everything. A little smoke rolls out the edges of the workpiece. A cross draft in the shop takes care of that, but my next step is to take an industrial curtain (with a tinted translucent center panel horizontally) and enclose it with an exhaust fan to create a negative pressure. Then I run the laptop from outside of that area. I would like to do a downdraft for welding though.
I love your content . You have some good mildly dry humor and a good way of explaining what's going on . Not even halfway through the video I subscribed. Great job man 👍
Very cool solution! I'm curious though, if the steel particles fuse into the aluminum somewhat due to melting it a bit, and potentially oxidizing due to being iron, is there a risk for a thermite reaction after this sees a bunch of use? Could perhaps be remedied by replacing the filter after a bit, but I dunno if it's actually something worth considering or not.
I’m tempted to build one to avoid cleaning my water tray out. Thanks for the great idea. Also ... have you thought about HVAC squirrel cage blowers? Much quieter
Did you consider sending the air flow up to see if gravity helps separate particles from the air?
I work in commercial dust collection. I feel uneasy about explosive steel dust hitting your aluminum coated pvc. The likelihood of something happening is low in your case but mixing those two metals together is very dangerous
Just a small detail on your airflow calculations: Be aware the cross-section of the fan is a circle, it won't have the same airspeed equally. The airspeed will be bigger thr bigger your diameter.... But since you always measured your airflow at the same distance from the center, it won't matter.
Seems to work really nice. I wonder if there's a component of just slowing the air flow down to give the particles more time to fall out. Which should work too, but would probably require a large volume.
I'm guessing you've already considered this... many ages ago I worked for company that build componets for the space shuttle. The specs included recoverable/reusable componets. One of our departments did "tons" of sand and glass bead blasting. Our filtration process was water based. Much like an evaporative coolers, the filtration was 'curtains of water' flowing over screens (the screens were had some sort of electronic charge -I'm not sure what or how). I've known some auto paint shops that use water curtains for filtration as well. FWIW -Thanks for simulating creative thinking processes!
itmores.info/player/video/yKh7ua2dq8hnqdo i think this is what OP was talking about if you were confused
@He Ka I think there's far too high a flow rate for a water tank less than like 5ft tall, and those get really stinky after a while, he'd have to keep refilling the water, etc, etc. I think a simple solution is often the most effective. Even if its not the most efficient, it doesn't take up space like a water tank does. That adds to its value.
I just found your channel the other day - absolutely loving all of your content. Settling the score with engineers via data is way too accurate (speaking as an engineer who's dating a data scientist...)
just a thought, maybe putting adhesive inside bends can help catch more particles and as the filter is easily accessible it can also be replaced after every cutting session, although idk how heat resistant the adhesives are.
To use a flo-meter, it's best to get several measurements at various positions, from the outer edge to the center. Air flow through a pipe/tube varies quite a lot. You should get a more accurate "average" if you take the average air velocity
very cool project! some future upgrades to further improve your dust collection could be to use the new data of dust reduction that you collected to measure the size an mass of the particulate exhausted and calculate a smaller effective radius for those particles to be stopped similar to a cyclotron separator. maybe even implement progressive vanes to collet the smaller stuff. great project tho!
Maybe add magnets to increase collection efficiency? Bonus points for the switchable kind like machinists use.
@Shawn Pitman "If you'd actually listen to the video"... Smartass. Ok back at you. One doesn't listen to video. You listen to audio. Which I did. I'll clarify my question. Water pans are commonly used as a means to suppress fumes and capture the molton metal expelled by the plasma cutting process. I asked for greater clarity on his research, decision and thoughts around the use of a water pan vs his choice of capture and/or extraction. If, Shawn, you have first hand experience in the matter please respond.
@John S He's using it as a fume extractor as well, if you'd actually listen to the video he explains that you don't want to be inhaling the metal that will be vaporized by it.
