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.

PGA117: PGA Needed

Part Number: PGA117
Other Parts Discussed in Thread: PGA460, PGA308

Tool/software:

I filtered the list on the TI webstire for PGAs & VGAs and it seemed like the PGA117 has many of these features such as hardware/software shutdown, daisy chaining, and cost it cannot autonomously utilize a lookup table or internal set of control registers to adjust gain over time without continuous microcontroller (MCU) intervention so I could use more advice if this is the best product fit. It would be nice to have a custom gain profile saved to the PGA or even a linear/curved gain profile saved on board which is not a requirement but would be a nice to have. Unforunately a AGC will not work for my application, it must be TVG. I'm looking for something that has similar functionality or better to the TVG in the PGA460 but I don't want to use the PGA460 for this application.

Core Requirements:

  • Programmable TVG Curve: Ideally, the TVG should have control registers or another programmable method to define the gain line/curve, allowing it to operate autonomously after an initial configuration. If there's another way to accomplish this with external circuity so the MCU can sleep that's an option too.
  • Minimal MCU Involvement: The MCU should only need to set up the gain profile once, or update it periodically, rather than actively controlling the gain in real-time.
  • Cost Target: Preferably a max $10-$15 range. I looked at AFEs but they are significantly more expensive but that's an option too
  • Gain Range: TBD, but similar applications use 32 dB to 90 dB. If a single TVG stage doesn’t provide sufficient gain, I’m open to cascading two stages.

Application Details:

  • Input Signal: Single-ended input from a Piezo transducer used for water sensing (can make it differential if needed)
  • Output: Single-ended preferred (but differential is possible if required). The TVG output will be fed into a 0-3.3V ADC.
  • Power Constraints: Battery-powered system with a 3V to 4.2V supply range.
  • Low-Power Operation: The system receives signals for only 20ms at 1-2 Hz, with everything else in sleep mode to maximize power efficiency. The TVG should allow the MCU to remain in deep sleep for as long as possible.
  • Operating Frequency: Currently ~40 kHz, though this could increase in future iterations.

Nice-to-Have Features:

  • Enable/Disable Control: A dedicated enable pin would be useful to conserve power, but I could use a FET to switch ground if needed as long as I don't have to rewrite the gain values and account for startup time.
  • Voltage-Controlled Alternative: If no register-controlled TVG is available, I could consider a voltage-controlled option but would prefer to avoid the complexity of external timing circuits in order to keep the MCU sleeping.

I’d appreciate any recommendations on suitable TI parts that meet these requirements. Thanks!

  • Hello Pete, 

    Maybe the PGA308... I need additional information about the input signal. 

    Input Signal: Single-ended input from a Piezo transducer used for water sensing (can make it differential if needed)

    Could you please provide what the voltage differentially would be? How will it change (input range)? 

    All the best,
    Carolina

  • Thanks Carolina, that's a challanging question to answer right now but based on research we expect it to be in the 500uV to 200mV (mas) peak to peak range, with 1-10mV range being most likely.

    It depends on how hard we drive the transducer as a transmitter (excitation voltage) and how long we listen for...the longer we listen the weaker the signal gets and we won't have data until we build a prototype which is in process now. We expect the recieve signal will have an exponential/logarithmic decline in amplitude over time due to physical and acoustic reasons but might be able to get away with a linear gain if needed but a stepwise gain is more likely to cause discountinuity throughout the receive pulse which might be problematic for our design as it's less consistant and we need a deterministic response for amplitude anlysis when comparing one receive ping to the next.

  • Hello Pete, 

    Thank you for the additional information, the PGAs I cover and the ones listed in that product page are a little simpler than what you are requesting (instrumentation amplifiers with choice of gain or bridge sensor linearization devices - PGA308/9).

    The devices you are describing are analog-front-ends which I think are better covered by this product page: Ultrasonic sensor AFEs product selection | TI.com

    I have aligned with the position sensing team and have reassigned this thread to them. They will answer soon. 

    All the best,
    Carolina

  • Pete,

    We have a portfolio of ultrasonic front ends that may help.

    Please take a look at the landing page for our Ultrasonic sensor AFEs and let us know your feedback.

    Regards,
    John