Hardware parts sorter & storage robot

Hardware parts often come in packages of fixed quantities, so over time one could have quite a few hardware parts, so separating the spare parts by type is a good practice, but I believe is a little hard to do.
Ideally a machine should sort the parts and store them by part type. In the event the user needs a part he should be offered a drawer with all the parts of that type. Every good mechanic has this sort of a drawer and automating most of it should make it easy for beginners to use RepRap. See on Google Image what I mean for hardware part storage: [images.google.ro]
Of course examples there are by no means automated...

Edited 1 time(s). Last edit at 03/16/2008 02:06AM by Bogdan Bivolaru.

I saw some machinery that was for part storage once.

It was like a vertical system with trays suspended from belts that ran around each end (Verticaly of the machine).

To get the parts you used an up/down button until the parts tray you wanted rotated (Like a Ferris Wheel or even a Paternoster Lift) to where you could reach.

Items to print then would be the Belt/Track/Chain components, sprocket wheels to drive the linked Chain/Track items and brackets etc for fixing the trays to the Chain/Track in a way that allowed them to hang and stay the right way up as the assembly rotated over the top.

aka47

---

Necessity hopefully becomes the absentee parent of successfully invented children.

Since "a Ferris Wheel or even a Paternoster Lift" are to my point of view a transport system, they do consume a lot of energy even if one needs just a small number of different parts.

I agree they are a lift.

loaded evenly (or relatively evenly) you are not actualy lifting any weight you only need to overcome the inertia of the stored mass.

Think of a lift with a counter weight that is equal to the mass that you are lifting each cancels out the weight of the other and to move the system you are accelerating/decelerating the mass of both.

You should actualy with a little help form gears (and preferably a brake) be able to wind the storage trays round by hand.

If you considered using a Permanent Magnet motor (Brushless is good) you could also apply regenerative braking to bet beck a quantity of the energy you used to accelerate the mass as you decelerate or brake it's movement.

In reality though the prime motivator for this type of storage system is the savings you make from taking up less useful space.

Consider that most commercial properties are rented out by the Sqr Meter. and there is a lot of wasted space between head height and the roof.

Additional floor space then is at a premium but the space to the roof is already paid for.

If you want to avoid Ladders and all those things but want to make use of that expensive but otherwise wasted space then a rotary storage system is a pretty good way of doing it.

It also lends itself to Automation, with the storage vehicles (Robots) running about at floor level but being able to store things up to roof height without actually lifting very far themselves.

Putting it all under automated control also makes it more likely that clever software and a database can be used to store and retrieve anything you like efficiently. The software can also take care of the Load Mastering ie making sure that those rotary storage racks are loaded up so that they are as close to optimally balanced as it is possible to get.

You are right of course this is entirely a "Large" to "Huge" number of parts application.

I guess you were more interested in something like the plastic boxes/drawers used to store electronic components in ??

aka47

---

Necessity hopefully becomes the absentee parent of successfully invented children.

Actually I was thinking at exactly automating the process of storing & finding a particular part of a machine. I am interested in a machine that automates most or all of the production process from ordering the parts to a supplier to assembling the pieces together and testing the new prototype. The storage system is just a requirement in building the assembling machine.
I have a lot of ideas, but I lack the technical expertise to even solder two capacitors on a PCB - let alone choosing the right capacitors for the job.
I hope this will lower the entry barrier for reprap fan clubs in areas where technical expertise is scarce & expensive. OK, I understand full automation is not possible, and humans are still needed for testing & fine tuning the machines and the end product. But every process can be perfected, right?

the "plastic boxes/drawers" concept requires that all (most) of the metadata of the parts in the drawers sits in the users head, but there are several problems with this:
* that metadata knowledge is hard to transfer to a fellow repraper;
* does not scale well;
* requires a disciplined user to maintain order.

PS: I noticed some circuit parts do have a barcode on them (capacitors, resistors, transistors); can we use that in the automatic storage engine?

Interesting

Yes I think I can see where you are headed.

Consider using a rotory storage (stacker thing) you can actually access it on both sides.

This could be put to good use Rapidly Manufacturing components on the one side.

Storing them along with their type, position and quantity in the rotary storage machine.

An assembly robot or operative could actualy be taking the parts out at the other side, making up sub assemblies and putting them into another storage machine on the other side again.

You could feasibly continue in this Pipelined production method with the storage machines providing the necessary buffering between each operation.

The final operation being dispatch.....

fun eh.

aka47

---

Necessity hopefully becomes the absentee parent of successfully invented children.

Hey why not something along the lines of this: [www.molecularassembler.com]
A forklift robot that can repair other forklift robots with parts stored in a "warehouse."

Cool

---

Necessity hopefully becomes the absentee parent of successfully invented children.

take some ideas from tape storage robots
[www.pik-potsdam.de]
imho such storage robots are kind of overkill for keeping track of your mech parts, it might start making sense if you have like thousands of different tiny parts

i think a bot moving on xz axis in front of your shelves full of parts with boxes labeled with barcode labels with the reader should do the job. it doesnt have to fetch your part i think, it should be enough if it just points out the box to look into

If you are using a storage mech that is in effect the Z axis you only need X & Y to retrieve.

aka47

---

Necessity hopefully becomes the absentee parent of successfully invented children.

I think that somebody could do this with several of those sets of plastic drawers that they sell in hardware stores, and a fairly small robot that can move around the shop.

The robot has to be able to carry one or a few of the drawers, put them in and pull them out of the slots, pick up small parts and put them into drawers (or, I suppose, get them out of drawers and hand them off to a pick&place bot) and climb up and down (and maybe across) the rows of plastic-drawer cases. It can drag a cable behind it as long as it's smart enough to mind that its cable doesn't get tangled, or it can be battery-powered and WiFi communicating.

And doing it this way scales; you can add more sets of drawers and the system still works, and you can add more identical copies of the robot and the system works faster. So you could have a small system with one robot and a few hundred little plastic drawers, or a large system with a few dozen robots and a few tens of thousands of little plastic drawers, and there are smooth, incremental transitions all the way up the size scale, where nothing you've built for the smaller versions of the system is unusable in the larger versions and you don't need additional designs for larger size systems.

And compatible systems (same type of robots and same type of drawers) can be easily merged or subdivided, so they can become a commodity that can be moved between shops with a purchase or sale. Also, it's robust. A single failed robot doesn't cause the system to stop working as long as the others can go around it.