force sensor to determine can weight

[Hester EE]

Hey guys I'm new to the site, it looks very interesting and I know I will learn a great amount of information from you all.  Anyway i'm a Senior in Electrical Engineering at University of Louisiana and I need some input on a couple of design ideas.   

I'm working on a senior design robot project, and it involves determining the weight of 3 cans.  The competition requires us to determine the weight of 3 cans, and then orgranize them on a competiton board based on 3 distinct weights:  63g, 128g, and 198g.  We want to be able to determine the approximate weight of the can by only using a sonar type or some other system.

The main idea is to determine the weight of the can without having to drive up to the can and pick it up; ie save critical time in the competition.  Do you think that a sonar type system can actually determine the difference in density level  between the 3 cans?  I don't know much about sonar, it was just an idea. 

what do you guys/gals think? thank you in advance!!

Keith

[Hester EE]

I need to be able to distinguish between 3 can weights: 63g, 128g, and 189g.  Do you think i'll be able to develop a system to weigh these cans using a force sensor and a pic microcontroller?

thanks

[paulstreats]

Since there are many commercial digital scales on the market, i cant see why you cant?

[Admin]

its been asked before

And no sonar won't work . . . ultrasound probably will but you'd need the cans beforehand to test/calibrate.

Its not dependent on weight but of density and material (metal, plastic, etc.).

Im assuming all 3 objects are the same shape/size?

[HDL_CinC_Dragon]

Using a force sensor would require you can pick it up... or at least push it but using the pushing method would require a lot more mathematics (determining friction force, breaking the friction bond, etc, etc) where as the picking up method would simply require 98N of force is about 10Kg mass with a gravity of 9.8m/s/s.... simply using the F=mg equation solving for 'm' by dividing 'F'(in Newtons) by 'g'(Acceleration due to gravity - typically 9.81m/s/s)

But as you said, you cant/dont want to have to pick it up right? What kind of cans are they? Are they coffee cans with just some weights put in them? or are they soda cans?

[SomeSaba]

ok this is pretty random but i have a Q:

what if u have a current sensor that checks the current draw on motor lifting the can, the current would increase with a heaier load right?

would a force sensor be better in this case?

[Hester EE]

Yes I have to also pick up the can to weigh it, I will use a forlift type system with a scale on the forks for accurate weight measurement.

We want to use a sonar system to quickly determine the weight of the cans.  The can will be a campbells condensed soup can oriiented sideways and it has a 4" square piece on each side so that the can is .68" off the ground.  The competition cans will have 3 different amounts of sand in them...each totaling 63g, 128g, and 189g. 

Where can I find additional information about the ultrasound idea?  Can it give an accurate measurement 1-2 ft away from the object?  Thank you!! 

[Admin]

Wait so you have multiple materials with multiple densities? (a metal can with stuff inside)

This is the best I can find:
http://scitation.aip.org/getabs/servlet/GetabsServlet?prog=normal&id=JFEGA4000126000002000189000001&idtype=cvips&gifs=yes

Its possible, but waaaay to complex for you . . .

Id just go with the current or force measurement ideas . . .

[ed1380]

IDk if soup cans are magnetic, but what about using a spring and a pot to measure how much the can moves teh spring

like this but with weaker spring and a pot to measure movement

[Hester EE]

I'm pretty sure the force sensor/scale idea will work, I just needed some input on that ultrasonic weight detection idea. 

[Hester EE]

I need some additional help developing my can grabbing assembly.  As before I need to pick up 3 cans (62g, 128g, 189g) and be able to determine which can is which.   I want to try and use a servo motor with a current sensor to detect which can I am picking up.  Is this feasible? should I use a different motor?  What kind of current differences am I looking at for the 3 different cans (roughly)?  I need the current sensor to output a voltage then I will perform a/d to figure out which can is which.  thanks

[Admin]

What you want is called a current sense IC (scroll halfway down):
http://www.societyofrobots.com/sensors_currentsensor.shtml

You then connect the output to the ADC of your mcu.

[Hester EE]

Is a servo motor easy to get a current reading for different outputs? will it be reliable?

[Admin]

Its easy to get the current . . . the current use from a servo is non-linear and unpredictable without prior testing. It does this strange power stepping, which I suspect is an artifact of its control system.

When you try it, you will see.

[ed1380]

will the cans be magnetic?

[Hester EE]

No just campbells soup cans filled with different amounts of sand.