• TI Thinks Resolved

AWR1243: TX-RX antenna isolation

Intellectual 570 points

Replies: 10

Views: 256

Part Number: AWR1243

Hello,

My Customer designed a cascade radar using Preproduction AWR1243P. They tested that the TX-RX antenna isolation was 15~30dB, and they learned in the following link that the isolation should be greater than 25dB:
e2e.ti.com/.../652520
They designed three editions, and all the measures were very strict, they thought their boards were all right.They wanted to know why the TX-RX antenna isolation could not reach 25dB, Is it due to the chip?

Looking forward to your reply.

Thanks~

  • Hi,

    We are checking with the hardware team and will get back to you later this week

    thank you

    Cesar

  • In reply to Cesar:

    Hi Cesar,

    Is there any update?

    thank you

  • In reply to user5144327:

    Hi Cesar,

    I am waiting for your reply. Thank you!

  • In reply to user5144327:

    Hi 

    Is this measured isolation in your customer's system causing any specific problems? Have they been able to test the antenna pattern isolation in isolation? For example, on an isolated antenna test PCB?

    Thanks, 

    -Randy

  • In reply to Randy Rosales:

    Hi Randy,

    This problem results in poor imaging focusing effect and close moving target detection distance.Even with equipment, it is very difficult to measure transceiver isolation on a single board.Now my customer has made a two-piece cascade board, but there is still the problem of isolation.

    They want to know when the  AWR1243 cascade radar (TIDEP-01012) can be purchased. They want to use it for comparison.

    Thank you!

  • In reply to user5144327:

    Hi 

    Have a few follow-up questions: 

    • Can you please describe the method that they are using for measuring TX to RX isolation in their system?
    • Have they developed a separate antenna-only PCB for antenna parameter measurements or are they trying to do this all in a single PCB + AWRx system?
    • What transmission line type was used for the RF routing?
    • What antenna design type is used?

    Thank you,

    -Randy

  • In reply to Randy Rosales:

    Hi 

    The Cascade EVM will be on ti.com very soon. Any follow-up messages so far? 

    Thank you,

    -Randy

  • In reply to Randy Rosales:

    Hi Randy,

    My customer reply is as follows:

    1. We observed the coupling problem between TXs in the chip, by cutting the traces to a transmitting antenna, and measuring the residue power from the receiving antennas.
    2. We do all the measure on the AWR1243+antenna board.
    3. We use grounded coplanar wave guide (CPWG) microstrip lines.
    4. The antenna is similar to the following pic.

    Do you have the exact date for the cascade EVM availability for the purchase?

    Thank you !

  • In reply to user5144327:

    Hi 

    Thank for the additional responses. The Cascade EVM will be launched on ti.com within a month's time. I do not have an exact date, but it will be very soon. 

    Let me see if I understand the TX/RX coupling test procedure. Please correct me if I am not understanding anything here: 

    - Start with custom AWR1243 PCB with embedded antenna structures. Antenna fan-out from AWR1243 BGA done with GCPW lines

    - a TX GCPW line is then cut at some point along the fan-out

    - AWR Chirp is performed through TX channel with cut

    - Resulting RX channel power adjacent to this cut TX is then measured with AWR IF/ADC path

    Follow-up questions/comments: 

    -  Any measurement would still include the cut GCPW radiation (unknown gain) and RX antenna gain. It would also include the (hopefully smaller) contribution from the GCPW lines themselves. I don't see how that can isolate the AWR TX/RX package/die coupling. 

    - For this type of measurement, what I would recommend is something closer to our own procedure. In our case, we measure these kind of effects with a PCB design we call a "probe card". This probe board is effectively our AWRx Boosterpack design, but we replace the antenna elements with VNA microprobe launches. We also include GCPW calibration structures on the PCB to calibrate out the launches as well. This allows us to take calibrated, multi-port S-parameter measurements of each of the TX and RX paths with the antenna completely out of the picture. The 25dB typical TX/RX coupling measurement was quote from such a system. 

    Thank you,

    -Randy

  • In reply to Randy Rosales:

    Hi Randy,

    Thank you for your patient answers and comments, but my customer said that they have done all you mentioned, so it is not helpful.

    Thank you all the same!