• State
  • Locked Locked
  • Replies 1 reply
  • Subscribers 30 subscribers
  • Views 103 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.

Original question:

LDC1614: Network between coils and LDC1614

LDC1614: Switching sensors on channel 0 using analog MUX - TMUX1136

Matangk
Prodigy 10 points
Part Number: LDC1614
Other Parts Discussed in Thread: TMUX1136

We are working on an application that identifies inductances in the order of ~200nH, With series resistances of ~0.5Ohm.

Due to the relatively low quality factor we need to use the LDC1614 in 'high current' mode. We use ~1nF capacitors in the parallel tank to obtain oscillation around 10MHz. Each of the inductors can be measured when connected standalone to the LDC1614, with an amplitude around 1.8V as per the chip spec.

We need to switch between two of those sensors but the LDC1614 can only provide 6mA on channel 0. To get around that, we connect two tanks (i.e. 2 capacitors and 2 inductors, see figure) via a dual SPDT analog MUX, TMUX1136, operated at 3.3V. The mux bandwidth is 250MHz which should be plenty. Since the mux is connected in series to the whole tank, its resistance is small compared to the equivalent parallel resistance of the tank (~300Ohm) in resonance - shouldn't be a problem.

We see a very odd behavior, where we get a watchdog timer error instead of measurement when the tanks are connected through the switch. If we fiddle with the pins that connect the tanks to the switch we see sudden measurements, which stop and start sporadically. We checked connections several times and they look fine. Is it possible that there is some strange latch-up or transient event that happens under those conditions? Is there something going on with the measurement initialization sequence?

Any insight will be helpful.

Cheers

  • 0
    TI__Mastermind 34505 points

    Matangk,

    A couple of questions so we can try to identify the root cause:

    1. Does the presence of the MUX change the sensor waveform if there is no switching between channels?
      In other words, if a single path thru the MUX is always enabled, and that LDC channel is always active, does the Tankx waveform look different than if there is no MUX and the Tankx is connected directly to the LDC pins?
    2. If the answer to (1) is 'No':
      As the MUX switches from Tankm to Tankn what does the Tankn waveform look like during the time just after the MUX completes switching? 
      There should be an interval where the Tankn start-up transient settles out and the steady-state oscillation starts.
      What is the duration of the start-up transient?
    3. Does the circuit behavior change if you use the following sequence:
      a. Set the LDC to Sleep Mode.
      b. Set MUX to channel Tankm, 
      c. Set the LDC to Normal mode and note any errors.
      d. If no errors, record data - does it agree with expectations?
      e. Set LDC to Sleep Mode.
      f. Repeat steps b - d for Tankn.
      g. Does the LDC successfully complete the measurements?
          Do you still see the watchdog timer errors?

    We have found that  even a high-impedance o-scope probe can load & distort the tank circuit waveform so it frequently helps to place a leaded 1k resistor between the probe tip and the PCB test point.