Printing airlock parts for my dust collector

[ddmckee54]

About 20-25 years ago I made a pair of Shop Notes dust collector cyclones. I used an electric leaf blower for the DC fan. My calculations showed it was moving about 300-350 CFM. I made a prototype cylcone for my shop to see if it worked. I used the air filter from an '89 Aerostar as the final filter, calculations showed that it was capable of 350 CFM plus.. It was a noisy DC but it worked surprisingly well. I then made a nicer version that I gave to my Dad for his shop as a Christmas present. When my Mom and Dad moved off their acreage into a condo in 2007, for health reasons, he gave the DC back to me since he'd never be able to use it again.

A house move later and I getting the DC cyclone set up in my shop again. I decided to improve it a little this time, I'm upgrading from the noisy leaf blower to a slightly quiter 600 CFM dust collector blower, and I'm adding an airlock to the cyclone discharge - airlocks are also sometimes called rotary valves. Before we had to seal the discharge bin and that made it a PITA to check the level in the bin - or to empty the bin.

Since I've got a 3D printer now I decided I could print some of the airlock parts and get a lot neater airlock than anything I could cobble together otherwise. The airlock body is a piece of 3" schedule 80 PVC pipe that's 150mm long. The drive motor for the airlock is a 20RPM 12VDC gearmotor, I just happen to have a 12V wall wart that will work nicely for a power supply. I'll wire in a switched outlet that will power both the fan and the wall wart and run the airlock whenever the fan is running.

Here's an in progress picture of the print of the part I'm calling the adapter. The adapter is the transition piece that goes from the 6" round DC cyclone discharge to the rectangular-ish opening on the airlock body. It was about 7 hours into the print when this picture was taken. The entire print was estimated to take 18 hours, I started it at night when I went to bed and it was complete when I came home from work the next day. I'm guessing the print time was in the 18-19 hour range since the bed was still warm when I got home.

The priming strip on the left is 2mm from the Y axis of the printer so the print is pushing the print envelope of the printer.

I've got most of the parts printed, and I'm working on cutting the inlet and discharge openings in the body of the airlock. If anybody's interested I can take some more pictures and show you what I've got done so far.

Don

 

[gld]

Yes, more pictures. I would like to see the rest of the story.

 

[ddmckee54]

OK, there's been a couple that have looked at this so I'll doument it a little more.

This is a picture of the fully printed part from the first post.

The following picture is what the complete airlock will look like. Just remember that it's upside down in the picture, the ring at the bottom is the mounting flange that bolts to the discharge of the DC cyclone.

The next picture shows the airlock body on the transition piece.

This picture shows the rotor and vanes added into the mix. The rotor will actually be centered in the opening, and the vanes will be extended a few mm. The vanes will be spring loaded so they maintain contact with the airlock body to minimize air leakage. Too much leakage will actually prevent the material from being discharged from the airlock - I've seen it happen.

This is the rotor with a vane removed and laying in front of the rotor. If you look closely you can see the pockets in the rotor and the vane where the springs will live.

As you can see I haven't got the inlet and discharge openings cut in the airlock body yet, that's the next step. This picture shows the jig that will be used to drill the mounting bolt holes and cut the openings.

My initial plan was to use a flush cutting router bit with the guiide bearing mounted on the top of the bit. The only problem is that all of my flush cutting bits have the bearing at the bottom.

On to Plan B, I'll use a guide bushing for the router. If I use a guide bushing with the current jig it will cut the opening about 1mm smaller all the way around which will leave a lip where crap can hang up. To avoid that I'm printing another part with the opening 1mm larger all the way around and about 5mm thick. I'll stick that part to the part shown above with double sided tape, Then Bob's you uncle and Fanny's yourf aunt, because I'm back in business and cutting the corrrect sized opening. One minor glitch, the opening in the plastic base plate of my trim router is too big for my router bushings. No big deal, I took that plate off and printed a plate with the correct sized opening. 3d printers are wondermous for doing crap like that.

Don

 

[ajoeiam]

OK, there's been a couple that have looked at this so I'll doument it a little more.

This is a picture of the fully printed part from the first post.

Don

Sorry there mate - - - - but for me at least - - - - there just ain't no pics (except for the one attached at the end).

I'm a thinking I want me one of those 3D printers - - - soon!

 

[awake]

No pics showing for me either, except the last attachment. But that's enough to say, well done!

Any thoughts about whether static build up is a concern, or how to mitigate it? I know there have been endless debates about using PVC pipe for dust collection with regard to static, seemingly split evenly between for and against.

 

[ddmckee54]

That one is listed as an attachment, all the rest aren't. I'll look at as see if I can figure out what I did differently on all the rest of the pictures. I'm kinda new at the posting pictures, I usually just lurk in the background. The images show up on my computer, but I guess thaat doesn't mean a lot.

Don

 

[ddmckee54]

When I preview the post everything shows up fine on my computer, I don't know about yours. Just in case I screwed up again, I'm going to attach all the pictures here too. Maybe even a couple extra pix - just not in the same order.

