• State
  • Locked Locked
  • Replies 8 replies
  • Subscribers 64 subscribers
  • Views 1147 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

TPS62122: TIDA-00690: Unable to send BLE broadcast

Rahil Jain
Prodigy 80 points
Part Number: TPS62122
Other Parts Discussed in Thread: CC2650, , TIDA-00690

Hello wonderful people on the TI design forum, 

I followed the Wireless Switch Power Reference Design to create a custom board. I am able to generate regulated 1.785V from the TPS62122 (DCN-8pin package). Without any load, the pulse lasts for about 2 secs. With the BLE CC2650 launchpad connected, I verified that the pulse lasts about 11 ms, which is similar to reported in the reference manual. However, unlike the reference manual, I am not receiving any BLE broadcast using the CC2650 USB dongle and Packet sniffer. The launchpad is flashed with the stack and application for the Simple BLE broadcast example located in: C:\ti\simplelink\ble_sdk_2_02_01_18\examples\cc2650em\simple_broadcaster. I have removed the jumpers on the launchpad to cut power to the debugger chip, but to no avail.

Using an external supply, I verified that my launchpad sends BLE broadcast if the pulse duration is at least 23 ms. Clearly, the issue is that the pulse generated doesn't last long enough. However, in the TIDA-00690, a 10.8 ms pulse was enough to send the BLE message, so I wonder if there is something missing in my setup that's not covered in the reference design. Do I need to disconnect any jumpers on the launchpad to reduce the power lost to any peripherals? Or, is the CC2650 still in debug mode even though the debugger is not turned on, perhaps a setting in CCS prior to flashing. There is something missing and I hope someone who has worked with this demo can help me find it.

I am a newbie on this platform and I apologize for the simple question. Thanks for your help. 

Rahil

  • 0
    Prodigy 80 points
    - I got inspired by your blog post to try this. Hopefully, you have some insights for me. Thanks!
  • 0
    TI__Intellectual 1680 points

    Hello Raphil,

    You need to disconnect the jumpers to power off the LEDs of the launchpad. 

    How often you send a broadcast events? I suggest you to reduce the period you send out broadcast events.

    Thanks

    Giovanni

  • 0 in reply to Giovanni Campanella
    Prodigy 80 points

    Hi Giovanni,

    I appreciate your reply. I made some progress based on your inputs.

    I did disconnect the LEDs but that didn't help.

    I was sending broadcasts events every 100 ms - that's the default. I changed it to 1 sec. But still, I need a pulse > 25 ms to broadcast (see attached picture of my external supply setup).

    Interestingly, I am now able to send BLE broadcasts when powered through the generator. However, the RF packets are only sent when you let the switch release forcefully. When the switch is released forcefully, it generates a regulated 1.785V pulse that lasts > 30 ms - enough to boot up the MCU and send the RF packets. No packets are sent when you normally actuate the switch, which generates a pulse of 10 ms, like in the reference manual. Here's a video demo:

    1drv.ms/.../s!AlcOLOGWkdvggogaDmRMiEcIaES81A

    This makes me wonder that there may something I am missing on the board that may have reduced the clock speed or similar. Any ideas?

    Thanks for all the help.

    Rahil

  • 0 in reply to Rahil Jain
    TI__Intellectual 1680 points
    Hi Rahil,

    This TIdesign has been designed to send a broadcast event every time the button is pushed and soon after released as you shown in the video.

    Let me know if there is anything else I can do for you.

    Best regards
    Giovanni
  • 0 in reply to Giovanni Campanella
    Prodigy 80 points

    Hi Giovanni

    Unfortunately, no broadcast is sent when the button is pushed, but only when the button is let release forcefully. The difference between the two actions is the pulse length of the regulated output. The pulse length is 10 ms for button push and 30 ms for release. The TI design was sending a broadcast with even the 10 ms pulse but I am not able to replicate this. Any thoughts what I might be doing wrong.

    Thanks

    Rahil

  • 0 in reply to Rahil Jain
    Prodigy 80 points

     - not sure if you saw my last response, do you have any further inputs on why I am able to broadcast when the regulated output is at least 25 ms long, which is unlike the TI reference design? Is there any debug or boot settings that i need to change? Thank you so much.

  • 0 in reply to Rahil Jain
    TI__Intellectual 1680 points
    Hi Rahil,

    The second pulse(button release) is longer than the first one(button push) because when the button is released there is still some energy left in the capacitor.
    For this TIdesign a broadcast event is transmitted after the button has been released as you shown in your video.

    Thanks
    Giovanni
  • 0 in reply to Giovanni Campanella
    Prodigy 80 points
    Hi

    Thanks for the reply. It is true only if the time difference between the push and release is short enough. I experimented by letting the capacitors discharge after the push activation. It typically takes under a minute. I was able to send the broadcast when I let the button release forcefully (pulse duration = 30 ms) and not when I naturally release the button (pulse duration = 10-15 ms). I think the biggest differentiator is the force (=flux) generated in the release and not the time from last actuation. If latter was true, we would see a broadcast event from a push actuation if it came right after a button release, right? I have never seen that.

    I am not sure if that's the case. The TI design does not indicate that the broadcast event is transmitted only after the button release. In Table 3 and Fig. 15 in the TI design, you note that the broadcast event takes less than 10 ms. That means, the voltage pulse generated from button push (=10 ms) should be enough to send the broadcast event. I have not been able to repilcate this. In my setup, I need at least 25 ms of regulated supply to transmit broadcast event, which makes me believe that there is something in the firmware that I need to skip to save time. I am able to generate a >25 ms pulse if I let the button release forcefully. A natural release only generates a 10-15 ms pulse depending on the time from last actuation, as you correctly point out.

    Thanks for your help so far. I very much appreciate the support.