I had a similar thought as you when I saw the sorting pics from the raindrop gb.
One thing I'm not clear on is do the key come from the manufacturer pre-sorted by key type (ie do you get a bag of tab keys and a bag of each alpha key?) Looking at the pile of keys from the raindrop GB it looks like that's how the came, but it's hard to tell and they may do it differently each time. But if that is the case it would make it easier because then you don't need the machine to figure out what key is what, you just need it to feed it keys and have it create sets.
Yes, they come sorted by key. So you have a big back of "A", "B", etc. I'd imagine the effort to build a machine to do it would be more than the effort to do it by hand. (Unless perhaps you were the manufacturer...)
Yes, they come sorted by key. So you have a big back of "A", "B", etc. I'd imagine the effort to build a machine to do it would be more than the effort to do it by hand. (Unless perhaps you were the manufacturer...)
Oh I'm sure it would be more work building a sorter for one group buy. But it's a type of problem that programmers/engineers like to think about because if you solve it once it could be reused over and over again. Of course in our case with different people running GBs it wouldn't be a huge help.
Post it?
The machine.Post it?
Post what?
Just as an exercise... what ways could you use to sort keycaps? Weight? Shape? How does a company like GMK or SP handle this? I'm sure they don't have people in rooms putting one keycap at a time into bags? Do they? Or is that an issue for the end consumer to handle? Do they send these keycaps out to major keyboard manufacturers as described earlier -- a bag of 'A', a bag of 'B', etc.
I always imagined that they injected a mold that encompassed an entire keyboard set, and then de-sprued that finished molding, and then all the keys were polished, etc -- but they traveled together, perhaps?
A keycap sorting machine... I see a conveyor belt with a weight sensor, and/or an optical scanner w/OCR of some kind. The weight might limit the possibilities, then the OCR determine which cap you're looking at, and diverts to the proper hamper. OR, a sweatshop. Either one. Second one is likely cheaper and easier to set up.
Someone build something like this ?More
Someone build something like this ?More
But you'd have to have 300 little cups at the bottom or something. :)
I envision it as a machine that you pour a bag of ESC keys into, and the thing has like 20 chutes with bags at the end. If the machine could send one key down each chute, you could just let it run for each key and boom 20 sets.
Ah! Here you go.
It's a very manual process compared to what I had in my head. Which is why we pay $50-80 a set at cost vs $10 or something if they were typical massed produced plastic pieces. :)
Were not trying to take a set of jumbled keys and sort them out. That part is done.this is not that hard to build, but it's definitely a project, and it will be very large. basically you have 87 vats, and you empty out a bag into each vat. at the end of each vat is a device that lets exactly one "unit" out per cycle. run the machines in synch and ta-da! the two problems you need to solve are:
We need the opposite of that machine, one that takes one of each color M&M and puts it in a bag
Were not trying to take a set of jumbled keys and sort them out. That part is done.this is not that hard to build, but it's definitely a project, and it will be very large. basically you have 87 vats, and you empty out a bag into each vat. at the end of each vat is a device that lets exactly one "unit" out per cycle. run the machines in synch and ta-da! the two problems you need to solve are:
We need the opposite of that machine, one that takes one of each color M&M and puts it in a bag
where do you find the space to put such a machine
how do you let exactly one unit out per cycle?
1) just requires money i think. rent an industrial space somewhere.
2) chute + gate + optical sensor. open the gate and snap it shut when the optical sensor fires. then, weigh what came out. you'll either have 0, 1, or more units, easy to discretize the weights. if you have non-1 units, dump what you exhausted into a refuse bin to be recycled and cycle again. if you have 1 unit, dump your output into the set and give the all-clear. all units proceed to the next cycle if and only if all units give the all-clear.
We've actually got something here in the office that I plan to use for doing the counting of blank sets.
Basically, the way it works is this:
- Item is dumped into a vib bowl at the top of the unit. through some adjustable peices, they eventually end up in a single row.
- They pass through an optical sensor. The sensor on our device has to be trained to the size & shape of what you're counting. Once you've done that the profile can be stored for later use.