@TechTroppy When you try to come with "any kind of metal is slowed by magnetic fields" don't then list the two most ferromagnetic metals and that's it and act like you've proved your point.
Most plasma tables I’ve seen use a pan of water below the cutting plane to collect the byproducts of the cutting process. Curious about the reasoning to use this dry evacuation, versus the water pan. Would you share your thoughts around this? Thanks.
For all of those saying magnets wont work because the sparks are above the curie critical temperature and won't be magnetic.... what if you created a system inside the half-cut tubes that re-magnetizes them? As in force them to be magnetic as soon as possible (as soon as they fall below the temperature)
I'm late to the "party" by almost a year, however, if you do read this, have you considered adding an electromagnet as well. oooh or statically charging the aluminum-coated PVC. This might decrease metal dust output even further if the material you are cutting is magnetic.
I always wondered if we could do something like this for car intakes, while the car is driving at speed use the high speed of the air to knock dirt debris and water down a separate path and clean air into the engine
pretty nice. i would have recommended lifting it up, to let the particles go into a collection tray. But hey, there's always room for try number 2.
Might be able to use magnets to catch the magnetic particles. For easy cleaning, just remove/ turn off magnets
I've seen that particular separator configuration used in steam boilers to separate droplets from the gaseous steam. I think it's called a scrubber. More stages makes it more effecting especially if spanwise edge traps catch the particle scooting along the concave surfaces. Cyclone separat oi r have the advantage of handling chunks in the dust stream. Scrubbers are restricted to smaller sized. Lots of engineering info available for scrubbers.
It’s really cool to see how much progress you’ve made in video production over the past half a year. Still making cool stuff, just now with better video presentation
Locomotives use this concept. They are called inertial filters. They work well enough to get about 90 percent of the dirt out of the air coming into the air room where the air is filtered further. There is a thirty horse vacuum cleaner that ejects the dirt out of the top of the locomotive for the next locomotive to ingest.
When I was in the industrial gas turbine engine industry, we used a couple of variants. Both were based on the same principle referred to as Particle Impact Separation. One device was for grading minute oil droplet sizes from various parts of an engine ... so small it looked like a fog. Another was an inlet air filter for a gas turbine APU intended for operation in a dusty, desert environment. It consisted of a maze of twists and turns of varying complexity. Micron-size particles entrained in the air fell out upon smacking into a wall at some of the turns then falling down into a collection chamber to be later purged. I recall the pressure drop to be surprisingly low. That's how I remember it.
As an (aerospace) engineer my intuition tells me that your "obstruction" would not have a very significant effect on airflow, at least for your purposes. Obviously something like a 10% reduction isn't enough to explain an 80% reduction in particle ejection except maybe if the particles just aren't moving as fast so they have more time to cool -- making them harder to see. I say intuition though because saying for sure requires math or at the very least a simulation that does it for you. However, it is important where you hold your flowmeter against the fan. Depending on the geometry of the blades (and this is still usually true), the velocity of the air from the fan will be faster towards the outside rather than the inside. Basically, the velocity diagram across the diameter is more like a V than a flat line. Again, not really that important for your purposes (you were holding it at roughly the same spot), just something to consider. Cool project.
Agree on the air flow test. Taking multiple evenly distanced velocity measurements and average them might give a better result.
@Programmatistís Adiáphoros ah, let the dude have his fun c'mon bro
@Programmatistís Adiáphoros c'mon guys let the little dude have some fun 😋
can you make two sets of the filters to lower even more of the particles? or does that lower air flow too much?
Different size Banks of tubes in series would also work to change the Dynamics of the air flow. You may find you'l get zero partials in the end.
You might look into installing them at an angle. You could maintain air flow rate while getting a boost to efficiency by using gravity to further accelerate the particles. You would see a slight uptick in build complexity but it could be interesting.
