• TI Thinks Resolved

PGA460-Q1: PGA460-Q1: Multiplexing multiple transducer, using only 1 transformer and some MOSFET

Part Number: PGA460-Q1

Hello,

I'm developing a ultrasonic range detector (from 1m to 10-15cm or so) using 3 different transducers (Murata hermetic / closed top) and a PGA460-Q1; the questions are:

1. Is it possibile to "multiplex" these 3 transducers to the only PGA460-Q1 using MOSFETs or something? There's a MPU onboard which has some free output pin to drive the MOSFET gates;

2. The transducers are connected to the PGA460 thru a 10m shielded cable, so I think it'll be better to drive them with a hgh voltage (transformer) - is it possible to use only 1 transformer to power up the 3 transducers (one at a time, btw)

Does anyone has a (working / basic) schematic? Thank you very much!

- Marco Dassi

  • Update:

    required range is from 30 ... 40 cm to 1 ... 1.5 m

    I saw a device with what I need, but there's no schematic and I don't need the microprocessor of PGA450 (in fact, I'm using a PGA460-Q1); does anyone recognize some circuit topology?

    Thanks!

  • Hi Marco,
    I have successfully tested a multiplexing solution for the PGA460 which will readily accommodate your requirements. The only difference is that the solution uses a half-bridge driver, rather than the transformer driver. Given your range requirement of 30cm to 1.5m, the half-bridge will be able to drive nearly any open or closed top transducer to generate sufficient Sound Pressure Level (SPL) to reach this maximum required range of 1.5m. Even the Murata MA58MF14-7N can detect an ISO-pole up to 1.5m in half-bridge mode, and larger targets at even greater distances. Ultimately, this half-bridge approach will reduce complexity and save you in solution cost and PCB size since transformers can be both expensive and bulky in size.

    Unfortunately, the DECPL pin is of no use for multiplexing since the user does not have absolute control, so you will need to use an external micro controller with GPIOs to control which transducer is to be fired.

    I have just created a separate E2E discussion post showing the schematic:
    PGA460-Q1 Multiplexing in Half-Bridge Driver Mode for Mono-static or Bi-static Transducers at
    e2e.ti.com/.../603257

    If you don't need control of individual transducers, and are simply looks to burst+listen multiple transducers simultaneously, there is a solution for that as well (which is even simpler). What type of application are you designing, and why are you need to switch between multiple transducers, rather than allocating each sensor to a dedicated PGA460 device?
    Akeem Whitehead
    Texas Instruments // ASC-SENSING-SSC // Applications Engineer
  • In reply to Akeem Whitehead:

    Akeem,

    Thank you very much for your clever schematic, you really saved my day!

    There's no problem to use half or full-bridge driver, as long as the device works! I like your solution because there's no transformer (bulky, expensive, and needs to be tuned).

    There are some problems here (in Italy) to find Murata transducers in small quantities, so I've chose a MCUSD14A40S09RS (closed top, sealed) from Multicomp, which is quite similiar (I hope so).

    No problem for the microcontroller, it's already on the board with a lot of free output pins.

    This application is for a moving stair, which has to "gently" land on the ground, hence the necessity to know the distance in real time from 3 different points (L, R and center); since the speed of landing is relatively low, I've chose to multiplex the transducer for cost-saving (the evil of today's engineers...)

    Just one more question: the board will be powered by 12 ... 24 V DC, what do you think about a MC33063A in step-up mode to make a regulated 30 V DC, independent from the supply voltage? The cables (coaxial) from the PGA to the transducers are 10 m long, I think it's better to be safer than sorry so 30 V DC excitation voltage seems OK to me (and it's PGA460-Q1 limit too)

    Again, thank you very much!

    Marco Dassi

  • In reply to Marco Dassi:

    Hi Marco,

    Yes, the MC33063A will suffice for the PGA460-Q1 operation. You only need upwards of 150mA for proper device operation in direct-drive mode.

    However, you may not need the step-up regulator to properly drive the Multicomp transducer. Before investing too much time in the step-up mode, I recommend that you experiment with the 12 - 24 VDC supply first. My testing at 8V shows 1.5m detection is possible, so a minimum of 12V at the supply should be even more robust, especially since your targets will have stronger sonar cross section values.

    Although, I have never performed when the PGA460 and transducer is separated by cables as long at 10m, so you may be on to something... I'd be interested to hear what type of result you are able to achieve with and without the step-up.
    Akeem Whitehead
    Texas Instruments // ASC-SENSING-SSC // Applications Engineer
  • In reply to Akeem Whitehead:

    Hi Akeem,

    thanks for your suggestions!

    However, as you have noted, the cables are really long, and I have to complete the PCB design as quick as possible, so I'm going to integrate the step-up circuit just in the case 12V will not suffice (with a 0 ohm resistor or so to bypass the circuit completely).

    I'll keep you informed! But I think I need another 2-3 weeks to start writing the software (need board prototypes first!...)

    Thank you again and best regards
    - Marco
  • In reply to Marco Dassi:

    Just another question and I'll quit for now, I promise!

    I've checked out the MC33063A datasheet, and I've noted that I have to change some component values for different voltages (my board will be installed on 12V OR 24V automotive), but I'll prefer a general solution to avoid different PCB versions (one for 12V, one for 24V, maybe another for 18V...) - do you know a versatile step-up IC that fit my needs without component changing?

    Assume that the board will be powered by a 9 to 27 V DC.

    Thanks! Ciao!
    - Marco
  • In reply to Marco Dassi:

    Hi Marco,
    The E2E Power Management forum could probably provide a better recommendation for your variable supply condition. Using the WEBENCH Power Designer, several solutions seem to be available using the TPS55340, LM5022, LM34878, ... (comes down to your preference of efficiency, cost, size, etc.)

    Here is the resulting link:
    webench.ti.com/.../webench5.cgi

    Keep in mind that the voltage at VPWR can only be as high as 28.3V. Otherwise, the device will assume there is an over voltage condition, and prevent the burst drivers from exciting the transducer. If you supply the VPWR pins with the 9 to 27 VDC directly, you can still connect the high side reference of the bridge drive to 30V. The VPWR supply does not need to be equal to the voltage sinking through OUTA and OUTB. Thus, you could supply the PGA460 with the minimum of 6V, and still sink 30V through OUTA/B as an extreme example.
    Alternatively, you could have your step-up set to 28V to supply both the PGA460 and high side of the bridge driver.
    Akeem Whitehead
    Texas Instruments // ASC-SENSING-SSC // Applications Engineer
  • In reply to Akeem Whitehead:

    Perfect!
    Thanks for the 28.3 V reminding, I forgot that!
    Best regards
    - Marco