How best to protect op-amp inputs from 100V. Analog switch?

Discussion in 'Analog & Mixed-Signal Design' started by nanok66, Jan 9, 2019.

  1. nanok66

    Thread Starter Member

    Jul 14, 2016
    47
    2
    Hey,

    I've got a project that is using one piezo transducer/sensor to both drive a vibration in a resonator and to generate a signal from the resulting vibration, which would be later amplified by an op-amp. To drive the piezo I decided to go with TI's DRV2700. It has a boost circuit that will give me a +-105V output that the piezo needs. Once the piezo has been vibrating for some number milliseconds I will shut the DRV2700 drive signal off and receive the ringing signal from the piezo and amplify it using an op-amp. The obvious issue is that I cannot simply have the op-amp input pins directly connected to the piezo since it will be driven at 105V. The good thing is the 105V signal will take place at a different time (previous) to when I want to amplify the resulting signal with the op-amp (post).

    Btw I should say that I'm trying to accomplish this without adding a second piezo to receive the signal.

    I searched mouser/digikey and found analog switches that can handle up to 200V but they are too expensive. I am looking for a more clever solution. For example, I do not need the 105V signal running through the switch. I could simply put the op-amp inputs through the switch. But as I am searching, I'm not sure if there is a difference between being exposed to (but not conducting) 105V. Some analog switches like DG2733e explicitly say the pins cannot block voltage much above the 5V supply.

    Are there analog switches I am missing or possibly another solution you guys can see for this?
     
  2. Bordodynov

    Well-Known Member

    May 20, 2015
    2,031
    616
    Usually this problem is solved with the help of diode limiters and a resistor. You can use a resistor and a low-power MOS transistor. Transistor open during strong signal.
     
    kubeek likes this.
  3. kubeek

    Expert

    Sep 20, 2005
    5,347
    1,009
    Can you show your schematic? It is a lot easier then to give proper advice.
     
  4. nanok66

    Thread Starter Member

    Jul 14, 2016
    47
    2
    The diode idea seems like it could work great! Thanks I think I'll try that first.

    I am unsure of MOSFETs. The piezo can produce an AC signal. Would MOSFETS affect the AC signal from the piezo?

    Kubeek, here is the schematic of the non-working circuit. I also drew another circuit with my analog switch idea if that helps you as well.

    IMG_2529.JPG IMG_2528.JPG
     
    Last edited: Jan 10, 2019
  5. Alec_t

    AAC Fanatic!

    Sep 17, 2013
    9,032
    2,131
    Use whatever signal gates the pulse generator 'on' to switch the opamp inputs to ground. I understand a similar procedure is used in radar/sonar systems.
     
  6. nanok66

    Thread Starter Member

    Jul 14, 2016
    47
    2
    Alec_t do you mean switch them to ground using an analog switch? (kinda like a modified version of my 2nd drawing?)

    Another concern that I just thought of - if I used clamping diodes to protect the op-amp, does that take away from the total power of the 105V signal that is used to drive the piezo?
     
  7. Alec_t

    AAC Fanatic!

    Sep 17, 2013
    9,032
    2,131
    I envisage the setup as being something like this :-
    ReceiverClamp.PNG
    The control input would be a logic-level signal to switch on all three transistors simultaneously, so while M1 allows the piezo to be driven the receiver input is clamped to ground by Q2.
     
  8. AnalogKid

    AAC Fanatic!

    Aug 1, 2013
    7,174
    2,020
    Can you post an image or sketch of the signal out of the DRV2700?

    Also, if the signal you want to amplify is a small AC ringing wave on top of a large DC pedestal, are two coupling capacitors an option?

    Also also, TI (Burr Brown) and Analog Devices make instrumentation amps that can handle high voltages.

    ak
     
Loading...