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.

LMK04828: ADC14x250EVM

Part Number: LMK04828
Other Parts Discussed in Thread: TEST, ADC14X250EVM

I am working with the ADC14x250 and associated configuration GUI tool (TSW16DX370EVM GUI) to study the configuration, operation and performance of the LMK04828 JESD204B clock conditioner. The GUI allows me to select one of 3 pre-defined configurations to load into the LMK04828. My  question pertains to "configuration 2" which is denoted "DEFAULT" on the soft button appearing in the MACRO CONFIG tab of the LMK04828 pane. I made all hardware modifications required as per the instructions in Users Guide (SLAU625) paragraph 5.1.1)  . I observed (with an o'scope) a 245.76 Mhz signal being present at the ADC regardless of the presence of an external clock connected to J7! (I.e. turning the signal generator on or off had no effect on the clock going to the ADC) I would not expect this to occur, I would expect that, given the description of "configuration 2"  the ADC clock to have some dependency on the external reference, and should go away or otherwise show signs of degradation when the ext ref is altered or removed.  There are no oscillators onboard my EVM, all three Y1,2,and Y3 did not come installed.  Please help me understand why this is occurring, I spent a lot of time using all tools available (including codeloader) but still don't understand why this is happening.

Thank you!

John

  • Hello John

    I need to contact the relevant engineer. Did i understand right that you power up the EVM with LMK04828 input clock not present and you still see correct frequency output clock? This can happen if the device is configured in dual PLL mode with VCXO installed for PLL1.

    Best regards
    Puneet
  • Yes, that is what happens!

    Using Configuration 2 i observed that the signal across R40 (DCLKOUT2) is a free running 245-247 Mhz. Yesterday i confirmed that it is indeed free running by examining the close in phase noise of the signal. It is indicative of a free running oscillator. This signal did respond to changes in the 61.44 Mhz reference on J7. I.e when i added it or removed the reference the DCLKOUT2 signal shifted frequency by a few tens of Khz. Very wired. So i examined the PCB and noticed R30 was open (It is labled DNI on the schematic) With R30 open there is no signal entering OSCIN pin 43. I didn't get to test this yet, probably not until next week as i have to travel today-friday to a vendor. 

  • Hi Puneet,
    I'm still troubleshooting this and have more info. First i confirmed that i am using the latest GUI SW, just downloaded and reinstalled morning from the ADC14x250EVM website. Here's the thing: While operating the GUI and connected to the EVM, I observe that the "Program LMK04828" button in the "INTRO" page has the same effect as the "Configuration 1" button in the "LMK04828" page despite the text in the GUI that says "Program LMK04828" should execute "Configuration 2" Now if i run "Configuration 2" i see a free running signal at/around 247 MHz. I measured the DCLKout2 output on a spectrum analyzer to confirm these observation. This inconsistency makes me suspicious about board mod guidance and other info i need to develop my design. Let me know what you find out, thanks!
  • Hi John,
    Did you check the clock at OSCin pin? You said in your previous post that there is no clock applied!
    Best regards
    Puneet
  • Hi Puneet,

    Yes i did check to see if the reference signal is present at pins 44 and 43. I installed cap C67 and resistor R30 which were required to get the Reference signal to the OSCin.

    Here's what i see on the spectrum analyzer. With the reference signal set to 61.44 Mhz, +12dBm on J7, i manually click "configuration 2 (DEFAULT)" from within the GUI. I probed the pads of R18 with an RF probe connected to a spectrum analyzer and observe a signal at/around 247 Mhz .  Upon pressing the "configuration 2 (DEFAULT)" button i noticed that the signal moves down to ~190Mhz for an instant Then two signals appear at/around 93Mhz and 202 Mhz for a second before dropping away. The 247Mhz signal reappears indefinitely. This occurs REGARDLESS of whether the reference is present or not. Weird, right? 

  • Hi guys, have you had a chance to look into this yet? 

  • Hi John,
    Can you read back registers from LMK04828 and post it here?
    Looks like the PLL is not locked. The clock will be there regardless of the input clock.
    Best regards
    Puneet