Correct way to model a pulsed load current in LTSpice

Thread Starter


Joined May 22, 2019
I have been trying to model my load in LTSpice but seem to be getting different results for different methods. The first suggested to replace my resistors with pulsed current sources, and that was it. Another tells me to keep my load resistors, but place pulsed current sources in parallel with them with the characteristics of my load.

The characteristics of my load is a travelling wave tube transmitter, so for example can be operating at a duty cycle of 1% @ 100kHz or 25% @ 10kHz, etc. I will attach the second method that I mentioned in a schematic in hopes the question will be clear. If there are alternative methods that are more accurate please let me know.

Output 1 is 15kOhm resistor, 3kV voltage, 600W and output 2 is 600kOhm resistor, 6kV voltage, 60W output power (continous pulse). Average pulsed powers are much lower.

The load pulses from full current to substantially ZERO current at the aforementioned rates. Only current in the secondary side is due to the resistive divider for feedback. This is what I have read. Others have told me I must include a minimum load resistor, so if anyone has advice on whether to do that(and how?) then that would be additionally very helpful.



Joined Jul 11, 2016
you can combine current sources/sinks with resistors the way you feel suitable for you
to mod a resistor dynamically - enter it's value as R=FnR() , then define for example
.func FnR(){2u+1k*u(time-1)}
.func FnR(){2u+1k*u(sin(8*atan(1)*time*1k))}
both of which have an issue that the R value would change at an instant
to smooth the trasnsition
you may want to use a (say 1V) pulsed voltage source define it's Trise , Tfall
then use
.func FnR(){2u+1k*V(VpulsedSource)}

with the current sinks/sources - make sure - you set the This is an active load [✓] check-box


Joined Mar 31, 2012
I don't think LTSpice (or most other SPICE-based simulators) are intended for these kinds of circuits. Which is not to say that they can't be used, but you are likely to have to spend a considerable amount of effort developing suitable models for all of the pieces parts of your system. Parasitics are likely to be very important.

There are other tools that are specifically designed to model and simulate "unusual" (as in not your run-of-the-mill electronic circuit) systems. You might spend some time doing some research and contacting people that work with these systems to see what they recommend.