Hey Ruud, you did it! You made it on camera... nicely done! "...my brain is super excited" 😀 Thanks for the work you are doing here and for continuing to share your work. It's incredibly impressive!
Just a thought about the particle counter you mentioned. Dyson vacuums have these sensors built in, and have no doubt spent a lot of time and money in developing it. Would it be possible to hack or straight up incorporate one of their units into your separator design somehow I wonder?? Looking forward to hearing more about the design development of this project 😊 👍
Looks amazing! Hooked on Wood has some really good videos on building a system that works well with the CamVac. Including upgrades, and a custom HEPA filter box
Hi Ruud, I'm also working on a Harvey gyro style separator but instead of using blades to spin the air it's just injected into the chamber at the desired angle (15*). With no blades to get clogged this removes the need for the additional separation stage making it simpler and more compact (maybe better performance?) Designing and modeling the ramped intake however was not simple...
I also have a version tested with the intakes at a 30 degree angle to get the most of the bigger stuff directly aimed into the bins. It works really well for high quantities of dust. While planing it was not measurable on my cheap kitchen scale. Interested to see yours design. Is it possible for you to share a photo of your work with me? You can send it to maakhet@makerr.nl
Great video! Thanks for sharing your know-how and cad files!!! Little idea to safe some money: Use 3mm rubber seal from a spool instead of the o-rings. You can cut them according to your needs and carefully glue the ends together using super glue. Saves quite a few bucks...
I've given a lot of thought to these horizontal style dust separators and I have one issue I can't get past. If you suck up anything bigger than chips made by a table saw, planer, jointer, router, etc., you are very likely going to clog the separator. In a small workshop (I also currently work in a one car garage) I don't have the space for multiple dust collection/vacuum systems, so whatever I install (I don't currently have a dust collection setup) has to be a do it all, one and done system. The problem is in my shop, I suck up small blocks of wood, paper towels, hand plane shavings, etc. while I'm cleaning up. If anything like that is sucked up into that separator, it's just going to get stuck on the first turbine blade it comes in contact with. Thein baffles have a similar problem. Normal conical cyclones are 100% immune. If it can fit through your hose, it'll get separated out and end up in your dust bin. That aside, if you are going to do side by side tests of the different systems, can you also measure the CFM of your blower with and without the dust separator?
I found out that the first stage needed to be more efficient in separation at hi dust volumes. I had some leaves and long ships clogging up. I made the Y-peace of the inlet run inwards at 30 degrees so the large parts were directly aimed inside the bins. That was working very well, only no difference while running the planer. I think this can be optimized even more in my design! I need to have a better testing setup to optimize it even more and be able to test the CFM and pressure drop. I am working to solve that!
Well Done. I am excited to see how your work progresses and hope to follow along and possibly make your design to test in my own shop as well. someday. Good luck.
Random thought: fine water mist(s) prior to exhaust. You'll want to have an exit path that allows any water to condense or otherwise drop out prior to going back into the room, but it should catch all of the micro particulates. Can recirc the water, or use a sterile water filter if you really want.
A wet scrubber is something I considered as a third stage. Only my CamVac is running all the exhaust air through the extractor motor for cooling. Water vapor residues must then have completely disappeared. I will give it a try later on if I change the extractor motors for a type that has a separate chamber for the air pump. Also the particles needed to be able to stick onto the water or get absorbed by the water. Don't know how long the particles needed to be in contact with the water? Maybe it is also possible to pull the dust through the water? And is the resistance it will generate less than running it through a HEPA filter. Many questions that I would like to have answered. So maybe later on I will deep dive into that as well :D.
You should look at a SEN5x on an Arduino as a dust sensor. They're affordable enough that you could use several if you wanted and track ambient dust and differences between stages.
I had a lot of trouble printing the mirrored impeller to get it smooth. It was due to the bambu printer default to print counter-clockwise causing issues with curling for the leading/overhang edge of the blade as it is printed first. Their slicer program does not allow you to change the direction of print. The solution was to use OrcaSlicer which does allow you to change the wall loop direction to clockwise.
Good to hear that you have found a solution to solve the issue! I will print the mirrored impeller this evening again with my settings on my Bambu Lab X1C. I can not remember there was a differents in the prints results the last time I printed them.
I can see what you mean now! The tips of the mirrored impeller are rough and have printing imperfections. I sended them away last time. Now I know why they only ware visible on the mirrored impeller! Fascinating how you solved that issue by making the connection between counter-clockwise printing! I will look into the OrcaSlicer to become familiar with it as well. And I will report it backup the community! Thanks for your input 💪
@@CapturingDust Orcaslicer is based off Bambu's software so will be familiar. Also found that I needed to keep my printer door open to keep the ambient air cooler so that curling was less likely to occur. Granted I was printing in a pretty hot room.
For those trouble shooting their prints, I added a pause in my g-code after ~?150 layers to inspect the print quality, so not to waste too much filament if it was going to be a dud.
I sent my CAD model to pilsons guitars and my thoughts on improvement after his 1st or 2nd video. I am not sure how much he used. I can send my files to you as well.
@@CapturingDust I'm building mine using Fusion_Woodworking's file too. I'm using 200mm diameter tubes and four Festool dust extractor motors for suction.
