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.

MSP430FR2355: SAR ADC Acquisition time

Part Number: MSP430FR2355

Dear *,

can you tell what value is the tacq(Acquisition time) of the SAR ADC?

tacq:

Acquisition time. This is duration that the sample and hold switch is closed. A longer acquisition time makes it easer to settle.   The data sheet provides a minimum acquisition time that corresponds to the maximum throughput (samples / second). 

Is it the tacq= tSettling = 100ns?

Best Regards,
David.

  • Hi David,

    The Sample and Conversion timing diagrams for the integrated 12 Bit ADC can be found in the Family User's Guide:

    https://www.ti.com/lit/ug/slau445i/slau445i.pdf 

    starting in section 21.2.5. There are two different sampling modes(extended and pulse) which have slightly different timing diagrams.  

    BR,
    Leo

  • Dear Leonardo,

    so i'm using 10bit mode and @ 6MHz. And the reference is 3.3V DVDD.

    so tacq minimal is tsettle_max + tsample_min from DS = 0.1us * 0.52us ?

    Best Regards,

    David.

  • David,

    so the 100ns define the time from ADCON until your sampling starts which has to be considered when you start you sample via the corresponding trigger e.g. ADCSC bit. This is something you need to consider in your code structure.

    The other one is the sampling time which is defined by your ADC frequency and sample clock number ADCSHTx bits. However which time you need depends on your external circuit as well as described in the users guide (posted it below) the datasheet numbers consider dedicated values for RS of 1000 Ohm. In your application you either need to implement 1kOhm or less or need to calculate the settling requirements based on your application.




  • Dear Dietmar,

    thank you for fast replay, if i understand correctly the sampling time eg acquisition time will change with respect to Rs and C_ext , is that correct ?

    Also in SLAA89 page 6 the formula can be used to determine the sampling eg acquisition time? 

    The datasheet value of 0.52us is achieved with Rs=1kOhm and Cext= 8pF.

    Because i don't want to load the MSP SAR ADC input i want  to buffer my voltage divider with op-amp, but going through TI material after the buffer i should use a RC charge bucket filter. 

    But if i want fast sampling time then i need fast Opamp is that correct ( picture 1)  ? if i use so that the Rs is 1kOhm as in Ds with opamp than i get only 1.9us sampling time ( picture 2) 

    But if i do calculations with the upper formula for pic1 i get 3.4us and not 520ns as showing in calculator , and for pic 2 i get 4us and not 1.9us as showing in calculator, What am i missing??

    pic1

    pic2

    Best Regards,

    David.

  • David,

    so if I use the values from the datasheet and calculate the minimum sampling time I exactly come to 0.52 us.

    tsample = ln (2 ^ 10+1) * (1000 + 4000) * 13.5exp-12 = 0.5146 us

    I do not now know how your calculations in your tool work and I also have no insight into the OpAmp but obviously the OpAmp influences the sampling time significantly.

    So in pic1 you get the 0.52us but in pic2 you get 1.9us is because you have smaller GAIN bandwidth set up for your OpAmp?

  • Dear Dietmar,

    this is a tool from TI it is called Analog Engineer's calculator from  TI 

    I also get this tsample = ln (2 ^ 10+1) * (1000 + 4000) * 13.5exp-12 = 0.5146 us

    but when i add the opamp i want that tsample stays the same , so per TI calculator  i need to change the Rs from 1000Ohm to 277Ohm and Cext from 8pF to 110pF ( picture 1) , but if i then use the formula from DS then i get 

    tsample = ln (2 ^ 10+1) * (277 + 4000) * 115.5exp-12 = 3.7 us and that is different from what the TI calculator calculates 520ns (picture1) ?

    Best Regards,

    David.

  • David,

    ok but then this is a calculation problem in the calculator because you manual calculation makes sense.

    Can you please post this in the Simulation, hardware & system design tools forum

    Would be ok for you to route this thread to a different forum?

  • Dear Dietmar,

    how can i rout this thread to different forum? Or do you mean i should post a new thread in the other forum?

    Best Regards,

    David.

  • David,

    rerouted to the corresponding forum.

  • David, sorry the reroute function changes the responsibility but will not help to show up the thread in the corresponding forum.

    Can you please generate a new thread in the forum mentioned above

  • Dear *,

    can you help to solve the questions from this thread ?

    Best Regards,

    David.

  • Hello,

    I think this is a misinterpretation of the calculator, and possible some differences between how standard standalone ADCs are designed versus the integrated ones.  

    In the calculator you enter the supply voltage, resolution, input sampling capacitance, and the acquisition time.  The tool then computes a conservative range of values for the external R/C charge-bucket filter and also the minimum suggested op amp bandwidth.

    There will be combinations with lower values of bandwidth, capacitance, etc. than what are shown in the tool that will still work, but the goal of the tool is to provide an optimized conservative estimate of the required external components.  More information can be found in the videos here:

    https://training.ti.com/ti-precision-labs-adcs-introduction-sar-adc-front-end-component-selection?context=1139747-1140267-1128375-1139106-1128643

    https://training.ti.com/ti-precision-labs-adcs-selecting-and-verifying-driver-amplifier?context=1139747-1140267-1128375-1139106-1134078

  • Dear Collin,

    how can we apply this to integrated SAR ADC that has internal Rs= 4kOhm and Csh = 5.5pF ?

    Best Regards,

    David.

  • Hi David,

    The ADCs this tool was meant to work with usually have switch resistances in the <50Ω (usually less) so we neglect it from the calculations.  For something as high as 4k, you'll need to build a model and run simulations in SPICE as recommended in the video series I previously provided, or work with the MSP430 team on what you're trying to do. 

**Attention** This is a public forum