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.

TLV3202: Common-Mode Range Query

Part Number: TLV3202

Hi Team,

I understand TVL3202's Vcm is Vee-200mV to Vcc+200mV. Our customer intends to use single supply (+5V Vcc and 0V Vee), and have TLV3202 in a window comparator configuration detecting voltages outside the range of 0-5V.

This is part of a voltage controlled oscillator circuit, and the window comparator is part of a feedback loop that tells the switch to reverse direction of integration. So it is detecting Vp and -Vp and only generates pulses with widths in the hundreds of nanoseconds range. The frequencies this VCO generates are in the audio range, so the input is able to turn around long before it reaches a value outside of TLV3202's absolute maximum ratings.

Is it okay to set detection thresholds at the rails as long they are not exceeding the absolute maximum ratings or Vcm? Will TLV3202 work reliably in this configuration? Thanks in advance!



Kind Regards,

Jejomar

  • As long as the input voltages do not go actually go beyond the specified limits, everything is OK.

  • Hi Jejomar,

    the comparator will only work properly within the specified common mode input voltage range which is 200mV beyond the supply voltages. For higher input voltages the comparator is not guaranteed to work properly and phase inversion can occur. The maximum input voltage range specified in the absolute maximum ratings of -0.5V ... Vcc + 0.5V only guarantees that no damage occurs but not that the comparator is working properly. So for proper operation the input voltages should not exceed the supply voltages by more than 200mV.

    As you say that the input voltages are short and narrow spikes, any unwanted inductances or other complex impedances in the signal path can cause signal ringing and overshots making the input voltage easily exceed these 200mV. Unwanted signal delays within the loop can also make the input voltages rising all too much. Not even talking about power-up issues, when the oscillator might never want to start because of exceeding the common mode input voltage range with the first ramp. No, in my eyes 200mV is just too little headroom.

    But why not decreasing the input voltages by the help of a simple resistive voltage divider? All oscillator circuits I know using a comparator to reverse the polarity of oscillation signal work with a voltage dividier.

    Kai   

  • Jejomar

    Thanks for your post and thanks Clemens and Kai for your excellent input.  

    To Clemens point, setting the thresholds at 0V and 5V to create your window is a warranted use case.  So no issue with that.  My only suggestions would be to consider adding a current limiting resistor, say in the 1k to 10k range, and to consider schottky diodes to VCC and GND just in case the system doesn't respond quickly enough.  

    Chuck

  • Hi Clemens and Kai,

    Many thanks for your valuable inputs!


    Hi Chuck,

    Regarding the possible solution of using current limiting resistors and Schottky diodes to protect the inputs. Is there a specific Schottky diode you can recommend? One with suitable Vf and low capacitance that doesn't affect the response time significantly? Are the extra diodes not strictly necessary? The internal clamping diodes do the same job do they not?


    Kind Regards,

    Jejomar

  • Thanks Jejomar for following up.  I would say they are not totally necessary but always something to consider when someone has an application that will drive the inputs beyond the rail.  In regards to the internal ESD protection, they are not the same.  Our internal ESD diodes start conducting at higher voltages and are fully conducting when 0.7V beyond the rails.  Schottky's would start conducting more in the 0.3V range.  It really comes down to what the waveforms at the inputs look like to determine how needed they are.  We warrant our parts to a certain level.  Going beyond that is always possible, we just don't warrant the operation.

    Chuck

  • Hi Jejomar,

    I still find the voltage divider a good idea Relaxed

    Kai

  • Thanks Kai for the support. 

    Jejomar, 

    If there isnt any further questions, I am going to close this post. Feel free to reply if you do.