I'm simulating the following comparator circuit using TINA-TI (Version SF-TI). The behavior, however, is not as expected:

enter image description here

Hopefully it is clear that V- remains constant, while V+ swings above and below it. The output stays at 5V though, it only slightly nudges in the right direction.

As a reference, here's the comparator LPV7215 datasheet.

Common mode input voltage is within range, inputs are within range, power supply is within range. The output is a push-pull. What gives?

  • \$\begingroup\$ The schematic looks OK, and I agree, it should be giving you a rail-to-rail output swing. At this point I would check the netlist, since you might have a missing connection that looks good on the schematic. Or perhaps the comparator model has a problem? \$\endgroup\$ – Paul Elliott Feb 2 '17 at 6:06
  • \$\begingroup\$ The output doesn't stay at 5v, there is some small activity, just not enough. Is the pin mapping between the part symbol and the underlying spice model correct? The pin order matters, and if the file defining the behaviour has different pin order to the diagram ... I'm not suggesting you try swapping the pins on the schematic, but open the definition file for that symbol, and check the pin assignments against the spice model. The model will usually say where the pins are, in comments. What's the provenance of the model, and of the symbol? Does a different comparator work? \$\endgroup\$ – Neil_UK Feb 2 '17 at 6:30
  • \$\begingroup\$ @paul Netlist seems fine. I suppose the model may have a problem. The circuit works normally if C2 is increased to 1 nF. Made me wonder, can comparators latch up due to input slew rate? \$\endgroup\$ – apalopohapa Feb 2 '17 at 7:45
  • \$\begingroup\$ @Neil_UK The model comes with TINA-TI. So Texas Instruments provides it. If I increase transition times on power source from 1 us to 100 us then it works. If I slow down signal by increasing C2 from 22pF to 1nF it starts working. If I reduce the amplitude of the swing (effectively reducing slew rate) then it works normally. I don't know if this is a simulation artifact, or device is supposed to latch up, or model is inaccurate. \$\endgroup\$ – apalopohapa Feb 2 '17 at 7:50
  • \$\begingroup\$ Why did you add R5? It looks like you gave the comparator a positive feedback. I'd try the circuit without the feedback resistor. Would love to know the result. Good luck. Moshe \$\endgroup\$ – Moshe Feb 7 '17 at 18:42

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.