• State Resolved
  • Locked Locked
  • Replies 6 replies
  • Subscribers 52 subscribers
  • Views 1965 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.

CC2564MODN: TX frequency shift by -1 MHz for LE and Classic

Martin 52
Prodigy 160 points
Part Number: CC2564MODN
Other Parts Discussed in Thread: CC2640, CC2564MODA

Hi all

We have a CC2564MODN module connected to a STM32F412 CPU. The BT module is connected to the following PCB BT antenna: link.

We are faced with an issue, that the transmit frequency is shifted by -1MHz for each channel.

For example, when selecting the BT LE Transmit Test Mode (as described in the TI wiki) we see the following center frequencies:

HCI_BLE_Transmitter_Test 0x201E, TX_Channel , Data_Length, Payload_Type
  • TX_Channel: 0 -> 2401 MHz, instead of 2402 MHz
  • TX_Channel: 19 -> 2439 MHz, instead of 2440 MHz
  • TX_Channel: 39 -> 2479 MHz, instead of 2480 MHz

TX Frequency: 2402 + 2*k, where k is the channel (max val is 39)

We get 2402+2*k-1MHz.

What could be the reason for this shift?

SP used: v1.6 (link)

HCI Version Information:

16:06:29.341 --
16:06:29.341 Packet "HCI_Command_Complete_Read_Local_Version_Information_Event", Opcode 0x1001
16:06:29.341 Parameters:
16:06:29.341 | Number HCI commands : 0x01
16:06:29.341 | Command Opcode : 0x1001
16:06:29.341 | Status : 0x00 (Success)
16:06:29.341 | HCI Version : 0x06 (Bluetooth HCI Specification 4.0)
16:06:29.341 | HCI Revision : 0x0000
16:06:29.341 | LMP Version : 0x06 (Bluetooth Core Specification 4.0)
16:06:29.341 | Manufacturer Name : 0x000d (Texas Instruments Inc.)
16:06:29.341 | LMP Subversion : 0x1b90
16:06:29.341 --

HCI_VS_Read_Patch_version

16:20:36.567 Packet "HCI_Command_Complete_VS_Read_Patch_Version_Event", Opcode 0xff22
16:20:36.567 Parameters:
16:20:36.567 | Number HCI commands : 0x01
16:20:36.567 | Command Opcode : 0xff22
16:20:36.567 | Status : 0x00 (Success)
16:20:36.567 | Enabled Mask : 0x2bac00000000
16:20:36.567 | ReleaseMajor : 0
16:20:36.567 | ReleaseMinor : 0
16:20:36.567 | PackageID : 0
16:20:36.567 | Build Number : 16

A similar issue has been reported here, but for a different device CC2640. This uses an external 24MHz crystal.

Best regards,
Martin

  • 0
    TI__Expert 6335 points
    Martin,

    We have assigned your post to the relevant engineer and will get back to you.
  • 0
    TI__Mastermind 29250 points
    Hi Martin,

    The CC2564MODN has the fast clock (38.4 MHz XTAL) included in the module. So it is not likely that the clock is the problem.

    Please address the following questions to identify if it is a measurement error.
    1. Is this issue only observed on one board?
    2. The span of the spectrum analyzer seems to be pretty wide in the attached screenshot. Please set the span to ~2MHz with center frequency set to the channel frequency (2402+2*k) to see if you still see the same behavior.
    3. Is there any predefined offset on the spectrum analyzer? Or is the spectrum analyzer out of calibration?
    4. If you have another spectrum analyzer, do you see the same behavior on the second equipment as well?

    Best regards,
    Vihang
  • 0 in reply to Vihang Parmar
    Prodigy 160 points

    Hi Vihang

    Thank you for your help! I will certainly raise the questions on Monday. However, I do not believe that the spectrum analyzer is the issue. The screenshot is coming from a certification lab for European regulation approvals.

    I did request to verify the BT RF FCC Modes, Continuous TX as described in the TI wiki - same result.

    They did execute the BT RF SIG Mode. With the result, that the Bluetooth Tester wasn't able to configure the BT module. They think, it's due to the frequency shift.

    I did request to test another unit - waiting on the result.

    We do use the HCI Tester, as described in the TI Wiki in the CC256x_Testing_Guide, with the following script. Anything wrong with it?

