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.

ADS5560: Clock input signal integrity and maximum ratings

Part Number: ADS5560


Tool/software:

Hi,

I am working with the ADS5560 and I have concerns about the input clock signal when operating in single-ended mode with AC coupling (CLKM grounded through a 0.1 µF capacitor).

Issue:

  • According to the datasheet, the absolute minimum voltage for CLKP is -0.3V.

  • Since my clock signal is AC-coupled, it is centered around 0V, meaning that it inevitably crosses the -0.3V limit.

  • I simulated the received signal using HyperLynx, and it looks like this:

            As you can see, it crosses the -0.3V limit.

Analysis:

I assume that the -0.3V limit accounts for the internal 5kΩ pull-up resistor to VCM = 1.5V. So, I repeated the simulation, now including the effect of this pull-up resistor:

As expected, the signal is now centered around 1.5V, but it still violates the -0.3V limit. To keep the signal within the specified range, I found that I need a minimum series resistor of 250Ω:

However, adding such a large series resistor degrades the slew rate, which in turn increases jitter, potentially affecting ADC performance.

Questions:

  1. How strict is the -0.3V limit on CLKP? Given that AC coupling inherently results in voltage excursions beyond this limit, is this specification meant to be an absolute restriction?

  2. Is there a recommended way to AC-couple the clock signal while staying within the voltage limits without degrading slew rate?

  3. Regarding Figure 53 of the datasheet, I tested connecting and disconnecting VCM to CLKM as suggested, but I noticed little to no difference (since CLKM already has an internal 5kΩ pull-up to VCM). Is there any practical benefit in connecting VCM externally to CLKM?

Any insights would be greatly appreciated.

Thanks!

  • Hi Ivan,

    For #1&3, Please follow the datasheet recommendations, otherwise, we do not guarantee the device.

    For #2, use the HSMS-2812 (or similar) to clip the clock signal coming into the ADC, this will actually increase the slew rate.

    Regards,

    Rob

  • Hi Rob,

    I understand that following the datasheet recommendations is necessary for guaranteed operation. However, my concern is that the datasheet does not clearly explain how to properly AC-couple the clock signal while keeping it within the specified voltage limits.

    • Should the internal 5kΩ pull-up resistor be taken into account in simulations?

    • What is the intended benefit of externally connecting VCM to CLKM, given that CLKM already has an internal pull-up to VCM?

    I’d appreciate any additional clarification.

    Regards,
    Iván

  • Hi Ivan,

    See my comments:

    the datasheet does not clearly explain how to properly AC-couple the clock signal while keeping it within the specified voltage limits. RR: I don't understand this statement, it clearly does on page 31, figure 53. As you plan to drive the clock single-ended as stated above.

    • Should the internal 5kΩ pull-up resistor be taken into account in simulations? RR: yes, if you plan to simulate this interface.

    • What is the intended benefit of externally connecting VCM to CLKM, given that CLKM already has an internal pull-up to VCM? RR: this is to maintain the common mode of the CLKM pin. Even though it seems to be stable by not connecting it.

    In general, I would not recommend driving any HS ADC single-ended. This decreases the swing and integrity of the incoming sampling clock by only using one input leg. SE CMOS also has poor slew and more jitter.

    What are you planning to use for a sampling clock?

    This is a 16bit ADC, so ensuring the best clocking approach with low phase noise available should be a top requirement.

    Regards,

    Rob

  • Hi Rob,

    Thank you for your response.

    Now that you’ve mentioned I need to consider the 5k ohm internal resistor in my simulation, it’s clear to me.

    Regarding the sampling clock driver, I will be using an FPGA.

    Could you review my simulations and let me know if you spot any issues or potential errors?

    CLKP simulation on die: fast (red), typical (yellow), and slow (green) modes.

    Differential CLKP-CLKM simulation on die, fast (red), typical (yellow) and slow (green):

    Regards,

    Iván

  • HI Ivan,

    Looks fine to me. Keep in mind the FPGA will lower the quality of the ADC's performance vs. the datasheet specs.

    Regards,

    Rob