2
\$\begingroup\$

I have downloaded LTspice PWM switch from christophe basso website http://powersimtof.com/Spice.htm file name PWM-CM2-MD2 by Mark Dimattina

I am doing 5.1V converter based on LTC3890 IC so I tried to compare frequency response using IC spice model and PWM switch average model but I need some clarification

LTC3890 Spice model based FRA enter image description here

PWM Average Switch circuit enter image description here

FRA from Average Model enter image description here

I am doing the bode plot for Cm-out/Cm-in and then Gain crossover frequency is 235KHZ as per average model (Fsw=440KHz) and phases are negative ...

if I plot CM-out/CM-Vc then I am getting gain crossover frequency around 16KHz and phase Margin -58 deg (Idont know why I am getting neg sign).which is pretty much same with LTC3890 Spice model based FRA except phase is negative. enter image description here I am not able to understand which one right and why phase is negative ..Can anyone please help

\$\endgroup\$
3
  • 1
    \$\begingroup\$ First off, I would recommend to have symmetrical ranges in Y axes for dB and degrees, like 50/-50 and 180°/-180° for instance. That way, it is easy to spot the 0-dB point and read the phase margin as the distance to 0°. In your second plot, it seems that the crossover is around 30 kHz (please, lowercase k and space inserted) and the phase margin is greater than 90°. In the third plot, you measure the power stage control-to-output transfer function (TF) so the phase starts at 0° and then drops towards negative values at it should. What you read is the argument or phase of the power stage TF. \$\endgroup\$
    – Verbal Kint
    Commented Mar 6 at 14:26
  • \$\begingroup\$ Many Thanks sir ...I will try to follow symmetrical axis recommendation. But which plot is actually a frequency response of converter second one or third one ? As per my understanding when we measure bode plot in lab that gives control to output plot (please correct me if I am wrong) in the second plot in image it looks like like gain crossover frequency is 30 kHz but it's crossing 0 dB at 234 kHz before it's completely flat line of 1.2 dB. what should we call this second plot ? \$\endgroup\$
    – Power Path
    Commented Mar 6 at 14:52
  • \$\begingroup\$ As per my understanding when we measure bode plot in lab that gives control to output plot Well, only if you measure the control-to-output response. If you use the normal technique of a small resistor at the top of the feedback divider for injection, and measure across that, it's the entire open-loop response of the converter. You can (usually) measure either response, and both can be valuable. \$\endgroup\$
    – John D
    Commented Mar 6 at 17:00

0

Reset to default