Awesome video!! Pro tip: (seriously, I'm a dust collection engineer) Make sure to earth that collapsing flex hoze. The metal wire in there is not to make it springy, but to function as earth wire to avoid static buildup and discharge... I live in a country of merely 11 million people and each year we have 34 dust collection explosions (and consequent total shop losses) caused by failure to properly earth sections that build up static charge... :) I thought that might be a stat that could spark your interest ;) =P (swidt? haha)
I use a bucket top cyclone device to clean out my wood burner. The ash make so much static electricity, I get zapped pretty hard core. The system is all plastic, so there's nothing really to ground.
And some flex hose like the 4" I'm using is antistatic without a metal wire, just has a rigid polymer helix instead and all the polymer is doped to have a surface resistance for dissipating charge to a ground wire.
@@usertogo no. having static electricity will always remain a huge explosion risk. It's fairly simple: every mixture of air and a combustible product (powder, granulate, gas, flakes,... and some materials that you don't even consider combustible like sand or flower are equally combustible anyway) has a certain concentration where it gets explosive. In the collection container where the mass sticks together, the concentration of the product is 100%, before the inlet of the collection system, the concentration is 0%, so by definition, somewhere in between there's a point at the perfect explosive concentration. ANY spark at that point will ignite the shit. So you absolutely positively want to get rid of all spark sources...
We used separators extensively in the lab and found through our experimentation that an inlet angle declination of 7 to 10 degrees made for the greatest debris drop-out.
William can you recall what pressure and air volume /min was used and did you try varying those? For some uses high suction, low vol can be ok while in others you want high air volume and lower suction is ok.
just to not misunderstand you, when you say declination, you are talking from the airflow's point of view, or the filter's point of view? (so is the air blowing downward or upward at the entrance?) It's good to share any results from experimentation, because there is a lot of trial-and-error in these systems, simply because there are so many variables, too many to run simulations or base solutions on calculations alone.
Hey! i've actually done these experiments myself a few months ago. I found that, as you mentioned, a steeper slope was better for larger particulates. But my results were also greatly improved when I elongated the cone and allowed the air more time to circulate. Additionally, I found that sanding the inside to be as smooth as possible helped. Great video, love ur channel.
Yeah, I was coming here to say the same thing - make the inside as smooth as possible. Also, it doesn't have to actually be a cyclone - I built a Thien baffle for my wood shop. Straight sides, but very smooth (bent acrylic), and as mentioned elsewhere, the exit tube comes *way* into the chamber - I think my chamber is about 24" diameter about more than 12" tall, it's 6" inlet and 6" outlet, and the outlet starts about 3" above the base of the chamber. It works very well. But, I've not actually tried to optimise it - very much a case of "well that's good enough then!".
I own the Makita cyclone separator with stick vaccum and my friend has the Dust Deputy they both have a steeper slope. They work well. I think the air velocity also matters in the cyclone chamber. The larger and longer the cyclone is, the more air volume and slowing of the air velocity in the top of the cyclone giving smaller particles a chance to fall rather than getting sucked back up. When you look at Dyson it does really well keeping the filters clean but the CFM on those are only around 60 on turbo so it can be a small cyclone system. A decent 5-6 hp shop vac could be 150cfm and those industrial separators are huge as was shown in the video which would be really high CFM. There would be a point where a cyclone separator could be too big and the vacuum CFM cant form a cyclone to. Another thing would be the way the air enters the cyclone, in the video its a really sharp edge and I would think that is causing a lot of disturbance right by the exit hole The Dust Deputy has a much more gradual way of taking the air and smoothly entering the cyclone. I'm thinking small particles would better make it to the walls of the cyclone with controlled airflow.
The velocity of the air stream, size and density of particles, and slope and diameter of the cyclone matter a LOT more than you'd think. The multi cyclone is essentially optimized for three different particle sizes, since it was done more or less randomly there's no telling what exactly those particles might be.
How well do these cyclone filters work with slower speeds with small particles? Ive got a small mushroom farm and during fruiting you get a lot of spores being released, would be nice to be able to use something like this on my air exchanger to cut down on buying filters.
A feature of many cyclone separators is what is termed a 'vortex finder' or 'outlet length'. The outlet pipe dips down into the inside volume of filter. This forces air from the inlet to orbit around the separator surface before exiting the top of the filter. This length is important for adjusting filter performance for pressure drop vs. fractional efficiency. Many papers use terms designated as 1D2D, 1D3D etc to describe the cyclone separator dimensions. The numbers represent the ratio between the cyclone diameter and cone length. Many papers are published which compare filtration fractional efficiency (as a function of particle diameter), pressure drop, and flow rate through the filter for a given Mean Particle Diameter. In general, a higher aspect ratio of cyclone with a large outlet length will filter better because it exposes the air flow stream to longer linear lengths of surface area. The principle is that at the boundary layer, air velocities are slow enough to drop out material against the surface, and the slope of the surface keeps the material from re-entering the fast air stream. Lower cyclone diameters will filter out smaller particles better, at the expense of significant pressure drop. According to ASHRAE, saw dust has an approx mean diameter of 600 micron, and flour dust has an approx mean diameter of 50 micron, for your reference. www.sciencedirect.com/science/article/pii/S0307904X06000291 This was a paper we used to guide our design of these separators. You should check out papers on the subject if you really want to optimize your filter.
Yeah for some strange reason most of the 3Dprinted project on internet get this ratio "wrong", I wondered if it was because of empirical tests that for some reasons makes the result different when 3Dprinting the thing (rough surface, PLA leakage, ...), or just a social reproduction thing where the first who did it just made it this ratio without thinking much (which is totally fine, sometimes you just want to make a quick and dirty thing without research and/or calculation) and the rest of the 3d printing community just followed it as a design trend ?
@@lolaa2200 In one hand i really start to need cyclone filter asap since i have lots of routering coming up near future, but then again i would like to get optimized as possible filter to save those shopvac bags, but then again i really dont have time or interest at this moment to go and learn about these filters that indebt, so has anyone optimized this and is there 3D model somewhere?
Point to note: the multi-stage design _should_ allow you to tune the first stage to remove big chunks at a cost of letting the fine stuff through, to be dealt with in stages 2 and 3.
I think the particle size being targeted with each stage is probably backwards, probably smallest to medium to largest, but I'd think you'd want to go the other way around.
Yes! It seemed the smaller series version did well with large particles. So maybe have that first and a large cyclone as a second stage, something like that.
@@Broockle There are losses on each stage. Since the airflow is equal, the pressure drop is about the same at each (equal) stage. The main advantage of multiple stages is probably getting the filtration efficiency really high. Each stage filters a certain fraction of dust so you can cut it down exponentially with more serial stages.
Also I'm not sure what's the relationship of size and pressure loss and filtration efficiency, there might be some tradeoff involved (again for different particle sizes as well).
"It became clear I needed to vacuum my floor more often" Even a genius needs a little help from technology to learn life skills. Love your content and your passion for everything you do.
If I remember my lectures correctly, you usually have the outlet not at the top but a bit further down the cyclone so the particles are slower. Maybe adding a small tube in the center will help. Would also be a quick print, if not just a cutted pipe is enough.
This is why I made two separate units in series. The first is your typical dust collector, the second is literally a giant bong. The air goes into water this making it super clean.
If you look at a Dyson vacuum, each tier of cyclones has a separate collection chamber separated by a seal, also rather than multiple similar sized cyclones each one is reduced in size in order to filter different sized debris, the main shroud for large debris the subsequent in size for finer particles. The diffusion method might be effective as lower airspeed but you have to keep in mind you want maximum power rather than an obstruction on purpose as that not only reduces cleaning performance but also puts strain on the motor which in a non-shop vac vacuum relies on pass through air for cooling.
Actually, most of those motors have external fan blades to circulate air around the motor for cooling. A centrifugal fan can't run the aspirated air over the motor because it's in an enclosed cavity. I understand what you're trying to say in the second half of your comment, but the reality is more complex than that. (the bit about Dyson is correct tho)
I appreciate this. I've been using a 20$ Oneida cyclone for almost 10 years and it's nice to see why it works. The other attempts don't add to the science but they really do help to see why the original design is so effective.
I did cut an entire flying wing on my mpcnc, it broke at the maiden flight, but still fun to do! I'll need to improve the design! Here is the video! ruclips.net/video/P0NrUnZaeNY/видео.html
Simply routing the air through a larger tube will drop the air velocity since Q/A=V where V is velocity, A is area and Q is flowrate. If you bump up the area the velocity goes down proportionally. The middle of the tube will have a much higher velocity (especially since this is 3D printed and doesn't have a smooth surface) so it would require a series of diffusers to make sure the velocity field is uniform. You'll need the bottom to be open to the bucket for the dust particles to descend and accumulate. If you get an anemometer (they are like 20 bucks) you can measure the velocity of the flow in your vacuum tube which combined with the cross sectional area can give you an estimate of how large the dust collector tube would need to be to drop the air velocity. The surface area of the opening in the diffusers should be equal to the original (non enlarged) tubing size. Basically the particles fall downward at a fixed rate decided by gravity and their drag. The velocity inside the collector must be low enough that a large portion of particles (>95-99%) have plenty of time to fall downward out of the moving air stream.
