I have two AD9648s that exhibit the same symptoms:
- no DCO output but the clock input is good
- Vref output is ~0.5V instead of 1V
It looks like the part is not starting up or something, but the SPI chip identification reads correctly.
What is the cause of this and how can it be fixed? Any suggestions on what to try?
We are using these parts in our product and so far only 2 of ~15 parts have shown this problem. One of the parts has always had this issue while a second one was working for a few months but now has the same problem. So, I'm wondering what could have happened to cause the second one to change it's operation. I initially thought the first one was just a bad part, but now having two with the same symptoms seems interesting. Especially since the second one didn't have a problem initially.
The power and clock input are within the data sheet specifications. I looked at the power on to the part with an oscilloscope and it shows a nice clean rise time to the final voltage. The 1.8V regulator does put out ~0.4V for a time and then switches to the full voltage. But other boards that work do the same, so I don't think that's the problem. Since the clock is not running when the power is applied, SPI is used to reset the part (see below)
The ADC is running at 1.8V for AVdd and 1.8V for DVdd. The clock is 100MHz single ended with 0.1uF series caps on pins 1 and 2.
The following sequence is used to initialize the ADC via SPI:
DCO delay = 1.12ns
Output Test Off