Building a High-Flow Peristaltic Pump: 3D Printed & 10 Liters/Minute

[@bubsbuilds]

Corrections and such: - In the voiceover, I say the wall thickness is 1mm, but as you can see from the measurement, it's actually 1.75mm. The nominal design leaves a gap of 3mm (added 0.5mm of interference to account for the compliance in things. - It was recommended that using a single oscillating lobe, and instead having multiple loops of tubing will result in longer tube life. I’m also curious if this can result in a smoother pulsation in the flow. (thanks @anothermidlifecrisis). - Pointed out that with the large cross section tubing and high flow rate, there may be potential for the vacuum pulled on the inlet side to collapse the tubing. After some quick tests, it doesn’t appear to be a consistent problem in my set up, but the tubing I’m using is relatively stiff compared to some. So something to keep an eye out for if using nice, soft, silicone tubing! (thanks @AlexRay0) - More confirmation that I am not to be trusted :) it was confirmed by another commenter (thank you @SianaGearz) that I shouldn’t be using Loctite around these PETG parts! But @seanconant3218 suggests VC3 as a polymer-friendly option. I haven’t used it (yet), but definitely sounds like it’s a smarter choice than my Loctite Blue! -- Additional +1 on regular Loctite not being compatible with many plastics from @AMakerDad, who offered up Loctite 425 as a cyanoacrylate (super glue) based thread lock, which is compatible with printed plastics. - A good option for extending tube life was suggested by @jimk8520. If positive displacement isn't necessary (which definitely not for this high flow fella), then not compressing the tubing fully will significantly reduce the stress it sees. - @ducktaperules pointed out that if you rotate the housings relative to each other, so that the tubes have an angular offset, it will stagger the load spike that occurs when the posts engage the tube, reducing the max torque required. I haven't tested to confirm, but the logic seems sound to me! - @Pyromancers suggested the idea of wrapping the contact posts in tef tape (or other similar low friction tape/wrap) to reduce friction. While I haven't tried it, it seems like a great idea to test out!

[@AMakerDad]

I can confirm that commonly used red and blue loctite attacks plastics - they cause the plastic to stress fracture over time. Years ago as a young engineer we had a military contract building chargers for L3. It had an expensive LEMO connector going into a plastic housing. The assembly drawing calls out Loctite 425, which is a cyanoacrylate, but the folks in manufacturing ran out and did a substitution for Loctite 271. About 6 months later we started getting calls that the units are failing - the plastic around the connector literally started to disintegrate. Because 271 is red, it was really easy to see what happened. We ended up having to build new wiring harnesses (with a ~$100 LEMO connector) and having to remount the mold and shoot new parts to replace the plastic housings. I am pretty sure we lost the shirt off our backs on that contract. As an interesting side note, cyanoacrylates also degasses - and the fumes from it reacts with fingerprint oils and forms a white milky film. After applying the CA we have to let the parts sit out for minimum of 24 hours to degas before closing up the enclosure.

[@bubsbuilds]

@AMakerDad Ouch!! Gotta love the, "eh, this seems close enough" production floor decisions! Definitely an understandable 'oops' though, I suppose, like you said, red is red. Was the root cause review fairly quick, or did it take some time to track it down? I could see that being a somewhat tricky one to catch (unless maybe it left some sort of obvious signature on the failed parts). And how bout that! I've definitely encountered plenty of that hazy white coating alongside CA joints, but never knew that was the reaction causing it (insert "the more you know" GIF here :) ). Does that 24hr window depend on the viscosity of CA you're using, or is 24hrs generally enough for any of the common offerings? Thanks for the confirmation and great anecdote!

[@TomcoDesigns]

The ball bearing printed bearing assembly is very clever.

[@bubsbuilds]

@TomcoDesigns thanks!

[@davidpinnington213]

Well the YT algorithm must be mind reading again - for the home production of stronger adult beverages on a continuous basis I need 1 or 2 pumps - I’ll need to tweak some details but your ground work has saved me some £€¥$ - many thanks

[@bubsbuilds]

All hail the algorithm! :) Glad to hear it! In case it'd be helpful for your tweaking, I'll throw a link to the native CAD files...it's not a pretty model tree, but may be easier than trying to mod STLs. Would love to see the finished product when you're done!