These are the airlock end caps.

This is the rotor with one of the vanes removed.

Another shot of the rotor and vane.

This is the transition piece.

This is the airlock body sitting on top of the transition piece.

.

This is cut jig sitting on top of the airlock body.

This is airlock body, end casps and transition piece.

This shows the rotor and vanes in the airlock body.

 

[ddmckee54]

Now that I think I've got that issue taken care of, I'll go on with an update. I discovered last night that my Harbor Fright variable speed control is useless on my Bosch Colt single speed router. I also discovered that my double sided tape is mostly usless, and while attempting to remove the router before had completely stopped it chewed up one corner of the cut jig. That 4 flute end-mill did make an impressive pile of plastic chips though.

Plan C is to print out out a couple more versions of the cut jig, only with the opening expanded by 1mm on all sides. I'm printing 2 of them just in case I do something stupid again.

Don

 

[skyline1]

Hi Don

Yes these 3D printers are amazing things. So amazing in fact that I've now got 2 of them (and thinking about a third if the budget will allow). My newest one is an Anycubic Chiron which is huge and will even print life sized model heads ( if I wanted to ! ).

The only limit is your imagination and If I want a weird shaped "widget" then I simply have to draw it in CAD, slice it then it's "fire up the Chiron"

If you don't want to do all that or simply want inspiration then try thingiverse Thingiverse - Digital Designs for Physical Objects thousands of ready made STLs to print (including a few of mine) and all for free

Even with PLA the easiest to use material for printing you can make really strong structural components, like your airlock,

So it's not just toys and Benchy's #3DBenchy but you can make those if you want, I made a 3x scale Benchy for a sea fishing friend, he loved it !
You can make real engineering components with them and PLA makes super foundry patterns especially for the lost wax process (lost PLA in this case)

One word of warning though, they are just so mesmerizing to watch (and listen to) that you can spend hours doing so when you should be doing other things.

Best of luck with your project and keep us informed on progress.

Best Regards Mark

 

[ddmckee54]

I was able to get most of the airlock assembled over the weekend. I realized that I needed to do a minor redesign of the vanes, and while they were printing my printer crapped out. It was 3/4 of the way through a 2 hour print when I realized that something didn't sound quite right. I checked on the printer and it wasn't feeding any filament. The PTFE liner in the heat break has failed and siezed to the filament. This happened once before so I've got the parts to repair it. But as long as I've got to tear the extruder apart, I might as well do some maintainence of the rest of the printer too. It'll be a couple of days before I get the printer back in action.

Don

 

[awake]

Don, it may be a good time to consider changing to a hot-end design that stops the PTFE before the heat zone. There are a lot of aftermarket extruders / hot ends on the market, both high end and cheap knock offs. My home-brewed 3d printer uses one of the ubiquitous import copies of the e3d design for the hot end, and in general it has performed flawlessly ... and it cost < $10, including spare nozzles and heat break tubes. I have had to replace the nozzle a couple of times, but as best I can tell, that is a normal part of filament printing. I once clogged up the all-metal heat break tube, but was able to drill it out and keep going - and I have a couple of spare heat break tubes on hand if/when that repair doesn't work.

 

[ddmckee54]

Andy:

I don't ever print anything but PLA so the PTFE liner really isn't an issue. The last time this happened was a year or two ago, probably a couple hundred print hours ago. I really should start to keep better track of the print hours, I just know it was many spools of filament ago.

This is my go to printer, it's a Wanhao D6 clone. I also keep a considerable stockpile of spare parts, I HATE waiting for repair parts to be delivered. The funny thing is that my standby printer is also down for the same reason, but it's the first time in 6 years that I've had the liner fail on that printer. I've had plenty off nozzles plug, but this is the first time the liner failed.

Don

 

[awake]

Gotcha. I probably print PETG and PLA equally, with occasional use of other filaments.

 

[ddmckee54]

I had the repair parts in stock, so I got my printer repaired and printed out the new vanes. I had this brilliant idea brain fart, that I could use silicon tubing as the springs to keep the vanes tight against the airlock body. It turns out that it doesn't compress too much axially, radially yes, axially not so much. I've got some 3mm OD compression spring stock that I originally got for the springs in the rubber band guns. It didn't work so well for them, but I'll try making some custom springs out of that.

I've also discovered that when I cut the openings in the PVC airlock body, the stresses in the PVC body apparently caused it to contract. As the vanes go across the opening they catch on the sharp edge of the opening. They wouldn't be catching on this edge if it hadn't moved in a little. I put a 1mm radius on the edge of the vane when I designed them to try and avoid this type of problem, but it didn't help much. I can probably take a sharp knife or chisel and bevel the inside edge of the opening a little bit to fix that problem.