I might have read this wrong, but doesn't the high velocity of the air generate the centripetal force that separates the heavier particles. This also might just be one factor, or I could be completely wrong.
@@terminus9897 ummm. Centripetal means striving to the center... Since this comment is about axial flow, not about any kind of rotation, centripetal forces are not at play. @Entropic0 is right that the slowing of the air will make the dust get caught by gravity if the speed drops slow enough. The smaller the particles, the slower it needs to be. But also, the smaller the particles, the slower they fall because of the ratio between friction, drag and the load capacity (which increases with speed). So it's not enough to calculate a lower speed, but you also need to calculate a minimum length of this wider section. For fine dust, these numbers would go up so high that it becomes unrealistic to implement. So it's a great idea for the larger size particles (except the diffusers, as they will capture particles and eventually clog up the sides, increasing the speed in the center until a point where it's fast enough to prevent the particles for getting caught, which will also be where the particles will no longer drop out of suspension.)
We’ve done a fair amount of research on these cyclones for our projects. There are a lot of academic papers out there which pretty thoroughly describe the performance characteristics of these things given different geometries and boundary conditions. A given geometry tends to preferentially separate out particles with a particular density. In other words, a given cyclone might perform great for steel grinding dust but not so well for sawdust. So your results will depend on what you’re trying to separate out. Fun project, thanks for sharing!
id like too see one used on a sub-7L engine and or in slowmo as i 🤔 if the cylinder cycling would let it work 🤔 or with water 💦/humidity. as for his i think 💭 the lines need to be smooth out as the one video frame looks like the parting line or lines is getting in the way and undoing the point of filtering out the junk and adding resistance for the vacuum cleaner
The slow motion sexy CNC and cyclone shots to the music was absolutely hilarious and I replayed it about 5 times, great video! I love 3D printing but never thought I'd sit through a video on a DIY shop vacuum, you made what should be a relatively boring thing into engaging content! Great vid!
When it comes to cyclones, bigger is always better simply because you have to match the angle of the cone with the speed of the air, so having one thats bigger means you dont have to worry too much about the angle as it will slow down the air alot. At my work we have a few of them and they do very well with both very fine metal dust and metal shavings simply because they are about 7 feet tall and 3 feet wide.
Awesome job, I really like the idea of them in series! I actually designed one of these for a project at work. I used it as part of a system that allowed us to lift sand to an elevated hopper. This was placed in the elevated hopper to force the sand to separate from the air. I found that having a tube extending into the chamber a few inches from the exhaust port helped the efficiency dramatically. I was also using a chamber made from a piece of acrylic tub so the walls were vertical with a PVC reducer at the bottom.
cyclones are a massive help in this use case. I made one myself and had some issues with the container seal... so I put clingwrap on the container as a separator, sprayed eurathane foam into the lid and clipped it on. Cure, cut off excess and now have perfect seal. Increased rigidity to prevent vacuum collapse and perfect seal.
So cool, how 3D printing gives "garage inventors" ability to design and test ideas. Daniel, I have a feeling that 3D printing combined with the Stepcraft router will be taking you to a whole new creative level. Go for it!
Great video! If you look at the 2nd generation Dysons, they have over a dozen on tiny cones. The tinier the lower end of the cone the higher the speed, but also the higher the resistance to the airflow, thus the massive parallelism. As mentioned elsewhere the inside of the cones should be as smooth as possible, to keep the speed up and reduce resistance of the filter (acetone vapor might help here). Also both the low cost and industrial filters I know of, have the air outlet at the top hang down a bit, for you that would be about 2". I like your serialization, have been thinking about that, but haven't tried it yet. I think two phases with the second being much smaller might give better performances: one big cone surrounded in Dyson fashion by a dozen of tiny ones.
I had the same issue with noise. I replaced a vacuum hose, with one designed for sucking water, pool use. The pool one was unbearable with noise. I went back to using two vacuum hoses connected together to get my desired length.
not only ribs cause noise, also the air speed. In my case I eliminated the noise of a small tube by using a bigger diameter one. I initially used a 1-1/4 (IIRC) electrical flexible tube ribbed and noise was terrible. I moved to a metal one, with external rubber lining that is 1-3/4 and noise went away. I guess the metal and rubber stuff might have something too, but air travel way slower.
This whole video I was screaming out loud "WHY ARE YOU TRYING TO REINVENT THE WHEEL????? I took "Air Pollution Control" in my Master's program for Civil Engineering. Cyclone separators are very simple and easy to engineer, build, and understand. I'm sure you can find stuff online about how to design and build them and what performance to expect based on particle size and density. You could even dissect nearly any bagless vacuum cleaner to get ideas on how they work and the proper proportions. One feature I noticed that was neglected in most if not all the designs, is that the exit pipe in the top center should extend into the cyclone chamber, a certain number of diameters of the pipe, like 1 or maybe a half, I don't remember. I think it's called a "vortex finder." This is a great practical video, I'm glad you made it, and I mostly enjoyed watching it, but you could have started much further ahead than you did with just 5-10 minutes of reading.
Some enjoy experimenting and observing on their own. Many learn better this way, and often new discoveries come from this kind of tinkering and observation discipline.
Stealing stuff from the past makes sense. But doing it by learning is GREAT FUN and you might invent new stuff not even heard of before. So keep doing by experimenting! Love the video's.
@@simon-pp5sm I get that. Personally, I like to start with something that already works and iterate it into something better, rather than just bash at it until it works kinda, which is what happened here. He stopped before he even had a really good result.
Irrespective of Brilliant being an excellent sponsor - thank YOU for recording, editing and producing this content for us for free. It's among some of the most enjoyable content that the internet has to offer these days and I, for one, certainly appreciate it.
It's so fun being an engineer! You go down all these rabbit holes you wouldn't have thought of and you end up learning so much! That's such an important part of being an engineer!
Always appreciate your videos!!! The collection efficiency of cyclones varies as a function of particle size and cyclone design. Cyclone efficiency generally **increases** with (1) particle size and/or density, (2) inlet duct velocity, (3) cyclone body length, (4) number of gas revolutions in the cyclone, (5) ratio of cyclone body diameter to gas exit diameter, (6) dust loading, and (7) smoothness of the cyclone inner wall. Cyclone efficiency will **decrease** with increases in (1) gas viscosity, (2) body diameter, (3) gas exit diameter, (4) gas inlet duct area, and (5) gas density. A common factor contributing to decreased control efficiencies in cyclones is leakage of air into the dust outlet (EPA, 1998).
I'd use first a gravity based (maybe mechanical theough some big mesh) filter and then the monocyclone for the rest. So basically into a bucket for the heavy big stuff, out of it and through a tube to the monocyclone. Pretty cool tests you did there 👌
Great video! It would be interesting to see if there is a noticeable difference between a smooth surface in the cone, compared to the ridged cone due to the printing layers. Also, it would be interesting to see if there is a noticeable difference if the top cylinder section was a bit taller, before it becomes the cone. I've seen designs where the cylinder height is actually taller than the cone section.
I was wondering the same, as far as smooth vs stepped, or even if you had ridges spiraling down to direct the airflow? Anyway, very cool, just got my first printer for xmas, so good project ideas👍
Awesome analytic and iterative descriptions. I'd worked this very topic about 10 years ago and was struggling with many of the same issues that you were able to visually expose. The biggest reason for 'short circuiting' or bypassing the centerfuge effect is not having the proper pressure drop in the chamber. Adding an extension tube to the clean air exit that's past the tangential entrance helps quite a bit. Another is having the correct aspect ratio of height to width based on your entrance velocities. Scaling the cyclone up or down works to a point...when debris is being passed through the cyclone or even being sucked up the center vortex you have too little of a pressure drop...you need a larger diameter cyclone to compensate for the velocity of your system. If you play with Dyson vacuums you'll find they have precisely scaled cyclones for their velocities. They don't have near the flow or suction of a shop vac but have near perfect separation and filtering because they tuned the entire system. Your shop vac is your control point... The cyclone needs to be tuned for your shop vac.
Lazy? Nah, you were smart. To quote myself, “I can do anything but I can’t do everything.” Just because you can do something doesn’t mean it’s worth the time to do it.
While watching this video it kept reminding me of one core concept, the scientific method. What a great job! You proposed your idea, researched it, made your hypothesis, tested it, analyzed the results, refined your idea, and repeat. Gotta love the scientific method!
