• State TI Thinks Resolved
  • Locked Locked
  • Replies 9 replies
  • Answers 3 answers
  • Subscribers 47 subscribers
  • Views 1514 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.

Original question:

ADS1256 strange problems

ADS1256: ADS1256 strange problems --> DRDY remains high for longer time after issuing the SYNC command

Ahmad Suliman
Intellectual 320 points
Part Number: ADS1256
Other Parts Discussed in Thread: ADS124S08, ADS1258, ADS131E08

I am using the ADC with the "Settling Time Using the Input Multiplexer" to read from all the channels one after the other. This has to repeat after every 8ms.

According to Figure 19 and table 14 of the ADS1256 datasheet, t_19 should last for up to a couple of hundred microseconds. But this does not necessarily mean that t_18 should also be in that range. So I would like to know what is a healthy range of t_18. For my application, as I am reading all 8 channels so I am following Figure 19 but my t_18 timing is in the range of more than 100 microseconds which is too long I guess. Is there a way to shorten the t_18 so I can reduce the overall conversion time of 8 channels?

Thanks

  • 0
    TI__Guru* 92715 points
    Ahmad,



    The time for t_19 is the time required to set up the multiplexer and then take the conversion.

    I believe that t_18 is the time it takes for a conversion to complete after indicating a SYNC to start the conversion. Times for t_18 are given in Table 13 in the datasheet and aren't changeable given the specific master clock frequency and the data rate.

    I'm not sure of your exact needs, but if you need to shorten the overall conversion time for the ADC, you'll likely need to choose a faster data rate. Another possibility would be to speed up the master clock, but there may be performance degradation with that.



    Joseph Wu
  • 0 in reply to Joseph Wu
    Intellectual 320 points

    Joseph,

    Thank you for this information. Pardon me if it is not a valid question but are there alternative 8 channel 24-bid ADC with same footprint to ADS1256 that will have shorter settling time? I am already running it with its recommended clock i.e. 7.68MHz.

    Thanks,

    Ahmad

  • 0 in reply to Ahmad Suliman
    TI__Mastermind 44550 points

    Hi Ahmad,

    How fast of a data rate are you trying to achieve while multiplexing?

    We don't have anything with the same footprint and pinout as the ADS1256; however, if you need an ADC with a faster settling filter then the ADS124S08 might be a potential option. It has a low latency filter that will settle faster than the ADS1256; however, its data rates don't go as high as the ADS1256 either, so that would be a trade-off.

    For even faster channel cycling and data rates, the ADS1258 is probably the next option I would recommend. It can multiplex at about 23 kSPS per channel (so with 8 channels you'd sample each one at about 2.8 kSPS).

    Best regards,
    Chris

  • 0 in reply to Christopher Hall
    Intellectual 320 points

    Hi Chris!

    Thanks for the message.

    My actual sample rate is 250 Hz. But that is how often I really read a channel when it is the time. And then when it is time to sample I should sample all channels such that sampling time for each will be as short as if one would sample all at the same time. I have set the ADC sampling setting to 30kHz which gives me the shortest t_18 but it is still above 200 uSec.

    The ADS1258 is good but it is in VQFN package is there an alternative package for that? And is it used the same way ADS1256 is used?

    Thanks,

    Ahmad Suliman

  • 0 in reply to Ahmad Suliman
    TI__Mastermind 44550 points

    Hi Ahmad,

    It should be possible to read from all 8 channels within a 250 Hz (4 ms) period; however at the 30kSPS data rate, each channel's data will be sampled one-by-one with about a 0.21 ms period between each conversion result.

    With the ADS1258, the time between conversions can be as short as 42.125 us. Is that okay, or might you need to consider using a device with 8x simultaneously sampling ADCs, to sample each channel concurrently?

    The ADS1258 is similar in functionality but SPI communication is a bit different in the following ways:

    • The auto-scan mode allows you to define which channels to enable in the channel conversion sequencing, so you wouldn't need to program the input multiplexer between channel conversion. However, you would still need to read the data after each channel conversion.
    • SPI commands have different values and are formatted differently.

    • To read data at the faster data rate you may need to use a faster SCLK and make sure your process has the bandwidth to service a more frequency /DRDY interrupt.

    Another important difference is that the ADS1258 does not have a built in PGA, so you would have to implement a discrete buffer or gain stage externally.

    Best regards,
    Chris

  • 0 in reply to Christopher Hall
    Intellectual 320 points

    Hi Chris,

    Thanks for the information.

    I would very much prefer the simultaneous options and especially if it would come with more than 8 channels (right now I am using 2xADS1256 and that is why 4ms is not enough plus first and last channels are sampled with a significant delay during a sampling instant).

    Also, I wouldn't worry about the built-in PGAs as I am anyway putting a gain of 1 on them with ADS1256 which means I am not using it I believe.

    The other thing that I would prefer to have is a non qfn package as I am hand soldering but it won't be a deal breaker.

    If concurrent 16 channel is not possible, then I can use 2 adcs with 8 concurrently read channels and would be still much better than multiplexing.

    I can see 1258 is a 16 channel but still non-concurrent so that will be my last option but another drawback on that is the qfn package.

    Thanks,

    Ahmad

  • 0 in reply to Ahmad Suliman
    TI__Mastermind 44550 points
    Hi Ahmad,

    In that case would something like the ADS131E08 work for you?

    I think this would be our lowest cost option that would meet your requirements. It is an 8-channel simultaneously sampling ADC and it comes in a leaded TQFP.

    Best regards,
    Chris
  • 0 in reply to Christopher Hall
    Intellectual 320 points
    Thank you Chris,
    So nothing that would match ADS1256 pinout and package?
  • 0 in reply to Ahmad Suliman
    TI__Mastermind 44550 points

    Hi Ahmad,

    You can search for parts that have the same package as the ADS1256 here: http://www.ti.com/packaging/docs/searchproductbypackage.tsp?orderablePartNumber=&packageDesignator=DB&pinCount=28&results=results#resultspage

    However, I'm not aware of any any devices in this list that match your application's requirements, and most likely the pinouts of these devices will be different from the ADS1256. If there are pin-to-pin functionally equivalent devices you will usually see links to these devices near the top of the product pages on TI.com. It is more common to see pin-to-pin functionally equivalent device with op-amps and LDOs and much less common with ADCs which tend to be much more diverse in function.

    Best regards,
    Chris