• State TI Thinks Resolved
  • Locked Locked
  • Replies 21 replies
  • Answers 7 answers
  • Subscribers 33 subscribers
  • Views 1036 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.

PGA460: PGA460: PGA460 USE Bi-static mode，RX floating，there is noise on RX

user6060721
Expert 1015 points
Part Number: PGA460

if we add parallel resistor to RX transducer，this resistor will change transducer resonant frequency？

Will it lead to inaccurate test results？

How to  optimize the value of this resistor？

Thanks

  • 0
    Expert 1015 points

    add question：

    How to design the value R(RU78) and C(CU82) and optimize the value？

    Thanks

  • 0 in reply to user6060721
    TI__Expert 6935 points

    Hi user6060721,

    If you are seeing the initial burst and ring-decay time on the receiver, you can use a damping resistor on the receiver to help reduce the ring-decay time. Section 3.4 of PGA460 Ultrasonic Module Hardware and Software Optimization app note provides more info on how to choose the ideal resistor for this.

    Regards,

    Mekre

  • 0 in reply to Mekre Mesganaw
    Expert 1015 points

    in this app note,i think the damping resistor is added on the transimitter.

    is it right add on receiver?

    if add on receiver,it will damp receive echo amplitude？

  • 0 in reply to user6060721
    TI__Expert 7599 points

    Hello,

    The app note principle of dampening resistor applies to receiver as well because the dampening resistor on transmit is for monostatic mode and the principle applies similar to bi-static and now the receiver path will need extra components. 

    Adding resistance will impact impedance and its a design trade off based on the system and the component selection will depend on the noise immunity needed, transducer impedance characteristics and receive signal signature. 

    As the ultrasonic solution is greatly dependent on the system outside of PGA460 we have provided the excellent GUI and EVM tools to help you evaluate the system and fine tune the component based on the system needs. 

    So the RU76 needs to be tried and tuned several times to get the desired optimum performance and you also need to understand the transducer impedance model and that can help in the resistor selection. 

    See below information from the app note https://www.ti.com/lit/an/slaa732/slaa732.pdf 

    Let us know if you need any further help here. 

  • 0 in reply to Arjun_Prakash
    Expert 1015 points

    1.I don not understan the meaning of minute-loading effects：

    2.RX signal amplitude is very small on oscilloscope，how can I jude the ring dacay on RX in bi-static mode？

    Thanks

  • 0 in reply to user6060721
    TI__Expert 7599 points

    The minute loading effects are in reference to transducer characteristics. Can you get the transducer model from the vendor ? What is the RT, LT, Ct and CPT ? 

    CTUNE is 3.3nF and Rdamp will be the tuning resistor. 

    The RX signal on the Oscope will be really really small uVs or less and cant be measured and needs a high gain output and the signal needs to be measured at the output of the PGA411. 

  • 0 in reply to Arjun_Prakash
    Expert 1015 points

    3.3nf is not the CTUNE.

    we use bi-static mode.

    Below is our schematic,CU75 is CTUNE,we are tuning this.

    Ru71 is Rdamp.

    I think RU76 is not Rdamp,but you think RU76 is Rdamp.

    I am confused.

    BVD model is below：

  • 0 in reply to user6060721
    TI__Expert 7599 points

    Hello User,

    I understand your confusion here and the CTUNE parameters apply to RX as well in mono static mode primarily because the RX AFE is also connected to the same terminals. For a bi-static mode similar components needs to analyzed and you may have to replicate the Ctune and Rdamp to the RX path as well. 

    There is a reference post that I can point you to so please reference to this e2e post on managing the resistor and guidance on steps to tune the system. 

    The system performance needs to be tweaked and upon several tries with different resistors and PGA460 settings you have a path to an optimal solution. 

  • 0 in reply to Arjun_Prakash
    Expert 1015 points

    when we adjust Ctune,we use the inductance of transformer.

    But there is no transformer on RX when we used bi-static mode.

    How can we adjust the value of Ctune on RX.

    Or we can use the same value as TX?

  • 0 in reply to user6060721
    TI__Expert 7599 points

    Hello User6060721, 

    The adjustment of the components depends on the system noise and the application and the performance you like to achieve. As discussed earlier please refer to this post. 

    For the RX path if there is no transformer please assume the transformer impedance is near zero and continue with the tuning process. If too low of a resistor is placed it will significantly load the RX input and you may not be able to measure any signal. The tuning process really needs to be performed via trial and test method. We dont have any specific tuning components for a specific transducer and have provided the GUI and the software to tune your system.  

  • 0 in reply to Arjun_Prakash
    Expert 1015 points

    Arjun_Prakash：

     I also have three questions：

    1.where does this noise echo come from？How to eliminate the noise echo？

    2.What is the relationship of quantity of pulses and SPL？More pulses lead higher SPL？Is it a completely positive proportion？

    3.When is the start time of RX echo record？from TX pulses start(point1)？ or TX pulses end(pint2)？

  • 0 in reply to user6060721
    TI__Expert 6935 points

    Hi user6060721,

    1. This noise echo may be coming from the transmission. Although a bistatic configuration is being used, you may be able to still sense the transmission bursts. The following post has more info on this topic and how to potentially address it by adjusting the time varying gain: https://e2e.ti.com/support/sensors/f/1023/p/841260/3118206?tisearch=e2e-sitesearch&keymatch=pga460%252520bistatic#3118206  
    2. The more pulses that are sent, the higher the SPL. The higher the SPL, the further you can detect. The disadvantage of using more pulses though is that it could increase the decay time, which may make it more difficult to sense closer objects.
    3. When looking at the graph of the echo data dump, the transmission energy is shown. The following video shows the different regions of an echo data dump (you can skip to around the 6:30 minute mark of the video): Please note though if you use the object detection feature, the returned time measurement for detected objects is with respect to when the burst stage is complete and not from when transmission started. Figure 32 of the PGA460 datasheet shows over what two durations is the time measurement of detected objects taken.