This video was actually very interesting. You make it seem so simple. Since you do planes and boats.. I would LOVE to see a GPS guided submarine, heck even a remote controlled one would be amazing (I don' think you've made a sub yet as far as I'm aware)
@@radpugguy I think the idea isnt gonna work. The wavelength of visible light makes it hard to concentrate to a small area. Even if you use laser, the projection on earth would be a few meters in width. A floating repeater above water could be a good idea, then it could use i2c or 1 wire to deliver the data to the sub. Or you could use sonar to detect the bottom of the sea/lake, then use that as location data? Or use 3 lasers onshore to shine a beam to the sub. If they could rotate freely, they could know how far and at what angle the sub is. Using three could triangulate the location. Or use sonar boardcasting on the sub, then three recievers to measure the delay, like the black box of planes if they fell in water.
Commercial particle separators have aggressive vertical sections before the cyclone allowing more heavy material to drop out, have an exhaust tube intaking below the cyclone intake port, have a plug in the dust port producing annular discharge (like a turbine), and sometimes use a smooth downward curve in the ceiling to introduce the vortex into the vertical segment. In the extreme cases this warps it from a "funnel" shape to a sort of fungal bloom, with a perforated internal riser instead of a conical segment, and a ball at the top allowing recirculation before dropping into the separator proper. I'd love to see more science and less art here, but I appreciate your time and funds wouldn't fit testing 200 extremely similar designs either. Great work so far!
Interesting. Given volume in equals volume out(roughly), I would suggest your air flowing in should have a smaller hose and the bottom of the cyclone should have a bigger hole, also the hole exiting the top should be larger as this will reduce the escape velocity of the material and hopefully minimise the amount of really fine particles remaining suspended. Another really interesting way to go is utilising water. Dust hits like a water wall and then is captured. I use this for sprays. Edited - elaborated what was implied by v originally. Mawyman below gives more of an explanation for volumetric flow. In essence your vacuum cleaner generally always pulls the same volume of air through it. If you have the straight pipe with no attachment on it you pick up some stuff, i you change it for one with a smaller opening it pick up most stuff.
Well technically the continuity equation is Ain*Vin = Aout*Vout or in other words Qin = Qout where q is volumetric flow rate. It’s clear you understand the areas impact but I figured I’d just clarify for future peeps.
@@JeronimoStilton14 haha yes, rightly put. In my haste I've suggested in the opening line that v in equals v out. I meant to imply volume but see where there could be a clear issue. Just wanted to imply that what goes in as volume must for the most part come out. Been a while since ive done fluid dynamics haha 😄 thanks for clearing that up though 😀
@@JeronimoStilton14 congrats mate. Finished civil about 10 years ago. So a vary bare bones shell left of the knowledge. I did smash Laplace Transform though haha
My suggestion for a future cyclone would be to print a series of spiral splines that force the particles down as they travel around the inside of the cyclone. Kind of like a thread pattern inside the cone.
RUclips has been recommending this video to me for two weeks now. Finally took the hint and watched it :D Definitely worth it and really enjoyed the video!
I use a commercially made cyclone on the input to my shop vac. Some really small stuff makes it through, but it saves on filter bags as most of the waste ends up in the bucket and not in the shop vac. Some big bits still orbit the cyclone without falling down, I wondered whether some vertical fins around the edge to stop them rolling would work, however since my cyclone does not come apart I cannot test this idea. However if I use it to sweep the floor and the intake get blocked, it sucks the bucket flat, so I had to make a vacuum relief valve to stop this, it also jeeps the circulation going in the cyclone. I note that Dyson do something similar.
im not sure what happened where i stumbled on a 14 minute video of dust separator tests by an rc enthusiast at 3am but i have no regrets. intriguing content!
Большая работа проделана!!! Спасибо большое за познавательное видео! Думал так же делать, но Вы хорошо проверили мои идеи и они оказались неэффективными, Вы сохранили мое и не только мое время, спасибо!
I absolutely love that you are testing cyclone separators! Your design skills are insane! I use a handful of these in my little machine shop, and a two cyclone setup for my floor vacuum. One thing that I've found that helps is at the port that goes from the cyclone top to the shopvac hose, to have a portion of the tube protrude a half inch to an inch into the cyclone top (down into the cyclone chamber) rather than it being a clean port into the shopvac hose. To be honest, you may have done this already, but I wanted to share this in case you hadn't tried it already. I think the idea is to give the solids a wall that needs to be hopped in order to make it into the shopvac hose. Granted, it looks like your designs are working pretty good, especially the large one. I would imagine that the shape of this barrier could play a role in how much gets past the cyclone too, as the solids mainly stick to the surfaces. (I think the original clear plastic dust cyclone has this little barrier, if you need a better visual than what I'm explaining) It may not improve the solid collection a whole lot, but any little bit that helps keep it out of that shopvac filter keeps suction as close to 100% as you can! I have managed to keep my shopvac filters nearly pristine for a few years now. Also not machining foam either though.. which is probably the toughest thing to keep out of your vac filter! Awesome stuff!!!
Great video! Other than increasing the steepness, consider making the subsequent cones with a smaller radius to increase the centrifugal force on smaller particles.
Great vid! I was just learning about James Dyson today! Too bad he didn't have a 3d printer, would have made prototyping his original design much easier! He apparently went through 5,127 prototypes! P.S. Idea for next video: "Testing 5,128 Different Cyclone Air Dust Separators"
You are missing a crucial part. The outlet needs to have a tube reaching down into the cyclone . About 1,5-2x the diameter of the intake . Also : the smaller the diameter of the cyclone is , the finer are the particles these get separated . Don't make it so small though that the air out of the intake smashes into the outlet tube . The upper part needs to be cylindrical and only a couple of outlet diameters under the outlet tube it needs to taper in . There must not be an airflow from the solids-outlet into the cyclone or it won't work, airflow out of the solid outlet needs to be kept to a minimum . There are some simple Formulars for dimension ratios to get decent results out of them and there are some way over the top complicated methods of modelling them . I did a cyclone separator to separate oil from. A stream of CO2 in my thesis . If you want to go into detail on how they actually work , I would gladly give you some info .
Seconded. I did some r&d on a separator several years back, but settled on a dust deputy. I drill bolt holes in concrete quite often and the dd in front of a hepa filter does the trick. Now, I just so happen to be searching for the perfect air/oil separator on a turbofied car engine. Any detailed info you might have would be welcomed here.
Not a high-tech-tip - but : if you put the wastebag inside bucket with the opening clamped into the sealing of the bucket, you don't have to decant the mess into the bag, when the bucket is full. I'm really amazed by your creativity!
I find this topic fascinating. Would be interesting to see if restricting airflow to the vacuum could filter out different particel sizes and densities (like metal dust).
I've watched & read 3,247 dust collection videos & articles, & wound up building a Thein baffle (a flattened cyclone). 1. Your dust and everybody else's may differ, much like carpet sweepings vs shop debris. 2. Air velocity vs particle size is one variable of separation (which is why large sawmills have multiple separators on a single air path, collecting chips & chunks in the primary & then smaller stuff in the subsequent stages. Particle density is another variable. 3. Cyclone size matters vs time for gravity separation. Horn style is likely to be more efficient by a teeny margin vs a well crafted Thein. But the Thein on my dust bin fits nearly under the work bench, just my preference.
For an airspeed separator, make a triangular chamber. The air enters at a lower corner, blowing up half the angle (i.e. at a point below the outflow). The outflow is in the upper corner and on the other side from the inlet. The outflow diameter must be bigger than any previous diameter including the nozzle. The bottom of the triangle opens into a bucket. Now the trick is to make the air path long enough that the dust has a chance to drop out of it before the airflow turns up into the outflow. Inside the chamber you have 2 air flow shaping systems meet: The incoming air will want to try a slightly inclined line and the outgoing air will then pick that air up and lift it to the outflow. Kind of like a hockey stick.
Love the video. Since the pitch of the cone changed which particle sizes it gathered most efficiently, have you thought about making a multi-stage separator, similar to the series separator except the pitch of the cones change to accommodate for the size of particles most likely not caught by the prior stage(s)?
Great video. I use 2 different cyclone pre-collector systems in my work shop. I have both a 4" high volume low pressure dust collection system with a 40 gallon steel can with a cyclone lid as the pre-collector and a a 2" shop vac system with the 5 gallon bucket and a cyclone on the top. I found the 5 gal bucket lid a bit of a pain to get on and off to empty the bucket. So I use a double bucket system where the top bucket has the bottom cut out of it and the lid always stays attached to the top bucket. It still seals when in use but to empty I just pull the buckets apart and it's easy to empty the bottom bucket. the system is also a few inches deeper.
If you print the intake to the cyclone to where it transitions into an oval and the radius of the oval is perpendicular to the cyclone wall instead of being formed out of the wall then you’ll get much better results as you won’t be blowing some of the particulate more towards the center and it will stick to the walls a bit harder
For what it's worth the serious defusers have water spray bars in them too. We have some of them where I work. The water spray helps to basically catch all of the dust and wash it out of the bottom. They're great for industrial scale when you start needing filter bags and the kind of maintenance that comes from abrasives. Probably not so much for what you're doing.
