ADS1232: Temperature drift correction

Simone Edilsite SRL said:

Part Number: ADS1232

Hi, I'm building a load cell, I'm having some trouble trying to understand how to fix this issue:

Well not building a load cell .. building a weigh scale using a load cell ....sorry

Simone

Hi Simone,

Welcome to the E2E forum! To fix issues with drift we first need to understand precisely what is being affected by drift.  Can you send me the schematic you are using?  Also, can you specifically tell me what the pin settings (voltage levels) are for gain, speed, etc.for the ADS1232?

For your load cell, can you give me information as to the maximum weight that can be applied to the load cell?  And also the sensitivity of the load cell (in mV/V)?

Are you issuing the self-offset calibration at power-up?  Have you tried issuing the self-offset periodically during operation?  Are you using the same voltage for both the reference voltage for the ADS1232 and for the excitation of the load cell?  Doing so helps limit the drift effects of the supply by making the measurement ratiometric.

Best regards,

Bob B

Bob Benjamin said:

Hi Simone,

Welcome to the E2E forum! To fix issues with drift we first need to understand precisely what is being affected by drift.  Can you send me the schematic you are using? 

i use the schematic from TI with one change

I connected the AVDD and DVDD to arduino VDD  so I'm using only the 3.3v from arduino

Bob Benjamin said:

Also, can you specifically tell me what the pin settings (voltage levels) are for gain, speed, etc.for the ADS1232? 

gain is set at 128

Bob Benjamin said:

For your load cell, can you give me information as to the maximum weight that can be applied to the load cell?  And also the sensitivity of the load cell (in mV/V)?

maximum weight 200kg 

sensitivity 2mV/V

Bob Benjamin said:

Are you issuing the self-offset calibration at power-up?  Have you tried issuing the self-offset periodically during operation? 

self offset only at first POWER ON, then I' don't use self offset anymore, i use a transistor to shut off the load cell, i use POWERDOWN for the ADS1232.

Bob Benjamin said:

Are you using the same voltage for both the reference voltage for the ADS1232 and for the excitation of the load cell?  Doing so helps limit the drift effects of the supply by making the measurement ratiometric.

same 3.3 voltage for ads1232 and load cell coming from Arduino

Simone

Thanks a lot

I tried without power down and original code from Amid as linked in the first post, but I have same bad drift temperature related, this is will end in a useful project and in a useful weight scale, hope someone can help, i have a weight scale on with same weight in the outdoor on 24 h a day, is it possible to read the same weight even if the temperature changes? this is should be a common problem, is there a common resolution? It's so frustrating, really thanks to everybody wants to help

Hi Simone,

Can you draw precisely how and where you have the transistor connected in your circuit?  You want to make sure that the switch is not a part of the measurement, so if the switch is on the low-side of the load cell, then the switch should be between REFN and ground.  Also the load cell EXC- lead should connect to the junction of REFN and the transistor where the other side of the transistor connects to ground.

Best regards,

Bob B

Thanks Bob I will try removing the transistor. And then I will post the results.. really thank you for your support. I will let you know as soon as possible

Simone Edilsite SRL said:

Thanks Bob I will try removing the transistor. And then I will post the results.. really thank you for your support. I will let you know as soon as possible

Ok I replaced the load cell and kept everything else the same. This is the result, still a temperature related drift but this time in inverse related, when the temp goes up the weight goes down, but this time is just 40 grams, and this is ok for my project. It seems that the only problem here is the quality of the load cells, not the circuit itself. I would like to implement a more stable power source, but don't know how, I would like to power all the board with a reference stable voltage of 3.300V with a maximum power load of 70mA, so LIPO battery > voltage regulator> circuit. But really don't know if I will get any better.

Bob Benjamin said:

 

Hi Simone,

You actually may be seeing combinations in drift sources.  I would recommend periodic offset calibration for the ADS1232 to reduce that drift error.  A common question is with respect to how often is periodic.  I really can't give a specific answer as this will depend greatly on environment and system application.

I use offset calibration every two minutes, don't know if it's correct, every time I "powerON" the system from "sleep" take the reading, trasmit with lora module, powerOFF

Bob Benjamin said:

As far as the issues related to excitation/reference drift, this should not be a factor if the measurement is ratiometric (reference and excitation voltage is the same).  As far as load cell drift, some load cells are temperature compensated, but many low cost load cells are not.  It sounds like the ones you are using are not compensated.

I understand that using the regulated voltage from arduino 3.3Volts is OK because i use it as reference and excitation, thanks. For load cell TEDEA will give me the solution but the cost is so high that will be usefull in our project that is a low cost low budget one

Bob Benjamin said:

 I don't know the battery voltage you are using, but you should be able to find a boost converter to help maintain a steady voltage.  I would recommend that you boost slightly greater than 3.3V and then use an LDO for the ADC and load cell supply to clean up any switching artifacts.

https://www.ti.com/power-management/non-isolated-dc-dc-switching-regulators/overview.html

Best regards,

Bob B