Yes, and I appreciate you sharing. Since my primary efforts in my shop are around building guitars (and my 3D design skills are hack-ish compared to Ruud's), I didn't take it much further than the results I ended up getting. I was happy with the collection level, though Ruud's design is able to achieve much higher volume. I do appreciate the contributions though!
Love what your doing. French cleat install is perfect for me. How to cut out the pipe? Thinking a jig might be needed. (3d printed of course). My usage is cnc.
I add tree cutting templates to the files you can use to tape on the tube. I used a multi tool for cutting the PVC. I am working on a build video to show you the entire process of building the separator. Hope to finish it in about two weeks.
hé Ruud, heel mooi ontwerp, ik kom al enkele weken of er nog aanpassingen zijn voor ik begin te printen 🙂 a very nice design Ruud, I have 2 questions. The Festool collection bins are not that large, is it possible to provide an output that goes to drums? The design does not have to be so compact, it can have some length, as long as the diameter still fits the printer. Personally I would make two of them, 1 high against the ceiling with the collection drums and the hose on a boom arm underneath for a sander, milling machine, etc. and 1 lower for the devices with larger chips.
Ik ben momenteel druk doende om het EU ontwerp goed te krijgen met ook een alternatief voor de Festool bins met meer volume. Daar is meer vraag naar idd. Er staat nog veel op de planning qua optimalisaties/versies maar het filmen van alles vertraagd het proces meer dan ik dacht 😅. Het is een vrije tijd project dus kan er helaas geen fulltime aandacht aan besteden.
On smaller particles, don't expect to get close to 100% just with the centrifugal separator. And here is a good video about particle filtering, HEPA or not HEPA, amount of clear air filters can supply, and why it might not matter that much, if you are continuously filtering the air in your workshop: ua-cam.com/video/gaQTYrisieA/v-deo.html It's a long video, but really full of useful information! Of course, for a workshop you will need a bigger filter than something sized for regular room use :)
I love your implementation of the festool "buckets" they are just insanely expensive for what they are, a clear bucket, I wonder if anyone has ideas for a cheap solution to replace it
I also looked for closed Euronorm bins as a alternative: norah-plastics.com/stacking-boxes-and-plastic-storage-crates-40-x-30-x-32-cm.html You need to make a clear window in it to see how full the buckets are. Here is a Issue page opened for this topic on my GitHub site: github.com/Makerr-Studio/CentrifugalDustSeparator-/issues/6
@@CapturingDust great, is there a vacuum on the bins? I was wondering if I could find something in ikea to use, but those bins are relatively soft, so I don't know
If you do it with a dual motor CamVac like mine there is quitte a lot vacuum when your hose is blocked. When running freely a cheap bin from ikea can do the job I think. You can try to stack two bins together for more strength. When you run a blower motor in front there is no vacuum. Maybe that is beter for thinner bins.
That would be great! I am looking into the Trotec PC220 as a baseline test device. The Dusttrak is way above my budget :D Like to hear you results and what kind of material range you run.
@@CapturingDustI don’t have any experience with the Trotec, seems like an ok little device. Dusttrak is kind of the go to in my line of work. Where are you situated in the Netherlands, we could compare the Trotec with a Dusttrak (once you get it).
I downloaded and built pilsons guitars design and got it working this week, I'm really impressed with the design. I used your design for the bins that clamp in instead of hoses to them, and it works very well. My only issue I think is that my air pressure is still too high (I'd upgraded the motor on my dust extractor to a bigger stronger one), I think I might need another two seperators. I'm curious about the design changes you did to the original.
Nice you have your separator running! I use 1 motor on my CamVac to try out the different between 1 and 2 motors. In my case the dust settles inside the second stage. Not reaching the second dust bin. What is it you’re expecting with the to high pressure? Is it your second bin that is filling up to quickly? It’s need to sit in the 80-20 range. So 20% in the second bin. That’s what the Harvey separator is aiming for I guess with the frequency drives.
@@CapturingDust I put a decent amount of smaller dust thought it yesterday, it was about 50/50 in each one. Worked well, but I think the flow is too fast through it, not letting the final stage settle.
@ADHDbuilt sounds like you have more the pressure for 8 inch separators I guess? Adding a other 2 of 6 inch will increase the friction to much I think. And to be able to distribute the flow evenly over 4 separators is also a challenge. With two it was even though, I noticed. Do you have any idea of CFM and pressure your new fan is able to produce?
@@CapturingDustI don't have any measuring equipment, but I went from a .7kw motor to a 3kw motor. I built a cyclone separator on the first round, but it never worked very well. Kept on getting clogged at the base and was hard to clean it.
Thank you for your compliment! Be very appreciated. A electrostatic filter is also something I was thinking about. But do you need high voltages for that? I have no clue (I'm not an Electrician :D)
@@CapturingDust Je zou de deeltjes moeten opladen door ze door een hoge spanning gaas te laten gaan en af te vangen in pijp met een tegengestelde lading. Het nadeel is aanrakingsgevaar en brand. misschien is afvangen met natte sponsfilters beter, maar het gaat wel om duizenden liters lucht en dat geeft weer luchtbevochtiging. Misschien met een oliebad. (ik ben een soort blij, dat dit niet mijn probleem is :)
On hi dust feed rates it did not perform well as the dust had not enough centrifugal force to spin up to the end. It clogged up on the bottom of the second stage tube. I need to do more testing to see how “bad” the separation is then compare with the two motors.
