Other Parts Discussed in Thread: TINA-TI, TPS54310, TPS40200
Hello,
I'm trying to design a power supply but I am having difficulty getting my simulations to work correctly. I am designing a 2-Stage Buck converter and I have successfully built and simulated both the primary converter and secondary converters separately in TINA-TI. Both stages work fine when simulated stand-alone. The issue is when I try and cascade the stages, I get zero volts out from all converters. Both the primary and secondary converters act as if they cannot start up. I have eliminated my circuit design as the culprit by breaking the connection between the primary converter and secondary converter. I then placed a simple resistor load on the primary converter and shorted the supply input to the secondary converter and tried the simulation again. So with the stages isolated from each other and zero input to the secondary stage, I expected to see the primary converter come up and the secondary converter stay at 0. Unfortunately, I got the same result as before where neither converter came up. This tells me that the issue is with the simulation and not my circuit. To further test this theory, I tried again. I isolated the converters like before except this time I supplied the secondary converter with a constant DC supply. This time both converters came up, but the primary converter was way out of regulation (2V and climbing over the setpoint and VERY noisy compared to the standalone simulation) thus confirming my theory. I never let the simulation get to the point to find out where the secondary converter was going to stabilize at because the simulation takes quite a while.
I think what is going on is the SPICE models for both converters may duplicate net names or something so the converters are not operating independently in the simulation. Unfortunately, the SPICE models are encrypted so I can't open them up and change internal net names to test my theory. Is there a place where I can find an editable TINA SPICE model? If not, can I request a model that has had net names altered so I can further test my theory? The converters I am using are the TPS40200 and TPS54310 and I would like both the transient and average models.
If somebody has any other thoughts as to why this is happening please let me know.
Thanks,
Shawn
