A few months ago we showed you a bar bot built by [GreatScott] that uses peristaltic pumps to food-safely move the various spirits and mixers around behind the curtain. The bar bot uses three of them, and at $30 each for pumps with decent flow rate, they added a lot to the parts bill. These pumps are pretty much the ideal choice for a bar bot, so what do you do? [GreatScott] decided to see if it was worth it to make them instead.
Peristaltic pumps are simple devices that pump liquids without touching them. A motor turns a set of rollers that push a flexible tube against a wall. As the motor turns, the rollers move liquid through the tube by squeezing it flat from the outside in turns. Typically, the more you pay for an off-the-shelf peristaltic, the higher the flow rate.
[GreatScott] figured it was cheaper to buy the motor and the control circuitry. He chose a NEMA-17 for their reputation and ubiquity and a DRV8825 controller to go with it. The pump is driven by an Arduino Nano and a pot controls the RPM. After trying to design the mechanical assembly from scratch, he found [Ralf]’s pump model on Thingiverse and modified it to fit a NEMA-17.
The verdict? DIY all the way, assuming you can print the parts. [GreatScott] was trying to beat the purchased pumps’ flow rate of 100mL/minute and ended up with 200mL/minute from his DIY pump. Squeeze past the break for the build video and demonstration.
Is there a bar bot build on your list? No? Is it because you’re more of a single-malt scotch guy? Build a peristaltic pachyderm to pour your potion.
Why we do DIY activities? To make somethings easy and to boost our knowledge. This is a Good DIY activity. Getting involved in the creation of DIY activity is what the creator in my opinion called DIY scientist because lots of thinking require to create new DIY activities.
In less words:
the pump used in the previous project, the $30 one has a flow rate of 525mL/minute;
his DIY 15$ pump has a flow rate of 200mL/minute;
the 10$s pumps are up to 90mL/minute (at 24v while both of the above are 12v).
Props for using ball bearings! But why the expensive stepper? Is that really cheaper than buying a 30 $ pump?
$13 stepper, $2 plastics, $2 tubing, $2 bearings is already in the range of 20 bucks. Not sure whether that’s worth it…
You need the tubing anyway. There’s no way that’s ~150 grams of plastic for those parts even at 60 percent infill. Steppers are $10 on amazon, probably less on ebay.
Even if we use your numbers – after just one failure, you’re ahead, because you can rebuild yours. And stepper motors are far more reliable than brushed DC motors.
What’s a shame is that they didn’t figure out why their design didn’t pump as much as the commercial product. In theory they should be capable of performing the same…and their design should work better since they’re using full-on bearings.
You have to also include the time and cost (and education needed) to design the model, do trial and error testing, the price to buy the printer and the time and effort needed to run the printer, among a few other items. Almost none of those are amortized or have any kind of economy of scale either though some could theoretically be skipped by others who download the files and replicate the output.
All the things you mention should be considered benefits, not costs. Education and tools, sounds like win-win to me!
Haven’t done a proper test but this is much better IMHO.
12V DC motor with a gearbox + encoder for 7$! A steal IMO!
https://prnt.sc/ihorh4
https://www.aliexpress.com/item/Peristaltic-pump-DC-inkjet-printer-Mabuchi-24V-365-dc-Motor-with-speed-metering-Hot-sales/32788123449.html
I’d be rather suspicious to pumps that don’t specify flow rate. Is it 500mL/min, is it 50mL/min ? It makes a huge difference in project like one Great Scott makes, cocktail maker that is. You probably don’t want to save few bucks and then have cocktail maker that takes 10 minutes to make a single drink. For watering your plants when you’re not home on the other hand, cheap and low-flow solution is good enough.
I’ve just tested it.
At 24V it’s flow rate is almost 100 mL/min
At 12V it’s flow rate is almost 50 mL/min
Tested also with liquid coming from above, also about 100 mL/min @12V
[youtube https://www.youtube.com/watch?v=EUiZvg8K4uQ&w=560&h=315%5D
https://www.youtube.com/watch?v=EUiZvg8K4uQ
Forget the pump project – that looks like it should be the start of every cheap robot/arm project.
In the lack of a stated max flow rate just looking at the size it’s going to be horrible. It’d work but with very low flow rates it’d take way too long to make a drink.
I have made cheap peristaltic pumps using three ball bearings and a 300RPM geared 12V DC motor. The flow rate is somewhere around 500ml/min but those cheap geared motors are quite noisy and the noise is further amplified by the pump housing.
Why not use gravity and a flow meter?
Interesting thought, but a flow meter internals would have to come into contact with your drink, which is bad for many reasons.
To qualify that previous statement, I might add that cheap flow meters (turbine based) come into contact with your fluid. I know you can get external and ultrasonic ones, but they’re pretty expensive.
It makes sense, those pumps are an easy way to limit contact to the tube only.
Many bars use inexpensive “Measured Liquor Pourer” devices – they look like pour spouts except have a ball valve to limit each pour to a specific volume (usually can be ordered from 1 to 2 oz per pour). Flexible tubing + gravity/siphon feed + measured pourer + servo and off you go (or stumble, as the case may be).
That is an even better idea, considering that most cocktails have standard dosage o alcohol. But you might need some dosage thing for the rest of the ingredients.
This bot works exactly like that: https://github.com/sexycyborg/BarBot/
This girl did just that last week: https://www.youtube.com/channel/UCh_ugKacslKhsGGdXP0cRRA
What about pumping air into the container and forcing the liquid out?
I feel like Perisaltic pumps are excessively complex for booze. It’s not like blood or gravy where you have to be careful not to agitate it too much.
I’m thinking measure weight for feedback.
Well done! (As someone who has spent too much time fighting with unnecessarily expensive lab peristaltic pumps), there’s potential for an nice analog to the fluid model of electricity with these pumps: wrap the tubing in a helix around the pump stage and you increase the “voltage” (stopping pressure); multiply branch/parallel-ize the tubing through the pump stage to increase your “current” (flow rate).
But overall, to pump more, you need more power, so a bigger motor. Can you can squeeze anything more from the existing motor?
The rotor he constructed will come apart because he didn’t mount the bearings properly. Place a thin shim between the inner races of the two bearings, and the outer races won’t get compressed together, which is what causes the bearings to jam when tightened together. A piece of paper will do.
Leaving the bolts only semi-tightened will have them loosen over time.
Also, to save on cost with the motors, there’s no troube adding more rotors on the same motor shaft and simply pinching the tubes which you don’t want to pump at the input side. If your motor is powerful enough, you can run the entire show on a single pump.
this makes me want to test the flow rate of one of those 9$ automatic soap dispences..
real real slow. they do okay pushing thick soap thru them, but even that was a lot slower than i was happy with. Gimme a pump like normal and I’m washed up and out. But for usable volumes, gonna need some serious hacking.
I should open the old one in my closet, its probably clogged solid now
I tried to make one of these and while transferring liquid at above or the same level works, Getting it to draw a liquid from a even 6″ below the target doesn’t.
I have never used one of these so you can take my opinion with as much salt as you like.
This type of pump depends on the rebound of the silicone hose to get more liquid into the pump, it will not be able to draw liquid against much pressure.
It should work best if the source is above the pump so that the weight of the fluid helps reopen the tube.