I really like the idea and execution. I know this was just a prototype and I don't know if you have made any other video on this but what I would personally do is use metal tubes and make them magnetic. I guess this would slow down the particles even more. In case the metal tubes can't stay magnetic for long enough there are thin and bendable magnetic sheets that can be applied to them. Great content btw :)
It would be nice to see the full setup and how it looks installed at multiple angles. I am interested in using your idea because I also have a downdraft system with the same fans you use and the side of my yard and part of my neighbor's yard looks like a bomb went off!
With that kind of airflow I'd be tempted to try putting a second layer in, made with smaller diameter tubing to try to catch the smaller particles.
What was your air movement on the side before your deflector plates before you install them versus after...curious if the actual capture velocity on the face of the downdraft was greatly reduced...
Am I missing something? Water-tables are pretty awesome. You could probably make it out sheet metal and paint it really well, that or make it out of plastic sheets etc with metal inserts so long as it was deep enough. By the way, love the channel, I happened to stumble by and man, its just awesome keep it up!!! 🇨🇦
If you reduced it to 5% I am curious if that will stack linearly with 2 of your filters in series. 5% of 5% would be a damn good filter.(99.75% reduction?)
@Jurre Volkers I was curious about how far it can be taken, not if it would work for his setup. You know, for science.
This would probably reduce the airspeed too much making the system less efficient which he was trying to avoid
probably wont be 5% of 5%, the particles that made it through likely made it because they are lighter and have less momentum, thus can go through the filter... Although I bet he would still see maybe a 20% of 5% so 99% reduction
I wonder if it would help to add some kind of texture to the surface of the pipes to increase friction on the particles as they slid around the curves?
That was incredibly cool to watch and I found it very I retesting. Thank you for making the video and sharing it so I could watch it!
If you used a conductive metal on the PVC and passed a safe current through it, you could ionize or magnetize the area and make the filter attract the other particles. A bit of plastic clips to insulate instead of screws would keep it separated from the rest of the power system
I just want to say that your videos inspire me to be a better 'maker'. Thanks for sharing your experience. Super interesting and makes me want my own lab.
Cool video! Maybe you're well aware of this, but if you wanted a quantitative measurement of the effectiveness of your filter, you could use free image analysis software (ImageJ comes to mind) to count the particles on your particle glue trap. There's instructions online, but off the top of my head it's just a matter of thresholding the image, then selecting Analyze->Analyze particles... super quick and easy. In fact, I just went ahead and did it using a screenshot from the video... The lighting/focus aren't ideal for this purpose, and if you set out to do it you could photograph a larger area for better statistics, but from my quick and dirty calculation your filter is catching about 88% of particles. The average size of particles in the filtered exhaust also seems to be somewhat smaller - makes sense.
Do you have any plans for your downdraft system on the plasma table? I plan on building something similar with a filter system like you made. Any plans or parts lists would be really helpful to speed up the process.
Should try an experiment with some electromagnets that have an oscillating magnetic field as part of a filter. Any electrically conductive material will be repelled by it, regardless if it's magnetic or not.
One objection about the setup and one tip: before removing ffp1 mask I’d double check if accumulated dust in filter is ejected after some accumulation, particularly at start time. The tip possible electrostatic or chemical « glue » . If are charged or if possible to charge them that is. A possible way to glue particals is on thin film of colofon ( resin) : diluted in alcohol and applied thinly on surfaces. Cleaning pine resin is easy ONLY with oil.
loved all the analysis at the end, especially the sticky tape for a real visual comparison. Data really is the way shut up doubters and keyboard experts.
some magnets glued on the outside of the pvc, may help with slowing down ferromagnetic and diamagnetic metals... depends on how much is oxide and and if its cooled down enough to become magnetic by the point its passing through. Considering how cheap magnets are its not an expensive test if it doesn't work out. If it works will work even better with sheet metal.
You could try applying an electrical charge on the metal dust filter so it holds onto the really fine particles while powered. Or an electro-magnet if you make em from something with iron.
