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.

MSP430FR2422: RST/NMI/SBWTDIO - RC recommendation

Part Number: MSP430FR2422

Hello,

After studying the MSP430FR2422 datasheet and the family datasheet I'm still in doubt how to interpret the RC recommendation for the RST/NMI/SBWTDIO pin. It's mandatory for me to fully understand what( and why) the requirements for this pin are, this to save parts/costs and board space. Even if only to save ONE resistor or ONE capacitor.

Conditions:

- SBW programming interface will be used. This to save I/O pins.

- NMI functionality will not be used.

Datasheet RC recommendation for this pin:

- Pull-up resistance 47k

- Pull-down capacitance 10nF.

When using the SBW programming interface the maximum value for this capacitance is: 1.1nF (otherwise programming tools are not able to drive this 10nF).

1) So why is it allowed to use a smaller capacitance when using SBW programming? contradicts with the 10nF recommendation.

2) Is it important that this pin has a specific Reset timing (minimum RC time)? I was not able to find something about this in the datasheet.

3) Or is this RC combination only used to filter EMI on this pin. Meaning that the RC value's are actually more based on: trace length, nearby switching signals, shielding, ground return path etc.

4) It seems beneficial to me to use the internal RST pull-up resistor (costs). Are there any drawbacks when using this resistor. For sure the tolerance of this internal resistor is big compared to an external pull-up resistor.

Thanks for your help,

Patrick

  • It depend how noisy is environment (and signals) related to used device. I never had any problems with (start and) running F5xx/6xx or FRAM devices in room (room temperature, no noisy environment), without any external components connected to RESET pin.
  • Hi Patrick,
    please let me elaborate a bit more in detail on this topic, as I assume you're not asking due to some student activities.

    1. The recommendation of 10nF/47kOhm is a legacy from times, when there was no Spy-Bi-Wire programming, but only JTAG available within the MSP430 portfolio. At that time even BOR logic was not integrated, but only simple POR. Thus the relative long RC time constant was sort of necessary also from this perspective. This recommendation has been kept since then, as it serves also protection against EMI/ESD events.
    For the pure RST functionality, the C is not required at all, with the more modern parts with BOR and integrated pull-up. This applies also for the MSP430FR2422.
    2. There is certainly a limit in respect of detection of a low pulse, but for the programming and HW resets, I would stay with the technically inaccurate statement not having a requirement. If you go with just a pull-up and use the recommended/official tools for programming and usual devices for RST/voltage supervision, if desired, you will not face an issues, with the pulse lengths, these devices apply.
    3. The recommendation to apply an RC component is triggered mainly by requirements on the noise immunity, addressing protection versus undesired reset events, as you indicated. Also as mentioned, the better your layout and thus noise protection, the smaller the trace structures connected to your reset pin the less energy the RST pin will face and the lower the probability will be, facing issues from that area, when going without a C at all. So for programming, the point of electrical contact on the PCB should be as close as possible to the RST pin, to keep the trace and antenna structure for the normal operation as short as possible. If you'd manage contacting the RST pin for programming directly, of course that would be the optimum, using than the integrated pull-up only. This is the shortest antenna...
    On the other hand using an external resistor allows you going to lower impedance, which increases the robustness versus disturbance as well. Here the limitation basically is the limitation of the drive capabilities of the logic, which needs to pull the RST low, either for programming or reset events. Here with Spy-Bi-Wire you need to keep in mind, the RST is bi-directional. That's why the respective programming tool's pin has a serial resistor in the 100Ohms range, to prevent a short.
    4. As indicated above, the main concern using the internal pull-up only, is the resulting noise immunity in the respective application. This means one needs to decide based on the environments the application will be operated in, the noise levels, and the integrity and protection of the design as such, whether lower resistance pull-ups and potentially a C is required.

    I hope these explanations are helpful to make your decision.

    Best regards
    Peter
  • Hello Peter,

    Thank you very much for your answer to the smallest details.

    Best regards,

    Patrick

**Attention** This is a public forum