So, I use these Cyclone vacs all the time. One thing you might want to take into consideration is flow rate of your air. I use them to vacuum grain, and the flow rate is EVERYTHING
Thats more or less his entire problem, flow volume is way to high for the size of cyclone, and the single cyclone is way, way too wide and short. All he had to do was search Kice, Walinga, Brandt, or any company that makes grain vacs or commercial dust collectors to get a list of scalable dimensions that would work without fucking around with so many half assed designs.
Speechless for that effort. Try to add spiral wings to the inner of the mono cyclone, with the same arrangement in the direction of air rotation, to force the dirt to the hole. Congratulations
You can modify a serial cheme. Inside bucket you can place some walls like in your diffuser setup, to prevent leaks. Also by varying input and output nozzle you can adjust speed. Adding a tube halp length of a cone to output can limit bypassing. Also idea for single cone variant. Add a spiral cone in middle. So ascending air also generate additional vortex colinear in its tangential motion, but opposite in axial.
Thank you for this research and sure that many rolls of filament you spent just for testing! I'm still lucky with my china cyclone, but I had to calculate that the fine dust will be filling up the vacs bag sooner or later. The advantage is that I can use a quiet household vac instead of an industrial, noisy one. Just the problem of the bucket squeezing has to be solved by inner fins or an under pressure (self designed, 3d printed) valve.
best DIY cyclone and testing I've seen yet! I want to share my thoughts how to improve it.......but I think I want to keep it to myself as Dyson's latest vacuum designs are old or quite flawed, can easily be challenged.
I don't know if anybody else mentioned it but be careful with static electricity, especially in the winter. All that air friction and saw dust can be a bad mix. Festool even sells some anti static hoses I think. I mention that for design ideas not for purchase considering how great you are at rapid prototyping this kind of stuff, keep it up. Can't wait to see all the foam projects.
The squeal is coming from the first section of tubing. Same thing happened to me when I tried to use a cheap flexible tubing originally designed for a sump pump. Vacuum tubing is specially designed with a smoother internal bore AND anti static additives in the plastic. Spend the extra money for proper vacuum tubing sold as a replacement for name-brand vacuums.
I work in a crisp factory and we use cyclones to remove potato starch from the water. Water is used to push the crisps from the slicer, through various systems. This water needs to be recycled through cyclones otherwise startch builds up around the tanks and causes motor failures etc. The cyclones we use are very tall and narrow. The end nozzle is roughly 2mm while the intake and outtake pipes are around 3 inches. Intake is where the water goes in, outtake which is at the top is where it sucks the water up. The intake and outtake pipes are adjustable. If we close the outtake pipe, it creates more pressure and more water comes out the nozzle. We do this if we have a potato that has more startch content. If we close the intake pipe, less water goes on but this can help remove more startch but at the cost of making the tanks hold back dirty water. The slicer water tank uses 4 cyclones, the peeler water tank uses 2 cyclones. In total they are 3ft tall. They are quite thick so I don't now the internal diameter at the top but I assume it would have to be the same diameter as the intake / outtake pipes.
If you look at those industrial designs you see a straight cylinder above the cone so when the particles spin they fall with gravity, the slope of the cone + spinning allows them to stay in orbit. If they orbit at the same level as the outlet then they can jump inward and into the outlet, if they orbit lower the middle cyclone will have time to spin them outward.
Make sure to crimp a ground wire to the metal wire in that flexible hose and screw it to the metal frame of the router to prevent static build up in the vacuum system.
Some of these handheld battery powered vacuums are just using a tiny HEPA filter that can clog quickly. I was thinking on making a cyclone separator for it, this is some great stuff!
I'm not sure if this has been recommended yet, but you should try a two-stage system with the first being a defuser and the second stage being the cyclone. In the defuser you should take the half tubes across the entire section alternateing their direction with a little bit of overlap (like full tubes that have been cut in half and one half moved down so its edge is in the middle of the other half, and repeat until it covers the entire thing. I'm not sure if I'm explaining this well enough). The air will be able to move quickly around the curves but any larger items will fall down because they can't make the quick turns. and the dust will then be picked up through your cyclone design.
I have to cyclones in series in my private workshop. On large for larger debris and a minor one for smaller particles. The intake pipe is 7 cm in diameter angled 45 degrees downwards inside the funnel on both however, on top of the first cyclone i have a small reversed funned, so the air loose some velocity near the lid. On top of the second one I have a 35 cm wide 45 cm pipe with a very steep angled, long, reversed funnel on top. For 6 months I have been using the same vacuum bag even though I vacuum everything from ashes to sawdust. The greatly reduced airflow from intake to vacuum in these two stages helps gravity collect almost all particles.
Try using a Diffuser with a checker board blocker with smaller grids as you get closer and air pulls into a Cyclone or 2 stage chamber. With the bucket glue a divider to prevent outlet from mixing air flows once it deposits in bucket from both sides. Larger dust balls and dirt will be sorted in the first chamber, where the second chamber will filter small particles. have the air intake be pulling from the small particle side. Have a cone attached to the intake pointing into the bucket on the cyclone side similar to a wasp trap this would make it harder for dust to puill up into the intake hose and swirl around the edges of the lid.
Cool man, I did something similar those experiments make it fun. I will share my last version is to just use two buckets with the hose angled to the side on them in series. Some anti-static dryer sheets help too.
Your video is awesome, and also are the comments with tips and tricks for improving your dust cyclone separator design backed up by scientific and real world data. Just reading the comments already made me wiser by a bit. Keep up the good work and I'm also looking forward to see what stuff did you made with that nice CNC router. Cheers!
Nice concept, It seems the single cyclone is pretty good, though I believe the best option would be to make sure each cyclone is in series and dumps the waste into a separate container. It all comes down to how much space you wish to use
It's important to consider air volume vs air (vacuum) pressure. FOr a single tool - a high suction lower airflow may be ok. For longer runs etc and some timber dust, high air volume at lower pressure becomes more useful. Tube diameter plays a big role in this.
I feel like the boundary layer deceleration is more important than the outward force. Definitely interesting ways to test this with different surface textures and straight sections to increase the boundary layer thickness
Awesome video!! Pro tip: (seriously, I'm a dust collection engineer) Make sure to earth that collapsing flex hoze. The metal wire in there is not to make it springy, but to function as earth wire to avoid static buildup and discharge... I live in a country of merely 11 million people and each year we have 34 dust collection explosions (and consequent total shop losses) caused by failure to properly earth sections that build up static charge... :) I thought that might be a stat that could spark your interest ;) =P (swidt? haha)
@sourand jaded Same here, I grounded the sandblaster nozzle directly to earth via wire around the hose, it was constant sparks!
Could there be a way to control the high voltage somehow and use it for additional electrostatic dust retention?
I use a bucket top cyclone device to clean out my wood burner. The ash make so much static electricity, I get zapped pretty hard core. The system is all plastic, so there's nothing really to ground.
And some flex hose like the 4" I'm using is antistatic without a metal wire, just has a rigid polymer helix instead and all the polymer is doped to have a surface resistance for dissipating charge to a ground wire.
@@usertogo no. having static electricity will always remain a huge explosion risk. It's fairly simple: every mixture of air and a combustible product (powder, granulate, gas, flakes,... and some materials that you don't even consider combustible like sand or flower are equally combustible anyway) has a certain concentration where it gets explosive. In the collection container where the mass sticks together, the concentration of the product is 100%, before the inlet of the collection system, the concentration is 0%, so by definition, somewhere in between there's a point at the perfect explosive concentration. ANY spark at that point will ignite the shit. So you absolutely positively want to get rid of all spark sources...
We used separators extensively in the lab and found through our experimentation that an inlet angle declination of 7 to 10 degrees made for the greatest debris drop-out.
Any experiments on the smoothness of the walls? It seems how rough the wall is may be counterintuitive.
@@timramich All of ours were made by the sheet metal shop on base so they were smooth walled to begin with.
Did you try a step in the angle?
William can you recall what pressure and air volume /min was used and did you try varying those?
For some uses high suction, low vol can be ok while in others you want high air volume and lower suction is ok.
just to not misunderstand you, when you say declination, you are talking from the airflow's point of view, or the filter's point of view? (so is the air blowing downward or upward at the entrance?) It's good to share any results from experimentation, because there is a lot of trial-and-error in these systems, simply because there are so many variables, too many to run simulations or base solutions on calculations alone.
Hey! i've actually done these experiments myself a few months ago. I found that, as you mentioned, a steeper slope was better for larger particulates. But my results were also greatly improved when I elongated the cone and allowed the air more time to circulate. Additionally, I found that sanding the inside to be as smooth as possible helped. Great video, love ur channel.
