Oh buddy am I watching this with a lot of interest. This would reduce the kit price a bit, but increase my enjoyment and productivity by more than anything else I can think of right now!
So far a sorter shown in the beginning of both vids,
And a counter,
I am fine manually packaging them since I do a hundred or two at a time, not thousands. But I am sure some sort of packaging is next!
Stackable rackable machines that increase productivity. Nice. Would be Interesting concept to see in person…
Happy to try and adhere to a foot print that enables stacking MP3DPs for a public event. Larger to smaller machines could be stacked with adaptor plates/brackets between them even.
Well shoot this is popping up other new vids. I built a machine that counts screws - YouTube
Yikes, they counted by hand…A counting scale is much faster and better for your mental health.
Haha I just watched a screw counter machine yesterday the one in your post above, and thought ohh I bet Ryan has some cool machine.
At work I wanted to make a machine for our stores person.
Rivet counting isn’t fun lol.
The video I saw had a decent failure rate of like 5% tho.
Watched that the other day and thought you should see it. He hasn’t shown the main part yet though.
so you can make a simple arduino scale from parts at Amazon and have a weigh counter, for like 10 bucks. This is not the one I was thinking of, but you get the point :
The biggest issue with that is when you get 10,000 bolts they show up in multiple boxes. Each box’s bolts can weigh a bit different. makes double checking your work difficult.
The nicest part of that other one is it does the sorting for you. Just dump the bolts in the bucket and go. It will not pick up incorrectly sized hardware and has them ready to dump directly in a bag. Currently, I have to dip my hand into a box of bolts 100+ times, then funnel them into a bag. It does not take a huge amount of time but it is very manual. To the point my fingernail starts to pull away from my finger a bit from opening so many bags. I learned to try and keep it under 100 per day. (this is why I do not sell hardware separately)
I did design a mechanical sorter but just like he found my metal sensors are not accurate enough for a target that moves like hardware does.
Optical, or that wheel counter is the only way to go.
Youtube gave me this as a random link today. Sounds like a kindred spirit:
Yup the error really is a bummer, he had it on that video as well.
I watched this the other day too and really love the “cassette” design so you can have the different bolts and nuts in the cassette’s and setup for what ever the bag requires, I believe that he was saying that he was going to introduce a screen to say have a button that give you 20 or 30 or 40 etc.
So in your case that would save you a bit of time, and effort. The stackable design is really nice too.
Air task out the labelling and “opening” of each bag and you are golden. Pick up a bag, put the already opened end over the shoot and press the button for 50 units (or what ever number), seal it and drop it into your storage bucket.
.Very slick, but wow the work to build one!
Also there was no mention of a unit and cassette designed for the matching nuts…
I can see why individual detection may be imperfect, but also weighing a large number of screws would be subject to error.
Perhaps you could use a counting machine to count out small batches onto a scale and then automatically dump from the scale into the output.
If you count out 5 screws onto the scale, maybe once in a while you get 4 or 6, but the scale should tell you how many you actually got, assuming the weight variation is less than 10%.
Or, if you can periodically tare the scale, or virtually by just recording the weight, you might be able to achieve the same effect without having to mechanically dump the small batches. You just need to pause the counting long enough for the scale to settle.
You might not even need to count the screws onto the scale. If you can deliver them approximately one at a time and you can pause, that should be enough.
The hard part might be integrating the scale with an arduino, unless someone is selling a hacker-ready scale.
The hard part might be integrating the scale with an arduino, unless someone is selling a hacker-ready scale.
There are lots of Arduino builds of scales. They use a load cell sensor and an amplifier, often sold together. Here is one link.
I can see why individual detection may be imperfect, but also weighing a large number of screws would be subject to error.
Personally, this is where I’d focus my design efforts…how to accurately grab exactly one item. I would think that it would be far easier for a load cell to detect that exactly one item is selected (vs 0 or 2 or 3…), then to determine if a bag has 99 or 100 items.
Yeah, that is the idea, except it might be faster to use small batches instead of weighing them one by one, but either way, it should be possible to get perfect counts.
Seems like it wouldn’t be too hard to fix the under issue using multiple sensors and discounting single triggers. It probably would be worth giving someone 66 M5 screws instead of 65 to save the time. 64 however is a big problem.
Well the errors I encounter might help define the issue.
Random miscellaneous hardware, not related to what I need. (Ex M8 bolt in a M5 box)
Chunks of scrap metal, from the manufacturing process I am sure.
Shorter or longer hardware, Need M5x30 get M5 by 10, or 40.
So that initial mag wheel in the first videos is very appealing because of the sorting abilities. Just like a vib bowl you can easily tune it not to pick up large things, and sort out longer and shorter things pretty easily as well.
I like the cassette idea, that is how they do it with vib bowls, I just would not do a vertical cassette. I understand he is concerned with space but I do not want to vertically stack 70 5/16" bolts the load would cause issues. I ramp or curved ramp would be a better option and make counting optically easier.
For bagging there are “autobaggers” pretty expensive but use a special roll of bags, blow them open and heat seal them. The other kind is like a roll of sheet material that gets roll formed into a tube heat sealed on the bottom, filled, heat sealed on the top and cut, rinse and repeat. You can import these commonly as loose tea baggers from china, from there you can see the different styles.
These do not need to be fast if they are fully automatic and reliable. If there was a cheap and easy way to do this a lot of small companies hardware prices would surely drop significantly.
From what I see in industry it is a custom tuned vib bowl ($3k+ each, before counting and packaging) for each different piece of hardware, optically counted as they slide down a shoot then dumped into a box or bag.
I think once you get the hardware into a slide, counting and getting them to a package is the easy part. It is the initial sorting and getting them to the slide that is not easy. I thought it would be.
This was my last attempt. The vibrator plate was loaded and would turn on when need to slowly load the plate. The plate had all sorts of attachments to arrange and space the hardware (worked with M2.5 all the way up to the 5/16x1 1/2"). Once everything faced the right way and had a bit of space to them they were kicked out over the sensor and counted. I had a web app that counter teh current number of items, and total number of bags counted, and it would stop when I hit the goal on both. for certain things it worked pretty well, for others the error was unpredictable. I tried 5 or more different sensors and all sorts of different chute designs. It has finicky. You could also pretty easily get hardware jams on the plate.
Look how well this worked, drop things in the middle and they slowly work there way out getting further and further apart, then booted out the shoot.
Wouldn’t need to have that many at once. If it can fill the tube as fast as shown in the video, you could only keep a dozen in the mag at a time.
There are counting scales designed for this. Not automated but definitely save a ton of time counting things. Used to use them for inventory were I used to work.