@@CapturingDust i have a 3 motor camvac, and I use the levels differently depending on my need, table saw is all 3 and cleaning my table is one, so I am interested to see if it would do the diffent tasks associated with the different levels
I'm watching with bated breath! I've seen and been intrigued by the the Under Dunn and Pilsonguitars videos on this subject. I'd like to share my thoughts on the location of the fan. Push vs Pull? I believe performance will be governed by pressure drop along the sytem? Pushing will cause high pressure at the start of the system that falls off through the system. Pulling will cause the opposite, low pressure at the output that rises through the system. This seems to amount to the same thing. However, it will only be the same thing if the pressure gradients through the system are identical. I feel that Pushing will cause a high gradient at the input and lower gradient at the output. Conversely, Pulling will cause a steep gradient at the output and lower gradient at the input. If this is true, ultra small particles that experience very little centrifugal force at the output of the centrifuge stage are more likely to get drawn down the steeper gradient of a Puller and end up in the hepa filter, whereas, the lower gradient of a pusher will give these particle more chance to find their way into the small particles bin. I hope this makes sense and any fluid dynamics experts out there- please, fire away
I started this journey because I couldn’t find any good research on this topic and had my doubts about the pull setup. It worked for my setup with the CamVac. Only I need two motors. With one the centrifuge did not have enough power to pull all the dust to the outlet of the second bin. Large dust particles ended on the bottom of the tube. I try to do more research on this topic. Only first I need to have a good measuring setup so I can do repeatable and accurate measurements on the different setups. My feeling is that push and pull both will work only not with the same power input to the blower motors (efficiency ratios).
After watching the original video from Under Dunn two years ago I have built a smaller, low flow version from PVC and bristol paper. It was supposed to filter small particles and cement dust from grinding but ultimately had efficiency of a mere 70% with that particle size and I lost interest in the project. They are really hard to remove. I was trying to maximise the centrifugal force while keeping air velocity moderate to get them out of the stream but there seems to be very little data on how much force one actually needs for given velocity, size and material.
It is true that there is really very little research or user experience available on this. There is a lot to find for vertical cyclones. Hopefully we can gain more experience with this type of separators together. Cement dust is also much heavier than wood dust, so it needs more force to spin. Thanks for sharing!
I would be happy to participate in your testing by printing one and sending you my results. I’ve got a handful of 3d printers in the shop right now with nothing lined up to print.
That would be very helpful! Also like to know what your blower setup is and dust types. I am focusing on building a testing setup to get repeatable results to optimize my MK2. Maybe you can start the printing process with the impeller. That is the most difficult part to print. Link to the files are in the description below this video. I added the Bambu Lab print files I used for the impellers. Like to hear your results on the print.
A "stupid-simple" option would be to have a light-curtain "lichtgordijn", either with a lens or a mirror-based scanner, and have a photodetector observe for light "flashes". A green one should be able to detect particles of ~500nm diameter and larger. for smaller particles, you'd need violet/UV. 395nm "laser pointers" are available, as spinoffs from the blueray factories. How many residual particles do you expect? ..and what size?
I am not sure what and how many to expect as it is depending on the type of separator it needs to test. I am going to try the Trotec PC220 as a baseline testing device. Six channels from 0.3 - 10 um. That is what I like to be able to measure real time in a cheap way. And subtract the measurement from the surrounding to get the parts that are not separated out. Someone suggested the sensirion sps30 or SEN5x sensors? Building one myself is a big step to take and not my cup of tea 😅
@@CapturingDust The Trotec is reeally expensive, I'm starting to worry a bit about your wallet Ruud! SPS30 seems like a nice sensor, at least from the datasheet and how it outputs 16-bit integers for each PM size, mass and number concentrations.
Hello! The MK2 are designed for the 6" Powertec pipes you can buy in USA on Amazon. When you live in Europe and have the Acrylic tubes from kunststofforte.nl you need the MK2_EU. There are scaling rings available for the different version to finetune the measurements because of production tolerances. So you need to order the tubes first and than select the right files for it. Hope this info helps!
If your goal is to measure the micro particles, how precise do you want to get? Can you device a method to compare a shitty separator with a good one? My money would also be hacking the Dyson vacuum cleaner. Because I suspect they measure reflected light and they probably trained ai to distinguish small from big particles. Then they probably calibrated it using known concentrations and you get the benefit. However, for prototyping, you don't need an accurate quantity, but a reliable measurement in performance differentiation. If you colour your known particles, you could potentially simply look at where the colour went and how saturated it is. Ps. Loving the stone coal english especially combined with the dutch dry humour/nuchterheid (frietpan frank is another good example of this)
Thx you can appreciate my stone coal English! It needs many takes to get it so far but that is part of the learning curve. About the testing. I need to get a testing stategy that is repeatable without me needing to run the planer every time for a hour or so. It needs to reflect the real life situation and measurements needed to be done accurately with automatic logging to analyze the data. Not like this how Oneida shows there testing; m.ua-cam.com/video/qPxKwzYCMiw/v-deo.html&pp=ygUOT25laWRhIHRlc3Rpbmc%3 I agree on you that for optimalisation there is no need for very accurate recordings. They need to be repeatable to show the difference between the different designs.
@@CapturingDustha interesting, their approach was my line of thought. But you will have to have a repeatable mixed size suction material. For automatic logging i would still focus on the known start amount and end result. Perhaps pictures and postprocessing can help? Continuous would be quite challenging since particulate spends varying amount of time until it is captured. So normalizing the stream of output data against input data is not realistic. I look forward to see what you come up with. Heel veel succes!