Don

 

[ddmckee54]

I talked with one of the mechanical engineers at work yesterday about my airlock problem, I'm an electrical engineer - not mechanical, and he was wonderfing if maybe I wasn't trying to over-compress the silicon "springs". So, I did a little bit of measuring last night as I took things apart. With the vane tight against the airlock body I have 10mm from the rotor to the airlock body ID. After I had taken everything apart, with the vane fully seated in the rotor, I have 9mm from the rotor to the the top of the vane. That means under normal conditions there's only 1mm that the vane can move before it bottoms out, When I measured the uncompressed height of the rotor/vane/"spring" combination I had over over 16mm from the rotor to the edge of the vane. OOOPS - guesss I seriously over-estimated how much those "springs" could compress. No wonder it fealt like the "springs" were folding over when I assembled the airlock, they were!.

I measured the depth of the pocket in both the rotor and the vane, added in the 1mm clearance, and then added another 2mm for "spring" compression. This total stackup amounted to about 11mm, which is the length that I cut the "springs" to. I reassembled everything, except the drive motor and motor mount, cobbled up a hand crank, and SHAZZAM... it still didn't work. The vanes would still hang up of the edges of the openings.

A little sharp knife, chisel, and sandpaper work later... and Bob's your uncle and Fanny's your aunt because it worked. The airlock rotor goes roundy-roundy. The vanes still click a little as they go past the openings so I need to figure that issue out.

I talked to the same mechanical engineer this morning and he recommended that I have more than 1mm clearance with the vane bottomed out. Which makes sense when I think about it, the airlock needs to be able to clear itself if a chunk of crap gets hung up between the vane and the airlock body as the vane is going around. I'll probably print a new set of vanes and give myself about 2-3mm of clearance. As he said, this is engineering on the fly - it's what we do.

I still need to make the mounting flange to bolt this thing to the dust collector cyclone. I've also ordered the parts I'll need to cycle the airlock on and off while the DC (dust collector) fan is running. I got to thinking about it, and there is NO WAY that cheap little gearbox that I bought will run continuously for possibly hours at a time. (A CNC router can make a HELLUVA mess if you don't have dust collection, and some really complex carvings will take hours of run-time.) So I ordered the bits and bob to run the airlock for 1-3 minutes, then turn it off for up to 10 minutes, using adjustable timing relays. This ON/OFF cycle will continue as long as the DC fan is running. Until I get that done I'll just leave the airlock hand crank in place and the airlock motor/gearbox off.

I'll make the mounting flange then take a picture with everything hanging off the DC cyclone where it will live from now on.

It'll be a while before the DC can actually be used. I've still got to get the DC ducting in place, install the DC fan with its' final filter, and actually build the CNC router. (I still need to build the final filter for the larger CFM DC fan too.) Once I've got the DC itself usable, I can hook the DC up to the drill press, the sander, the bandsaw, the tablesaw and all the other dust producers I've got. I'll just install the DC ducting drops for all the "FUTURE" stuff, at least the stuff that has an assigned spot.

I know the way that I'll duct things together may not be optimum, but I've always thought that even crappy dust collection is better than no dust collection.

Don

Edit: Damned dyslexic keyboard spelled of as if, and spell-checker let it go since it was spelled correctlly.

 

[awake]

Don, I am trying to picture how this works, but not quite succeeding. Why will you need to cycle the airlock on and off while the DC is running? I think I must not be correctly understanding what the airlock is for - limited imagination on my part

 

[ddmckee54]

The airlock motor is an inexpensive 12VDC gearmotor that is in no way rated for continuous duty. It is light enough and compact enough that I can hang it off the airlock body instead of having to build a separate frame to hold the motor/gearbox. By cycling it on and off I'll get a lot more hours of life out of the motor.

Don

 

[awake]

I was thinking you would turn the motor on, open the airlock, and turn it off - while using the DC. Then when the DC is done, turn the motor on, close the airlock, and turn it off again. But I guess I'm thinking of this like a typical gate; you are clearly doing something different ... I am sorry that I am not understanding it (yet), but again, that is on me!

 

[ddmckee54]

You described the funtion of the blast gates that are opened and closed to connect or disconnect a piece of equipment to the dust collection system. The airlock is a rotary valve that will be located at the bottom of the cyclone. The cyclone separates the particulate out of the dust collector airstream and the airlock/rotary valve will empty the particulate out of the cyclone.

I'm used to how the dust collection equipment runs at work, where it has to run 24/7 and we need to worry about duty cycles. Setting things up for continuous operation is overkill for my workshop, but it's what I'm used to.

Don

 

[awake]

Don, thanks so much for the clarification. I was indeed getting confused with a blast gate. So now I think I get it, but let's make sure: I'm guessing that the debris will collect in section of the airlock that is open to the cyclone, but that closes off the cyclone from the lower part of the system. Periodically the airlock will rotate, at which point it will dump the debris out, meanwhile rotating another section of the airlock into place to receive debris - and all of this without ever allowing the air to leak past the airlock. Did I get it right, or close?

In the cyclone systems I have used / seen, instead of an airlock, there is simply a sealed lower section into which debris can fall and accumulate - a 50 gallon drum or something along those lines. But I'm guessing the difference is that the DC has to be shut down to empty the drum, whereas the system you are building can run continuously, and you can just swap one drum for another between airlock cycles.

I will be interested to see how this project progresses!