• State Resolved
  • Locked Locked
  • Replies 15 replies
  • Answers 1 answer
  • Subscribers 31 subscribers
  • Views 3504 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.

AWR1243: DCA1000 and 1243 hardware trigger: STS_NO_LVDS_DATA

Yuji Sugimoto
Intellectual 280 points
Part Number: AWR1243

We are currently using a DCA1000 and AWR1243 Boost in mmWave Studio. The goal is to be able to capture data on the AWR1243 using hardware triggers from an external source.
The procedures we followed are:
1) Set as Software Trigger and then hit DCA arm and trigger frame.
2) Set the trigger select to Hardware Trigger and hit trigger frame
3) After a second we armed the DCA1000 and activated our external trigger.
We have gotten this to work using TSW1400, but when using the same procedures on the DCA 1000 we keep getting the message: STS_NO_LVDS_DATA Async event recieved(0)
The AWR1243 is in Dev Mode.

Thanks!

YS

  • 0
    TI__Guru 58035 points

    Hello,
    The DCA1000 has a 30sec timeout after the DCA arm. So it expects the sensor starts sending the LVDS data out within the 30sec time. If its larger than that it would time out.

    Regards,
    Vivek

  • 0 in reply to Vivek Dham
    Intellectual 280 points

    Hi Vivek,
    Thanks for the reply.
    1) I followed your previous post to try to set up the hardware config. When doing this, how do you specify how the trigger will come in. I could not find anything in the DCA1000 manual about setting it as hardware trigger using the switches.
    2) This may be because of the above, but I keep getting a red light on LVDS_PATH_ERR. Wireshark doesn't show any data packets from the DCA 1000 to the computer.
    3) I don't think the timeout is the issue. We are making sure the push the buttons in quick succesion, 

  • 0 in reply to Yuji Sugimoto
    TI__Guru 58035 points
    Hello Yuji,
    The red LED on LVDS_PATH_ERR indicates that the fpga did not receive any data over LVDS. This could be due to the time out.
    Just to clarify, it works all fine if you use "software triggered " mode in the frame config , is that right? You have the issue only in HW triggered mode?


    Regards,
    Vivek
  • 0 in reply to Vivek Dham
    Intellectual 280 points

    Yes.

    We have had no problems in software trigger mode, but can not get hardware triggers to work with the DCA1000. 

    The light turning on is definitely inevitably due to a time out, but I am trying to send in the triggers to the AWR1243 via SYNC_IN pin before the light turns on to no effect. 

    Just to make sure SYNC_IN is pin 10 on JTag 6 and I have made sure to populate R165. The trigger I am using has worked previously when using the TSW1400. 

    Thanks,

    YS

  • 0 in reply to Yuji Sugimoto
    TI__Guru 58035 points
    Hello YS,
    SYNC IN signal is on pin 9 of J6 connector, can you double check if you are feeding the pulse on the right pin?
    There should be no difference in the operation between DCA1000 and TSW, the hardware trigger and SW trigger is on the AWR device and should not relate to the capture device used.

    Regards,
    Vivek
  • 0 in reply to Vivek Dham
    TI__Guru 58035 points
    Hello YS,
    Can you confirm after you have started providing the SYNC IN pulses you are then doing the "DCA1000 ARM" and "trigger frame" in the mmwave studio? What is the periodicity of the SYNC IN pulses you are giving and what is the programmed frame periodicity in the frame config?

    Regards,
    Vivek
  • 0 in reply to Vivek Dham
    Intellectual 280 points

    HI Vivek,

    I've tried both pin 9 and 10. The AWR1243 booster documentation says that SYNC_IN is pin 10, but the AWR1243 schematic sheet says pin 9. 

    I have tried providing the SYNC_IN pulses while/before doing the ARM and trigger frames.

    The characteristics of the SYNC_IN pulse is 3.4V, 30ms period, 11us width. 

    I have trued frame periodicity in the frame config of 30ms and 40ms

    I've also been trying to measure the SYNC_OUT pin to see if it matches the SYNC_IN. I have been unable to see anything on the SYNC_OUT pin even though I have soldered the R108. Is there anything I'm forgetting?

    Thanks!

    YS

  • 0 in reply to Yuji Sugimoto
    TI__Guru 58035 points
    Hello YS,
    The SYNC OUT would not be enabled for a single chip device, thats the reason you dont see it.

    In the configuration you have provided the programmed frame periodicity cannot be greater than the SYNC IN pulse periodicity, hence a 40 msec frame periodicity and 30msec SYNC IN pulse periodicity is not valid.
    Can you increase the pulse periodicity to 35msec or so and keep the frame periodicity to 30msec?

    Regards,
    Vivek
  • 0 in reply to Vivek Dham
    Intellectual 280 points

    Hi Vivek,

    We have tried changing the frame periodicity to 30msec and keeping the pulse periodicity at 35msec to no avail. 

    YS

  • 0 in reply to Yuji Sugimoto
    TI__Guru 58035 points
    Hello YS,
    Can you confirm if with the same hardware (same EVM) , same mmwave studio , same firmware and settings you are able to get the HW triggered working using TSW1400?


    Regards,
    Vivek
  • 0 in reply to Vivek Dham
    Intellectual 280 points

    Yes this hardware works with the TSW with the exact same settings. 

    We are able to put the AWR1243 in Dev Mode. Open up mmwave studio, use the steps outlined on my first post and have it capture data using hardware triggers. 

    For some reason, the DCA1000 does not allow me to do this. 

    Please advise what to maybe try.

    Yuji 

  • 0 in reply to Yuji Sugimoto
    TI__Guru 58035 points
    Hello Yuji,
    The only issue I can see is if you the delay between "DCA1000 ARM " to "Frame trigger" + SYNC IN pulse is larger than 30sec. Could you confirm if you are making sure these are done within the required time period?

    regards,
    vivek
  • 0 in reply to Vivek Dham
    Intellectual 280 points

    We are definitely pushing the buttons within a 30 second time frame.

    Yuji

  • 0 in reply to Yuji Sugimoto
    Intellectual 280 points
    Hi Vivek,
    We have been able to fix this issue by getting rid of R120 on the DCA1000. This resistor connects the SYNC_IN pulse to one of the DCA's FPGA's.
    We were seeing that the SYNC_IN signal that we were feeding in had a pk to pk of 3.3V usually, but when the DCA was connected it went down to 500mV. Something about connecting it to the DCA was greatly attenuating the signal. When the resistor was taken away the signal once again returned to 3.3V pk to pk and the triggers started to work.

    YS
  • 0 in reply to Yuji Sugimoto
    TI__Guru 58035 points

    Hello YS,

    Thanks a lot for the update. I did double check and found that the FPGA was driving this pin since its also used for another interface. 

    Your fix seems to be the right thing to do. Other option is to remove the SYNC IN connection to the 60pin HD connector on the AWR EVM itself.

    Regards,

    Vivek