I like to print the final version out of aluminum or something. To see if it makes a difference in efficiency and static energy. The material needs to be very smooth as a surface and strong to last long. Do you have any suggestions on that?
I know I have made plenty of comments already, but could you measure the "loss" how much less airflow than without, and compared with the other methods, oneida cyclone, then baffle etc
Yes that is exactly where I am curious about to see and test the difference between the separator types. Could not find that online. So need to test it myself. I suspect that the simple waste bin has the most pressure and CFM loss. My interest is to discover the prise/performance ratio of the different designs. And also take account of the power consumption of the blower (CamVac in my case) and fine dust separation performance.
This is a really interesting project, touching an important subject for us hobby makers/small shop pro's. One aspect i think is to be clear about the vacuum generator that is driving the system. You are using a Camvac, others may need to consiser transferability to their vacuum. This guy makes some useful comments ua-cam.com/video/KPU1Vi4XgJY/v-deo.htmlsi=1wEcGWj8Rgzf48Kk
If you shoot a laser at a glass rod it creates a strip of light.. if you shine that through the air flow with talc powder you see slices of the air flow. Most CCD are 2x as sensitive to green so using a green laser's probably best you can average a long exposure of that slice with a sensor to get an idea of how much stuff is passing through. You can do it quickly if you tune it on a black background and count all the pixles with just the laser value of light in it with something like imagemagick. It won't detect micron/HEPPA particles well if at all (though if you have good optics and a large sensor maybe) but it'll give you an idea of particle count and let you test & visualize how the airflow is working all at the same time. ua-cam.com/video/TILjzuBGkRc/v-deo.html shows how the laser stuff looks/works.
That is very generous! I am in the market to buy a Trotec pc220 myself as a baseline for the cheaper module that I am currently exploring. What is your brand and type of model? And experience with it. Thx!
@@CapturingDustMet One HHPC 3, it is precise and used for clean room management, but the one you’re looking at would be much more convenient with all its functions. The Met One only counts particle numbers and doesn’t do any conversions. I think you’ll be a lot happier with the one on you are looking at.
When I used to measure engine exhaust gas particulates (prior to the days of Smart Samplers) we used to take a known proportion of the exhast gas and draw it through a filter (Watman GFA in those days). Prior to use the filter was weighed on a microbalance (Metler Toledo) and also after the test (there was some conditioining time involved too but tht was to deal with the unburnt hydrocarbon fraction, something that you don't need to worry about). You could use a simple orifice plate for the total flow and a gas meter on the sampler side to get the flows in the main air stream and the sample stream. Post test weighing would tell you the mass of particles collected and the two flow measurements would allow you to scale the sample mass to the total flow. If you want particle sizing that gets a lot more complicated (and expensive) but you could always stack filters of different sizes in the sample line and measure the individual mass collected on each filter. You also need to check where you take the sample from and be aware of any features that may create dead zones or act as edge filters leading to erroneous results.
Haha klopt! Dat is iets wat ik ook nog verder moet ontwikkelen 😅. Leren door het veel te doen zeggen ze. En ik sprak het nooit 😜. Kan dus alleen maar beter worden!
I did a quick test with 1 and 2 CamVac motors. And 1 was for hi dust rates not enough CFM. Dust in the second stage was down on the bottom of the tube. Don’t know the exact CFM rate. I ordered a CFM meter for that.
@@CapturingDust Thanks for the answer! i have 4HP blower and it would be interesting to know if the same setup would be good for me as well. Thanks for the hard work!
Yes of course: I linked the files in my other video. Just added them under this video as well. MK2 is available on github: github.com/Makerr-Studio/CentrifugalDustSeparator-
Hey Ruud, you did it! You made it on camera... nicely done! "...my brain is super excited" 😀 Thanks for the work you are doing here and for continuing to share your work. It's incredibly impressive!
Just a thought about the particle counter you mentioned. Dyson vacuums have these sensors built in, and have no doubt spent a lot of time and money in developing it. Would it be possible to hack or straight up incorporate one of their units into your separator design somehow I wonder?? Looking forward to hearing more about the design development of this project 😊 👍
Looks amazing! Hooked on Wood has some really good videos on building a system that works well with the CamVac. Including upgrades, and a custom HEPA filter box
Hi Ruud, I'm also working on a Harvey gyro style separator but instead of using blades to spin the air it's just injected into the chamber at the desired angle (15*). With no blades to get clogged this removes the need for the additional separation stage making it simpler and more compact (maybe better performance?) Designing and modeling the ramped intake however was not simple...
I also have a version tested with the intakes at a 30 degree angle to get the most of the bigger stuff directly aimed into the bins. It works really well for high quantities of dust. While planing it was not measurable on my cheap kitchen scale. Interested to see yours design. Is it possible for you to share a photo of your work with me? You can send it to maakhet@makerr.nl
Love channels like this, fun to watch and you actually can learn things. Good luck on your goal!
Great video! Thanks for sharing your know-how and cad files!!!
Little idea to safe some money: Use 3mm rubber seal from a spool instead of the o-rings. You can cut them according to your needs and carefully glue the ends together using super glue. Saves quite a few bucks...
