• State Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 1 answer
  • Subscribers 47 subscribers
  • Views 2414 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

Recommended ADC for real time load cell?

acdr
Prodigy 20 points
Other Parts Discussed in Thread: ADS1231, ADS1230, ADS1131, ADS1235EVM, ADS1231REF

I'm at my wits end with the existing ADC setup (MCP3561 + MCP1501) due to poor performance and documentation, so I'm looking to find a replacement part which is easy to get working.

The device is measuring a 1kg-5kg load cell in real time for a control / measurement application. The samples per second will need to be more than 10, with as many noise free bits as possible. I moved the setup to a separate board from the rest of the electronics, so I'm open to redesigning the ADC board entirely.

The relevant details:

- the ADC board is powered by 5v over a 6 conductor cable, and has logic level shifters / buffers to convert from 3.3v -> 5v on either side. The 5v goes through a 3.3v regulator with good noise rejection on the ADC board, which powers 3.3v port of the buffer, and is available for electronics.

- CK, SS, and MOSI are seen as inputs on the ADC board and cannot be changed / driven on the ADC board

- MISO is an output on the ADC board and cannot be changed or driven on the main board

- a full duplex SPI interface is currently used, and UART (only tx/rx) via RS232 is also available

- only one differential channel is needed, ideally with a PGA of 64 or more (or whatever gives best accuracy)

These are the conductors on the cable (shielded mini-din6):

(1) 5V

(2) GND

(3) CK

(4) SS

(5) MOSI

(6) MISO

All I really need is gram accuracy with a 1kg load cell. The easier the part is to interface with from a software perspective, the better. I don't want to spend another week writing and debugging drivers again, only to have the device display undefined behavior. The price is no longer much of a concern since it's a low volume production and I have wasted a lot of time already. If there are samples available that would also be good.

Thanks,

an exhausted intern

  • 0
    TI__Mastermind 40891 points

    Hi ACDR,

    You might consider the ADS123x family of devices for your system, as they are designed for measuring load cells / resistive bridges.

    All of these devices are pin programmable, so there are no registers to program with just a read-only DOUT signal. As a result, these devices are very simple to use.

    One such device is the ADS1231: https://www.ti.com/product/ADS1231

    • 24-bit ADC
    • AVDD & DVDD = 3V to 5V (so no level-shifting required if you choose 5V)
    • 1 differential channel
    • Fixed gain of 128
    • Two speed modes (programmed via the SPEED pin): 10 SPS or 80 SPS
    • Internal oscillator
    • External VREF inputs for a ratiometric measurement

    Below is the block diagram for this device. We also have an 18-bit ADS1131 and a 20-bit ADS1230, if you do not need the 24-bit ADC.

    Finally, we have an EVM if you are interested, or if you just want to review the schematic in the user's guide to see how the device is setup

    Let me know if you have any questions.

    -Bryan

  • 0 in reply to Bryan Lizon86
    Prodigy 20 points

    Thanks for the info, that looks like a good choice. I noticed there is a significant difference between the example circuit in the datasheet and the schematic in the EVM. Is the schematic in the datasheet sufficient to achieve the listed performance? Currently I have some lowpass filters on the load cell lines.

  • 0 in reply to acdr
    TI__Mastermind 40891 points

    Hi ACDR,

    The circuit in the datasheet is a simplified version of what you would actually use in a real design. You will need lowpass / anti-aliasing filters on the inputs to the ADC and VREF (these are included in the EVM schematic). Below is a snapshot of the input structure to the ADS1235EVM, which is a bit simpler compared to what is on the ADS1231REF board. This is the type of filter structure you want to have.

    Also, please consider that your system can add errors and noise to the signal chain, and this will impact the performance of the ADC. For example, if you had a very noisy sensor it would not really matter how precise your ADC is. Just keep this in mind as you design your system moving forward.

    -Bryan

  • 0 in reply to Bryan Lizon86
    Prodigy 20 points

    Perfect, I updated the filters in the schematic. When I was reading the datasheet it didn't mention if a pullup resistor is required on the data output pin, would you be able to clarify? I am still using the one way buffers so it would be difficult to add afterwards. I noticed the last ADC I used had several pins requiring pullups.

  • +1 in reply to acdr
    TI__Mastermind 40891 points

    Hi ACDR,

    You do not need a pull-up resistor on DOUT for the ADS1231.

    -Bryan