• State
  • Locked Locked
  • Replies 4 replies
  • Subscribers 83 subscribers
  • Views 935 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

TMS320F28069 /XRS behavior and Vddio decoupling caps

Greg Berling
Prodigy 20 points

Other Parts Discussed in Thread: TMS320F28069

Hi,

I am planning on using your TMS320F28069 controller on my upcoming design and I have a couple questions about this part that I am having trouble finding in the datasheets and application notes. My first question is how to size the decoupling capacitors for the Vddio pins (I am planning on using the internal regulator on this chip for the vdd pins if that makes a difference). I was also curious if I could get a little better explanation on the behavior of the /XRS pin. In the datasheet it specifies that an R-C circuit should be connected to this pin for noise immunity, but I have had toruble finding any notes on how to size this filter.  I was also curious about the POR and BOR function on this chip. On the previous design I have a reset chip that is 2.93V, open collector, and active low connected to this pin and I plan to do the same thing on this one. I was wondering how the reset chip would react with the POR and BOR and if I use a 10K pull up resistor, could this result in too much current flowing through  this pin?

Thanks

  • 0
    TI__Guru 60865 points

    Hey Greg,

    For decoupling capacitors, I would recommend using a 2.2uF (0805 or smaller) ceramic capacitor per supply pin, both for VDDIO and VDD.  This setup is used on all the ControlCard development boards and the C2000 launchpad successfully.  

    If you are particularly concerned with EMI for your application, you could also add a 0.1uF cap in parallel with the 2.2uF (as well as making sure that you have good power and ground planes), but this shouldn't be necessary for reliable operation.

    The active low open collector supervisor is what you are going to want on XRSn on the device;  The POR and BOR are active low open collector through XRSn.  I would caution you that 2.93V is a little below the recommended minimum VDDIO voltage level of 2.97V.  

    I am not sure how to quantify the filter/pull-up design requirements, I will try to look into it some more and get back to you. We do have a TI internal test board that has a 1.5k pull-up plus a 2.2uF cap to ground which has seen plenty of use without issue.     

  • 0 in reply to Devin Cottier
    Prodigy 20 points

    Hi Devin,

    Thanks for the response. I was looking at the datasheet for this DSP and it says the Vddio BOR trip point is typically 2.78V (page 61 in the piccolo microcontroller datasheet) so I figured sizing the supervisor to be higher than this would be sufficient. So my question is, if the minimum recommended Vddio supply voltage is 2.97V and the BOR trip point is 2.78V, is the BOR really a reliable reset for the chip?

    Thanks again

  • 0 in reply to Devin Cottier
    Intellectual 830 points

    Hi Devin,

    I am also have the same concerns about XRSn pin. However, I am checking with concerto control card design, it is used 0.1 uF instead of 2.2 uF which you mentioned. However, I am not sure which value is the appropriate one. 

    I will appreciate if you clarify it. Because based on threads I followed in E2E and other forums it is mentioned an slight deviation in the correct design of Reset pins, the entire operation results in incorrect behavior. 

    Thanks in advance,

    Masoud.

  • 0 in reply to Masoud Zamani
    TI__Guru 60865 points

    Masoud,

    I agree that it is important to get the reset pin design correct. This is what I have been able to determine:

    On all devices, a strong pull-up resistor is required on XRSn.  There is a weak pull-up inside the device, so the pin could theoretically be left floating, but a strong pull-up will give better noise immunity.  Use something in the range of a couple k-ohms. 

    On older devices without a POR circuit, we required that the R-C on the XRSn pin be sized such that reset was held low for some time after the power and clock were good. This generally resulted in a somewhat large value for C (R should always be in the few k-ohm range).  On newer device with a POR (Concerto and Piccolo series) this requirement has been obviated by the POR circuit, which will hold the reset low for at least 400us after the supply voltage reaches the POR threshold.  It may still be necessary to put a circuit with a larger RC time constant on the XRSn pin if something else in the system has a longer power-up time (say a canned oscillator that supplies the device clock).  

    Because Concerto and Piccolo devices have the POR, you can safely omit the C on the XRSn pin, this is the case for Piccolo ControlCards, which have a 2.2k pull-up only. You can also add a small capacitor in an attempt to suppress HF noise, this is the case for the Concerto ControlCards, which have a 2.2k pull-up and a 0.1uF capacitor.  You can always increase the size of the capacitor, although this may increase start-up time (it could also reduce ability to suppress HF noise if larger capacitors use larger packages or different materials with higher ESR).

    Greg,

    The POR/BOR should be a reliable reset for the chip coming out of reset assuming that the power supply doesn't ramp too slowly.  Very regrettably, I am not sure how good the BOR circuit is for guaranteeing operation of the device if the supply rail temporarily dips too low after power-up (a true brown out).  Overall, if you have a safety critical system, I don't think the BOR can guarantee that reset will be asserted under all conditions outside of the recommended operating supply range, so you may need an external supply voltage supervisor circuit.