In need of a conveyor belt built for me. NYC

[grins]

Hello All,

I am in need of an automated conveyor belt that passes cards vertically in front of proximity card scanners.  The reason I chose a conveyor belt instead of any other possible solution is because I believe this would give me the most flexibility. I even included a rough sketch of my idea below. Is this in stone? Absolutely not. I am open for any suggestions as long as it gets the job done; specifications follow. If you feel you can do this, please let me know what your price is and how long it would take. If you can suggest a company I can go to for this, please provide a URL or phone number. Your help is greatly appreciated. I would love to do this on my own but I simply don't have the time for the trial and error. Taking my electronics and physics experience into account, there would be lots of trial and error.

My specifications:
The cards that I have are approximately 2.25" wide by 3.5" tall and the bottom of the card must be 8" - 9" from the surface as they are passing in front of the card scanners. I’m looking to have 50 – 100 cards on the belt at all times and the belt must move extremely slowly, between 4 – 6 rpm, in order for the scanner to be able to scan the cards as they pass. But I will also need it to move at a faster rate, up to 60 rpm, for other types of card readers. Furthermore, the cards will need to have a 3" gap between them so that the reader has some time to re-initialize. Ideally, the belt would move 3" then pause for 2 seconds and repeat this process. This would have to be working non-stop for as long as 2 weeks at a time, maybe more, so it can’t be battery operated.

I understand that based on 50 cards and their 3" gaps, I would require a belt that has a 262.5" circumference and a 7' diameter. This is far too large of a footprint for my space. My idea is that since I can't build out, I must build up. Below is a rough yet descriptive sketch of my thought on this; click on the link beneath it to see it in full size. But, again, I am open to any suggestions you might have.

Another thing is that the weight of the belt with the cards on them and any supporting columns will introduce resistance that the motor must be able to handle.


http://grebnirg.com/Demo1.jpg

[Admin]

I just want a few clarifications . . .

Is this like a business card scanner? What material is the card made out of (paper, plastic?)? How thick is it? How stiff is it?

The only purpose of the belt is to pass the card under the scanner, correct?

How are the cards placed onto the device? How are they removed? Do they still need to be in order after they are removed? Can the conveyor belt just dump them into a box?

What does the scanner look like? How close must the cards be on the scanner? How would the scanner know if a card is in front of it or not?

Does it need a speed control or just an on/off switch?

I think a cd changer like device, or printer feeder like device, would do the job. Put a stack of cards manually into the feeder. The device would then take a card from the stack, hold it in front of the camera, then discard that card. Repeat.

How much would you pay for such a device? It sounds like $200 or so in materials, alone. Probably 50 to 100 man-hours. How many do you need? When do you need it by?

Back in college I made a robot arm that can shuffle and deal a deck of 52 playing cards. It doesnt sound like that different of a problem . . .

[grins]

These are hard plastic cards that vary in thickness and stiffness slightly, starting from approximately credit card thickness/stiffness and increasing a little. These are the type of cards you would use to get into a building by passing the card along a reader installed on a wall or door.

The same set of cards will be taped to the belt as it rotates. So they would just repeat the journey over and over again.

The purpose of the belt is to move the cards vertically in front of the readers.

The readers will be placed, upright, no more than 2 inches from the moving cards. The cards dont need to actually touch the reader.

The readers are somewhat tall, so the card would need to travel at approximately 8 inches off whatever the surface (floor/table) is. You can think of the reader as the set of lines that make the tall and narrow "F" on the left side of my sketch. I will have a bunch of them placed on the surface, around the belt.

The readers look almost like the devices your fedex or ups delivery guy uses to scan the bar-code on the label. Maybe slightly bigger. But these dont require buttons to be pressed or any other user interaction. They just stand there and read, re-initialize, then read again...

Speed control is needed as the motor needs to turn from 4 rpm to 60 rpm.

This is a lot easier than a robotic arm that can shuffle and deal cards. Its just a rotating belt with a decent amount of torque and some speed control. The only catch (at least to me) is that the belt has a course that it needs to follow, while carrying the cards, without snagging or stopping due to the cards being attached to it, which I believe would be guaranteed by making sure the circumference is big enough on all the turning points.

Please tell me your total price and how long you think it will take based on what I have said so far. If you have something in mind, other than the device I sketched, please explain how it would work so that I can decide if it would work for me.

