• TI Thinks Resolved

ADC3424: Test Pattern Confusion

Prodigy 20 points

Replies: 3

Views: 114

Part Number: ADC3424


I am running an ADC3424 at 100 MSPS and am having some difficulty understanding the test patterns. After powering up the ADC, I pulse the RESET pin, enable the test patterns by writing a 1 to bit 1 in register 06h, and then configure various test patterns such as zeros, ones, toggle, deskew, custom, and sine. I am not writing to any other registers except for when I set a custom pattern in registers 0Eh and 0Fh. 

The zeros, ones, toggle, deskew, and custom patterns all read as I would expect. However, I don't quite understand what I'm seeing from the ramp and sine patterns. The ramp pattern repeats each value four times before incrementing.  For example, I might read 0, 0, 0, 0, 1, 1, 1, 1, 2, 2, 2, 2, 3, 3, 3, 3, .... I expected to see 0, 1, 2, 3, 4, 5, 6 based on what I read in the data sheet.

The sine wave pattern gives me a repeating sequence of eight samples, as stated in the data sheet, but the eight values I see do not match those stated in the data sheet. The data sheet reads: "8-point sine-wave: data are a repetitive sequence of the following eight numbers that form a sine-wave: 0, 599, 2048, 3496, 4095, 3496, 2048, and 599." When using the two's complement data format, I see the repeating sequence 0, 978, 1384, 978, 0, -979, -1385, -979. So, I see a sine-wave-like pattern, but with an amplitude different from what is specified in the data sheet.

Do these patterns make any sense? Am I misinterpreting the data sheet, or am I doing something wrong here?

Thanks a lot for reading!

  • Hi Aaron,

    For ramp mode, you should be seeing 0,1,2,3,4,... as you expected.

    Can you please confirm that you are enabling the patterns by writing the following [Address Data]:

    0x06 0x02 Enable Test Pattern

    0x0A 0x40 Enable Ramp CHA

    Are you operating in 2 Wire mode? Can you also please ensure that the DCLK is toggling at the same rate as the data?

    Best Regards,


  • In reply to dBrock:

    Hi Dan,

    Yes, that is how I'm enabling the pattern, and I'm operating in 2-wire mode. I believe that the DCLK is toggling at the same rate as the data, but I'll check again for sure on Monday. The toggle and sine wave patterns do change values on each new sample, and the sine wave pattern has the expected period of eight samples (but, as previously mentioned, the eight values are not the expected ones).  I should also mention that we were able to digitize a sine wave produced by a function generator, and our recorded waveform had the expected period given a 100 MSPS digitization rate.

    Thank you very much for your help,


  • In reply to Aaron Fienberg:

    Hi Aaron,

    Have you been able to check the data on all 4 of the channels when you provide an external sinewave to one of the analog inputs? 

    Can you provide screen shots of your captured data or Firmware signal tap/ probe?

    Best Regards,