• State Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 2 answers
  • Subscribers 30 subscribers
  • Views 676 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 tramsmit issue

Roger Green
Prodigy 65 points
Other Parts Discussed in Thread: TDC1000

I have found something interesting....

The TDC1000 transmits pulses ie: square wave.

When I connect up a piezo disc (the 1 MHz one supplier with the eval board) and attach the disc to metal conduit I get very strange results.

The recieving sensor picks up the tx signal (with almost no delay) along with the signal as attenuated and delayed by the liquid betwen the sensors.

It seems that there is more than one signal path.

Now a square wave is a fundemental and infinite odd harmonics and I think that these higher frequencies are causing problems.

I tried the exact same sensors with a signal generator and with a square wave I get identical results but with a pure sine wave I only get a signal with the correct delay time on a sensor in line with the tx.

So question, why does the TDC1000 ouptut a square wave when I believe (for much better results) it should be a sine wave?

  • 0
    TI__Expert 4725 points
    Roger,
    Can you clarify your question?

    If you're seeing some specific behavior with your EVM that you'd like help with scope shots and EVM settings are required to get a better understanding of what you're seeing.

    With regards to excitation pulse shape, yes the TDC1000 drives the piezo with a sqaure wave and not a sine wave. Square wave excitation is sufficient for the applications the TDC1000 was targeted at.

    Best regards,
  • 0 in reply to Matthew Minasi
    Prodigy 65 points

    Hi Mathew,

    I cannot give scope shots, not on my current one.

    However, what I gather is that piezos not only have multiple radial resonant frequencies but also axial resonance.

    A square wave could be exiting the piezo in more than one frequency.

    With a square wave I am seeing a signal with almost no delay on the receiver, could this be due to axial resonance through metal pipe itself?

    I do not see this with a sine wave - I only see the correct signal with the right (expected) delay.

    I cannot understand why an additional opamp LPF stage was not put in the TDC1000 so you only exite the piezo as one one desired frequency.

  • 0 in reply to Roger Green
    TI__Expert 4725 points
    For some good basic understanding of piezos, how they work and how to select them I recommend reading my AppNote on transducers www.ti.com/.../snaa266.pdf

    Yes, every transducer resonates in both radial mode (~ to radius dimension) and axial (thickness) mode (~to thickness dimension). Depending on your material etc a 2.1mm thick 10mm diameter disc will resonate ~1Mhz (axially) and ~215kHz (radially) .

    For TOF measurements we are utilizing axial resonance.

    The immediate signal you are probably seeing is ringdown of your transducer and not an echo but without a scope shot its impossible to tell.

    What exactly is your application?

    Best regards,
  • 0 in reply to Matthew Minasi
    Prodigy 65 points

    Hi

    Thanks for that....

    Our application is flow measurment.

    Ringdown is where the transducer rings for a time after exitation - correct?

    Well it might be this and if it is then interestingly this is much much smaller when you use a sine rather than a square wave as an input

    I will be getting a much more modern oscilloscope shortly and will be able to grap traces.

    In the meantime if I remove the resistor that joins the rx and tx together I will be able to introduce a filter on the tx and with a high o/p impedance enable on the tx side will be able to also receive - by making the tx high Z.

    Another thought is that the start signal only occurs for the first 3 cycles of the tx, so I cannot use this as a high Z disable - will have to use an additional circuit to detect the tx pulses - buffered sample and hold...

    Regards,

  • 0 in reply to Roger Green
    TI__Expert 4725 points
    Roger,
    I suggest you read our flow meter app note www.ti.com/.../snia020 to get a better understanding of how to use the TDC1000 for your application.

    Best regards,