Yeah, I was coming here to say the same thing - make the inside as smooth as possible. Also, it doesn't have to actually be a cyclone - I built a Thien baffle for my wood shop. Straight sides, but very smooth (bent acrylic), and as mentioned elsewhere, the exit tube comes *way* into the chamber - I think my chamber is about 24" diameter about more than 12" tall, it's 6" inlet and 6" outlet, and the outlet starts about 3" above the base of the chamber. It works very well. But, I've not actually tried to optimise it - very much a case of "well that's good enough then!".
I own the Makita cyclone separator with stick vaccum and my friend has the Dust Deputy they both have a steeper slope. They work well. I think the air velocity also matters in the cyclone chamber. The larger and longer the cyclone is, the more air volume and slowing of the air velocity in the top of the cyclone giving smaller particles a chance to fall rather than getting sucked back up. When you look at Dyson it does really well keeping the filters clean but the CFM on those are only around 60 on turbo so it can be a small cyclone system. A decent 5-6 hp shop vac could be 150cfm and those industrial separators are huge as was shown in the video which would be really high CFM. There would be a point where a cyclone separator could be too big and the vacuum CFM cant form a cyclone to. Another thing would be the way the air enters the cyclone, in the video its a really sharp edge and I would think that is causing a lot of disturbance right by the exit hole The Dust Deputy has a much more gradual way of taking the air and smoothly entering the cyclone. I'm thinking small particles would better make it to the walls of the cyclone with controlled airflow.
Trust in Jesus Christ
@@juanit0tackit0tackito2 ok
@@juanit0tackit0tackito2 I dont remember a passage in the bible where he used a vacuum cleaner
The velocity of the air stream, size and density of particles, and slope and diameter of the cyclone matter a LOT more than you'd think. The multi cyclone is essentially optimized for three different particle sizes, since it was done more or less randomly there's no telling what exactly those particles might be.
Yes, I agree, try the Stairmand geometry, you can find the sizing table with google ^^
How well do these cyclone filters work with slower speeds with small particles? Ive got a small mushroom farm and during fruiting you get a lot of spores being released, would be nice to be able to use something like this on my air exchanger to cut down on buying filters.
A feature of many cyclone separators is what is termed a 'vortex finder' or 'outlet length'. The outlet pipe dips down into the inside volume of filter. This forces air from the inlet to orbit around the separator surface before exiting the top of the filter. This length is important for adjusting filter performance for pressure drop vs. fractional efficiency.
Many papers use terms designated as 1D2D, 1D3D etc to describe the cyclone separator dimensions. The numbers represent the ratio between the cyclone diameter and cone length. Many papers are published which compare filtration fractional efficiency (as a function of particle diameter), pressure drop, and flow rate through the filter for a given Mean Particle Diameter.
In general, a higher aspect ratio of cyclone with a large outlet length will filter better because it exposes the air flow stream to longer linear lengths of surface area. The principle is that at the boundary layer, air velocities are slow enough to drop out material against the surface, and the slope of the surface keeps the material from re-entering the fast air stream. Lower cyclone diameters will filter out smaller particles better, at the expense of significant pressure drop.
According to ASHRAE, saw dust has an approx mean diameter of 600 micron, and flour dust has an approx mean diameter of 50 micron, for your reference.
www.sciencedirect.com/science/article/pii/S0307904X06000291
This was a paper we used to guide our design of these separators.
You should check out papers on the subject if you really want to optimize your filter.
was about to say it
:')
Wonderful explanation.
Yeah for some strange reason most of the 3Dprinted project on internet get this ratio "wrong", I wondered if it was because of empirical tests that for some reasons makes the result different when 3Dprinting the thing (rough surface, PLA leakage, ...), or just a social reproduction thing where the first who did it just made it this ratio without thinking much (which is totally fine, sometimes you just want to make a quick and dirty thing without research and/or calculation) and the rest of the 3d printing community just followed it as a design trend ?
@@lolaa2200 In one hand i really start to need cyclone filter asap since i have lots of routering coming up near future, but then again i would like to get optimized as possible filter to save those shopvac bags, but then again i really dont have time or interest at this moment to go and learn about these filters that indebt, so has anyone optimized this and is there 3D model somewhere?
Excellent stuff, I took a class on this in college but I don't remember enough of it to explain it that eloquently.
Point to note: the multi-stage design _should_ allow you to tune the first stage to remove big chunks at a cost of letting the fine stuff through, to be dealt with in stages 2 and 3.
I think the particle size being targeted with each stage is probably backwards, probably smallest to medium to largest, but I'd think you'd want to go the other way around.
You should test combining the big mono and a smaller mono, since they both filter out different size particles.
Yes! It seemed the smaller series version did well with large particles. So maybe have that first and a large cyclone as a second stage, something like that.
An unfortunate side effect of a combination of separators is reduced air flow. With light weight foam it may not be as big of an issue.
don't u need a stronger vacuum then?
I think the ideal design takes out all the particles but also leaves you as much power on the vacuum as possible.
@@Broockle There are losses on each stage. Since the airflow is equal, the pressure drop is about the same at each (equal) stage. The main advantage of multiple stages is probably getting the filtration efficiency really high. Each stage filters a certain fraction of dust so you can cut it down exponentially with more serial stages.
Also I'm not sure what's the relationship of size and pressure loss and filtration efficiency, there might be some tradeoff involved (again for different particle sizes as well).
"It became clear I needed to vacuum my floor more often" Even a genius needs a little help from technology to learn life skills.
Love your content and your passion for everything you do.
How to make a genius vacuum his floor? "You have the right to design and build the vacuum!"
If I remember my lectures correctly, you usually have the outlet not at the top but a bit further down the cyclone so the particles are slower. Maybe adding a small tube in the center will help. Would also be a quick print, if not just a cutted pipe is enough.
true he should test it, the result will be much better than a top sucking.
Yes, it should extend down a few inches and have a defuser that resembles a hydroponics growing cup that breaks up the laminar airflow.
This is why I made two separate units in series. The first is your typical dust collector, the second is literally a giant bong. The air goes into water this making it super clean.
If you look at a Dyson vacuum, each tier of cyclones has a separate collection chamber separated by a seal, also rather than multiple similar sized cyclones each one is reduced in size in order to filter different sized debris, the main shroud for large debris the subsequent in size for finer particles. The diffusion method might be effective as lower airspeed but you have to keep in mind you want maximum power rather than an obstruction on purpose as that not only reduces cleaning performance but also puts strain on the motor which in a non-shop vac vacuum relies on pass through air for cooling.
Actually, most of those motors have external fan blades to circulate air around the motor for cooling. A centrifugal fan can't run the aspirated air over the motor because it's in an enclosed cavity.
I understand what you're trying to say in the second half of your comment, but the reality is more complex than that. (the bit about Dyson is correct tho)
I appreciate this. I've been using a 20$ Oneida cyclone for almost 10 years and it's nice to see why it works. The other attempts don't add to the science but they really do help to see why the original design is so effective.
Can't wait to see what you cut out with this magnificent thing - level up!
Wow amazing project thumbs 👍, may I ask you what was the Creality 3D printer was used for this project?
SO MANY BOTS WTF
I did cut an entire flying wing on my mpcnc, it broke at the maiden flight, but still fun to do! I'll need to improve the design!
Here is the video! ruclips.net/video/P0NrUnZaeNY/видео.html
@@f_2476 I think he use a modified cr-10
Upload a new video,
Simply routing the air through a larger tube will drop the air velocity since Q/A=V where V is velocity, A is area and Q is flowrate. If you bump up the area the velocity goes down proportionally. The middle of the tube will have a much higher velocity (especially since this is 3D printed and doesn't have a smooth surface) so it would require a series of diffusers to make sure the velocity field is uniform. You'll need the bottom to be open to the bucket for the dust particles to descend and accumulate. If you get an anemometer (they are like 20 bucks) you can measure the velocity of the flow in your vacuum tube which combined with the cross sectional area can give you an estimate of how large the dust collector tube would need to be to drop the air velocity. The surface area of the opening in the diffusers should be equal to the original (non enlarged) tubing size. Basically the particles fall downward at a fixed rate decided by gravity and their drag. The velocity inside the collector must be low enough that a large portion of particles (>95-99%) have plenty of time to fall downward out of the moving air stream.
I might have read this wrong, but doesn't the high velocity of the air generate the centripetal force that separates the heavier particles. This also might just be one factor, or I could be completely wrong.
@@terminus9897 ummm. Centripetal means striving to the center... Since this comment is about axial flow, not about any kind of rotation, centripetal forces are not at play.
@Entropic0 is right that the slowing of the air will make the dust get caught by gravity if the speed drops slow enough. The smaller the particles, the slower it needs to be. But also, the smaller the particles, the slower they fall because of the ratio between friction, drag and the load capacity (which increases with speed). So it's not enough to calculate a lower speed, but you also need to calculate a minimum length of this wider section.
