• Locked Locked
  • Replies 7 replies
  • Subscribers 49 subscribers
  • Views 1286 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.

LMP91000 & sensor simulation

Jeremy Wilson
Prodigy 80 points

Other Parts Discussed in Thread: LMP91000, LMP91000EVM

I'm using a variable power supply and series resistor to simulate a galvanic oxygen cell (2-lead) with the LMP91000, but can't get the output to change.  This sim works great on our existing TIA circuits, and we can sim on our potentiostat circuit for 3-lead CO sensors by pulling the current out the WE and into ground.  Communication with the LMP91000 is good, I can set or read the registers with no problems.  What's the catch?  Do I have to do the ground-referenced 2-lead mode?  In this mode, is the parallel resistor necessary?  What value?

The reason we simulate is to verify/calibrate the amplifier circuits.  

Thanks.

  • Prodigy 80 points

    I tried the 2-lead ground referenced mode and I do have output voltage proportional to input current.  However, changing gain resistors (Rtia) doesn't have any effect.  But, changing the parallel resistor does have a big affect.  The datasheet says that the gain in this mode is 1+Rtia/Rload, but perhaps the parallel R should be in this equation.  Any help is greatly appreciated!

  • in reply to Jeremy Wilson
    TI__Guru**** 182921 points

    Hi Jeromy,

    We'll look into this and get back to you shortly.

  • in reply to Tom Hendrick
    Prodigy 80 points

    I've discovered that when wired as 2-lead ground-referenced, there is no current going through the sensor connections (simulating 200 uA by 2.00V through 10K).  The voltage out is essentially mirroring the sim power supply, and does not (hardly) change with resistance in the sim power supply line.  Why is there no current going through the TIA?

    Thanks. 

  • in reply to Jeremy Wilson
    Prodigy 80 points

    Circuit in previous post is NOT using parallel resistor because it just shunts the current to ground.  The lower the resistor, the lower the Vout from TIA.

  • in reply to Jeremy Wilson
    Prodigy 80 points

    Does it matter if the CE and RE are shorted while using the 2-lead ground-reference mode?

    Circuit is still behaving like a volt meter instead of current meter because no current is going through TIA for some reason.

    Please help!!!

  • in reply to Jeremy Wilson
    Prodigy 80 points

    Wow, response from the forum has been underwhelming!  Are there no TI engineers that know anything about the LMP91000?  I really need to get this chip running to replace the existing TIA's in our gas analyzers - need to ship product in January.  Please help!

    Obviously, 2-lead gnd-ref mode is a voltage amp, not current amp, but how come changing the gain has no effect?  This mode is unity gain buffer.

    It seems that for current amplification in this mode, sensor + should connect to Vref and sensor - should connect to C1, with WE,CE,RE open.  But that doesn't work.

    I don't understand this chip!  We've been using op-amp based TIA's in 2-lead and 3-lead analyzers forever and the sensor can be simulated with a power supply and current-limiting resistor.  I've tried everything to get 2-lead potentiostat and 2-lead gnd ref modes to work.  What am I missing?  

  • in reply to Jeremy Wilson
    TI__Genius 9655 points

    Hi Jeremy,

    In the picture below you can see the how the LMP91000 looks like when in 2-wire Sensor GND referred configuration.

    The TIA doesn't work anymore as a TIA, but it is an amplifier in non-inverting configuration, it means that the current generated by the O2 sensor flows in the R_O2_LOAD resistor and generates a voltage which is amplified by the TIA stage (1+RTIA/Rload). In your bench circuit (Power supply+series resistance) there is a mistake, in order to simulate a 200uA of current on a 100ohm (typical rload for Oxygen sensor) you need a supply voltage of 20mV with in series 100ohm.

    Connecting a 2V supply voltage wiht 10Kohm in serie you saturate the TIA.

    Please let me know if this helped.


    P.S. The LMP91000EVM, as written in the documentation, doesn't support the 2-lead GND referred configuration 

    regards,

    Domenico