• State
  • Locked Locked
  • Replies 5 replies
  • Subscribers 46 subscribers
  • Views 215 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.

AFE5816: Analog input to AFE doesn't work

Koo SeongWoo
Prodigy 120 points
Part Number: AFE5816

Tool/software:

Hi, everyone

Now i'm trying to test output of the AFE5816.

This is my register setting sequence for low-power mode operting.

01h / 0014h

41h / 8000h

42h / 8000h

41h / 0000h

42h / 0000h

03h / 2000h

04h / 0001h

00h / 0010h (for DTGC mode)

b6h / 8820h

b5h / 00c0h

Additionaly, my adc_clk has 20MHz frequency.

Case 1) Test pattern output

When i write 02h / 0080h for  testing and measuring DOUTN pin, then result is like the figure below 

output is same with desired one.

But, when i try to test with arbitary sine waveform generated by function generator, there are no output signals..

I made 1MHz frequency, 400mVpp amplide, No offset value sine waveform. And this is waveform measured at INP pin.

When i measure DOUTN pin with this sine waveform input, there are no output signal but only noise signal (very low amplitude compared to test pattern output) that have frequency same with fclk like the figure below

Is there any problem with my register setting? or is there anything I can test to deal with this problems?

Additionaly, if there are no input to AFE5816 device, then DOUTN pin always have '1' value?

Thank you.

  • 0
    Prodigy 120 points

    Additionally, This is AFE part circuit. Is there any problem with circuit??

    Do i have to make INP pin floating??

    Thank you

  • 0
    TI__Expert 6195 points

    Hi,

    In case of test pattern ( half frame one and half frame zero) you are seeing the doutn pattern at 20MHz which is as expected . This ensures dout pin is working properly .

    Now when you are giving analog input 

    This dout waveform is digital waveform with N bits ( Dn,Dn-1,....D1,D0) . You might not see any pattern . 

     What is the input signal swing ? What is the gain setting you are programming ? With these two you can estimate the expected output swing . This is sampled by ADC and giving you the digitized output . If the output swing is greater than ADC range you might see constant code in dout . You can vary the gain/attenuator setting to limit the output swing less than 2vpp. 

    Since this is digitized output why are you not using fpga to analyze the data ?

    If there is no input to device only noise will be sampled by ADC inside AFE and you should be LSBs of DOUT varying in the captured data on fpga .

  • 0 in reply to Sachin H K
    Prodigy 120 points

    Actually, there is a mistake in my question.

    My main question now is INM pin floating.

    In datasheet, INM has to be floating or connect to GND with capacitors.

    But my circuit INM is now connect to GND like you can see in figure above question.

    My input signal swing is 400mVpp swing and gain setting is programmable fixed gain & 40dB gain.

    My FPGA ILA also doesn't capture digital output, too.

    Thank you.

  • 0 in reply to Koo SeongWoo
    TI__Expert 6195 points

    In datasheet it is shown that Inm has internal 10nf cap which is connected to ground . External capacitor option is given to reduce the HPF corner . When you connect it to ground there will be some parasitic cap which comes in parallel with this 10nF cap . So HPF corner wont change much.

    You have 400mVpp swing and gain is 40dB(ie gain of 100) . This will definitely saturate the output . Can you reduce the gain so that output will be less than 2V and try ? 

  • 0 in reply to Sachin H K
    Prodigy 120 points

    Thank you, I'll try things you advice me and share the results.