Hello,
I work for a bakery and I need to make a report of how many baskets go to where. RFID is out of the question right now (breaks my heart!). The system used currently involves making people stack baskets better. I don't like that very much because it's boring so the results vary, but losing track of these things is too expensive.
Constraints:
The baskets are plastic, their weight is consistent. Their color is not. There are two varieties, each with different weights and sizes. People seperate each basket by type. The baskets have square holes on the bottom, like a shipping cart.
The basket collector can be instructed to pass baskets one stack at a time through the sensor area. Also, flow is very important. A solution that requires excessive time or care is not a good solution at all.
The final peice needs to cost less than $500 per unit. The cheaper the better though because the benefit is not easily quantifiable. I'd say at $501 a unit, this will not happen though. At $100 or less it probably will. This includes cost of manufacture and shipping. I'd say...500 peices total so there will be only a slight volume markdown I imagine.
The output can be simple such as a digital readout of a number.
My work so far:
I have two ideas. 1 is weighing the stacks. The problem with this idea is that the weighing mat would be hard to pull a stack over if it's any taller than a few millimeters. ( the wheels are 1in in diameter and the stack weighs about 100lbs). I'd need two different scales, one for each type of basket.
2 is the measurement of stack height via an ultrasound sensor. I wonder is the sensors can pick up the right height through the bottom grate. One basket is only an inch high so the disparity between a full stack (6') and a half stack may be too much for a precise measurement.
Please comment with ideas and suggestions. Thanks!
is the measurement of stack height via an ultrasound sensor. I wonder is the sensors can pick up the right height through the bottom grate. One basket is only an inch high so the disparity between a full stack (6') and a half stack may be too much for a precise measurement.
How about having the ultrasound sensor above the stack, measuring the distance to the top of the stack? (The height can then be calculated by subtraction.)
You should be able to make a light curtain with multiple beams -- the distance between beams equal to the height of a basket. Should be less than $5 per light beam, with the number of beams needed equal to the max - min number of baskets expected.
The system used currently involves making people stack baskets better.
You need to clarify what good and bad stacking are. I am presuming you mean they put too many or too few baskets in a stack.
If, on the other hand, they are not nesting the baskets neatly inside one another that makes a whole different problem.
You should also clarify what the minimum number of baskets that will be processed is, e.g. is it one?
The stacks are a maximum of 6' high, since each basket is 1" high that means there should be at most 72 baskets.
However presumably people sometimes sling a couple of extra ones on the stack.
The first step might be just to have a fixed bar under which the stack is pushed.
That means you know for sure a stack never has more than 72 baskets.
Next you could use the light curtain suggested by Chagrin.
Alternatively you might use a single beam to detect and count stacks and drop a moving bar onto the stack measuring how far it has to drop to count the baskets.
To handle the different basket types you have two big buttons, and two lights above the sensor area, colored the same as the two basket types. Pressing a button illuminates the light of that color. A light stays on till the other button is pressed. So the opperator pushes the button corresponding to the baskets being processed and presses the other button when he changes type. You could add further sensors to detect operator error pressing the buttons if need be.