Pages: [1]   Go Down
Author Topic: Dual scale project  (Read 737 times)
0 Members and 1 Guest are viewing this topic.
Offline Offline
Newbie
*
Karma: 0
Posts: 6
View Profile
 Bigger Bigger  Smaller Smaller  Reset Reset

Hello,

I am currently working on a project where I am connecting 2 sets of digital scales to an arduino and outputting the result (individual and total) to an LCD screen. It will be weighing people and measuring how balanced people are between their left and right legs.

The scales are basic models from Argos (UK), they have 4 load cells in each corner of the scale arranged in a full bridge. I am keeping this arrangement. From what I can tell (I don't have access to a scope) they used an alternating excitation voltage.

I am using a Texas Instruments ADS1232 for both the PGA and ADC stage.
I have it set to:
  • Speed: Fast (80SPS)
  • PGA: 128
  • Internal oscillator

The LCD screen is a 4D Systems uLCD32PT. This is working great!

I have it linked up pretty much as the diagramme in the data sheet.
A few changes:
  • The second bridge is connected to AINP2 and AINN2.
  • A0 is directly connected to the arduino
  • Excitation voltage is drawn from an ardiuno pin and REFP is tied to this as well.
  • Speed pin is high (80SPS)

The main problem I'm having, and hoping people might be able to help with, is getting consistant, repeatable output.

I have done a lot of averaging and smoothing in software and the numbers are looking ok, but if I weigh myself and then do it again I can get different results, up to a few kilos difference or about 4%.

Below is the code I'm currently using. The softSerial is there for debugging.

I have attached the code as the max forum post is 9500 chars.

* ChiroSAM_Production.ino (17.3 KB - downloaded 18 times.)

* 6404.ads1232_example.bmp (3075.52 KB, 1296x810 - viewed 57 times.)
Logged

Left Coast, CA (USA)
Offline Offline
Brattain Member
*****
Karma: 361
Posts: 17294
Measurement changes behavior
View Profile
 Bigger Bigger  Smaller Smaller  Reset Reset

Quote
Excitation voltage is drawn from an ardiuno pin and REFP is tied to this as well.

Could you be more specific about which arduino pin is being used for the excitation and REFP. These two points should be wired to the Vcc voltage of the ADC chip so that the reading will be ratiomatic. Not saying that is your problem but "arduino pin" leaves a lot of possibilities that could be less then desirable.

Lefty
Logged

Offline Offline
Newbie
*
Karma: 0
Posts: 6
View Profile
 Bigger Bigger  Smaller Smaller  Reset Reset

I'm using pin D8.

The system will likely be left on for long periods of time so I'm trying to find a way avoid have power across the scale until it's needed. I could look at powering the whole ADC and scale from a pin?

Thanks,
Logged

Left Coast, CA (USA)
Offline Offline
Brattain Member
*****
Karma: 361
Posts: 17294
Measurement changes behavior
View Profile
 Bigger Bigger  Smaller Smaller  Reset Reset

I'm using pin D8.

The system will likely be left on for long periods of time so I'm trying to find a way avoid have power across the scale until it's needed. I could look at powering the whole ADC and scale from a pin?

Thanks,

I would think that would not be a good idea, either what you are doing or proposed powering the whole module from a digital output pin. A high quality ADC setup requires a good stable noise free Vcc voltage not subject to the variations of digital output pin voltage with current draw. Suggest you at least testing your setup using the normal regulated +5vdc power bus for the module and bridge excitation voltage and see if your accuracy/variation problem is still apparent or not.

Lefty
Logged

Offline Offline
Newbie
*
Karma: 0
Posts: 1
View Profile
 Bigger Bigger  Smaller Smaller  Reset Reset

I am just getting started with Arduino and electronics, so I can't speak much about circuits, but I work for a digital scale company and have too much knowledge about scales. Here are some links that might help:

http://www.analog.com/library/analogDialogue/archives/39-12/weigh_scale.html

http://www.weighing-systems.com/TechnologyCentre/Strain1.pdf

Fundamentals of Weighing Technology - Attached. EDIT: Arduino forum attachment isn't working. When I download the attachment, I only get a partial file that is corrupted. Sorry.


There are many sources of error in weighing systems that get filtered out through hardware and software. Besides internal electrical noise, there are also errors from things like load cell hysteresis, vibration, temperature fluctuations, etc. Most digital bathroom scales also adjust the zero point each time they are turned on, and aren't meant to be left on constantly. Digital bathroom scales typically use four cheap steel loadcells which aren't the best for high accuracy, but do well in home use scales. More accurate platform scales will use one single-point load cell, but they cost quite a bit more than home bathroom scales and aren't as low-profile.

* Fundamentals_of_Weighing_Technology.zip (763.58 KB - downloaded 85 times.)
« Last Edit: April 05, 2013, 10:38:33 am by onyx86 » Logged

Pages: [1]   Go Up
Jump to: