By the second pump: it happens because you have double the diameter for intake then for the outlet. The vacuum is to low for one side. If you would use a connector for the intake-lines to bring them together, or you use a smaller diameter for the intakes, both sides would pull water.

Something critical you seem to have missed is that the first pump had a positive displacement - that is, there was essentially a fixed moving space that was pumping the water because of the meshing with the other spiral. The 2nd and 3rd pumps were really nothing more than a spinning fan, so it was easy for the water to just rotate with the rotating part without being forced along.

The main issue with the first one is that this type of pump is not made for high flow, but high pressure, therefore requiring very tight seals between the screws and the case and also between the screws themselves. This pump is pointless if you can't make this super precise which I don't think is possible with 3D printing yet. The problem with the other screw pumps you made is that they don't have a second screw making separate pockets that move the water to the output, but just one long tube that just kind of accelerates the water. You used a pump designed for pressure and removed its ability to make high pressure. Of course it didn't work. I loved the video though. Nice to learn that high flow is relatively easy to make with 3D printing but not high pressure because that requires more precision. Another critique I have which I've mentioned before is that you seem to not know the difference between flow and "performance". Just because it has low flow, doesn't mean its bad, as long as it has high pressure. Pressure x flow is effect. You need to check for both. Just like electricity. You have voltage and amperage. Voltage is the pressure and amperage is the flow where volts x amps = watts.

If your goal is to make a 3D printed water pump, your best bet will be an impeller based design, it will give you the best balance between flow rates and pressure without having the fine precision requirements of screw pumps. There's a lot less wear points as well, the impeller can be driven directly from the motor shaft so there's no gears or contact points to wear out. Still pretty cool that you can make such complex parts that function to some degree, it's amazing what 3D printing has brought to the world.

7:00 Typically pipe fittings have a tapered thread. This causes the threads to essentially have an interference fit when you tighten it down all the way. If you model that tapered thread into the print and then heat up the fitting a bit before inserting it (beyond glass temp, but not by so much it's going to just melt the print) the fitting should make and form a water tight connection. (in theory)

And here we see how important clearances are in screw pump performance - and why impeller pumps are more common. Impeller pumps are less sensitive to having larger clearances. I see this channel has done several types of pumps, so I will now go and see how well they did!

Wow, I'm actually working on twin screw pump for the last few days and now you upload this. What a coincident! Love your channel a lot, please keep making great content!

Brah your sense of humor is what every one needs. Low key insanely funny stuff as well as educational!

You can HEAR the pumps inefficiency where they create turbulence and air bubbles (cavitation), if you could make them so they are not creating all the bubbles, and being inefficient because of it, that would help I am sure.

My advice when printing things that other bits screw into is to print those parts first on their own to test fit before printing large units. If I ever print a box with a screw lid I print a really thin lid and the top of the box to check it all lines up and the holes are the right size. Saves hours and lots of material.

Hey ! I think your pumps are having a hard time starting because of the air in the feeding tubes. You can fix this either by using a secondary pump capable of sucking this air out or just by dumping your pumps into the water, thus eliminating the tube and air getting into the pump problems. I think that might help. Love your content so keep uploading great vids even when they fail of course 🙃

The first and second pump arevery impressive as they're the first pumps which are not spraying water everywhere while running. That's a huge progress for 3d printed pumps :D

Pump 1 - you only need one motor. The gearbox will cause both screws to turn as well as ensure they stay properly meshed. This setup is commonly used for precise addition of dry materials in industrial chemical processing. See Coperian for the commercial use examples Pump 2 - I agree with husky9104, air leakage is preventing the one side from maintaining prime. Secondary issue would be that with there is a significant possibility of overpressure if both sides actually pump against that small discharge Pump 3 - again a very common piece of equipment for transfer of dry materials, just typically bulk transfer vs. fine control

Those bumps are an awesome design!

The last couple pumps don't work by displacement, they are centrifugal. The crews are just spinning the water. I'm actually a bit surprised they moved water at all. Your first pump, with dual screws worked through displacement, which is why it worked better. I wish you has fixed the front seal. that thing could be awesome.

your tenacity is awe inspiring, your fast becoming one of my favourite channels, even your failures are great.

Isn't one of the greatest advantages of dual extruder setups that you can use water soluble material (PVA) as supports?

If ur having trouble with supports and you have a dual extruder 3D printer, water soluble pva is a great solution as you can print with ur desired filament, then swap to the pva for supports. After the print, you can resolve the supports in a bucket of water.

I love the accent! It's so relaxing to hear you slamming those words into the microphone.

Great videos. I dig that you mention and have fun with your mess ups. It's incouraging. I've been thinking researching these types of pumps for a project i want to put together, all your work and tests are suuuper helpful. Thank you.