For fine dust, these numbers would go up so high that it becomes unrealistic to implement. So it's a great idea for the larger size particles (except the diffusers, as they will capture particles and eventually clog up the sides, increasing the speed in the center until a point where it's fast enough to prevent the particles for getting caught, which will also be where the particles will no longer drop out of suspension.)
There was a Practical Engineering video on this for settlement bins, but it was for waste water treatment. Fascinating stuff!
We’ve done a fair amount of research on these cyclones for our projects. There are a lot of academic papers out there which pretty thoroughly describe the performance characteristics of these things given different geometries and boundary conditions. A given geometry tends to preferentially separate out particles with a particular density. In other words, a given cyclone might perform great for steel grinding dust but not so well for sawdust. So your results will depend on what you’re trying to separate out. Fun project, thanks for sharing!
id like too see one used on a sub-7L engine and or in slowmo as i 🤔 if the cylinder cycling would let it work 🤔 or with water 💦/humidity. as for his i think 💭 the lines need to be smooth out as the one video frame looks like the parting line or lines is getting in the way and undoing the point of filtering out the junk and adding resistance for the vacuum cleaner
The slow motion sexy CNC and cyclone shots to the music was absolutely hilarious and I replayed it about 5 times, great video! I love 3D printing but never thought I'd sit through a video on a DIY shop vacuum, you made what should be a relatively boring thing into engaging content! Great vid!
I appreciate the use of animation, clear acrylic, and inclusion of windows so we can literally see what's going on. Thanks!
When it comes to cyclones, bigger is always better simply because you have to match the angle of the cone with the speed of the air, so having one thats bigger means you dont have to worry too much about the angle as it will slow down the air alot. At my work we have a few of them and they do very well with both very fine metal dust and metal shavings simply because they are about 7 feet tall and 3 feet wide.
Awesome job, I really like the idea of them in series!
I actually designed one of these for a project at work. I used it as part of a system that allowed us to lift sand to an elevated hopper. This was placed in the elevated hopper to force the sand to separate from the air.
I found that having a tube extending into the chamber a few inches from the exhaust port helped the efficiency dramatically. I was also using a chamber made from a piece of acrylic tub so the walls were vertical with a PVC reducer at the bottom.
cyclones are a massive help in this use case. I made one myself and had some issues with the container seal... so I put clingwrap on the container as a separator, sprayed eurathane foam into the lid and clipped it on. Cure, cut off excess and now have perfect seal. Increased rigidity to prevent vacuum collapse and perfect seal.
So cool, how 3D printing gives "garage inventors" ability to design and test ideas. Daniel, I have a feeling that 3D printing combined with the Stepcraft router will be taking you to a whole new creative level. Go for it!
Great video!
If you look at the 2nd generation Dysons, they have over a dozen on tiny cones. The tinier the lower end of the cone the higher the speed, but also the higher the resistance to the airflow, thus the massive parallelism. As mentioned elsewhere the inside of the cones should be as smooth as possible, to keep the speed up and reduce resistance of the filter (acetone vapor might help here). Also both the low cost and industrial filters I know of, have the air outlet at the top hang down a bit, for you that would be about 2". I like your serialization, have been thinking about that, but haven't tried it yet. I think two phases with the second being much smaller might give better performances: one big cone surrounded in Dyson fashion by a dozen of tiny ones.
it resonates because the tube is ribbed, if the tube was smooth it wouldn’t have made sound. You can use a smaller diameter smooth tube.
But he needs to make sure its suction tubing, e.g. for water pumps. They still have ribbing, but only on the outside.
I had the same issue with noise. I replaced a vacuum hose, with one designed for sucking water, pool use. The pool one was unbearable with noise. I went back to using two vacuum hoses connected together to get my desired length.
not only ribs cause noise, also the air speed. In my case I eliminated the noise of a small tube by using a bigger diameter one. I initially used a 1-1/4 (IIRC) electrical flexible tube ribbed and noise was terrible. I moved to a metal one, with external rubber lining that is 1-3/4 and noise went away. I guess the metal and rubber stuff might have something too, but air travel way slower.
This whole video I was screaming out loud "WHY ARE YOU TRYING TO REINVENT THE WHEEL?????
I took "Air Pollution Control" in my Master's program for Civil Engineering. Cyclone separators are very simple and easy to engineer, build, and understand. I'm sure you can find stuff online about how to design and build them and what performance to expect based on particle size and density. You could even dissect nearly any bagless vacuum cleaner to get ideas on how they work and the proper proportions.
One feature I noticed that was neglected in most if not all the designs, is that the exit pipe in the top center should extend into the cyclone chamber, a certain number of diameters of the pipe, like 1 or maybe a half, I don't remember. I think it's called a "vortex finder."
This is a great practical video, I'm glad you made it, and I mostly enjoyed watching it, but you could have started much further ahead than you did with just 5-10 minutes of reading.
Some enjoy experimenting and observing on their own. Many learn better this way, and often new discoveries come from this kind of tinkering and observation discipline.
Okay dude
Stealing stuff from the past makes sense. But doing it by learning is GREAT FUN and you might invent new stuff not even heard of before. So keep doing by experimenting! Love the video's.
@@simon-pp5sm I get that. Personally, I like to start with something that already works and iterate it into something better, rather than just bash at it until it works kinda, which is what happened here. He stopped before he even had a really good result.
@12:45 you got bars!!! I see you also are channeling the spirit of Rick Flair Wooo!
Irrespective of Brilliant being an excellent sponsor - thank YOU for recording, editing and producing this content for us for free. It's among some of the most enjoyable content that the internet has to offer these days and I, for one, certainly appreciate it.
It's so fun being an engineer! You go down all these rabbit holes you wouldn't have thought of and you end up learning so much! That's such an important part of being an engineer!
Always appreciate your videos!!!
The collection efficiency of cyclones varies as a function of particle size and cyclone design. Cyclone
efficiency generally **increases** with (1) particle size and/or density, (2) inlet duct velocity, (3) cyclone body
length, (4) number of gas revolutions in the cyclone, (5) ratio of cyclone body diameter to gas exit diameter,
(6) dust loading, and (7) smoothness of the cyclone inner wall. Cyclone efficiency will **decrease** with increases
in (1) gas viscosity, (2) body diameter, (3) gas exit diameter, (4) gas inlet duct area, and (5) gas density. A
common factor contributing to decreased control efficiencies in cyclones is leakage of air into the dust outlet
(EPA, 1998).
Any books or texts that you recommend on the subject? I'm trying to learn more about it
I'd use first a gravity based (maybe mechanical theough some big mesh) filter and then the monocyclone for the rest. So basically into a bucket for the heavy big stuff, out of it and through a tube to the monocyclone.
Pretty cool tests you did there 👌
Great video!
It would be interesting to see if there is a noticeable difference between a smooth surface in the cone, compared to the ridged cone due to the printing layers.
Also, it would be interesting to see if there is a noticeable difference if the top cylinder section was a bit taller, before it becomes the cone. I've seen designs where the cylinder height is actually taller than the cone section.
I was wondering the same, as far as smooth vs stepped, or even if you had ridges spiraling down to direct the airflow?
Anyway, very cool, just got my first printer for xmas, so good project ideas👍
Cyclone being one of my favorite songs to this day. thank you for that!
I'd love to see more experimental videos on dust separators. This was really cool
Awesome analytic and iterative descriptions. I'd worked this very topic about 10 years ago and was struggling with many of the same issues that you were able to visually expose.
The biggest reason for 'short circuiting' or bypassing the centerfuge effect is not having the proper pressure drop in the chamber. Adding an extension tube to the clean air exit that's past the tangential entrance helps quite a bit.
Another is having the correct aspect ratio of height to width based on your entrance velocities. Scaling the cyclone up or down works to a point...when debris is being passed through the cyclone or even being sucked up the center vortex you have too little of a pressure drop...you need a larger diameter cyclone to compensate for the velocity of your system.
If you play with Dyson vacuums you'll find they have precisely scaled cyclones for their velocities. They don't have near the flow or suction of a shop vac but have near perfect separation and filtering because they tuned the entire system.
Your shop vac is your control point... The cyclone needs to be tuned for your shop vac.
Great project, Daniel! Thought about printing one a few years ago. Was lazy, and bought one :)
Lazy? Nah, you were smart. To quote myself, “I can do anything but I can’t do everything.” Just because you can do something doesn’t mean it’s worth the time to do it.
While watching this video it kept reminding me of one core concept, the scientific method. What a great job! You proposed your idea, researched it, made your hypothesis, tested it, analyzed the results, refined your idea, and repeat. Gotta love the scientific method!
This video was actually very interesting. You make it seem so simple. Since you do planes and boats.. I would LOVE to see a GPS guided submarine, heck even a remote controlled one would be amazing (I don' think you've made a sub yet as far as I'm aware)
That would be amazing, not sure about how to get GPS signal underwater though
GPS underwater? I dont think thats a good idea. You might need a floating repeater for that.
