Because of the Thanksgiving holiday in the U.S., TI E2E™ design support forum responses may be delayed from November 25 through December 2. Thank you for your patience.

  • State TI Thinks Resolved
  • Replies 8 replies
  • Answers 1 answer
  • Subscribers 16 subscribers
  • Views 137 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

RF430FRL152HEVM: ADC Issues and Firmware Sampling Error

Aditya Rao
Prodigy 10 points
Part Number: RF430FRL152HEVM
Other Parts Discussed in Thread: MSP-EXP430G2ET, DLP-7970ABP, , MSP-FET

Tool/software:

Hi everyone,

My lab partner and I are working on the RF430FRL152HEVM evaluation board and are encountering some issues. Initially, we tried connecting the board to the GUI using the MSP-EXP430G2ET + DLP-7970ABP setup, but we couldn’t get it to connect.

I wrote an iOS app to read and write data to the board and power the sensor chain. The app is getting data from the light sensor (ADC0), and we can see it changing to 0 when we cover the sensor. However, the voltage readings don’t match what we measure on the board with a multimeter. The multimeter shows 0.65V for ADC0, but the raw ADC values are 0x3FFF and 0x2080 which convert to 0.45V and 0.228V using a gain of 2. Adjusting the sensor configuration gain (tried 1 and 2) didn’t change the raw ADC values, though gain is part of the conversion formula. I have tested my app, and I'm sure the issue isn't on my end.

Setup for blocks 0–8:

Block 0: 0x0100040208010000
Block 1: 0x0000000000000000
Block 2: 0x0909090900000000
Block 3: 0x000000000000FFFF
Block 4: 0x0000FFFF0000FFFF
Block 5: 0x0000FFFF0000FFFF
Block 6: 0x0000FFFF0000FFFF
Block 7: 0x00000000000000FF
Block 8: 0xF80100000000A6A3

Results:

ADC0 enabled
Block 9: FF3F802080208020
Block 10: 8020802080208020

ADC1 enabled
Block 9: 0000000000000000
Block 10: 0000000000000000

ADC2 enabled
Block 9: 0000000000000000
Block 10: 0000000000000000



I read in the manual that the firmware might need updating. I tried loading all three firmware versions from slac691.zip onto the board, but I’m seeing the same results with each one. After this when I start sampling, I get a sampling error indicated by the status register (Block 0: 0x000B040004000000). Sampled data in blocks 9 and above read as 0xFFFFFFFFFFFFFFFF.

Any help would be appreciated.

Thanks,

Aditya

  • 0
    TI__Guru**** 175326 points
    Aditya Rao said:
    My lab partner and I are working on the RF430FRL152HEVM evaluation board and are encountering some issues. Initially, we tried connecting the board to the GUI using the MSP-EXP430G2ET + DLP-7970ABP setup, but we couldn’t get it to connect.

    What do you mean it will not connect?

    Can you show a screenshot on how you use the GUI to connect between the MSP-EXP430G2ET + DLP-7970ABP and RF430FRL152HEVM?

    I will suggest you get the reader using MSP-EXP430G2ET + DLP-7970ABP to work so you can compare with your phone reader. 

    Aditya Rao said:
    I wrote an iOS app to read and write data to the board and power the sensor chain. The app is getting data from the light sensor (ADC0), and we can see it changing to 0 when we cover the sensor. However, the voltage readings don’t match what we measure on the board with a multimeter. The multimeter shows 0.65V for ADC0, but the raw ADC values are 0x3FFF and 0x2080 which convert to 0.45V and 0.228V using a gain of 2. Adjusting the sensor configuration gain (tried 1 and 2) didn’t change the raw ADC values, though gain is part of the conversion formula. I have tested my app, and I'm sure the issue isn't on my end.

    Refer to Frequently asked questions for RF430FRL15xH devices (Rev. C) section 2.8 on how to determine the voltage from ADC value. 

  • 0 in reply to Charles Tsai
    Prodigy 10 points

    Hi Charles,

    I'm not in the lab right now, so I can’t get a screenshot of the GUI. I can send that once I go to the lab later today. I do have a picture of the launchpad and eval board setup.



    When I tried connecting to the GUI, I made sure we were using the correct COM port. However, when I hit the connect button the ISO15693 light turned off and the D3 LED changed from purple to red. I had to power cycle the launchpad to get it to detect the eval board again.

    As for the voltages, I double checked the calculations but still getting the same values.

    0x3FFF = 16383
    (16383 / (2^14 - 1)) * (0.9 / 2) = 0.45V

    0x2080 = 8320
    (8320 / (2^14 - 1)) * (0.9 / 2) = 0.229V

    Also, do you have any ideas on the firmware? Before updating I could record samples from ADC0, and ADC1 and ADC2 would show 0s. Now with the new firmware I’m getting sampling error instead. I’ve tried all the versions from the slac691.zip package but it's the same with each of them.

  • 0 in reply to Aditya Rao
    TI__Guru**** 175326 points
    Aditya Rao said:
    When I tried connecting to the GUI, I made sure we were using the correct COM port. However, when I hit the connect button the ISO15693 light turned off and the D3 LED changed from purple to red. I had to power cycle the launchpad to get it to detect the eval board again.