I've given a lot of thought to these horizontal style dust separators and I have one issue I can't get past. If you suck up anything bigger than chips made by a table saw, planer, jointer, router, etc., you are very likely going to clog the separator. In a small workshop (I also currently work in a one car garage) I don't have the space for multiple dust collection/vacuum systems, so whatever I install (I don't currently have a dust collection setup) has to be a do it all, one and done system. The problem is in my shop, I suck up small blocks of wood, paper towels, hand plane shavings, etc. while I'm cleaning up. If anything like that is sucked up into that separator, it's just going to get stuck on the first turbine blade it comes in contact with. Thein baffles have a similar problem. Normal conical cyclones are 100% immune. If it can fit through your hose, it'll get separated out and end up in your dust bin.
That aside, if you are going to do side by side tests of the different systems, can you also measure the CFM of your blower with and without the dust separator?
I found out that the first stage needed to be more efficient in separation at hi dust volumes. I had some leaves and long ships clogging up. I made the Y-peace of the inlet run inwards at 30 degrees so the large parts were directly aimed inside the bins. That was working very well, only no difference while running the planer. I think this can be optimized even more in my design! I need to have a better testing setup to optimize it even more and be able to test the CFM and pressure drop. I am working to solve that!
Well Done. I am excited to see how your work progresses and hope to follow along and possibly make your design to test in my own shop as well. someday. Good luck.
I just stumbled upon this and am immediately hooked. I'm exited to see where your channel goes.
Just ordered some of the clear and pvc pipe. Coincidentally I have 5 kilos of pla. Will give it a test asap
Random thought: fine water mist(s) prior to exhaust. You'll want to have an exit path that allows any water to condense or otherwise drop out prior to going back into the room, but it should catch all of the micro particulates. Can recirc the water, or use a sterile water filter if you really want.
A wet scrubber is something I considered as a third stage. Only my CamVac is running all the exhaust air through the extractor motor for cooling. Water vapor residues must then have completely disappeared.
I will give it a try later on if I change the extractor motors for a type that has a separate chamber for the air pump. Also the particles needed to be able to stick onto the water or get absorbed by the water. Don't know how long the particles needed to be in contact with the water?
Maybe it is also possible to pull the dust through the water? And is the resistance it will generate less than running it through a HEPA filter. Many questions that I would like to have answered. So maybe later on I will deep dive into that as well :D.
Under Dunn also built a turbine style separator. Oh you cited him in the other video. Cool!
Yes his build was genius without 3D Printing! Really enjoyed watching his video. Only safety is not his best skill 😅
Nice start! Ik ga jouw uitleg volgen!
Lekker gewerkt Ruud
You should look at a SEN5x on an Arduino as a dust sensor. They're affordable enough that you could use several if you wanted and track ambient dust and differences between stages.
Cool project, will be following
I had a lot of trouble printing the mirrored impeller to get it smooth. It was due to the bambu printer default to print counter-clockwise causing issues with curling for the leading/overhang edge of the blade as it is printed first. Their slicer program does not allow you to change the direction of print. The solution was to use OrcaSlicer which does allow you to change the wall loop direction to clockwise.
Good to hear that you have found a solution to solve the issue! I will print the mirrored impeller this evening again with my settings on my Bambu Lab X1C. I can not remember there was a differents in the prints results the last time I printed them.
I can see what you mean now! The tips of the mirrored impeller are rough and have printing imperfections. I sended them away last time. Now I know why they only ware visible on the mirrored impeller!
Fascinating how you solved that issue by making the connection between counter-clockwise printing! I will look into the OrcaSlicer to become familiar with it as well. And I will report it backup the community! Thanks for your input 💪
@@CapturingDust Orcaslicer is based off Bambu's software so will be familiar. Also found that I needed to keep my printer door open to keep the ambient air cooler so that curling was less likely to occur. Granted I was printing in a pretty hot room.
For those trouble shooting their prints, I added a pause in my g-code after ~?150 layers to inspect the print quality, so not to waste too much filament if it was going to be a dud.
Thx! I will try it out on Wednesday.
Ik ben nu al onder de indruk Ruud! Keep it up!
I sent my CAD model to pilsons guitars and my thoughts on improvement after his 1st or 2nd video. I am not sure how much he used. I can send my files to you as well.
That would be great! What is your exact use case and fan configuration?
@@CapturingDust sent.
@@CapturingDust I'm building mine using Fusion_Woodworking's file too. I'm using 200mm diameter tubes and four Festool dust extractor motors for suction.
Yes, and I appreciate you sharing. Since my primary efforts in my shop are around building guitars (and my 3D design skills are hack-ish compared to Ruud's), I didn't take it much further than the results I ended up getting. I was happy with the collection level, though Ruud's design is able to achieve much higher volume. I do appreciate the contributions though!
@@pilsonguitars Collectively, we will get this solution to the next level.
Houdoe 😂 Always great to see other Brabanders here. Maybe you can contact Henk Maas (Veen) to lend you a calibrated scale?
Love what your doing. French cleat install is perfect for me. How to cut out the pipe? Thinking a jig might be needed. (3d printed of course). My usage is cnc.
