• State
  • Locked Locked
  • Replies 5 replies
  • Subscribers 48 subscribers
  • Views 517 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.

TLE2426: SPICE model doesn't work in floating circuits

Antenna_Head
Genius 12055 points
Part Number: TLE2426

The SPICE model for TLE2426 doesn't work with floating circuit simulations, see below.  V+ should be 5V, and V- should be -5V.  There should be NO CURRENT in the output of the '2426.

I compared this with an equivalent circuit built from TLV9102- this provides correct results, also shown below

The TLE2426 model is ancient, the one included in TINA appears to be the same as the latest in the product folder (just wrapped up for PSPICE).  This bug has wasted a lot of time and very surprised this hasn't been revealed in the previous 25-30 years...

Please fix this, especially in light of the fact that this part will likely be updated with migration to RFAB (hopefully).

Attached simulation in case you would like to see it.

-Steve

TLE2426 compare.TSC

  • 0
    Guru 140200 points

    Hi Steve,

    I think the TLE2426 wants signal ground to be connected to its GND pin:

    Kai

  • 0 in reply to kai klaas69
    TI__Genius 12055 points

    Sure. If you hook up pin 2 to simulator GND, it works.  That isn't the point.

    I need it in a circuit not referenced to ground.  This is actually a pretty common application.  It is obviously a limitation in the model- in real life the circuit would work just fine, just like to other, discrete example shown.

    -Steve

  • 0 in reply to Antenna_Head
    TI__Guru 64290 points

    Steve,

    As you know TLE2426 is meant to be used as a "rail splitter" with supply voltage (referenced to ground) as its input and providing mid-supply "virtual ground" as an output.  Even though in the real world shorting the output to ground should result in +/-5V supply, TLE2426 is a 30 year-old part with legacy macro-model with its "limitations" and it would require a full re-characterization of the internal 30-year-old op amp to create a modern model based on our Green-Lis architecture with floating internal ground. Considering required resources to do so, this may not be feasible.  Any migration of the part to RFAB will require a full characterization of the new part and thus result in a modern macro-model being created.

  • 0 in reply to Marek Lis
    TI__Genius 12055 points

    Marek-

    I understand that re-doing the existing device is not really an immediate priority given everything else going on.

    I'm finding applications where the rail splitter can be used with an iso-watt type of device and provide a very nice isolated bipolar power supply when needed.

    Please make sure that any new models created are portable to simulators beyond PSPICE.  Generally the amps team is pretty good about this, but just a reminder.

    Thanks!

    -Steve

  • 0 in reply to Antenna_Head
    TI__Guru 64290 points

    Steve,

    Yes, any new rail-splitter macro-model will be based on our new architecture, which includes a floating internal ground - this will result in a proper operation of the circuit with output shorted to any voltage including ground.