- Once it's been trained, you set the size & number of sets your counting. When you start the machine, it pushes the items through, until it's reached the number it's been set to, and then pauses. You push a button, it fills the bag/cup/pill bottle your using for storage, and then starts counting the next batch.
LOL no, but we have made several accessories for it on the laser ;)
My boss found a fantastic deal on it, and I believe we aquired it for roughly 10K or so. Not a small chunk of change, but when it's saved us countless man hours at this point, it's earned it's title in the office off being one of the most useful purchases we've ever made. I'll see if I can shoot a quick video of it in action today or tomorrow.
As promised:
Menial tasks... isn't that what these are for?Show Image(http://i.imgur.com/jbI6qct.jpg)
LOL no, but we have made several accessories for it on the laser ;)i have talked to someone looking at making medium scale versions of these kinds of devices and described our problem to them. basically these kinds of machines are not actually that hard to make. you just design a workflow and then build a small machine that handles each piece of workflow to your desired tolerance. one of the reasons they cost so much right now is that they have to be made either with a machined metal or tooled up polymers. the guy i talked to also happens to be big in the hobbyist 3d printing community.
My boss found a fantastic deal on it, and I believe we aquired it for roughly 10K or so. Not a small chunk of change, but when it's saved us countless man hours at this point, it's earned it's title in the office off being one of the most useful purchases we've ever made. I'll see if I can shoot a quick video of it in action today or tomorrow.
The only problem would be the size, and mobility. Could the owner of the machine be a key cap sorting and packaging *company*? So SP would generally ship to them, they charge a fee to the GB participants and do the sorting and shipping in a shorter time frame.
The only problem would be the size, and mobility. Could the owner of the machine be a key cap sorting and packaging *company*? So SP would generally ship to them, they charge a fee to the GB participants and do the sorting and shipping in a shorter time frame.
The only problem would be the size, and mobility. Could the owner of the machine be a key cap sorting and packaging *company*? So SP would generally ship to them, they charge a fee to the GB participants and do the sorting and shipping in a shorter time frame.
But even if we have this super sorting machine (the one Acidfire use at job), how will you manage to know that you have a "K" and not an "A" (for example) since they both have the same size. If you put all the keys together, you need to get one of each in each bag instead of a number X of the same item. The problem seems really different to me unless I missed something big!
What would be nice is to have a big cone (pointing down) and with a lot of containers all around the big circle at the top of the cone. In each container you put all the keys from 1 specific key (so one container for the letter A, one for the letter B, etc.) and then you have a big button that, when you press it, lets 1 key from each container drop in the middle of the cone, and at the base of the cone you have a bag where the keys drop. So basically, you push the button, 1 key of each drops in the bag, you close the bag and put a new one.
The problem is that you will need as many containers as there are keys, and as many opening system as there are containers, unless you make one with like X number of containers and you do you bags with X keys at a time, them once you sorted all the keys from the X containers, you start again with X different keys, putting the bags with the previous keys again.
And then not all the keys are the same size, so you need a different opening for each size.
The only problem would be the size, and mobility. Could the owner of the machine be a key cap sorting and packaging *company*? So SP would generally ship to them, they charge a fee to the GB participants and do the sorting and shipping in a shorter time frame.the person i am talking to has a warehouse and does fulfillment for his own stock. he is interested in building specialized fulfillment machines to do medium scale fulfillment for organizations and companies who have bursty or weird fulfillment needs but can't justify a dedicated ff building
@Spam the problem with SP is that they charge an exorbitant amount like 5cents a cap I think which is understandable if you have to employ someone to do it by hand.
@Spam I would want the sorter to ship them to the end customer as well, im sure if everyone used the company they could do a decent price.
@mkawa Shweet!
What would one use a keycap sorting machine for? Does anyone really have THAT many caps?
What would one use a keycap sorting machine for? Does anyone really have THAT many caps?
What would one use a keycap sorting machine for? Does anyone really have THAT many caps?
I think it would be more meant for GB organizers. But the majority of the time, the GB adds in a sorting fee to have SP do it for them