I add tree cutting templates to the files you can use to tape on the tube. I used a multi tool for cutting the PVC. I am working on a build video to show you the entire process of building the separator. Hope to finish it in about two weeks.
hé Ruud, heel mooi ontwerp, ik kom al enkele weken of er nog aanpassingen zijn voor ik begin te printen 🙂 a very nice design Ruud, I have 2 questions. The Festool collection bins are not that large, is it possible to provide an output that goes to drums? The design does not have to be so compact, it can have some length, as long as the diameter still fits the printer. Personally I would make two of them, 1 high against the ceiling with the collection drums and the hose on a boom arm underneath for a sander, milling machine, etc. and 1 lower for the devices with larger chips.
Ik ben momenteel druk doende om het EU ontwerp goed te krijgen met ook een alternatief voor de Festool bins met meer volume. Daar is meer vraag naar idd. Er staat nog veel op de planning qua optimalisaties/versies maar het filmen van alles vertraagd het proces meer dan ik dacht 😅. Het is een vrije tijd project dus kan er helaas geen fulltime aandacht aan besteden.
On smaller particles, don't expect to get close to 100% just with the centrifugal separator. And here is a good video about particle filtering, HEPA or not HEPA, amount of clear air filters can supply, and why it might not matter that much, if you are continuously filtering the air in your workshop: ua-cam.com/video/gaQTYrisieA/v-deo.html It's a long video, but really full of useful information! Of course, for a workshop you will need a bigger filter than something sized for regular room use :)
I love your implementation of the festool "buckets" they are just insanely expensive for what they are, a clear bucket, I wonder if anyone has ideas for a cheap solution to replace it
I also looked for closed Euronorm bins as a alternative:
norah-plastics.com/stacking-boxes-and-plastic-storage-crates-40-x-30-x-32-cm.html
You need to make a clear window in it to see how full the buckets are.
Here is a Issue page opened for this topic on my GitHub site:
github.com/Makerr-Studio/CentrifugalDustSeparator-/issues/6
@@CapturingDust great, is there a vacuum on the bins? I was wondering if I could find something in ikea to use, but those bins are relatively soft, so I don't know
If you do it with a dual motor CamVac like mine there is quitte a lot vacuum when your hose is blocked. When running freely a cheap bin from ikea can do the job I think. You can try to stack two bins together for more strength.
When you run a blower motor in front there is no vacuum. Maybe that is beter for thinner bins.
I just ordered the clear pipes and will start printing next week. I can do some tests for you with Dusttrak 8533 devices if you would like
That would be great! I am looking into the Trotec PC220 as a baseline test device. The Dusttrak is way above my budget :D Like to hear you results and what kind of material range you run.
@@CapturingDustI don’t have any experience with the Trotec, seems like an ok little device. Dusttrak is kind of the go to in my line of work. Where are you situated in the Netherlands, we could compare the Trotec with a Dusttrak (once you get it).
I downloaded and built pilsons guitars design and got it working this week, I'm really impressed with the design. I used your design for the bins that clamp in instead of hoses to them, and it works very well. My only issue I think is that my air pressure is still too high (I'd upgraded the motor on my dust extractor to a bigger stronger one), I think I might need another two seperators. I'm curious about the design changes you did to the original.
Nice you have your separator running!
I use 1 motor on my CamVac to try out the different between 1 and 2 motors. In my case the dust settles inside the second stage. Not reaching the second dust bin.
What is it you’re expecting with the to high pressure? Is it your second bin that is filling up to quickly? It’s need to sit in the 80-20 range. So 20% in the second bin. That’s what the Harvey separator is aiming for I guess with the frequency drives.
@@CapturingDust I put a decent amount of smaller dust thought it yesterday, it was about 50/50 in each one. Worked well, but I think the flow is too fast through it, not letting the final stage settle.
@ADHDbuilt sounds like you have more the pressure for 8 inch separators I guess? Adding a other 2 of 6 inch will increase the friction to much I think.
And to be able to distribute the flow evenly over 4 separators is also a challenge. With two it was even though, I noticed.
Do you have any idea of CFM and pressure your new fan is able to produce?
@@CapturingDustI don't have any measuring equipment, but I went from a .7kw motor to a 3kw motor.
I built a cyclone separator on the first round, but it never worked very well. Kept on getting clogged at the base and was hard to clean it.
Fantastic! I'm thrilled that you were able to use the designs and make it work! I too like Ruud's design on the bins. (much simpler than what I built)
Mooi! Ben erg benieuwd!
Maybe an electrostatic filter for the smallest particles? You deserve a large audience!
Thank you for your compliment! Be very appreciated.
A electrostatic filter is also something I was thinking about. But do you need high voltages for that? I have no clue (I'm not an Electrician :D)
@@CapturingDust Je zou de deeltjes moeten opladen door ze door een hoge spanning gaas te laten gaan en af te vangen in pijp met een tegengestelde lading.
Het nadeel is aanrakingsgevaar en brand.
misschien is afvangen met natte sponsfilters beter, maar het gaat wel om duizenden liters lucht en dat geeft weer luchtbevochtiging. Misschien met een oliebad. (ik ben een soort blij, dat dit niet mijn probleem is :)
I definitely want to know how your separator works when the camvac is on 1 motor
On hi dust feed rates it did not perform well as the dust had not enough centrifugal force to spin up to the end. It clogged up on the bottom of the second stage tube. I need to do more testing to see how “bad” the separation is then compare with the two motors.
@@CapturingDust i have a 3 motor camvac, and I use the levels differently depending on my need, table saw is all 3 and cleaning my table is one, so I am interested to see if it would do the diffent tasks associated with the different levels
I'm watching with bated breath!