[@jimk8520]

I spent a whole lot of time many years ago fixing peristaltic pumps and one thing I can put out there is that smashing the hose too tightly over time will rupture the tubing. If your system doesn’t need to create high pressures, it’s better to reduce the pinch and let a little leak by happen if you want this to run without liquid surprises in somewhat short order.

[@Somun-a]

Especially with 3d printed (ridges) pins like that.

[@bubsbuilds]

@jimk8520 Great point, thanks!

[@oddspaghetti4287]

Yeah threadlocker corrodes PETG at least. I suggest just using nylock nuts or making tight threads into the plastic which is what I often do.

[@bubsbuilds]

@oddspaghetti4287 Do you find that the tight threads in the plastic give enough friction by themselves to avoid backing out, or is that more in parts that don't have a torque through the fastener during operation? Thanks for the suggestions!

[@oddspaghetti4287]

@@bubsbuilds Hard to say if printed threads are enough against vibration and tension on a screw. One of my models has M3 screws with 8mm of thread engagement (hole in Z direction) and I find that the amount of friction is excellent and surprisingly remains decent even after multiple reassembles. I've printed hundreds of this assembly and it works every time. Feels about on par with nylocks. Of course you will have to print some test pieces to find the correct dimensions there's a very small difference between too tight and too loose.

[@ducktaperules]

This might be useful to you. I found that when designing a dual / multi tube pump, if you stagger the inlet / outlet points then the load increase at the start point is spread to difrent points of the rotation making it easier to overcome the starting friction. For you that may just be a case of rotating one of our outer faces by 1/6th or 1/3rd rotation.

[@bubsbuilds]

@ducktaperules Useful indeed, thanks!

[@klausnielsen1537]

Great tip. I came here to suggest something similar 😊

[@danieljordan4320]

Custom make your own silicone tubing in wall thickness and diameter. Silicone caulk can be pushed through a tube forming mold and can be hardened with a Hydrogen Peroxide bath. 3D print the molds so they can be split open and cleaned

[@stefanguiton]

Excellent work

[@bubsbuilds]

@stefanguiton Thanks, much appreciated!

[@Ahnii]

"Planetary pump" was my first thought when you mentioned the gearbox in this context, I mean plantery gears have the perfect motion squeezing motion^^

[@bubsbuilds]

@Ahnii right?? From the couple of test builds I've made thus far, they seem like a great pairing indeed!

[@0JThomps0]

I wanna see that planetary peristaltic hybrid. Great idea!

[@bubsbuilds]

@0JThomps0 Thanks, I'll see about getting to a 'shareable point' with it sooner than later! The most recent iteration I made is just too big a PIA to assemble, but I have a couple of ideas for fixes to try. If you have any specific requests for features or capabilities you'd want to see, I'd welcome em!

[@cuisineefficace7903]

Best chanel on youtube

[@bubsbuilds]

@cuisineefficace7903 Mom?? Is that you again?!? For the last time, no more fake hype accounts!! :)

[@cuisineefficace7903]

@@bubsbuilds bro it’s my cooking Chanel 100% real 🔥

[@roboman2444]

Does the small diameter of the contact posts wear the tubing quicker than a large diameter contact point, like the outside of a roller bearing?

[@bubsbuilds]

@roboman2444 I have wondered the same thing myself. It makes sense that having a more gradual slope in the tube deformation would create less localized stress, which I would assume will extend the tube life. I would guess there is some ratio of contact diameter to tube diameter that it's good to stay above...unfortunately I have no clue what that ratio is :)

[@DaveEtchells]

I came here to say the same thing, My physical intuition sent up a flag when it saw how sharply the tubing was flexing due to the small roller diameter. *GREAT* project and design though! I think peristaltic pumps are a great, simple and highly accessible solution for hobbyists and makers everywhere 😊

[@bubsbuilds]

@DaveEtchells yeah, that tubing just doesn't look too happy, does it? :) Thank you! And couldn't agree more! They're pretty forgiving, and (at least in my opinion) oddly satisfying

[@napalmholocaust9093]

Move the motor away and use a pulley to slow it down. There's a pully like a chain with links if you want adjustable, probably print it if your plastic is durable enough. Also with pullies is that a stacked set like inside a drill press with a ratio range might be all you need to conrol speed.

