• State Resolved
  • Locked Locked
  • Replies 11 replies
  • Answers 4 answers
  • Subscribers 46 subscribers
  • Views 403 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.

AMC1210MB-EVM: AMC1210MB-EVM: data analysis from AMC1210 Eval tool

Bhargav Pendurthi
Prodigy 160 points
Part Number: AMC1210MB-EVM
Other Parts Discussed in Thread: AMC1210

Hello,

I'm working on a project with the AMC1210. At the input from the AMC1210, I apply the MDAT signal from the ADuM7701 becuase this is similar to AMC1203 function (MDAT).

The voltage applied is 5V input and Modulator clock signal is 10MHz but I used System Clock 32MHz in Eval board as Output, please see it in below cofig. When I dont apply any shunt input to the device 0mV than I can get approx. 50% of voltage (2,5V) can say Ref point. Now I configure the AMC1210 so that it runs with a sinc1 or sinc3 filter and experimenting with the oversampling ratios of the sinc filter unit.

My Question, Is there any way to calculate back the voltage from the AMC1210 Eval tool measurement output shown below. It is showing 0 to -1 change. and also the provided data is correct or not, please confirm

Regards,

Bhargav Pendurthi

  • 0
    TI__Guru 50151 points

    Hi Bhargav,

    Can you re-post your figures? 

    With the inputs shorted to ground, I would expect a 50/50 distribution of 1s and 0s from the device, which if a 1 is = 5V and a 0 = 0V, 2.5V average makes sense. 

    Perhaps with the figure I will have a better understanding of your question, but the best way to test functionality would be to input a known signal such as a sine wave to the input, then look at the output of the digital filter to see if you have a similar, albeit slightly more digitized, version of the same signal. I cannot recommend manually back calculating the voltage from the 1's and 0's distribution as this would be quite tedious.  

  • 0 in reply to Alexander Smith
    Prodigy 160 points

    Hello Alex,

    Thank You for your Response.

    Please find the figure in below link. My application is DC votlage.

    Please let me know, the file is opening or not?

    /cfs-file/__key/communityserver-discussions-components-files/73/AMC1210-tool-data.docx

    one more question:

     Eval board is with parallel data connection as per schematic, can I use with SPI configuration from AMC1210 by connecting M0 and M1 pin to '0'  and will it work with the tool?

    Thank You

    Regards,

    Bhargav Pendurthi

  • 0 in reply to Alexander Smith
    Prodigy 160 points

    Hello Alex,

    One mor equestion. Is there any way to select SOSR and IOSR value. When I change the IOSR value the code value also changing. If I want to traace back the actual voltage applied, I wont get with the change in value.

    In the Integrator Filter Registers, if I disable the demodulation Enable bit, than I can see the steady state Code value for applied DC volage.

    I really didnt understand how to calcualte back the applied voltage with SOSR and IOSR values.  W.r.t SOSR value, I can calculate Bit Width fir Sin3 filter (Log(X^3,2) and LSB Size (Vref/2^bits-1). is this correct. will it change with IOSR value? with LSB size based on the code value, I can get back applied voltage. is it correct.

    But my question is, is there any way to select IOSR value based on SOSR and how it will be different for 16bit and 32bit format. how to calcualte back the voltage.

    below is the example :

    /cfs-file/__key/communityserver-discussions-components-files/73/AMC1210-tool-data1.docx

    Thank You

    Regards,

    Bhargav Pendurthi

  • 0 in reply to Bhargav Pendurthi
    Prodigy 160 points

    Hello,

    Can any one please support ?

    and also how to calculate or to find out improvement of ENOB from SOSR and IOSR values? 

    Thank You

    Regards,

    Bhargav Pendurthi

  • 0 in reply to Bhargav Pendurthi
    TI__Guru 50151 points

    Hi Bhargav,

    I apologize for the delay in my response. 