I've seen and been intrigued by the the Under Dunn and Pilsonguitars videos on this subject.
I'd like to share my thoughts on the location of the fan.
Push vs Pull?
I believe performance will be governed by pressure drop along the sytem?
Pushing will cause high pressure at the start of the system that falls off through the system.
Pulling will cause the opposite, low pressure at the output that rises through the system.
This seems to amount to the same thing. However, it will only be the same thing if the pressure gradients through the system are identical.
I feel that Pushing will cause a high gradient at the input and lower gradient at the output. Conversely, Pulling will cause a steep gradient at the output and lower gradient at the input.
If this is true, ultra small particles that experience very little centrifugal force at the output of the centrifuge stage are more likely to get drawn down the steeper gradient of a Puller and end up in the hepa filter, whereas, the lower gradient of a pusher will give these particle more chance to find their way into the small particles bin.
I hope this makes sense and any fluid dynamics experts out there- please, fire away
I started this journey because I couldn’t find any good research on this topic and had my doubts about the pull setup. It worked for my setup with the CamVac. Only I need two motors. With one the centrifuge did not have enough power to pull all the dust to the outlet of the second bin. Large dust particles ended on the bottom of the tube.
I try to do more research on this topic. Only first I need to have a good measuring setup so I can do repeatable and accurate measurements on the different setups.
My feeling is that push and pull both will work only not with the same power input to the blower motors (efficiency ratios).
very nice i'd love to see it working.
subscribed :)
Maybe some sort of electrostatic collection of the super fine particles?
this guy is hooked on wood
I watched Dennis from @hookedonwood a lot 👍😉
After watching the original video from Under Dunn two years ago I have built a smaller, low flow version from PVC and bristol paper. It was supposed to filter small particles and cement dust from grinding but ultimately had efficiency of a mere 70% with that particle size and I lost interest in the project.
They are really hard to remove.
I was trying to maximise the centrifugal force while keeping air velocity moderate to get them out of the stream but there seems to be very little data on how much force one actually needs for given velocity, size and material.
It is true that there is really very little research or user experience available on this. There is a lot to find for vertical cyclones. Hopefully we can gain more experience with this type of separators together. Cement dust is also much heavier than wood dust, so it needs more force to spin. Thanks for sharing!
I would be happy to participate in your testing by printing one and sending you my results. I’ve got a handful of 3d printers in the shop right now with nothing lined up to print.
That would be very helpful! Also like to know what your blower setup is and dust types. I am focusing on building a testing setup to get repeatable results to optimize my MK2.
Maybe you can start the printing process with the impeller. That is the most difficult part to print. Link to the files are in the description below this video. I added the Bambu Lab print files I used for the impellers. Like to hear your results on the print.
My friend, you are egregiously Dutch.
A "stupid-simple" option would be to have a light-curtain "lichtgordijn", either with a lens or a mirror-based scanner, and have a photodetector observe for light "flashes". A green one should be able to detect particles of ~500nm diameter and larger. for smaller particles, you'd need violet/UV. 395nm "laser pointers" are available, as spinoffs from the blueray factories. How many residual particles do you expect? ..and what size?
I am not sure what and how many to expect as it is depending on the type of separator it needs to test. I am going to try the Trotec PC220 as a baseline testing device. Six channels from 0.3 - 10 um. That is what I like to be able to measure real time in a cheap way. And subtract the measurement from the surrounding to get the parts that are not separated out.
Someone suggested the sensirion sps30 or SEN5x sensors? Building one myself is a big step to take and not my cup of tea 😅
@@CapturingDust Hm. That makes sense. I look forward to hear about your findings!
@@CapturingDust The Trotec is reeally expensive, I'm starting to worry a bit about your wallet Ruud! SPS30 seems like a nice sensor, at least from the datasheet and how it outputs 16-bit integers for each PM size, mass and number concentrations.
Bonjour, j'aimerais savoir exactement la différence dans vos fichiers entre les fichiers MK2 et MK2 EU? Merci
Au passage, j'adore votre réalisation.
Hello! The MK2 are designed for the 6" Powertec pipes you can buy in USA on Amazon.
When you live in Europe and have the Acrylic tubes from kunststofforte.nl you need the MK2_EU.
There are scaling rings available for the different version to finetune the measurements because of production tolerances. So you need to order the tubes first and than select the right files for it. Hope this info helps!
If your goal is to measure the micro particles, how precise do you want to get? Can you device a method to compare a shitty separator with a good one?
My money would also be hacking the Dyson vacuum cleaner. Because I suspect they measure reflected light and they probably trained ai to distinguish small from big particles. Then they probably calibrated it using known concentrations and you get the benefit.
However, for prototyping, you don't need an accurate quantity, but a reliable measurement in performance differentiation.
If you colour your known particles, you could potentially simply look at where the colour went and how saturated it is.
Ps. Loving the stone coal english especially combined with the dutch dry humour/nuchterheid (frietpan frank is another good example of this)
Thx you can appreciate my stone coal English! It needs many takes to get it so far but that is part of the learning curve.
About the testing. I need to get a testing stategy that is repeatable without me needing to run the planer every time for a hour or so. It needs to reflect the real life situation and measurements needed to be done accurately with automatic logging to analyze the data.
