This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

  • Resolved

ADS1234 - weight measurement by load cell - Kelvin sensing

Hi,

I have previously tried ADS1231 with one load cell and took the readings through an Arduino micro-controller, and the result was very stable and no problem at all.

Then I started using a ADS1234 ADC to measure weight through 4 load cells. I am using Arduino micro-controller to read data from ADS1234. Unfortunately the values coming through each load cell jump all over the place and I don't get consistent readings. I was hoping if someone could help me on this matter.

I am using the following load cell:

Model: YZC-133
Rated Load: 10Kg /22lb
Rated Output: 1±0.15mV/V
Input Resistance: 1066 ± 20 Ohm
Output Resistance: 1000 ± 20 Ohm
Insulation Resistance: 2000 Ohm
Safety Overload: 120%F.S
Recommend Excitation Voltage: DC 5V
Max Excitation Voltage: DC 10V
Cable Length: 240mm/ 9.4 inch

Since the cable length was short for me, I soldered an extra 1 meter of wire for each Supply,Vout+,Vout-,GND. Then I hooked up the wires to ADS1234.

I read and followed the following document that points out kelvin sensing(six wire bridge) configuration, but still didn't work for me:

http://www.ti.com/lit/an/sbaa184/sbaa184.pdf

Then I read and followed page 21 of the following document that advises to have capacitors across the Vout+, and Vout- lines, but still didn't help me at all:

http://www.ti.com/lit/ml/slyp163/slyp163.pdf

I am using internal crystal of ADS1234, with 10SPS and PGA = 128, and I'm implementing everything on a bread board. I am supplying 5VDC to ADS1234 and load cells through a voltage regulator and powering the Arduino through it internal voltage regulator. So, AVDD and DVDD have separate supply sources.

Working with ADS1231 was easy, but somehow I cannot use ADS1234. Could please help me and point out what am I doing wrong?

If you need any more info please let me know.

Thanks

  • Hi Arash,

    Welcome to the forum! A couple of things come to think about.  One is making sure you have adequate bypassing and analog input filtering.  Also, breadboarding a basic concept may be fine for making sure you have the proper concepts and connections but is very difficult to achieve high levels of performance. 

    Adding additional wire lengths to the load cell may be part of the problem.  The cable should be shielded all the way from the load cell to the ADC and properly terminated.  The cable can act as a big antenna.  Maybe you can post a picture of your setup and provide a schematic of how you are connecting things up.

    Best regards,

    Bob B

  • In reply to Bob Benjamin:

    Hi Bob,

    I bought some shielded wires and after trying different methods to reduce noise, I gave up, and I decided to try ADS1232 and see whether I have the same problem. I noticed ADS1232 worked fine and without any problem! Since I had positive results from both ADS1231 and ADS1232, I was suspicious that maybe the ADS1234 chip that I had was damaged, and I was right. I decided to order another ADS1234 and it worked! Either the chip was already damaged before it was shipped or I may have damaged it somehow I'm not sure.
    Do you know what's the fastest way to test a chip to figure out it's damaged or not before using it?

    Thanks,
    Arash
  • In reply to Arash Ahmadi:

    Hi Arash,

    All TI components are factory tested.  If a device becomes damaged, it is usually due to improper handling which causes an Electrical Overstress (EOS) to the device.  The EOS can be a result of ESD, so proper handling procedures are required.  Also, ESD can occur from paper or plastics that come in contact or near contact with the device.

    Other forms of EOS can come from exceeding the device Absolute Maximum Ratings as given in the datasheet.  An easy way to exceed the maximum may happen by applying a reference or input voltage to the device before the supply.  Prototyping and adding and removing wiring while the device is powered can also be a problem. 

    So the best way to test the device is to prevent any problem that can damage the device.  Other than that, there is no easy way to test the part as there are so many options and configurations.

    Best regards,

    Bob B

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.