Diese Pumpe ist nicht für konstante Steuerungsdrücke mit Rückflussdrücken ( zum Beispiel aus einem Zentralheozungssystem ) zu gebrauchen. Dennoch konnte Sie eine tolle Anwendung bei der Feuerwehr haben wo man sie als Sprühnebelverdichchter einsetzt. Die Idee mit dem Schneckenprinzip ist schon sehr alt, aber hier von immenser efektivität. Mich erinnert so was an die Förderschnecke des Fleischwolfes, die meine Großmutter oft benutzte, wenn auch wieder zu Weihnachten Butterplätzchen gebacken wurden, ähnlich dem schottischen Shortbread!

I’ve never had problems printing with petg, I actually only use this plastic, you should try a bigger nozzle, like a 1mm or 0.8mm, it works well. With smaller ones is a lot more difficult, and you need a little bit higher flow rate, less printing speed and a slower fan speed. You can try it with 0.6mm nozzle too, 0.4 is maybe too small. I recommend you putting some lithium grease on those gears, it really reduces friction and heat. I love using fusion360, nice videos man

As I work as an engineer in a screw pump industry, I know what exactly is going wrong here! You are using water as the pumping medium. Which has a very low viscosity. And screw pumps are predominantly used in medium to high viscious media. As screw pumps has a clearance between the screw and housing, using low viscous media will just slip back, rather than going to the outlet. To improve the pumps performance, you have two options. Either you can reduce the clearance between the screw and housing, or you can try a higher viscosity media like lube oil or similar media.. Try it! All the best

The reason why they don't work as expected is because they cannot build up pressure, therefore, they can't suck water in, they just give speed to the water, which is not needed

For that last one, you need a vane at the inlet in the housing that fits into the form of the screw, otherwise you're just turning the water around inside the housing instead of propelling it forwards.

Автору: Последние две конструкции работать как насос не будут, их используют для транспортировки сухих сыпучих веществ. Первая конструкция используется в промышленных компрессорах, но для уплотнения и сжатия воздуха там используют масло. Попробуй сделать разный диаметр входного и выходного вала. Нужно создать зону разряжения.

If you are pumping water from above a reservoir it is best to use one that displaces air (much like an oil pump from an engine) rather than a turbine.

With those single screw pumps there is nothing preventing the water from simply spinning instead of being pumped. I think this is why you had worse results with those. Anyways, great video and always a pleasure to watch!

Nice engineering!

You can print threads without support as long as they have a 45 degree or less overhang. So make them shallower with 45-degree sides and you'll be able to have much tighter tolerances on the 2-screw design and better pumping.

Thought you might like to know that the auto-generated captions said [Applause] when you turned the motors on for the first time at 3:18.

I’m so fascinated by all these water pump designs and tests!

Couple things- a: leaking because you need a rotating face seal of some sort. Probably not something you can 3D print but maybe? It looks like you solved that in the second make, or addressed it with a proper seal. Other thing, on the second video, with a pump like this you really need a high level of precision a 3d printed part may not be able to achieve. You essentially would require both screws to be precisely identical, otherwise the pressure being generated by the higher performance screw will deadlock the other, especially with the velocity head being generated by the working screw being in line with the other's discharge. Same reason why it's no good putting two different pumps in parallel and expecting them both to generate flow

I'm all here for this water pump design adventure :)

interesting to see at least the guy is very honest about his endeavours. I think the last pump if you put a tube on the output nozzle and raised it up a bit it would help the pump stay primed. The previous pump with the two opposing threads I believe the reason the other side did not stay primed was because of air being forced from the dominant side which would make it difficult to stay primed for the less dominant side. Great to watch though even though these pumps are gonna be inherently inefficient. Great video.👌👌

So the title of this video could have been 3D printed screw ups 😊. Still interesting to watch. Thanks!

the main issue here is most likely the tolerances the lobe pumps are designed for high volume and low pressure, that's why they work so well as 3D printed parts. a screw pump is designed for high pressure, high viscosity or solid matter the latter are used in pellet or wood chip heating systems, on combines(harvesters) and plastic injection molding machines to name a few examples

Although your pumps didn't work that well they look great.

keep building pumps that don't really do anything, it's entertaining AF. :)

I think that the first inlet of the second pump is just blowing air,not sucking water.If you rotate the poles it will suck,but the second one will blow.Your screw has opposite directions on the ends.If you make them attached to different shafts with different rotation,then they will all work fine.