    I am able to open the word file. With 0V applied to the input, I would expect to see a 50/50 ratio of 1's and 0's which you can see in the first two plots. The third and fourth plots show the noise that is being converted, as expected when applying 0V to the input. In order to test this properly, and I think it would help you visualize the conversion, try inputting a +/-250mV sine wave instead of 0V. As I said in my previous post, I cannot recommend manually back calculating the voltage from the 1's and 0's distribution as this would be quite tedious.  

    SOSR is the oversampling ratio of the Sinc filter, and IOSR is the oversampling ratio of the Integrator. You will want to scale these values based on your output data rate, and accuracy requirements. This app note covers this in detail, please see section 5: http://www.ti.com/lit/an/sbaa144/sbaa144.pdf

  • 0 in reply to Alexander Smith
    Prodigy 160 points

    Hi Alex,

    Thank You for your respone. As per my understaing if I apply sine wave than the output also show Sine wave output in Code vs sample plot. similar way if I apply DC voltage than it has to show constant DC value in Code vs samples plot. and also I am assuming that there is a way to calculate or represent the data by SW. I hope its all calculation based to represent data in Code vs samples plot.

    Can you please see below file and provide the differences in it.

    /cfs-file/__key/communityserver-discussions-components-files/73/AMC1210-tool-data2.docx

    Thank You

    Regards,

    Bhargav Pendurthi

  • 0 in reply to Alexander Smith
    Prodigy 160 points

    Hello Alex,

    I had gone through the link you have provided.

    I think below are the setting need to be considerd with Sinc3 filter representation for 16bit and 32 bit format.

    SOSR value SH Value Data Format
    32 0 16bit
    64 3
    128 6
    256 9

    SOSR value SH Value Data Format IOSR value
    256 0 32bit 127

    as per the settings, the code value will be shown on tool..I dont know about the other settings and usage of it.

    can you please verify below and let me the max error % at lower input voltages will the device provide?

    /cfs-file/__key/communityserver-discussions-components-files/73/AMC1210-tool-data4.docx

    Thank You

    Regards,

    Bhargav Pendurthi

  • 0 in reply to Bhargav Pendurthi
    TI__Guru 50151 points

    Hi Bhargav,

    At lower input voltages the noise will have a more dominant factor in the signal to noise ratio calculation. Since noise is a random distribution, if you average several measurements you should be able to reduce the error caused by this noise. The max error will be dependent on the particular characteristics of the noise in your system.

  • 0 in reply to Alexander Smith
    Prodigy 160 points

    Hi Alex,

    I was an assumption that Digital filter will clear out the noise and provide the accurate result with integration funtion in picture. Integrator will take more samples from SOSR output and provide the accrate result.

    is my assumption is correct? but do you have any idea about the accuracy range of device with decent input to AMC1210.

    Thank You

    Regards,

    Bhargav Pendurthi

  • 0 in reply to Alexander Smith
    Prodigy 160 points

    Hello Alex,

    Thank You for your support and I understood the reason at 0V input. the output code value is due to noise and if we consider it as an offset value than the accuracy is better at lower currents.

    @Modulator Clock Frequency 16MHz (MClk= Clk/2)
    Actual Shunt Input@50mOhm (V) Calculated Code value Measured Code value using AMC1210 Tool Error%
        Min Max Min Max
    0.000003 0 -65 -45    
    0.006017 616 553 563 0.35% 1.27%
    0.01204 1233 1172 1181 0.36% 0.53%
    0.02028 2077 2016 2024 0.22% 0.35%
    0.05011 5131 5069 5078 0.06% 0.16%
    0.10058 10299 10239 10248 0.05% 0.06%
    0.15055 15416 15356 15363 0.03% 0.05%
    0.20001 20481 20421 20433 0.03% 0.01%
    0.25022 25623 25566 25574 0.03% 0.01%

    Please find the results.

    Thank You

    Regards,

    bhargav P