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.

Power consumption and other question about the ADS1248

Other Parts Discussed in Thread: ADS1248, ADS1278, ADS1274, ADS131E04

We are considering the ADS1248 as a top candidate for our application.

1) Before our final selection for 24bit ADC, we need to know the power requirement in detail, since which is the critical factor in our application. The datasheet shows the power dissipation of the ADS1248 at 20SPS is 2.3 mV (with 5V supply). Is this for single channel? or with all channels activated? Do you have more information about the power dissipation for higher SPS? --> the power dissipation is an important factor for wireless application.

2) Basically, our application is the structural vibration measurement. Expected sensing frequency range is up to 1Khz, and we are going to connect an 3-axes analog accelerometer to this ADC. From the title of the datasheet, "ADS1248: 24bit ADC for Temperature sensors", I have little concern about the application. It should not be only for temperature sensing. Do you see any problem in applying it to our case for dynamic structural vibration measurement with accelerometer? 

  • Hongki,


    Thanks for your post, I'm glad your considering the ADS1248. I do have some responses for you, but read through it all and you can post back with some follow-up if necessary.

    1. The ADS1248 is multiple channels, but the input is a multiplexer. In this way, you only are running one channel at a time. It does not measure multiple channels simultaneously. This will definitely affect your data throughput. Therefore the power dissipation is as listed when running the device regardless of how many channels are used.

    The current consumption versus data rate is described in detail in the Typical Characteristics section of the datasheet on pages 18 and 19. Look at Figures 27 to 32 and you should be able to get the information you need.

    2. The ADS1248 was designed for temperature measurement but it doens't mean that it won't work for your application. The device has extra features such as input voltage biasing for thermocouples or current DACs for ratiometric measurements used with RTDs.

    However this still might not be the best ADC for your application. The one thing that I would point out would be that the frequency response of the ADC measurement might be a problem for you. Look at pages 26 to 28 describing the digital filter of the ADC. Measurements of the ADC will have a frequency response as the measurement is filtered. As an example, using the ADC at a data rate of 2000 SPS, you will have a -3 dB bandwidth of 1465Hz as show in Table 9. If you were to measure a 1465Hz signal, the measurement would be attenuated by -3dB and the frequency response would be as shown in Figure 64.

    If this frequency response is a problem for you, the ADS1278 might work better for you as an ADC. It has been used previously for vibration measurement and has a frequency response that is better for the usable range. It has more samples per second, with a wide flatband frequency response. However, the big downside is that the current consumption is much much larger than that of the ADS1248.

    If you need to reconsider your ADC selection, post back and let me know. Maybe we can help you find a more suitable device.


    Joseph Wu
  • Thanks very much for such nice information, Joseph. 

    Imutaneous measurements of multiple channels is very important for our appication. Actually, we have used ADS1274 (4ch version) before, but we did not need such high sampling rate (and corresponding power consumption). That's why we considered the ADS1248 (lower power and lower sampling rate, and better ENOB). But it seems like ADS1248 may not be for our application. 

    So... would you recommend ADCs for our application? Some requirements are..

    1) As long as the bandwidth is higher than about 1kHz, then sufficient. 

    2) Simultaneous measurement of multiple channels (at least 4 channels) --> critical

    3) If possible, lower power consumption --> for wireless application

    4) Better ENOB (at least 18bit, or more at the best condition)

    5) We will measure the vibration responses of structure, such as acceleration, strain, displacements. 

  • Hi Hongki,

    our best option would probably be ADS131E04 in this case.
    However in contrast to the ADS1274, the ADS131E04 offers a SINC3 type digital filter response. That means the digital filter does not offer a flat passband region as in the ADS1274. This is usually not desirable for vibration measurements where an AC signal has to be measured. But you could in theory compensate the droop of the digital filter within your MCU.

    Regards,

  • ADS131 looks good to us. By the way, the power consumption shown in the page 5 of the datasheet is for single channel? or all channels activated?

  • The power consumption specified on p.5 under 'ADS131E04, normal mode' is valid when all 4 channels are activated.

    Regards,