I understand the desire to stick with 3D printed parts as much as possible, but there are some things that 3D printing just isn't good for and water tightness is one of those. With some cheap, commonly available parts, you could make your designs much more reliable and effective. A good pump requires tight clearances to maintain pressure, not just to keep the water in, but to keep air out. Using lots of screws to tighten the pieces together can help, but at some point you need a material that is more compliant and can fill gaps better. This is where gaskets and O-rings come in. You could 3D print these from flexible materials like TPU if you want, but it's probably more accessible to everyone if you use easily-available parts like standard O-rings. For gaskets, you can buy sheets of rubber and cut out holes where needed. You could even 3D print a template to help cut them nicely, but the cuts don't have to be exact, as long as the gasket seals all the way around the holes. For elements that need to be threaded in, you can either cut threads yourself or you can use a threaded insert. Taps for threading are relatively cheap and can be used for a lot of things, but PLA doesn't tend to make nice, tight threads. Threaded inserts will provide a much better interface between the threads. They can be securely fastened into the PLA by heating them with a soldering iron. This melts the PLA around them giving a very tight seal. You could instead use epoxy or some other gap-filling adhesive to ensure there are no gaps between the PLA and the threaded insert. Bearings are also important to avoid slop in the mechanisms. Any movement of the screws is going to create openings that prevent a proper buildup of pressure. Skateboard bearings are common and good for radial loads, but axial loads require thrust bearings. Printing tight tolerances is very difficult, but plastics like PLA have the advantage that you can easily machine them down. If you oversize things a bit and run them, they will naturally wear down until they aren't interfering anymore. This can give you quite good tolerances. You can accelerate the process by putting some lapping compound or polishing paste on the screws. Another option is what machinists will often do. Put some thin paint or dye on the screws, then run them for a little bit. Look at the areas where the paint is worn off: this is where the screws are making contact. Sand them down a bit, then paint them again and repeat the process. Continue until you don't get paint rubbing off. For even better seals between the components, put rubber or silicone wipers at the edges of the interfacing elements, such as the outside edge of the screw. You can buy sheets of thin silicone rubber, cut them into strips, and embed them in the edge of the screw. Be sure to design a groove in the screw to fit the wiper. All of these things are readily available at most hardware stores, and they are relatively inexpensive. You don't even need exact, specific parts; you can often make substitutions that accomplish the same task. I think these improvements would help a lot to make a working pump that is still easy for others to print and reproduce for themselves.

amazing channel, good luck !

Just seal the first pump. It's the only one that can actually pump water. Because the other two pumps only have one screw, there's nothing preventing the water from flowing backwards. If the drive shaft is locked in place, the only type of pump that should allow water to flow freely is a centrifugal pump.

Positive displacement pumps such as these are primarily for pressure, so maybe instead of a flow test, attach a pressure gauge to the end with a recirc back to the original bucket?

The effort going from the rirst designs to this video is great, reading through the comments alot of tips there, my 2 cents would be to design a channel along the top of the spiral to slot in a ptfe tube in order to compensate for the clearances in the housing, something like those vases with filaments for accent colors

its not about how bad the pump worked..... But the effort you put in to make videos.

I greatly enjoyed watching you make 3 pumps that didn't work very well. ❤ keep up the great work.

Pumping water is not easy, it’s low viscosity is a problem. Awesome design.

8:21 the vortex in the outlet flow is Amazing. Congratulazioni 👏 Why didn't you use X1C CMS to print multimaterial instead of the Qidi?

This is awesome 👏 👏👏. I may be wrong but you might be getting cavitation? Boiling of the water from the vacuum. I couldn’t make this. Kudos to you for having the skills.

thanks for your videos! but in this one there is a small thing that you miss... there is a difference between siphon and water pump and that is that the water pump raises the height of the water, the siphon only manages to pump downwards. I explain? I would like to see all your inventions according to the height to which they raise the water

Use silicon or liquid gasket for engines. Hot glue creates airleaks all the time. Put the inlet in the middle of the screw since you dont have a good motor axle seal, I think the the primed water would present enough resistance for the air to maintain some low pressure inside.

Excellent work, dude! Nicely done! 😃 But yeah, building water pumps is not an easy task. Every time I tried the water came through the motor and killed it. 😬 Anyway, stay safe there with your family! 🖖😊

If you want more pressure from your feeding screw try tapering the screw closer to the walls. I saw this on "smarter every day" when he first went to the film factory

If you have a dual extruder printer, I think it would be easier to print PLA + PVA. The PVA is water soluble, so you can just flush away the supports.

Oh man, I've been looking for this video for so long...

Your’re dedication is amazing. I really enjoyed your video!

good job! even when they arent too perfect, I mean they work a little bit, thats atleast something

The last pump in this video is like an auger it’s meant to pump solids like grain

I'm so glad Bambu sent you a Carbon X1. I felt like your previous printing capabilities were holding back some of your projects.

