[christianw.]

Welchen R meinst du?

Zitat:Note the 1-V zener diode associated with the PWM comparator ? s input from the error amplifier. This is not an actual diode in the device ? s design, but an indication that the maximum current sense input amplitude is 1-V typical. When this threshold is reached, regardless of the error amplifier output voltage, cycle-by-cycle current limiting occurs, and the output pulse width is terminated within 35 nanoseconds (typical). According to the UCC38C42 data sheet specifications, the minimum value for this current limit threshold is 0.9 V with a 1.1 V maximum. In addition to the tolerance of this parameter, the accuracy of the current sense resistor (or current sense circuitry) must be taken into account. It is advised to factor in the worst case of primary and secondary currents when sizing the ratings and worst-case conditions in all power semiconductors and magnetic components.

Quelle: http://www.ti.com/lit/an/slua257/slua257.pdf

Mir ist durchaus bewusst, dass der Spannungsabfall über Rs zum Abschalten des RS-FF führt, nur braucht es dazu 0.9V-1.1V. Bei 10mOhm, müssten dazu aber annähernd 100A fliessen.

Laut AN funktioniert die Abschaltung am CS Pin dann hier, weil eine Slope-Compensation eingesetzt wird.

Zitat: The two signals are summed together at the current sense input (CSNS) connection at the filter capacitor. To minimize loading on the oscillator, it is best to buffer the timing capacitor
waveform with a small transistor whose collector is connected to the reference voltage. The timing resistor
should be broken into two resistors in series with the center-point connected to the transistor?s base. This configuration develops more voltage across the slope compensation resistor than others techniques and produces better results.

Klemmt man diese ab, läuft das Teil (gezwungener Maßen) im reinen VCM (voltage control mode).



Was übersehe ich hier?