    4606.HCI_BLE_Testing.txt 
    # BLE Testing
# -----------

# 1. Initialize CC2564 device
# Load Core BT Service Pack (BT SP)to the CC2564 device
# ==> Automatically done during initialization of the device

# 2. Enable BLE mode 
# ==> BLE is enabled during initialization of the device
Send_HCI_VS_LE_Enable 0xFD5B, 1, 1
Wait_HCI_Command_Complete_VS_LE_Enable_Event 5000, any, HCI_VS_LE_Enable, 0x00

# 3. Set the BLE Test Parameters 
# Note: Power Level 1 is used for BLE. For more information on this command, refer to the CC256x_VS_HCI_Commands wiki 

# Command Parameter  	Size (bytes) 	Value 		Description 
# -----------------		------------	-----		-----------
# Packets to transmit  	2   			0x0000  	Unlimited - Continuous TX 
#										N			N number of packets to transmit
# Access code  			4  				0xXXXXXXXX  An access code to sync and transmit 
# 													Default = 0x71764129 (TestMode AC) 
Send_HCI_VS_Set_LE_Test_Mode_Parameters 0xFD77, 0x01, 0x0, 0, 0x71764129, 0, 0, 0, 0, 0, 0
Wait_HCI_Command_Complete_VS_Set_LE_Test_Mode_Parameters_Event 5000, any, HCI_VS_Set_LE_Test_Mode_Parameters, 0x00

# 4. Start TX test

# Command Parameter	Size (bytes) 	Value			Description 
# -----------------	------------	-----			-----------
# TX_Channel		1				0 - 39  		TX		Frequency: 2402 + 2*k, 
#													where k is the channel (max val is 39) 
# Data_Length  		1				0 - 37 (0x025)	Packet payload length up to 37 (0x25) bytes 
# Payload_Type  	1  				0 - 7  			0 - PRBS9 
#													1 - FOFO 
#													2 - ZOZO 
#													3 - PRBS15 
#													4 - All Ones 
#													5 - All Zeros 
#													6 - OFOF 
#													7 - OZOZ  
Send_HCI_BLE_Transmitter_Test 0x00, 0x00, 0
Wait_HCI_Command_Complete_BLE_Transmitter_Test_Event 5000, any, HCI_BLE_Transmitter_Test, 0x00

# 5. Take measurements
# Look at the Bluetooth tester results.

# 6. End BLE TX Test
Send_HCI_BLE_Test_End 0x201F
Wait_HCI_Command_Complete_BLE_Test_End_Event 5000, any, HCI_BLE_Test_End, 0x00, 0

# Note: Note: To re-start transmission, go back to steps 3 (optional) and 4

    Is there any other option, special HCI command or something else, what we can try to gain more information?

    General BT Classic (SPP) connection to Windows, Android and iOS is no problem at all.

    Best regards,
    Martin

  • 0 in reply to Martin 52
    Prodigy 30 points

    Hi

    I'm in Martin's Team and I'm also working on the "frequency shift" issue. I noticed, that there is a "crystal mode" option when generating a Service Pack file. Do I have to enable this option? Does it have an impact on the frequency? TI Wiki says, that this option is for 26MHz support.

    Best regards

    Thomas

  • 0 in reply to Thomas Liechti
    Prodigy 160 points

    Hi Vihang

    The latest results from this day:

    • Verified that the SP has no impact. Tested with changed "crystal mode", with the result that the BT stack does not start anymore. Verified with old SP, same results.
    • The spectrum analyzer is not the issue, no offset, not out of calibration.
    • Additional plots of the spectrum analyzer.
      • Channel 0:
      • And channel 19

    Open tests:

    • Verify with an eval board and CC2564MODA. If this works, the module itself should not be the problem.
    • Verify with a different BT antenna. If this works, the BT antenna is the issue.

    BT antenna: Any comments?

    Best regards,
    Martin

  • 0 in reply to Martin 52
    Prodigy 160 points

    Hi all

    This issue is resolved. Fortunately, it was not a HW related problem. The SP was not properly loaded, because it was assumed, that this is done by the application running on the STM32. After adding the SP loading to the test scripts as instructed here the frequency shift is not visible anymore.

    Best regards and many thanks for your support!

    Martin