Post Go back to editing

Inconsistency Between CN-0422 Design Files and Wiki Documentation (ADP5302 Input Voltage)

Category: Hardware
Product Number: ADN4654

Hello,

In my system, I am using the ADN4654 DVI isolator. When I refer to the CN-0422 reference design, I noticed that the ADP5302 used to supply the ADN4654 VDD2 has a 5V input, while the ADUM5020 in the same circuit is configured for 3.3V operation.

However, when I reviewed the CN-0422 documentation on the Analog Devices Wiki (link), I saw that this voltage is actually shown as 3.3V, not 5V. So, there seems to be an inconsistency between the two versions of the design.

Coming to my main question:
According to the ADN4654 datasheet, the typical IDD2 current is approximately 60mA. Since this design includes two ADN4654 devices, the total current drawn from the 2.5V rail is about 120mA.

When I check Figure 8 in the ADP5302 datasheet, I see that the efficiency is about 93% when the input is 3.3V and 90% when the input is 5V. Given the 120mA load, the current drawn from the ADUM5020 would be around 97mA with a 3.3V input, or 66mA with a 5V input. Since the ADUM5020 can supply a maximum of 100mA, it seems that using a 5V input for the ADP5302 would be a safer and more efficient option.

However, based on the discrepancies mentioned above, it is not clear which input voltage is actually used in the reference design.
Could you please provide clarification or guidance on which voltage level (3.3V or 5V) should be used for this configuration?

Thank you for your support.

  • Hi,

    The schematic confirms the following signal chain: 5V --> ADUM5020 --> 3.3V --> ADP5302 --> 2.5V --> ADN4654

    You can download the schematic from the CN0422 landing page: https://www.analog.com/en/resources/reference-designs/circuits-from-the-lab/cn0422.html

    If you need higher resolution, we have recently released the CN0571 for up to 1080p Full HD Resolution at 60Hz Refresh Rate: https://www.analog.com/en/resources/reference-designs/circuits-from-the-lab/cn0571.html 

    Kind Regards,

    Riccardo

  • Hi Riccardo,

    Yes, I’ve noticed that as well in the schematic. I was referring to the inconsistency between the schematic and the figure.
    My main question is, wouldn’t it make more sense for the ADP5302 input voltage to be 5V based on the current calculation?
    If it’s configured for 3.3V as shown in the schematic, I believe the ADUM5020 might run quite hot, and there could even be a drop in its output voltage.

  • Hi,

    As you noticed, the buck is more efficient at lower input voltages:

    As long as you pick the ADUM5020-3 which has 33% efficiency at 3.3V output voltage, you would get roughly 1 degree less temperature raise because of the higher efficiency of the buck.

    The temperature rise of the components depends on the power dissipated, not on the output current.

    Kind Regards,

    Riccardo

  • Yes, the heat generated is of course proportional to the power dissipated. However, I suspect any observed drop at the ADUM5020 output would be due to the device hitting its current limit rather than purely thermal effects.

    A quick recap of my calculation:

    • Two ADN4654 devices draw approximately 60 mA each from the 2.5 V rail → total I_out = 120 mA.

    • Output power from the 2.5 V rail: P_out = 2.5 V × 0.12 A = 0.30 W.

    • With the ADP5302 configured from a 3.3 V input (≈ 93% efficiency from the datasheet), the input power is P_in = P_out / 0.93 ≈ 0.3226 W, so the input current from the ADUM5020 would be I_in ≈ 0.3226 W / 3.3 V ≈ 97.8 mA.

    • For comparison, with a 5 V input (≈ 90% efficiency), I_in ≈ (0.30 / 0.90) / 5 V ≈ 66.7 mA.

    Since the ADUM5020 is specified for a maximum output current of 100 mA, the ~98 mA figure with a 3.3 V input is essentially at the device’s limit and leaves very little margin. That makes output voltage droop or current limiting highly likely under load (and also leaves little headroom for tolerances or additional leakage). For this reason, using a 5 V input to the ADP5302 appears to be the safer choice.

    Please check my arithmetic and assumptions (IDD2 ≈ 60 mA per ADN4654, ADP5302 efficiency figures from Figure 8, ADUM5020 max output = 100 mA). If any of these numbers are different in your design or datasheets, let me know and I’ll re-run the calculation.

  • Yes, the heat generated is of course proportional to the power dissipated. However, I suspect any observed drop at the ADUM5020 output would be due to the device hitting its current limit rather than purely thermal effects.

    A quick recap of my calculation:

    • Two ADN4654 devices draw approximately 60 mA each from the 2.5 V rail → total I_out = 120 mA.

    • Output power from the 2.5 V rail: P_out = 2.5 V × 0.12 A = 0.30 W.

    • With the ADP5302 configured from a 3.3 V input (≈ 93% efficiency from the datasheet), the input power is P_in = P_out / 0.93 ≈ 0.3226 W, so the input current from the ADUM5020 would be I_in ≈ 0.3226 W / 3.3 V ≈ 97.8 mA.

    • For comparison, with a 5 V input (≈ 90% efficiency), I_in ≈ (0.30 / 0.90) / 5 V ≈ 66.7 mA.

    Since the ADUM5020 is specified for a maximum output current of 100 mA, the ~98 mA figure with a 3.3 V input is essentially at the device’s limit and leaves very little margin. That makes output voltage droop or current limiting highly likely under load (and also leaves little headroom for tolerances or additional leakage). For this reason, using a 5 V input to the ADP5302 appears to be the safer choice.

    Please check my arithmetic and assumptions (IDD2 ≈ 60 mA per ADN4654, ADP5302 efficiency figures from Figure 8, ADUM5020 max output = 100 mA). If any of these numbers are different in your design or datasheets, let me know and I’ll re-run the calculation.

  • Hi,

    I confirm that the maximum output current is 100 mA for the 3.3 V version (ADuM5020-3, which also requires a 3.3 V input voltage).

    If you need additional margin beyond the maximum current limit and prefer using the ADuM5020-5 input voltage version, then the 5 V output is the best option.


    The efficiency of the ADuM5020-5 is approximately 33% for a 5 V output, but it drops to 27% for a 3.3 V output (unless the ADuM5020-3 is used, which maintains 33% efficiency at 3.3 V output).

    Kind Regards,

    Riccardo

  • The ADUM5020-5 device not only provides a fixed 5 V output but can also be configured for a 3.3 V output. In fact, when reviewing the EVAL-CN0422-EBZ design documentation, it can be seen that the ADUM5020-5 is used and its output is set to 3.3 V.

    Could you please calculate the estimated current drawn from the device with the U6 designator in the EVAL-CN0422-EBZ documentation? I believe this will help you better understand the point I’m trying to make.

Before You Switch


Switching languages will make ADI Explorer unavailable. Resume your session by switching back to English and reopening ADI Explorer.