Isn't DIY perks making this?
you're gonna need to launch you're own satellite constellation using visible light if u want to pierce the water. alternatively, buoy.
@@radpugguy I think the idea isnt gonna work. The wavelength of visible light makes it hard to concentrate to a small area. Even if you use laser, the projection on earth would be a few meters in width.
A floating repeater above water could be a good idea, then it could use i2c or 1 wire to deliver the data to the sub.
Or you could use sonar to detect the bottom of the sea/lake, then use that as location data?
Or use 3 lasers onshore to shine a beam to the sub. If they could rotate freely, they could know how far and at what angle the sub is. Using three could triangulate the location.
Or use sonar boardcasting on the sub, then three recievers to measure the delay, like the black box of planes if they fell in water.
Commercial particle separators have aggressive vertical sections before the cyclone allowing more heavy material to drop out, have an exhaust tube intaking below the cyclone intake port, have a plug in the dust port producing annular discharge (like a turbine), and sometimes use a smooth downward curve in the ceiling to introduce the vortex into the vertical segment.
In the extreme cases this warps it from a "funnel" shape to a sort of fungal bloom, with a perforated internal riser instead of a conical segment, and a ball at the top allowing recirculation before dropping into the separator proper.
I'd love to see more science and less art here, but I appreciate your time and funds wouldn't fit testing 200 extremely similar designs either. Great work so far!
Interesting. Given volume in equals volume out(roughly), I would suggest your air flowing in should have a smaller hose and the bottom of the cyclone should have a bigger hole, also the hole exiting the top should be larger as this will reduce the escape velocity of the material and hopefully minimise the amount of really fine particles remaining suspended. Another really interesting way to go is utilising water. Dust hits like a water wall and then is captured. I use this for sprays.
Edited - elaborated what was implied by v originally. Mawyman below gives more of an explanation for volumetric flow. In essence your vacuum cleaner generally always pulls the same volume of air through it. If you have the straight pipe with no attachment on it you pick up some stuff, i you change it for one with a smaller opening it pick up most stuff.
Well technically the continuity equation is Ain*Vin = Aout*Vout or in other words Qin = Qout where q is volumetric flow rate. It’s clear you understand the areas impact but I figured I’d just clarify for future peeps.
@@JeronimoStilton14 haha yes, rightly put. In my haste I've suggested in the opening line that v in equals v out. I meant to imply volume but see where there could be a clear issue. Just wanted to imply that what goes in as volume must for the most part come out. Been a while since ive done fluid dynamics haha 😄 thanks for clearing that up though 😀
@@cjcro6456 I’m a recent mechanical engineering grad so fluid dynamics is painfully fresh in my mind haha.
@@JeronimoStilton14 congrats mate. Finished civil about 10 years ago. So a vary bare bones shell left of the knowledge. I did smash Laplace Transform though haha
Omgg the Cyclone outro was a throwback! Awesome video!
My suggestion for a future cyclone would be to print a series of spiral splines that force the particles down as they travel around the inside of the cyclone. Kind of like a thread pattern inside the cone.
RUclips has been recommending this video to me for two weeks now. Finally took the hint and watched it :D Definitely worth it and really enjoyed the video!
I use a commercially made cyclone on the input to my shop vac. Some really small stuff makes it through, but it saves on filter bags as most of the waste ends up in the bucket and not in the shop vac.
Some big bits still orbit the cyclone without falling down, I wondered whether some vertical fins around the edge to stop them rolling would work, however since my cyclone does not come apart I cannot test this idea.
However if I use it to sweep the floor and the intake get blocked, it sucks the bucket flat, so I had to make a vacuum relief valve to stop this, it also jeeps the circulation going in the cyclone. I note that Dyson do something similar.
im not sure what happened where i stumbled on a 14 minute video of dust separator tests by an rc enthusiast at 3am but i have no regrets. intriguing content!
Большая работа проделана!!! Спасибо большое за познавательное видео!
Думал так же делать, но Вы хорошо проверили мои идеи и они оказались неэффективными, Вы сохранили мое и не только мое время, спасибо!
I love that the simplest design is also the most effective.
I absolutely love that you are testing cyclone separators! Your design skills are insane! I use a handful of these in my little machine shop, and a two cyclone setup for my floor vacuum. One thing that I've found that helps is at the port that goes from the cyclone top to the shopvac hose, to have a portion of the tube protrude a half inch to an inch into the cyclone top (down into the cyclone chamber) rather than it being a clean port into the shopvac hose. To be honest, you may have done this already, but I wanted to share this in case you hadn't tried it already. I think the idea is to give the solids a wall that needs to be hopped in order to make it into the shopvac hose. Granted, it looks like your designs are working pretty good, especially the large one. I would imagine that the shape of this barrier could play a role in how much gets past the cyclone too, as the solids mainly stick to the surfaces. (I think the original clear plastic dust cyclone has this little barrier, if you need a better visual than what I'm explaining) It may not improve the solid collection a whole lot, but any little bit that helps keep it out of that shopvac filter keeps suction as close to 100% as you can! I have managed to keep my shopvac filters nearly pristine for a few years now. Also not machining foam either though.. which is probably the toughest thing to keep out of your vac filter! Awesome stuff!!!
Great video! Other than increasing the steepness, consider making the subsequent cones with a smaller radius to increase the centrifugal force on smaller particles.
Great vid! I was just learning about James Dyson today! Too bad he didn't have a 3d printer, would have made prototyping his original design much easier! He apparently went through 5,127 prototypes!
P.S. Idea for next video: "Testing 5,128 Different Cyclone Air Dust Separators"
The small corrugated hose will sing even a low air flow. You can take a 2-5 foot section and swing it around like a lasso and it sings.
You are missing a crucial part. The outlet needs to have a tube reaching down into the cyclone . About 1,5-2x the diameter of the intake . Also : the smaller the diameter of the cyclone is , the finer are the particles these get separated . Don't make it so small though that the air out of the intake smashes into the outlet tube . The upper part needs to be cylindrical and only a couple of outlet diameters under the outlet tube it needs to taper in .
There must not be an airflow from the solids-outlet into the cyclone or it won't work, airflow out of the solid outlet needs to be kept to a minimum . There are some simple Formulars for dimension ratios to get decent results out of them and there are some way over the top complicated methods of modelling them . I did a cyclone separator to separate oil from. A stream of CO2 in my thesis . If you want to go into detail on how they actually work , I would gladly give you some info .
Wow, great info! I'd be interested in anything you wished to share on this subject.
Seconded. I did some r&d on a separator several years back, but settled on a dust deputy. I drill bolt holes in concrete quite often and the dd in front of a hepa filter does the trick.
Now, I just so happen to be searching for the perfect air/oil separator on a turbofied car engine.
Any detailed info you might have would be welcomed here.
Not a high-tech-tip - but : if you put the wastebag inside bucket with the opening clamped into the sealing of the bucket, you don't have to decant the mess into the bag, when the bucket is full. I'm really amazed by your creativity!
I find this topic fascinating. Would be interesting to see if restricting airflow to the vacuum could filter out different particel sizes and densities (like metal dust).
it does :) but you really don't want to experiment with mixed dust (metal and wood)...
@@lukearts2954 Oh, but I do >:-)
@@OperationDarkside hahahaha, I think you need a different channel. I'd suggest Explosions&Fire for example =)))
I've watched & read 3,247 dust collection videos & articles, & wound up building a Thein baffle (a flattened cyclone). 1. Your dust and everybody else's may differ, much like carpet sweepings vs shop debris. 2. Air velocity vs particle size is one variable of separation (which is why large sawmills have multiple separators on a single air path, collecting chips & chunks in the primary & then smaller stuff in the subsequent stages. Particle density is another variable. 3. Cyclone size matters vs time for gravity separation.
Horn style is likely to be more efficient by a teeny margin vs a well crafted Thein. But the Thein on my dust bin fits nearly under the work bench, just my preference.
Awesome job. I’m going to add one of these to my lab setup! What kind of glue do you use for the PLA?
For an airspeed separator, make a triangular chamber. The air enters at a lower corner, blowing up half the angle (i.e. at a point below the outflow). The outflow is in the upper corner and on the other side from the inlet. The outflow diameter must be bigger than any previous diameter including the nozzle. The bottom of the triangle opens into a bucket. Now the trick is to make the air path long enough that the dust has a chance to drop out of it before the airflow turns up into the outflow.
Inside the chamber you have 2 air flow shaping systems meet: The incoming air will want to try a slightly inclined line and the outgoing air will then pick that air up and lift it to the outflow. Kind of like a hockey stick.
Love the video. Since the pitch of the cone changed which particle sizes it gathered most efficiently, have you thought about making a multi-stage separator, similar to the series separator except the pitch of the cones change to accommodate for the size of particles most likely not caught by the prior stage(s)?