I wonder if this could be used for boat propulsion. Propellers sometimes injure sea-life. Although when the screw propeller was developed it was discovered that a shorter screw was more efficient then a longer one so perhaps my idea is stupid.

I think first pump with two screws should have had gears with screws inside pump and metal shaft in one screw with lip seal and bearing. Only inlet, outlet and sealed shaft to reduce leaking. Also what about printed threads? Or you can print coupling with thread on one side and cone on the other side, so you can hammer or glue them in mating cone hole.

On the last one The leading and trailing edges of the impeller need to be at a 30 degree angle. U R Getting impingement at the ends of it along the case.

Second and third pump doesn't wanna work because you have only one screw, so water can spin with the screw and not move, or spin faster than screw and go in opposite direction, to prevent that you need second screw, like you did with first pump. Roots and Lysholms superchargers also have two "screws".

the blue screw pump would be more at home with something like sand or feed or a similar granular solid going through it.

Aight next build, mechanical siren or a roots blower, if the siren is made, make it have a 8/10 port ratio or. 7/10 or a 10/12 port ratio

That design for the 2 facing opposite directions needs a choker module installed

Cool projects. I really enjoy your videos. Keep them coming.

Pack the shaft with grease not a seal… These will be great pumps! You just need better seal. If you have ever worked on a mechanical injection pump on a diesel you will understand how just the smallest of air leaks can cause a pump to never hit its curve.

It seems to be very usefull to cement 3D printing

It would be interesting to check the pressure as well as the flow!

you could split the screw in half and print with no supports. Then glue the two halves together?

One easy idea to seal the tube was to put a aquarium sealent around the outside of each joint.

Another good video. You don't learn if you don't try new things out 😎 I wonder how well these pumps, all of them really, would do if you were to line pump housings with wax of some sort to close the gap between the impellers and the housing. I doubt you could actually use a rubber seal on these because you can't introduce some lubricant other than water (I suppose you could use soapy water, might work from a testing perspective, just not practical application). The wax would serve two purposes, 1) closing the space between impeller and housing 2) be a kind of lubricant against the plastic moving parts. Might help make them run a little quieter. How long would the wax last, who knows! That's what testing is for 😉

twin screw need high precision fiting to minimize air gap between the screw and housing

You should look further into gear pumps, there are other types (internal gear pumps for example)

Man I'm jealous of that carbon x1

One interesting pump design you could try ist the mazda rx 7 oil pump sesign. It should create a very high pressure.

the second pump needs 2 screws like the first one. the screws need to mesh or the water can just leak back around the spiral

The first pump model looks like a star trek space ship.

Your video is amazing especially in 8:12, very smart pumping machines on that project.

Easyer to use (i dont know the english name) in Hungary called: lábszelep. Thus device use for water and other application, not need primeing every tine your pump. Not need so much stuff, some rubber, 3d printing and maybe a washer.

Use oil. Positive displacement pumps other than piston pumps are better at pumping viscous fluids

Try to use a variable spiral, more close in the outlet! Like an industrial air screw compressor!

Waiting for an explosion are we? I saw the loose screws in the gearbox at the end. I thought that last pump would do better. These pumps work best if they are actually in the water. Good video.

the last one would have worked better with the inlet of the screw submerged ... its basically an archimedes pump low pressure high volume when used right

I think the first design is very good. But, you need to add check valve in outlet and inlet.

maybe it would help if you sealed your designs with o-rings? there is so much air in the other systems as well. and thing: oilseal rings are cheap. after fdm printing just push a ring a plastic tube over the end for a better surface for the oilseal.

Ever think of making a spiral compressor pump? They use an oscillatory motion instead of spinning.

You should try sinking the pump in the water... without a inlet tube... It will give better results for all your tests...!!!

I assume this is why many water pumps use a centrifugal fan vs screw fan.

wow bro.I enjoy your perseverance and great spirit. I hope you will reach your goal soon. You will definitely succeed.

Cycloydal pumps are pretty snazzy.

Really nice print 👍

Your pumps suck lots of air. The first pump was a good pump that needs a better gear ration because the motor struggled. It's really easy to make orings from bulk material. Great prints! I look forward to seeing more.

There is water dissolvable filament, which is great for supports on dual extruder printers.

The destination bucket should be placed higher than the source bucket, to better demonstrate the pump doing some useful work.

Very nice 👍 next pump make it self primming....

9:36 these little screws dancing around in your gearbox give me anxiety - they are so close to getting caught in the gears 😅

Id say the reason the one one tube was not pumping, is the size of tube, That's allot of water to move. I think if you used a smaller diameter tube, you might see better results.. Plus I remember way back in school, being told about Cavitation - or something like that ! liquid dynamics and how the liquid can turn to vapour due to low/high pressure.. I can't quite remember how it went - but that word got stuck in my head !! lol