• State Resolved
  • Locked Locked
  • Replies 5 replies
  • Subscribers 31 subscribers
  • Views 262 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

DCA1000EVM: Emission field strength in combination with IWR6843ISK-ODS (EMC)

Lars Hornig
Prodigy 90 points
Part Number: DCA1000EVM



Dear experts,

I have a question concerning the combination of the DCA1000EVM evaluation board with the IWR6843-ODS Radar Board an and their electromagnetic radiation.
 
We used both boards for some time and had not issues with collecting data itself. Right now we are measuring the electromagnetic compatibility of both boards in reference to EU law, so that we can use the boards at different locations without putting someone in danger. We used the product norm ETSI EN 301 489-1. There two classes are defined: Class A for industrial applications and class B for residential, business and commercial areas.

The emission field strength of both devices on their own and in an active state are below the maximal class B levels. This means, on their own both could be used almost everywhere. When we combine the DCA1000EVM with the IWR6843-ODS using the 60-pin SAMTEC connector, the emission field strength rises above maximal class B levels with peaks in the frequency area around 800 MHz (+-100 MHz). This happens even when we don't measure anything (no data transfer led blinking on DCA1000EVM) and only when we connect both boards. Because of this, we can not use the devices together in residential, business and commercial areas. We assume that the data transmission through the 60-pin SAMTEC connector leads to the high emission field strength. We tried this using 3 different DCA1000EVMs and 3 different IWR6843-ODS Radars in different combinations but the field strength is still high in the mentioned areas. All measurements were done in a certified (DAkkS) measurement laboratory.

Is this interaction behavior known? And do you know any way to shield this emission radiation off?
The declaration of conformity for both the IWR6843-ODS and the DCA1000EVM do not specify the electromagnetic compatibility class that the devices have. Is there a document which has this information?

Kind regards
Lars

  • 0
    TI__Genius 16630 points

    Hello lars,

    Thank you for reaching out to Texas instruments. Please allow us some time to get back to you on this.

    Thanks,

    -Shareef

  • 0 in reply to E. Shareef
    TI__Mastermind 30775 points

    Hello Lars,

    Can you let us know what the operating state of the devices is when you see this phenomenon? Is the radar device framing and sending out any LVDS/CSI2 data? If you are sending out data over LVDS, there would be significant emissions from the samtec connector and cables.

    Also, the DCA1000EVM and IWR6843-ODS are not designed for a production system, these are just debug tools. Generally, when going to production and trying to pass EMC standards customers will design a board with a shielded enclosure to reduce the radiated emissions. Can you let us know why you are trying to certify the evaluation boards? Are you intending to go to production with the design as is?

    Regards,

    Adrian

  • 0 in reply to aozer
    Prodigy 90 points

    Hello,

    thank you for the fast reply!

    Our goal with the DCA1000EVM and IWR6843-ODS is not to produce any final product or to certify the individual parts. We want to evaluate if we are able to measure small motions with a 60 GHz Radar in an environment susceptible by even very moderate levels of electromagnetic radiation. To test if this works in general, we want to use the TI hardware. Everything after this will be based on our design. But to make a good design, we want to make sure that we know where every noise comes from. We were just not expecting these high spikes in these frequency areas when both boards are combined and assumed a known interaction behavior.

    Both boards were configured via the switches before powering them. The IWR6843-ODS was in functional mode and the DCA1000EVM in software configuration mode. The DCA1000EVM-FPGA was not configured after powering it up nor was a measurement started. I would assume that, because of the IWR6843-ODS functional mode,  LVDS data is send automatically via the samtec connector to the DCA1000EVM even without a started measurement.

    Do you have any recommended shielding for the samtec connector, which would reduce the radiated emissions?

    Kind regards,
    Lars

  • +1 in reply to Lars Hornig
    TI__Mastermind 30775 points

    Hello Lars,

    When we did EU RED certification on the DCA1000EVM we wrapped the radar board and connector in tin foil to reduce the emissions from the radar board and connector. We first place the radar board/connector in an ESD bag before wrapping in foil to prevent the foil from shorting anything on the radar board. With this setup we were able to meet the Class B limits. 

    Whether or not the radar will start sending our raw data in functional mode will depend on your software.  Are you using the OOB/TI metaImage or your own embedded application? If you are using OOB then the radar will not start framing/sending raw data until you send the configuration over the CLI. Can you try changing the LVDS data rate to see if this causes a shift in the emissions?

    The AC wall adapter can also a lot of radiated emissions so if you are not doing this already I would recommend to power the boards using a 5V power bank instead of through a wall adapter.

    Is it also possible to share your emissions plots?

    Regards,

    Adrian

  • 0 in reply to aozer
    Prodigy 90 points

    Hallo Adrian,

    for our measurements we will try your way of of emission suppression to get a better feeling for the device. Thank you!
    Unfortunately I cannot share the emissions plots.

    For the software on the Radar we used the  "Hard Coded Config for OOB Demo", which could mean that there should be no data coming from the Radar. Changing the LVDS data rate could be a good way to make sure if that's the case. Because of internal project reasons, this will take a while. Your comment answered by immediate question and I will reference this thread if anything comes up again.

    Thank you and have nice weekend!
    Lars