If there’s an unsung hero of manufacturing, it’s the engineer who figures out how to handle huge numbers of small parts. It’s one thing to manually assemble something, picking each nut, bolt, and washer by hand. It’s another thing to build a machine that can do the same thing, but thousands of times in a row, ideally without making mistakes.
Most of us don’t need that level of automation in our processes, but when you do, it results in some interesting challenges. Take this pneumatic screw accelerator that [Christopher Helmke] designed for his modular production system. One of the custom machines in his system is a screw counter, which uses a magnetic wheel to feed screws — or nuts or washers — from a hopper, orient them correctly, and drop them into an output chute. While the counting bit worked quite well, parts would only go so far under the force of gravity in the clear vinyl tube used to connect the counter to the next process.
[Christopher]’s solution was simple but effective. His first prototype simply injects compressed air into the parts feed tube, which pushes the screws through the tubing. It works surprisingly well, propelling the parts through quite a long length of tubing, handling twisting paths easily and even working against gravity. Version 2 integrated the accelerator and a re-orienting fixture into a single part, which mates with a magazine that holds a large number of screws.
There are a lot of interesting features [Christoper] built into these simple parts that are worth keeping in mind. Our favorite is printing channels to guide small cable ties around the tubing to clamp it into the accelerator. We’ll be keeping that trick in mind.
I used to work for Partylite candle corp. in the tealight department. We had 5 lines making 130 tealights per minute per line so high speed everything. the tealight cups would go into a wicker machine from a vibratory bowl then the sustainer (little metal part holds the wick) would come from another vibratory bowl into a chute. The wick would feed from the bottom of the sustainer and the get cut. The sustainer would get crimped in the same operation. the wicker would then carry the sustainer/wick and the cup around the inside of the wicker. At one point a drop of wax is dropped into the cup and after several more moves the sustainer/wick is pressed into the cup. The cup gets filled with wax and moves down a cooling conveyor. Throughout this whole process if anything was not in line the candle would fail. The moral of the story is one of the engineers had a saying that went “Put a blow on it.”. Sustainers getting stuck in the chute “Put a blow on it.”. cups not feeding down the feeder “Put a blow on it.”. cups not feeding into the machine properly “Put a blow on it.”. I have no idea how many time in my life I have fixed a problem because of those words of wisdom. “Put a blow on it!”
That wicked ! mind blowing.
An article about shifting complexity would be interesting.
A length of extrusion isn’t necessarily simple either – it’s just shifting the complexity to a different place where you don’t have to deal with it: the factory that makes them. It comes out of the box formed and cut to specifications you can rely on. A truly simpler part could be a plain round or box tube, or a flat bar, but then you have to deal with the uncertainties with your interfacing parts.
One could also take a view on loose vs. tight designs – how well the parts have to align and fit to make it functional, how rigid or flexible, and what that means in terms of complexity of dealing with the non-ideal construction. It’s easy to design something that works perfectly in a CAD simulation and then have a drop of grease or a piece of dirt jam the entire thing.
I had a good learning experience in building a prototype. It had some axle bearings simply sitting loose in slots, which were good enough for the prototype, but I was concerned that it might shift the axle out of place and cause problems – so the next version I specified captive pillow blocks. The machine became all crunchy and sticky – whoops – the original version allowed enough play that the bearings would not get twisted and pinched when other things shifted around, which was hiding rigidity issues elsewhere.
Sometimes when you fix one problem, you have to chase it through the entire mechanism. The good question is, was this a problem in the first place? Did you just complicate the entire thing by over-engineering one part?
In a different case, wise from the previous, I designed a “tolerant” fit. Well, the bearing went slightly askew and started swimming up and down the axle, winding tight like a screw and then popping back down. Crunch crunch it went again.
Vibratory sorters and air handlers were a mainstay of the plants where I worked. It was almost magical to watch parts sort themselves, march up a ramp, and leap into the product on a moving conveyer belt at exactly the right place and time.
I also enjoyed watching a pneumatic box packing machine. A box would fall off the end of the belt, puffs of air would rotate it into just the right orientation to drop into the shipping container. In a matter of seconds, the small boxes would be stacked in interlocking layers like bricks, just in time for the big arm to grab the now-filled box, seal it, and pass it on to go into the warehouse.
No computers, no electronics; just very clever mechanical engineering.
Not enough comments celebrating watching this thing shooting out machine screws, IMO. :)
https://youtu.be/hcJ7R7xvt44?t=332
It’s a screw machine gun!!! Everybody needs a screw machine gun!! High enough pressure could shoot screws through solid objects!!! YESSSSSSSS!
It´s your fault, you trained people with click-bait articles for years.
Change it to “machine gun spits screws at light-speed at moronic HaD editor” and you will see your ad revenue increasing.