• State Resolved
  • Locked Locked
  • Replies 6 replies
  • Subscribers 50 subscribers
  • Views 165 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.

TLV8544: Switching stage 1 and stage 2 sensitivity

Filip Halamka
Intellectual 320 points
Part Number: TLV8544

Hi, 

I would like to ask you, if do you have any experience with TLV8544 using for PIR element with 2-3 different sensitivities. I mean i you lf you look into schematics from BOOSTXL-TLV8544PIR reference design, would be possible to change values of R9 + C6 and R7 + C7 by either some mechanical switch or being controled by MCU with some switch circuit? Do anyone have any experience if is it even possible doing it without corrupting amplifier's parameters?

Thank you very much!

  • 0
    TI__Mastermind 20770 points

    Hi Filip,

    I don't have much experience in what you are trying to do. However, my experience with switches is that you have to be very conscious of parasitic that they have inherent to their system. 

    Would this switch be actively changing positions as the system is running or would it only switch once for each time the system "starts up"/begins to read the PIR?

    Best Regards,

    Robert Clifton

  • 0 in reply to Robert Clifton56
    Intellectual 320 points

    Hi Robert,

    Thank you for your answer, there were actually two ideas:

    1) use MCU controlled loadswitches for switching these passives in firmware (there would be one fw for one sensitivity and second for another).

    2) use simple dipswitch for setting sensitivity once in manufacturing process.

    What do you think about that?

    Thanks!
    Filip

  • +1
    Guru 140200 points

    Hi Filip,

    the gain of the first stage is 220V/V and the gain of the second stage is 220V/V as well giving a total gain of 48400V/V. This indicates that the input signal coming from the PIR is very very very small, I guess in the range of 10...50µV or so.

    Adding a µC controlled switch in parallel R9 (or instead of R9) would mean that glitches and digital noise being superimposed to the switch control signal and which is injected into the signal chain by charge injection (!) is also amplified by a huge gain. Because of this, I would not add a switch to the first stage. I would leave the first stage untouched and run it with the minimum gain which is necessary in your application and move the µC switches to the second stage or another later stage.

    In any case, since you have a mixed analog digital circuit then, you will need to low pass filter the switch control signal extremely well. Otherwise the digital ground return currents will contaminate the analog section with digital noise. In combination with the ultra high gain of the analog circuit this is super problematic and does not forgive even the least mistake.

    Another remedy is to design several parallel signal paths with different gains and different outputs which go to a multichannel ADC. Then you would not have any switchings and digital noise in the analog circuit.

    Kai 

  • 0 in reply to kai klaas69
    TI__Mastermind 20770 points

    Hi Kai,

    Excellent point about the large gains amplifying any error caused by the switch. Always appreciate the insight you provide!

    Hi Filip,

    kai klaas69 said:
    Another remedy is to design several parallel signal paths with different gains and different outputs which go to a multichannel ADC. Then you would not have any switchings and digital noise in the analog circuit.

    To add to the parallel paths idea, if you choose op-amps with shutdown, then whichever path you aren't using can be put into shutdown mode to reduce the overall power consumption of this system. If that's even a concern of course. 

    Best Regards,

    Robert Clifton 

  • 0 in reply to kai klaas69
    Intellectual 320 points

    Thank you very much, that is great perspespective!

  • 0 in reply to Filip Halamka
    Guru 140200 points

    Good luck Relaxed