Not like this how Oneida shows there testing;
m.ua-cam.com/video/qPxKwzYCMiw/v-deo.html&pp=ygUOT25laWRhIHRlc3Rpbmc%3
I agree on you that for optimalisation there is no need for very accurate recordings. They need to be repeatable to show the difference between the different designs.
@@CapturingDustha interesting, their approach was my line of thought.
But you will have to have a repeatable mixed size suction material. For automatic logging i would still focus on the known start amount and end result. Perhaps pictures and postprocessing can help?
Continuous would be quite challenging since particulate spends varying amount of time until it is captured. So normalizing the stream of output data against input data is not realistic.
I look forward to see what you come up with.
Heel veel succes!
Would be interested to see if materials other than PLA perform better, such as nylon.
I like to print the final version out of aluminum or something. To see if it makes a difference in efficiency and static energy. The material needs to be very smooth as a surface and strong to last long. Do you have any suggestions on that?
@@CapturingDust
Maybe you van print your finale version from carbon?
I know I have made plenty of comments already, but could you measure the "loss" how much less airflow than without, and compared with the other methods, oneida cyclone, then baffle etc
Yes that is exactly where I am curious about to see and test the difference between the separator types. Could not find that online. So need to test it myself.
I suspect that the simple waste bin has the most pressure and CFM loss. My interest is to discover the prise/performance ratio of the different designs. And also take account of the power consumption of the blower (CamVac in my case) and fine dust separation performance.
This is a really interesting project, touching an important subject for us hobby makers/small shop pro's.
One aspect i think is to be clear about the vacuum generator that is driving the system. You are using a Camvac, others may need to consiser transferability to their vacuum.
This guy makes some useful comments
ua-cam.com/video/KPU1Vi4XgJY/v-deo.htmlsi=1wEcGWj8Rgzf48Kk
If you shoot a laser at a glass rod it creates a strip of light.. if you shine that through the air flow with talc powder you see slices of the air flow.
Most CCD are 2x as sensitive to green so using a green laser's probably best you can average a long exposure of that slice with a sensor to get an idea of how much stuff is passing through. You can do it quickly if you tune it on a black background and count all the pixles with just the laser value of light in it with something like imagemagick. It won't detect micron/HEPPA particles well if at all (though if you have good optics and a large sensor maybe) but it'll give you an idea of particle count and let you test & visualize how the airflow is working all at the same time.
ua-cam.com/video/TILjzuBGkRc/v-deo.html shows how the laser stuff looks/works.
Just wondering if it would be possible for you to upload the stl files for printing. Thank you
I will make STL-files later this week for you! STEP files are already on my GitHub page: github.com/Makerr-Studio/CentrifugalDustSeparator-
STL-files are available now: github.com/Makerr-Studio/CentrifugalDustSeparator-/tree/main/MK2/STL
@@CapturingDust Amazing you are a legend, thank you so much.
I could lend you a 0.3um 3-channel particle counter if you’d like.
That is very generous! I am in the market to buy a Trotec pc220 myself as a baseline for the cheaper module that I am currently exploring. What is your brand and type of model? And experience with it. Thx!
@@CapturingDustMet One HHPC 3, it is precise and used for clean room management, but the one you’re looking at would be much more convenient with all its functions. The Met One only counts particle numbers and doesn’t do any conversions. I think you’ll be a lot happier with the one on you are looking at.
When I used to measure engine exhaust gas particulates (prior to the days of Smart Samplers) we used to take a known proportion of the exhast gas and draw it through a filter (Watman GFA in those days). Prior to use the filter was weighed on a microbalance (Metler Toledo) and also after the test (there was some conditioining time involved too but tht was to deal with the unburnt hydrocarbon fraction, something that you don't need to worry about). You could use a simple orifice plate for the total flow and a gas meter on the sampler side to get the flows in the main air stream and the sample stream. Post test weighing would tell you the mass of particles collected and the two flow measurements would allow you to scale the sample mass to the total flow.
If you want particle sizing that gets a lot more complicated (and expensive) but you could always stack filters of different sizes in the sample line and measure the individual mass collected on each filter. You also need to check where you take the sample from and be aware of any features that may create dead zones or act as edge filters leading to erroneous results.
Lekker steenkolen engels😅
Maar wel leuk project.
Haha klopt! Dat is iets wat ik ook nog verder moet ontwikkelen 😅. Leren door het veel te doen zeggen ze. En ik sprak het nooit 😜. Kan dus alleen maar beter worden!
@@CapturingDust
Heel veel amerikaans/engelse filmpjes kijken.
Leer je de uitspraak snel.
Can someone estimate how sensitive we will be for different CFMs (750 vs. 2000, for example)?
I did a quick test with 1 and 2 CamVac motors. And 1 was for hi dust rates not enough CFM. Dust in the second stage was down on the bottom of the tube. Don’t know the exact CFM rate. I ordered a CFM meter for that.
@@CapturingDust Thanks for the answer! i have 4HP blower and it would be interesting to know if the same setup would be good for me as well. Thanks for the hard work!
Are you willing to share the models?
Yes of course: I linked the files in my other video. Just added them under this video as well. MK2 is available on github:
github.com/Makerr-Studio/CentrifugalDustSeparator-
ja houdoe heh
And sub ❤
There's a typo in your thumbnail. It's either "I am" or "I'm", not "I'AM"
Thx! I changed it :D
@@CapturingDust Sh*t happens 😉👍