For the next iteration you might use the plasma cutter to slice and dice a sheet of corrugated steel roofing. The curves are not a pronounced, but you could probably set up multiple layers.
The only bummer I see happening is the fine particulate is the only media that mostly won't be caught. Such small debris will travel farther through the path and exhaust into your work space. On the bright side now that you have an obstruction wall you should be able to install a media filter after your S wall and catch more of the fine inhaleable dust.
Since you don't have a dust "collection" system, it's necessary that the dust which falls out of the air is ejected through the fans. You are removing a lot of energy from the dust and coalescing the smaller particles together when they drop down. It could be worth adding a trap at the bottom for the particles to fall out of, I think just the passive airflow will push the particles out the trap. You'll have to build the trap in a manner to prevent a venturi effect, but I think it could work.
@Stuff Made Here Look up how a Torit system works. That's what you're really try to replicate. The air flow is good but you do need to collect the dust somewhere. I've spec'd Torit systems for our three plasma tables. They work like your sawdust collection example. Dust collection is going to be your next hurdle. Good luck. I enjoy the channel.
That was one of my top risks when designing this but I did some testing before building anything and I was able to put a bunch of plasma dust right near the fans and almost none of it was sucked out by the fans. Presumably there just isn't enough flowrate to pick particles up and eject them so I cut out the trap.
Greetings, I know I am a little late to the video, but I am a filter engineer. I have two recommendations. 1) Vane anemometers are notoriously inaccurate, especially a pocket one like that, using a hot wire anemometer would work much better. 2) I would recommend an expanded aluminum, you can find these are ready replacements for a range hood filter, they come in a variety of sizes so you should be able to find one that meets your arrestance & pressure resistance needs. They are normally cleanable, but I am not sure in this case. (Bonus points if you use a pleated design rather than flat sheet form factor).
Love this filter, have an industrial one about 2 stories tall in my shop that saves me a TON of money :)
One question... where do the bits go when they get trapped? Cool design. 👍
This is literally what i do at work, we install units for industrial laser tables to catch the metal particles. Very cool video i wish i could have helped
Is there a place for the particles to drop into instead of simply collecting in the baffles of the filter? You should have a tray or something below it and then thats all you would ever need to empty apart from maybe giving the baffles themselves a once over and a clean for smooth airflow.
If you were to add one more layer of baffling, could you get the exhaust particles down to less than a fraction of 1 percent, maybe? Maybe the aluminum tape coated PVC becomes layer 2 and layer 1 gets to be made out of chimney flashing metal tacked together? Let the metal be the first baffle and the PVC be the second.
One modification you could do is have the pieces push the particles a bit more into the airflow instead of just into a wall, would give you better efficiency
We had a moisture separator in our shop that worked on the same principle as your filter. It worked a lot better when it was rotated 90 degrees so that the particles had to go from down to up instead of side to side.
Sounds like the perfect application for an axial flow cyclone. Basically a tube with a twisted vane in the tube, also called a “swirl tube cyclonic separator”. The air is made to rotate as it goes through the tube and the heavy particle fly to the outside, clean air stays in the centre. At the end of the tube you accept only fluid from the centre region, the outer region is blocked by a plate, the heavy stuff falls down out the bottom. They have very low differential pressure but you need many of these short tubes stacked on each other like a honeycomb pressed up close (but not touching or they won’t trap the particles) to a holey plate. The holes in the plate are smaller than the tubes because they only allow the air in the centre to escape. I’ve seen these on heavy ore moving trucks, about 1m^2, 2 or 3 stacked next to each other as the air intake filter to the diesel electric power plant. They were only about 10cm thick. They filter the dust from the mining operation from going into the power plant. Also used in coal mining equipment to filter coal dust. Although you do see the swirl idea used in reverse flow cyclone separators, the reverse isn’t necessary. The air isn’t redirected which is why the pressure drop is so small, it goes straight through albeit having been spun around on the way. Not uncommon In water filtration.