    On the J101 jumpers, did you select HW UART? You will need to put the jumpers for the RXD and TXD in their horizontal positions. 

    Aditya Rao said:

    As for the voltages, I double checked the calculations but still getting the same values.

    0x3FFF = 16383
    (16383 / (2^14 - 1)) * (0.9 / 2) = 0.45V

    How can you get 3FFF? I tried to shine a flash light directly on the light sensor I can never get more than 0x2786. It seems like you are not reading the sensor. 

    Aditya Rao said:
    0x2080 = 8320
    (8320 / (2^14 - 1)) * (0.9 / 2) = 0.229V

    Although 0x2080 looks more reasonable, how did you measure the input voltage with a multimeter? 

  • 0 in reply to Charles Tsai
    Prodigy 10 points
    Charles Tsai said:

    On the J101 jumpers, did you select HW UART? You will need to put the jumpers for the RXD and TXD in their horizontal positions. 

    I already had the jumpers set horizontally

    Charles Tsai said:
    How can you get 3FFF? I tried to shine a flash light directly on the light sensor I can never get more than 0x2786. It seems like you are not reading the sensor. 

    3FFF is what I'm seeing as the first sample in block 9. Here's the data again for reference

    Block 9: FF3F802080208020
    Block 10: 8020802080208020

    Aditya Rao said:
    I read in the manual that the firmware might need updating. I tried loading all three firmware versions from slac691.zip onto the board, but I’m seeing the same results with each one. After this when I start sampling, I get a sampling error indicated by the status register (Block 0: 0x000B040004000000). Sampled data in blocks 9 and above read as 0xFFFFFFFFFFFFFFFF.

    Also wanted to highlight the firmware problem again. After loading the new firmware I'm not able to get the data from ADC0. I have the MSP-FET connected to the computer with USB and connected to the eval board with JTAG. The eval board is not plugged in and the switch is set to supply. On CCS I imported the default firmware folder and hit debug. Let me know if there's anything off in this setup.

    Appreciate the help

    Aditya

  • 0 in reply to Charles Tsai
    Prodigy 10 points
    Charles Tsai said:
    how did you measure the input voltage with a multimeter? 

    I measured the ADC0 voltage across these two points

  • 0 in reply to Aditya Rao
    TI__Guru**** 175326 points
    Aditya Rao said:

    I already had the jumpers set horizontally

    ok.

    Aditya Rao said:

    3FFF is what I'm seeing as the first sample in block 9. Here's the data again for reference

    Block 9: FF3F802080208020
    Block 10: 8020802080208020

    I don't think the first sample is correct.

    Have you tried to increase the decimation ratio for a more accurate sample? It seems like in your below configuration for block 2, you have the decimation set to 64 as the default value. Why don't you try higher decimation ratio for more accurate samples. 

    Block 2: 0x0909090900000000

    Aditya Rao said:
    The eval board is not plugged in and the switch is set to supply.

    What do you mean the eval board is not plugged in? 

    Aditya Rao said:
    Also wanted to highlight the firmware problem again. After loading the new firmware I'm not able to get the data from ADC0.

    Do you see the firmware successfully loaded by CCS? Do you see any error message?

    Aditya Rao said:
    On CCS I imported the default firmware folder and hit debug.

    Do you encounter any errors after hitting debug? 

    Aditya Rao said:

    I measured the ADC0 voltage across these two points

    Can you please show me in which document that you can measure ADC0 using these two test points? I didn't find it in the EVM user's guide. 

  • 0 in reply to Charles Tsai
    Prodigy 10 points
    Charles Tsai said:
    What do you mean the eval board is not plugged in? 

    Eval board not connected to the computer with usb.

    Charles Tsai said:

    Do you see the firmware successfully loaded by CCS? Do you see any error message?

    In CCS it says the firmware loaded successfully, but it also says "no source available for reset_isr()". I searched on the forums and that seems to be fine https://e2e.ti.com/support/wireless-connectivity/other-wireless-group/other-wireless/f/other-wireless-technologies-forum/457248/rf430frl152hevm-default-firmware

    Charles Tsai said:

    Do you encounter any errors after hitting debug? 

    No errors when hitting debug

    Charles Tsai said:
    Can you please show me in which document that you can measure ADC0 using these two test points? I didn't find it in the EVM user's guide. 

    Page 30 in the EVM manual

  • 0 in reply to Aditya Rao
    TI__Guru**** 175326 points
    Aditya Rao said:

    Page 30 in the EVM manual

    Thanks for pointing this out which I missed. 

    I kind of think what might have happened is that by default the virtual ground is used which raised the ground by 0.4V. When you measure the voltage between the two test points, you are referencing the true ground. If you look at the measurements you got which is 0.228V and if you add 0.4V to it, it is about 0.63V. This is kind of close to 0.65V you measured at the test points. You also use the default decimation ratio of 64 and this gives about 9 bit of accuracy. Can you try a higher decimation ratio like 1024 or 2048 at the expense of slower conversion time. Another experiment is to not use virtual ground and see if my theory holds. I'm also learning this device as I speak.