Smart Bluetooth Forum Discussions

Has anyone got this working yet and gave give me some pointers. I get the windows can not load the VM error message.

Many thanks, in advance.

Hi,

I'm following the "Application Note BCM920732HW" and using its function to use SPI in Master mode and I'm unable to do for BCM20737S. I am using following PINs:

P3 as CLK

P4 as MOSI

P25 as MISO

and

P32 as Chip Select

(Just for information I'm using I2C as well)

What I find in  in spiffydriver.h is that combination of line I am using is not mentioned in header instead following is given.

MASTER2_P03_CLK_P04_MOSI_P01_MISO = 0x00030401,

MASTER2_P03_CLK_P04_MOSI_P05_MISO = 0x00030405,

How I can modify this header?  for my application.

Thanks in advance.

Kind Regards,

Ghalib 

Hello everybody,

I'm still working on the BCM20736S and its connection using the P_UART to a master MCU.

At the moment I'm trying to fix a bug that prevents the P_UART from sending and receiving data as soon as tracing is disabled. The issue is located inside this function:

static inline void enableRxInterrupt(BOOL enable)
{
#ifdef BT_BLE_TRACING
    if (enable == TRUE)
    {
        ble_trace0("enableRxInterrupt(TRUE)");
    }
    else
    {
        ble_trace0("enableRxInterrupt(FALSE)");
    }
#endif
    if (enable == TRUE)
    {
        // set watermark to 1 byte - will interrupt on every byte received.
        //bleprofile_PUARTSetRxWaterlevel(waterMark);
        //P_UART_WATER_MARK_RX_LEVEL(waterMark);
        // enable UART interrupt in the Main Interrupt Controller and RX Almost Full in the UART
        // Interrupt Controller
        P_UART_INT_ENABLE |= P_UART_ISR_RX_AFF_MASK;
    }
    else
    {
        P_UART_INT_CLEAR(P_UART_ISR_RX_AFF_MASK);
    }
}

This is called frequently while handling P_UART data. If I undefine BT_BLE_TRACING no data is sent by the BLE module.

I'm guessing this is a timing problem?

Or are the P_UART_INT... macros (originally only for internal use) having side effects?

It does also work with disabled traces inside the function enableRxInterrupt if I send traces before calling it.

Thanks for your help!

Best regards

Hannes Baumgart

Hi,

  I have some issues while programming BCM20736 Bluetooth Chip through Chipload.exe in Command prompt. Please see the attached picture for more reference.

While I sent the programming command through CMD, Command Prompt responds "Runtime R6002 Error" . We have tried same command with WinXP computer and it is working fine. So it seems my programming command is right. So I suspect some problem is in Chipload.exe which is not supporting for Windows 7. Can some one help me to solve this issue? Thanks in advance.

Regards,

Palanivel

For programming BCM20736 module, I'm following this link " http://community.broadcom.com/community/wiced-smart/wiced-smart-forums/blog/2014/04/23/factory-programming-example-programming-the-20732s-module-using-2073xtag-board

During FCC testing, it was discovered that there are wideband emissions around channel 36 (2402 MHz) for the 2.2.2 SDK build.

The waveform vacillates between the wideband (~4Mhz) and normal bandwidth (~2MHz).

It appears that the order of the test being sent to the module using the Bluetool has an impact on the modulation issue.

1) If you do Set_tx_carrier_freq_arm ​in Bluetool (ch0, prsrb9, LE mod, 4dB power) without doing another test first, the wideband waveform is not seen:

a.      For V2.2.2 you don’t get obvious shoulders (See blue line ignore green):

b.      For V2.2.0 you get shoulders but the higher frequency shoulder is not as sharp.

2) If  you do LE_Transmitter_test only , you get:

a.      V2.2.2: Shoulders are present:

b.      V2.2.0: Shoulders are present, but high frequency shoulder is not as sharp

3) If you first do LE_Transmitter_Test then LE_Test_end then do Set_tx_carrier_freq_arm:

a.      V2.2.2: No shoulder waveform is seen, but in about 1 minute, the very wide band waveform appears (green) and this vacillates back and forth 

b.      V2.2.0: Shoulder waveform is seen, but in about 1 minute, the  very wide band waveform appears

4) Doing a wide sweep of the advertising channels we see:

a.      V2.2.2 - Way wider Ch37 than Ch38 and Ch 39 (disregard Wifi signal)

b.      V2.2.0 – Looks like the correct BW for all 3 channels

Where can I go to find a list of the changes made and issues resolved between different revisions of the SDK?

Hello, is it possible to send advertising packets on one fixed channel?   (instead of three channels)

Also, is it possible for the master node to scan only one fixed channel whenever it scans?

Thank you.