Other Parts Discussed in Thread: TINA-TI,
Tool/software: TINA-TI or Spice Models
Dear colleagues,
since I was not able to run the THS4551 amplifier SPICE model designed for TINA in LTspice I did some investigation and found out that more people had similar problem with TINA models of another TI amplifiers too. LTspise displays several warnings about floating nodes in the model. It has shown that there are series combinations of voltage controlled current source and capacitor at several places in the model. Quite ugly combination -the node in between them seams to be floating. It can be fixed by simply checking the "Add GMIN across current sources" checkbox in the "Control panel" > "Hacks!" menu in LTspice. TINA probably behaves this way by default so the manufacturer did not recognize the problem.
It seams to me that the above mentioned voltage controlled current sources model just resistors. Maybe using resistors would be more robust solution? Maybe this tricky way how to express resistor behavior is a result of some partially automatic model synthesis ...
The question to TI is: shouldn't your amplifier model synthesis procedure be improved to produce more robust models easily usable by other SPICE based simulators too?
David