Great video. I use 2 different cyclone pre-collector systems in my work shop. I have both a 4" high volume low pressure dust collection system with a 40 gallon steel can with a cyclone lid as the pre-collector and a a 2" shop vac system with the 5 gallon bucket and a cyclone on the top. I found the 5 gal bucket lid a bit of a pain to get on and off to empty the bucket. So I use a double bucket system where the top bucket has the bottom cut out of it and the lid always stays attached to the top bucket. It still seals when in use but to empty I just pull the buckets apart and it's easy to empty the bottom bucket. the system is also a few inches deeper.
I was planning on making a cyclone for my dust collector. I'm so glad I found this video! Very helpful!
Rapidly becoming my favorite RUclips channel
If you print the intake to the cyclone to where it transitions into an oval and the radius of the oval is perpendicular to the cyclone wall instead of being formed out of the wall then you’ll get much better results as you won’t be blowing some of the particulate more towards the center and it will stick to the walls a bit harder
For what it's worth the serious defusers have water spray bars in them too. We have some of them where I work. The water spray helps to basically catch all of the dust and wash it out of the bottom. They're great for industrial scale when you start needing filter bags and the kind of maintenance that comes from abrasives. Probably not so much for what you're doing.
That cyclone parody could not have caught me more off-guard, absolutely mental. Subscribed!
So, I use these Cyclone vacs all the time. One thing you might want to take into consideration is flow rate of your air.
I use them to vacuum grain, and the flow rate is EVERYTHING
Thats more or less his entire problem, flow volume is way to high for the size of cyclone, and the single cyclone is way, way too wide and short. All he had to do was search Kice, Walinga, Brandt, or any company that makes grain vacs or commercial dust collectors to get a list of scalable dimensions that would work without fucking around with so many half assed designs.
You are just the coolest person. Clever, talented, funny, and capable. Shine on.
Speechless for that effort.
Try to add spiral wings to the inner of the mono cyclone, with the same arrangement in the direction of air rotation, to force the dirt to the hole.
Congratulations
This is the "doing the work" part of science. Great job dude
This helped me so much to define what type of cyclone I need to build.
You can modify a serial cheme.
Inside bucket you can place some walls like in your diffuser setup, to prevent leaks. Also by varying input and output nozzle you can adjust speed. Adding a tube halp length of a cone to output can limit bypassing.
Also idea for single cone variant. Add a spiral cone in middle. So ascending air also generate additional vortex colinear in its tangential motion, but opposite in axial.
i cant wait to start 3d printing. that router machine and stuff looks so cool how you can carve wood and all that.
Thank you for this research and sure that many rolls of filament you spent just for testing! I'm still lucky with my china cyclone, but I had to calculate that the fine dust will be filling up the vacs bag sooner or later. The advantage is that I can use a quiet household vac instead of an industrial, noisy one. Just the problem of the bucket squeezing has to be solved by inner fins or an under pressure (self designed, 3d printed) valve.
I have an Weird fascination with vacuums so please keep iterating!
A floor sweep could be a fun addition (vacuum assisted hole to sweep stuff into)
My autism loves dust collection so keep it coming man
This is definitely one of the most informative videos in RUclips!
best DIY cyclone and testing I've seen yet! I want to share my thoughts how to improve it.......but I think I want to keep it to myself as Dyson's latest vacuum designs are old or quite flawed, can easily be challenged.
I don't know if anybody else mentioned it but be careful with static electricity, especially in the winter. All that air friction and saw dust can be a bad mix. Festool even sells some anti static hoses I think. I mention that for design ideas not for purchase considering how great you are at rapid prototyping this kind of stuff, keep it up. Can't wait to see all the foam projects.
The squeal is coming from the first section of tubing. Same thing happened to me when I tried to use a cheap flexible tubing originally designed for a sump pump. Vacuum tubing is specially designed with a smoother internal bore AND anti static additives in the plastic. Spend the extra money for proper vacuum tubing sold as a replacement for name-brand vacuums.
I work in a crisp factory and we use cyclones to remove potato starch from the water. Water is used to push the crisps from the slicer, through various systems. This water needs to be recycled through cyclones otherwise startch builds up around the tanks and causes motor failures etc. The cyclones we use are very tall and narrow. The end nozzle is roughly 2mm while the intake and outtake pipes are around 3 inches. Intake is where the water goes in, outtake which is at the top is where it sucks the water up. The intake and outtake pipes are adjustable. If we close the outtake pipe, it creates more pressure and more water comes out the nozzle. We do this if we have a potato that has more startch content. If we close the intake pipe, less water goes on but this can help remove more startch but at the cost of making the tanks hold back dirty water. The slicer water tank uses 4 cyclones, the peeler water tank uses 2 cyclones. In total they are 3ft tall. They are quite thick so I don't now the internal diameter at the top but I assume it would have to be the same diameter as the intake / outtake pipes.
Using your highspeed camera you can see the debris angle. If you 3D print a screw spiral in the cone it may help "force" the larger debris flow down.
The carpet in your shop gave me the shivers. Awesome video btw.
If you look at those industrial designs you see a straight cylinder above the cone so when the particles spin they fall with gravity, the slope of the cone + spinning allows them to stay in orbit. If they orbit at the same level as the outlet then they can jump inward and into the outlet, if they orbit lower the middle cyclone will have time to spin them outward.
Make sure to crimp a ground wire to the metal wire in that flexible hose and screw it to the metal frame of the router to prevent static build up in the vacuum system.
the kitchen towel as a filter looks very useful.
Fun slow motion vids! It was interesting learning that Dyson didn't come up with the cyclone separators too.
made complete sense on how it works and ive seen this before but never completely understood it so cool vid i learned new stuff
Neat! Another dimension of tests you could run is to see how it affects the volume of air being moved / CFM.
Some of these handheld battery powered vacuums are just using a tiny HEPA filter that can clog quickly. I was thinking on making a cyclone separator for it, this is some great stuff!
I'm not sure if this has been recommended yet, but you should try a two-stage system with the first being a defuser and the second stage being the cyclone. In the defuser you should take the half tubes across the entire section alternateing their direction with a little bit of overlap (like full tubes that have been cut in half and one half moved down so its edge is in the middle of the other half, and repeat until it covers the entire thing. I'm not sure if I'm explaining this well enough). The air will be able to move quickly around the curves but any larger items will fall down because they can't make the quick turns. and the dust will then be picked up through your cyclone design.
I have to cyclones in series in my private workshop. On large for larger debris and a minor one for smaller particles. The intake pipe is 7 cm in diameter angled 45 degrees downwards inside the funnel on both however, on top of the first cyclone i have a small reversed funned, so the air loose some velocity near the lid. On top of the second one I have a 35 cm wide 45 cm pipe with a very steep angled, long, reversed funnel on top. For 6 months I have been using the same vacuum bag even though I vacuum everything from ashes to sawdust. The greatly reduced airflow from intake to vacuum in these two stages helps gravity collect almost all particles.
If I didn’t enjoy these videos so much, I’d still come for the sick beats.
Try using a Diffuser with a checker board blocker with smaller grids as you get closer and air pulls into a Cyclone or 2 stage chamber. With the bucket glue a divider to prevent outlet from mixing air flows once it deposits in bucket from both sides. Larger dust balls and dirt will be sorted in the first chamber, where the second chamber will filter small particles. have the air intake be pulling from the small particle side. Have a cone attached to the intake pointing into the bucket on the cyclone side similar to a wasp trap this would make it harder for dust to puill up into the intake hose and swirl around the edges of the lid.
Cool man, I did something similar those experiments make it fun. I will share my last version is to just use two buckets with the hose angled to the side on them in series. Some anti-static dryer sheets help too.
Your video is awesome, and also are the comments with tips and tricks for improving your dust cyclone separator design backed up by scientific and real world data. Just reading the comments already made me wiser by a bit. Keep up the good work and I'm also looking forward to see what stuff did you made with that nice CNC router. Cheers!
Nice concept, It seems the single cyclone is pretty good, though I believe the best option would be to make sure each cyclone is in series and dumps the waste into a separate container. It all comes down to how much space you wish to use
It's important to consider air volume vs air (vacuum) pressure. FOr a single tool - a high suction lower airflow may be ok. For longer runs etc and some timber dust, high air volume at lower pressure becomes more useful. Tube diameter plays a big role in this.
This is very cool!
Try inserting some inches of the clean air tube in the cyclone so the dirty air has some cycles before they meet
I feel like the boundary layer deceleration is more important than the outward force. Definitely interesting ways to test this with different surface textures and straight sections to increase the boundary layer thickness
You should make a spiral on the side to direct the dust into the bottom.
Love your channel keep open the good work!
Subscribed because making a song for your video shows some serious effort, good job man, two thumbs up!!
The greatest part of it is the song is the most replayed part in the video
Albeit short, that was one of the best renditions of cyclone in this website.