I'm relatively inexperienced with electronics, but I'm hoping somebody here will be able to help me. I need to make a device that will receive inputs from 3 digital calipers (length, width, height) and a strain gage based load cell (weight), calculate the specific weight, identify the material from a list of about 20 materials, and display the specific weight and material identification simultaneously on an LCD. The blocks will always be solid perfectly rectangular parallelepipeds. I'd like to make it as simple and cost effective as possible, and completely build it within the next month or so.
Would this be an application for an Arduino? What components would I need in addition to the digital calipers and a load cell (from a digital bathroom scale)? I've seen a couple similar devices, but they don't involve combining multiple inputs and comparing them to a database.
Are there LCDs with built-in processors that could take the inputs, calculate, identify, and display the information without the need for an Arduino board? I found these LCDs with an RS232 input, but would they be able to do what I want without a separate board and memory?
I thought this was a commercial project till I saw this...
a load cell (from a digital bathroom scale)
I had a good chuckle. I guess it is a (homework?) school project for an engineering class. ]
Look for sensors here:
Get some ideas and see what else you can think of.
I think lots of us build this stuff for a living. So yes it can be done with arduino -- but you said nothing about operating conditions -- hot, cold dusty, vibration???? Is this a prototype or a working unit? So if this really is an inspection station and it really is commercial the post the requirements specification document and I for one would be happy to look at it.
Yes, I'm a mechanical engineering student, and it's a project for school. I'm supposed to rig up a fixture to identify a brick by it's specific weight. Crazy, I know. It needs to work when I'm finished with it. I will be required to demonstrate it's functionality. Operating conditions are relatively stable (room temperature, clean, no vibrations, etc...) Block dimensions will not exceed 150 mm (6 inches) on one side and need to be accurate to 0.1 mm. Weight can go up to 5000 grams and needs to be accurate to 1 grams.
Well, the goal is to make it as accurate as possible. I believe those specs are feasible (see the amazon link above). That scale can measure 11 lbs with 1 gram accuracy.
As the materials are to be identified from a list containing specific weights.
The measurement will be inaccurate as the 3 dimensions and weight will all have their relative errors. Still the calculated specific weight will be near one from the list.
Can you provide this list?
Then you turn around the formula, given that you identified the material and its specific weight and know the 3 dimensions you can calculate a more accurate weight.
e.g.
size = 10 x 20 x 5= 1000
weight = 12000
==> specific weight = 12
In the table this "matches" with material X with SW = 11.85
Given that the size = 1000
the calculated weight = 11850 grams.
and a set of (2^14) 14 bit strain gauges might do it... a 12 bit read is a touch short, 13 bit is enough -- but you have to remember that the errors do accumulate for multiple gauges....
Maybe the prof is going into business and needs a design for a decent scale so he can corner the market. $)
A pad, four strain gauges at the corners. Add them up -- absolute error adds absolutely --- so I dunno about the accuracy. A brick is pretty craggy -- unless we are talking ceramic bricks -- but they are like cookies (kinda misshapen).
I think the guy is a sadist. He should be reported to the SPCA.
I was one of the first students to do this type of a project. (The profs told me the first and it was in 78, but maybe they were just thinking of my University) The Project Marker, a Mech Eng Masters student, was quite ticked off after he estimated the cost of the SPICE run time to check the circuitry as I had to use individual components... He was going to fail me on the grounds of impudence. ] I think I started a bad,bad trend.
It is not that tough. Just requires some thought. DOn't waste time on pretty packaging. Just test some things quickly as you think of them.
Put your feet up and think for a while. Most people skip that step and they pay for it.
PS: I am not doing your homework unless you work for me for free for six months. It''s a fair trade. There are lots of practicing engineers who stole my work when I was a student (and they were students) -- it's payback time. rotflmao! PPS: Don't walk on their bridges or buy their products. ] leh
A quick scan of that specific list learns it is ambigu for some values, you need to make your code robust to select all possible materials (bonus points or at least give a notice that the material cannot be determined for sure.
Would it be accurate enough to use a partially covered light sensor and a laser? You could clock the brick as it moves along a slow conveyor belt along the x and z axis for the length and width. Then measure it as it's lowered onto a scale for the height. Or use a servo to move a laser and sensor along the y axis.
leh
or at least give a notice that the material cannot be determined for sure.