• State Resolved
  • Locked Locked
  • Replies 3 replies
  • Answers 1 answer
  • Subscribers 30 subscribers
  • Views 907 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

TDC1000 Measurement Range

WAYNE FISHER
Prodigy 25 points
Other Parts Discussed in Thread: TDC1000

Hi,

The datasheet for the TDC1000 indicates a measurement range of up to 8 ms. The datasheet also indicates that the min operating frequency is 0.06 MHz.

Using a 1.28 MHz clock (for example), the TDC1000 can be configured for echo listening periods of up to 15 ms. (Based on Figure 38: Standard TOF Measurement.)

What is the source of the 8 ms value for the measurement range?

I would like to achieve a longer listening period than 8 ms - preferably something closer to 30 ms. (ie with a 0.64 MHz clock).

On our prototype board with a 1.28 MHz clock, I am able to detect signals up to 8.5 ms or so. When the input signal is detected, the signal (pulse) going in to COMPIN looks fairly strong. After making a very minor change in the distance to slightly increase the TOF, the pulse is gone.

I would expect that the pulse would degrade slower and not suddenly around 8.5 ms.

In parallel to my own debug efforts on this, I thought that I would ask if there is something in the TDC1000 that the 8 ms time is based on that might be at play here.

Thanks,

Wayne

PS. My application is for distance measurement through air. I have a boost circuit on my transmit signal to increase the voltage to my 40 kHz emitter to improve the range. And I have looked at the PGA450 but it doesn't appear to meet my need to synchronize more than one of them together.

  • 0
    TI__Expert 4725 points

    Wayne,

    That time measurement maximum has to do with the comparator input section of the TDC1000.

    During a measurent the TDC1000 goes through an "Autozero" where it charges some small caps on the inputs of the comparators up to VCOM. Over longer periods of time that charge dissapates which looks shifts the comparator reference voltage.Shifting the reference voltage results an incorrect threshold being triggered on and ultimately an incorrect measurement.

    So yes the clock can be reduced but the measurement time limit is still the same.

    Also for measuring longer distances (>8ms) require two major changes to the system:

    1. Higher transducer excitation voltage (there's a reason all the air based solutions out there use transformers) You'll want to hit your transducer as hard as you can to get any echo back.
    2. Signal gain. The TDC1000 only has 41db of gain. It was designed to work it a signal path with attenuation levels similar to liquids. You can add more gain to the TDC1000 system but beyond a certain point you will  only amplify your noise.

    Regards,

  • 0 in reply to Matthew Minasi
    Prodigy 25 points
    Hi Matthew,

    Thanks for the reply.

    I have already taken care of generating a higher transducer excitation voltage.

    And I have the benefit that my use-case is measuring the distance between the emitter and receiver. So I'm not looking for an echo that bounced off something, but am instead looking for the actual burst.

    I'll scope around my filter circuits some more, but looking at the COMPIN pin, the pulses just appear to vanish past TOF times in the 8.8 to 8.9 ms range.

    I was expecting to see the pulses get weaker with a small movement of the emitter away from the receiver - not to just disappear. Hence my question on whether there is something limiting its operation past the 8 ms in the datasheet. COMPIN is before the comparators so I don't think that a shifting reference voltage is my current issue. Its not that an incorrect threshold is being used, it is more the pulse may not be getting through earlier parts of the analog front end.

    For our prototype stage, if we could get the TDC1000 up closer to 15 ms, I would be quite happy. It looks like we will need to move to something else if/when we move to productize it.

    Is there anything else in TI's product catalog that might work for longer distances? I looked at the PGA450 but I don't see a way to closely synchronize more than one of them together similar to the TRIGGER input of the TDC1000.

    Thanks,

    Wayne
  • 0 in reply to WAYNE FISHER
    TI__Expert 4725 points

    Wayne,
    What I think you're seeing is the receive channel shutting down after the maximum timeout. This is set in the TOF0 and 1 registers. At present only our GUI limits the calculations of the time delays to a minimum input of 1Mhz.
    What is your transdcuer frequency? In your situation I'd do the following:
    1. get your setup working with a higher frequency External Clock input somewhere between 1-8Mhz (so you can have the GUI calculate all the timing parameters)
    2. Save the register setup using "Save Config" on the TDC1000 tab.
    3. Make sure "continous trigger" toggle on TDC1000 tab is red.
    4. Reduce your External clock by a power of 2 that gets you no less than 2x your XDCR frequency
    5. Change NUM_TX value to the correct divider so you are still exciting the XDCR at correct frequency.
    6. set "continous trigger" toggle on TDC1000 tab to green.
    7. Verify XDCR excitation is correct. check for your echos on Comp_mon.
    -Regards,