    Regards,

    Mekre

  • 0 in reply to Mekre Mesganaw
    Expert 1015 points

    I can not understand this BVD model.

    1.When the transducer resonant,equivalent LT、CT serial resonant？ When receive Echo，Lsec and Ctune+Cpt is parallel resonant？

    2.Lsec is the secondary inductance as below？3mH？

    3.I think Adjust Ctune is for receive echo，not for transimit，is this right？

    4.We adjust Ctune is for shortest decay time？

    below is our final result adjusted，the amplitude of decay is higher than burst，is this normal？

  • 0 in reply to user6060721
    TI__Expert 6935 points

    Hi user6060721,

    The items within the “Transducer BVD Model” are the components that are actually part of the Butterworth-Van *** model of an ultrasonic transducer. RDAMP is the dampening resistance that is used to reduce the decay time. LSEC is the secondary side leakage inductance that is from the transformer and CTUNE is something we add to compensate for this leakage inductance. CTUNE is most relevant to the transmit transducer since that is the transducer that is connected to the transformer and would therefore see the secondary side leakage inductance. Similar to the dampening resistor, you adjust CTUNE to reduce the ring-decay time.

    Do you have the echo data dump of the results that correspond to the oscilloscope graph?

    Also, just so you know, many of us supporting these forums will be out for the next two weeks for the holidays. Someone will be monitoring this forum during that time, but replies may be delayed.

    Regards,

    Mekre

  • 0 in reply to Mekre Mesganaw
    Expert 1015 points

    I also have the questions:

    1.We adjust Ctune and Rdamp are all for shortest decay time？

    2.How we know  secondary side leakage inductance of the transformer?

    Merry Christmas

  • 0 in reply to Mekre Mesganaw
    Expert 1015 points

    we ncountered consistency of ultrasonic transducer.

    influence short distance detect.

    Transducer 1 :short distance echo is high:

    Transducer 2:short distance echo is low:

    How we cover this difference of Transducers？

    By software？hardware？

    Thanks

  • 0 in reply to user6060721
    TI__Expert 6935 points

    Hi user6060721,

    You want to adjust Ctune and Rdamp to reduce your ring-decay time requirements while also still ensuring that the resulting decrease in echo amplitude from tuning still allows you to meet any requirements you have for the return echo amplitude. This technique doesn’t require knowing the value of the secondary leakage inductance or using an impedance-gain phase analyzer to extract the BVD equivalent component values.

    One way of dealing with the ultrasonic inconsistencies you mentioned is to first use a small time varying gain value during the time period where the initial false peak is observed. This will allow you to better differentiate false peaks from peaks that are actually from object detection, as is referenced in the below post:

    https://e2e.ti.com/support/sensors/f/1023/t/638088?BOOSTXL-PGA460-How-to-decouple-RX-and-TX-signal-path-in-bi-static-mode-

    During calibration, you can then find the sensor-specific threshold levels that can be used for differentiating false peaks from peaks due to object detection.

    Regards,

    Mekre

  • 0 in reply to Mekre Mesganaw
    TI__Expert 6935 points

    Hi user6060721,

    We just released an ultrasonic training series today.  The following video from that training series provides more details on how to adjust Ctune and Rdamp:

    Regards,

    Mekre

  • 0 in reply to Mekre Mesganaw
    Expert 1015 points

    If we use a small time varying gain value during the time period where the initial false peak is observed，then the long distance gain also be decreased.

    Because the gain of pga460 is sectionalized.

    If we we adjust short range gain from 52 to 46，the long range gain decreased from 84 to 78.this will influence capability of distance detect.

    How to solve this contradiction？

  • 0 in reply to user6060721
    TI__Mastermind 18545 points

    Hello, 

    Just wanted to give you a heads up that our ultrasonic experts are out of office this week and will be able to answer your question next week. 

    Best Regards, 

  • 0 in reply to user6060721
    TI__Expert 6935 points

    Hi user6060721,

    In the suggested scheme, you would actually keep the AFE_GAIN_RNG value (this is within the DECPL_TEMP register) constant. The time varying gains will specifically be applied by modifying the TVGAINx registers to apply different gains at different points in time. You can also modify the short range and long range digital gains using the Px_GAIN_CTRL registers.

    As an example, you can select AFE_GAIN_RNG= 11b to select a range between 32 to 64. Then you can modify the TVGAINx registers to ramp the varying gain, such as ramping the gain from 32dB to 62dB across the record length. In addition, you can apply a large digital gain multiplier at a specified time in the record window, such as starting with a short range gain of 2 and then using a long range gain of x16 for improved SNR resolution/scaling. 

    Regards,

    Mekre