• State
  • Locked Locked
  • Replies 1 reply
  • Subscribers 46 subscribers
  • Views 862 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

ADS1262: Conversion Latency clarification and help

Paulo Arruda
Prodigy 10 points
Part Number: ADS1262

Hello,

I'm using the ADS1262 to read two load cells (AIN0-1, AIN2-3) at a high rate (7200sps). According to the datasheet:

"To make sure that conversion are settled after changing channels, start a new conversion for each channel using the START pin or start command"

So, I'm using the pulse conversion mode, 7200sps, changing the input multiplexer register (AIN0-1), sending the "START1" command, waiting for DRDY go low, reading, sending the "STOP1" command, changing the input multiplexer register (AIN2-3) and repeating everything again.

The problem is, the maximum reading rate that I can reach is about 1900 sps (Using FIR mode at 7200sps) (Using SINC1  at 38400sps I can get 3200sps).

There's any way to get a higher data rate (at least nearest to 7200sps)? 

Thanks for the help!

  • 0
    TI__Mastermind 44680 points
    Hi Paulo,

    Welcome to the TI E2E Forums!

    When multiplexing between multiple channels it can be difficult to achieve a high throughput with delta-sigma ADCs. The digital filter is effectively performing a moving average calculation on the conversion results, so it can take multiple conversion periods for the filter to settle (particularly for the SINC2/3/4 filters, which take approximately 2, 3, or 4 conversion cycles).

    In the case of the effective throughput of ~1900 SPS at the 7200 SPS data rate, I don't think you will be able to improve the throughput by much, unless you focus on acquiring data from just a single sensor (no multiplexing)...

    If you take a look at table 17 in the datasheet, this table will tell you the time required between the START1 command and when /DRDY goes low for each filter and data rate. This is the time for the first conversion result. Ongoing conversions will take the expected amount of time (1/data rate). Unfortunately, the 38400 SPS data rate comes out of a SINC5 filter, so even though the data rate is faster, the filter's settling time require approximately 5 conversion cycles. I wouldn't recommend using this data rate if you're looking to achieve low-noise performance. The 38.4 kSPS data rate is MUCH noisier than all the other data rates.

    In general, most weigh-scale applications are concerned more about noise than bandwidth. Is there a particular reason why you would need more bandwidth/throughput than ~2 kSPS?

    Best regards,
    Chris