Post Go back to editing

AD9680 PN Sequence Issues

Thread Summary

The user was troubleshooting issues with the AD9680 ADC PN23 and PN9 test modes, observing correct JESD Interface PN sequences but unexpected ADC PN sequences. The solution was to recognize that the DDCs output 16-bit data, causing the 24-bit PN sequence to be split and repeated, leading to multiple occurrences of the same 16-bit word. The user successfully verified the sequence by triggering on two consecutive 16-bit words. In 14-bit mode, the PN sequences are correct in bits 15:2, as per the datasheet.
AI Generated Content
Category: Hardware
Product Number: AD9680

Hello,

I have the AD9680 setup in the following configuration: 

I am able to run and successfully verify all the ADC Test Patterns, except the PN23 and PN9 test modes from the ADC. I am even noticing correct JESD Interface PN23 and PN9 sequences at my ILA in the FPGA, but for some reason I am not seeing the expected sequences for the ADC PN test modes. I think I am not understanding something correctly...

Here is a snapshot of the ILA, where I am only looking at Converter 0 (m0) samples:

I triggered on 0x125B in the above image for PN9, but then I don't see the next value of 0x3C9A anywhere close. is the PN sequence being copied and sent out on each Lane and sample like below:

Or Should I expect it to be sequential like this:

Or something else? I am confused though, because I am able to trigger on other sequence values, but they don't look close to one another... 

Thanks

  • So, im noticing that once I put the ADC in 14-bit mode, which is the pure resolution of the ADC, I get the right sequences in bits 15:2, as noted in the datasheet. But with my configuration shown above its actually 16-bits since the DDCs output 16-bits. In this case, does the PN sequences change? Should I be using a 16-bit output for the PN9 and 23 sequences? Which also changes how many times to shift the PN9? 

  • I figured it out. While using the DDCs, the output is 16bits and the PN sequences use a 16bit output. I was having issues figuring out why the PN23 sequence still wasnt looking right...but it was because I was triggering on one of the first words in the sequence, but since we only see 16-bits of the 24-bit sequence, that 16-bit word I was triggering on actually shows up multiple times and was in a random part of the sequence, not the start like I expected. 

    Once I triggered on 2 back to back 16-bit words that I expected, the rest of the sequence looked right.