[grins]

Any help or suggestions would be greatly appreciated. Even if anybody here thinks I would have a better change of finding what I'm looking for at another form, a link would be great. Thanks

[Militoy]

To be quite frank – the concept of using a long belt and wrapping it up around a columnar spindle seems to me a little Rube-Goldberg-ish. If I were given the task of moving cards horizontally past a reader, and had a limited budget, I would likely take a slightly different approach. There are many ways of course to solve the problem – but for the sake of argument, let’s say we’re stuck with the conveyer belt approach. First of all – a 25-inch long belt will hold 100 cards, if the cards are overlapped, and spaced 0.25 inches apart. To hold the cards, I might use small tubes, made of a springy low-cost material like beryllium copper, and split lengthwise to hold the cards securely. The tubes could be easily attached to the belt by brazing another, maybe smaller, tube to the split tube. These would be held to the belt using pins on top and bottom, forming a hinge, and this would allow each card to “flap” over when we wanted it to. I would then mount the belt on two spindles, one of them driven by a small stepper motor with a built-in rotary encoder. The whole thing would be maybe 11 inches long, not including the cards hanging out the ends. It would then be fairly straightforward to design a little pin to be secured to the bottom of each tube; used to present the front of only the card being read to the scanner, as the pin passes a fixed pawl at the base of the scanner. The stepper would advance the belt so that each exposed card could be read, then back across the pawl, flipping over the card, so that the next card could be read. The cards before the “active” card would have their backs exposed, and the cards after it would be hidden behind the active card. Cards could be quickly pulled from or inserted into the “stack”, by sliding in or out from above. The stepper motion control could be handled by any PC, from an old 286 up – or an embedded controller could be built in. A microswitch could even be used to detect when a card was flipped, to simplify programming the repetitive motion control, and to allow for mechanical tolerances.

[grins]

Hey Militoy, Thanks for getting back to me. While your idea would probably work if I had only 1 scanner, that is if I understand your idea correctly, I'm trying to use up to 15 scanners simultaneously. This is why I figured it would be easiest to have the cards constantly in a scan-able position. Would your idea work for this many scanners standing side by side and scanning cards simultaneously?

As much as I want to get into connecting logical systems to motors, and I ultimately will later in life, I cant spend the time learning these procedures for this project. This is why I would like somebody with a bit more experience building it for me.

Taking into consideration that I am not even a beginner as I haven't ever begun anything involving motors, if somebody here can explain how I can complete the motor setup (based on the specifications in my previous posts) in a manner simple enough for a smart six year old to understand, I am willing to give it a shot. But I would prefer to pay somebody to do it.

[Militoy]

A version of the idea would work for multiple scanners - but only if the scanners aren't synced to each other as far as time is concerned - in other words, if a card can be read by each scanner in turn, then the next card, etc. (card 1 scanners 1->15; card 2 scanners 1->15...). A belt can be made to move rather quickly, so the dwell time in between cards can be as short or as long as you want. It sounds to me like the size of your setup will be determined mainly by the size of your scanners, and how they are placed. As I noted - there are lots of ways to solve the problem - my idea was just one example. Out of curiosity - are you able to share with us what this thing willl actually be used for?

(Edit) BTW - a Google search for Machine Design companies in the NYC area turned up >22,000 hits.

[grins]

In their stands, the scanners are 5"W x 11.5"H x 6"D. The scanners are not syncing with each other. They are syncing with a computer via a wireless connection. I'm doing some QA work for a company that is writing software for card scanners and I want to automate the scanning process.

Trying out a Google result, emachineshop, I downloaded their very cool CAD/pricing software and, just to get an understanding of pricing before I dive in and design my model, I got a price check on a thin-walled, cheap metal, 12"x4"x5" box. It came out to over $600! The thing is, I believe my required machine is not an expensive item when it comes to parts. A reasonable labor charge is understandable. But $600 for a small box? I'm not looking for something to display at CES. Functionality is key here.

I have to say that I can't visualize your idea. This is probably due to my lack of experience. Would you mind elaborating or showing me pictures of examples? Remember, the cards must be vertical and facing the scanner when being scanned. And the part of the scanner that actually does the scanning is 8"-11" off the surface.

[Admin]

another clarification . . .

The cards are added to the 'belt' manually, correct? If its just 50-100 cards, the cheapest way is to hire some guy to manually attach them. Robotics should only be used when units go into the range of a few thousand.

Can you tell us why there are 15 scanners, and why the cards must be re-read?

The attached image is a quick 15 min sketch in 3D of what I think you want. So there are 100 cards manually attached around the belt, and one at a time it drives the cards through the 15 scanners. The machine can be rotated in whatever direction the cards need to be read in.

Thinking about materials . . . the belt is like $50+, the stepper motor probably $100-$200, control electronics about $200, total structural materials $300-$500, clips probably $300 (if its just $3/each). Labor for a single cheap mechanical engineer at $30/hr rate, for about 60 estimated hours . . .  FYI, for it to be worth the time and effort, a typical company would triple the price of the device for a good profit margin above the cost (breaking even means a wasted effort) . . . well you do the math, thats the cost for the thing.  ($9k)

I do not know of any companies personally that would do this. But Im confident there are a lot. Try googling custom manufacturing machine design.

I would/could do it except Im in DC, and I wont build anything past the design on my CAD without being prepaid

[grins]

This company is not willing to pay that much for this machine and the sketch you provided wont work since it doesn't take into account the specifications I posted. I appreciate both of you trying to help me out with this. It looks like I have to approach this from another angle. Thanks again.