• State TI Thinks Resolved
  • Locked Locked
  • Replies 7 replies
  • Answers 3 answers
  • Subscribers 47 subscribers
  • Views 728 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.

ADS1158: is AINCOM connected to muxoutn in single ended mode?

Art Mecina
Genius 16655 points
Part Number: ADS1158
Other Parts Discussed in Thread: ADS1258, LM7705, OPA365, OPA2365

 Hi e2e,

Good day.

One of my customer is asking if the AINCOM pin is connected to MUXOUTN in single ended mode for ADS1158 ?

He is planning to use the following set-up:

16 single-ended inputs, 0-5V
AINCOM = 2.5V
He is expecting that the MUXOUTN pin will be 2.5V.
He will use the 2.5V as a common mode input reference for a single ended to differential lamp circuit. 
Appreciate your support on this.

Thanks in advance.
Art
  • 0
    TI__Guru**** 181001 points

    Hello Poster!

    Our support staff is taking time with their families for the Independence Day holiday here in the US.  It may be July 8th before we can get back to you with help on your query.  We apologize in advance for the delay and will get back to you as soon as possible.

  • 0
    TI__Mastermind 44680 points

    Hi Art,

    Yes, AINCOM is connected to MUXOUTN when selecting a single-ended input channel in auto-scan mode (i.e. When auto-scan mode is enabled and any of the AINx bits is set in the MUXSG0 and MUXSG1 registers).

    NOTE: AINCOM is not a selectable input when using fixed-channel mode.

    Please let me know if I can review you customer's schematic. The described configuration sounds okay, but I would need to see a schematic to provide more support or feedback on the implementation.

  • 0 in reply to Christopher Hall
    TI__Genius 16655 points

    Hi Chris,


    Thanks for the support!

    I have relayed the answer to the customer and inform him that the we can do a review of his schematic. 

    He agreed to provide a copy of his ADC portion. See attached.

    Also he wants further explanation of what does it mean by "AINCOM is not a selectable input when using fixed-channel mode" ?

    2063.ADC.pdf
    Thanks in advance.

    Regards,

    Art

  • 0 in reply to Art Mecina
    TI__Mastermind 44680 points

    Hi Art,

    • In auto-scan mode, channel are selected by programming the MUXDIF, MUXSG0, MUXSG1, and SYSRED registers. The MUXSG0 and MUXSG1 registers allow for selection of the "single-ended" channels that connect AINCOM to MUXOUTN.

    • In fixed-channel mode, the MUX is controlled by the MUXSCH register. The upper nibble of this register controls the MUXOUTP channel and the lower nibble controls the MUXOUTN channel (For example 0x01, would configure a differential measurement between AIN0 and AIN1). Due to this size limitation of this register, only inputs pins AIN0 through AIN15 can be connected to the the MUXOUTx pins in fixed-channel mode.

    Most of the time the ADS1258 is used in auto-scan mode, so it is typically not an issue that AINCOM is not selectable in fixed-channel mode. However if single-ended measurements are needed in fixed channel mode, you can select any of the AINx pins to be the common reference signal instead of AINCOM, OR you could consider bypassing the internal MUX and externally providing your 2.048V signal into your single-ended to differential amplifier circuit. ...It looks like you have already provided resistors to allow for the later case.

     

    Schematic Feedback

    • I would recommend routing VREFN of the ADS1158 back to pin 4 of the REF5040, before connecting it to ground. In this way, you keep VREFP and VREFN routed like a differential pair and keep some of the ground noise out of the VREFN signal.

    • It would also be a good idea to add a small differential filter capacitor between VREFP and VREFN, as close to the ADS1158 as you can place it.

    • I would consider using a a crystal oscillator instead of a crystal to drive the ADS1158's clock. The low power crystal driver of the ADS1158 is not compatible with all 32.768 kHz crystals and sometimes it can be a bit of a hassle to tune the crystal circuit to get it to oscillate. If you do use a crystal, I would recommend using a crystal with a low ESR (preferably 50kOhm or lower), as it is typically a bit easier to get crystals with a lower ESR to resonate.

    • The single-ended to differential amplifier circuit will be nonlinear for inputs voltage between 0 and 100mV. The reason for this is that the U117 and  U123 amplifier outputs will start to saturate as their outputs attempt to drive towards the negative supply rail. You might consider adding something like an LM7705 to the negative supply rail of the OPA365's to allow these outputs to drive to 0V.

  • 0 in reply to Christopher Hall
    TI__Genius 16655 points

    Hi Chris,

    Thanks for the excellent support!

    I will inform the customer of the answer and will inform you in case he has some feedback.

    Best regards,

    Art

  • 0 in reply to Art Mecina
    TI__Genius 16655 points

    Hi Chris,

    Good day and I hope you are having a wonderful one.

    The customer has some follow up inquiry in which i hope you could help us.

    Customer mentioned that he will be using the neg bias generator on U117 and U123 as you advise. But he is asking if wouldn't he
    need the neg bias generator on U118 and U124 as well? Because as Vin goes towards 4.096V the signal at the output of U118 and U124 drives towards 0V too.

    Also, he recently reviewed the Rail-to-rail input of OPA365 on section 7.3.1 of the datasheet: http://www.ti.com/lit/ds/symlink/opa365.pdf

    and is wondering now if the negative bias is really needed..?


    Thanks a lot!

    Art

  • 0 in reply to Art Mecina
    TI__Mastermind 44680 points

    Hi Art,

    When I did a quick simulation, I didn't notice U118 and U124 driven to ground since these opamps are biased a bit differently. However, if the output of U118 and U124 need to go to 0V (or close to it), I would definitely add the negative bias generator to them as well. If the customer is considering using the dual OPA2365 then it would be easier to connect V- of both opamps to the negative bias generator.

    Section 3.7.1 of the OPA365 datasheet describes the input range of the opamp... The input-common mode voltage is able to slightly exceed the supply rails; however, the output range is still limited.

    One really key detail to pay attention to which any op-amp datasheet, is the test conditions under which the open-loop gain of the opamp was tested... 

    This input range tells you that the open-loop gain of the op-amp remains high as long as the output voltage remains 100-200 mV away from either supply rail. Therefore, the opamp will behave linearly within this range!

    The "output swing to rail" specification only tells you at what voltage the op-amp output will fully saturate. In the case of the OPA365, it's output can drive to about 10mV from either supply, but between 10mV and 100mV the op amp is partially saturated and you shouldn't expect it to be linear.