[@bubsbuilds]

@napalmholocaust9093 Not sure why a belt drive never even occurred to me! Thanks for the solid suggestion!

[@anothermidlifecrisis]

One big oscillating lobe, with multiple passes of hose is better for the life of the hose. Use a header for the inlet and outlet.

[@bubsbuilds]

Ooo, that's really interesting! I don't think I've encountered a pump with that config, I'll absolutely give it a shot, thanks for the suggestion!!

[@karyjas1]

Why do you use many rollers? Why not use the minimal amount? Do they add any benefit if you dont need precise amounts of fluid being moved? If i understand correctly, you just need your tube to form a full cycle for the roller, so that there is no back flow, but the decreased amount of deformation would save you on stress and energy spent, right? Same flow still. Maybe a slight drop, but not by 3x, im confident. And of course, would need a support bearing, but one that would just ride against the solid chasis, and not the tube

[@bubsbuilds]

@karyjas1 Very true all around! I think the main reason I use multiple is more about being 'stuck in my ways' than any really valid reason :) I do like that it balances the loading on the main bearing, but, as you point out, this could definitely be accomplished by balancing the load with a bearing on the solid chassis as well. Having the tube make a full wrap also makes for a slightly trickier track and inlet/outlet configuration, but far from impossible to resolve.

[@evren.builds]

hey, nice build Curious if the silicone hose wears down?

[@bubsbuilds]

@evren.builds Thanks! It unfortunately most definitely does! It can last quite a while, but in my experience they eventually embrittle enough to split. So I typically just give them a feel from time to time and just swap them out any time they start to feel...different? For some of the pumps I've made for hydroponics setups that are just flowing water, I'll usually just let them run till they fail.

[@Pyromancers]

Would some uhmwpe tape wrapped around the rollers that squeeze the tube be a good idea?

[@bubsbuilds]

@Pyromancers I hadn't thought of that one, and I would think that would definitely help! Seems like a much less messy option than lathering it in grease like I did :) I'll definitely give that a try on my next pump build, thanks for the suggestion!

[@SianaGearz]

Threadlocker. Haha. I have had both PETG and HIPS start crumbling apart in my hands from blue threadlocker fumes. These are the only materials that i print, i don't know how the rest behaves.

[@bubsbuilds]

@SianaGearz Ahhh, so confirmation, thank you! PETG is my filament of choice for most builds, so definitely good to know. @seanconant3218 recommended VC3 as a good alternative for use with plastics. Do you have any experience with that one or other PETG-friendly loctite alternatives?

[@SianaGearz]

@@bubsbuilds Nope not familiar at all. I don't think availability of VC3 in my part of the globe is all too great, i've never seen it, but perhaps i can get it somehow :D

[@bubsbuilds]

@SianaGearz ahhh, that does indeed make that tricky :) Well, if it helps, someone else suggested Loctite 425 as a cyanoacrylate-based option that's plastic friendly...if that one's not around your area either, guess you're just gonna have to glue everything together :D

[@seanconant3218]

VC3 is a good loctite alternative for plastic

[@bubsbuilds]

@seanconant3218 Good to know, thanks for the recommendation! Just put some on order :)

[@Pyromancers]

Maybe using propylene glycol would be better than IPA?

[@bubsbuilds]

@Pyromancers Is propylene glycol a viable solvent for common UV resins?? Consider me intrigued :)

[@Pyromancers]

@@bubsbuilds it is and it seems to work rather well. Issue is boiling it off I suppose. It doesn't evaporate as readily as alcohol. But I'm trying to find a way to switch to it.

[@bubsbuilds]

@Pyromancers insert "the more you know" GIF here! That's awesome to know, thanks! I'm absolutely going to have to get me some on order to test out.

[@Simonthadude]

Love the visuals, hope u reused the toothpaste :D

[@bubsbuilds]

@Simonthadude Thanks! And hey, I'm recently unemployed, so you better believe I scooped that minty goodness right up for later!! :)