Fast PWM IC

[elki]

Is there any IC that could provide a PWM at 100 MHz or so? I would love to use atmega ICs but these go up to 20 MHz, if I am not mistaken. The idea is to use it for a voltage control in sample-and-hold circuit for short input pulses. Thanks!

[strawberry]

custom circuit with digital logic gates or FPGA

[langwadt]

Is there any IC that could provide a PWM at 100 MHz or so? I would love to use atmega ICs but these go up to 20 MHz, if I am not mistaken. The idea is to use it for a voltage control in sample-and-hold circuit for short input pulses. Thanks!

100MHz pwm frequency or 100MHz time resolution?

[elki]

Frequency.

Is there any IC that could provide a PWM at 100 MHz or so? I would love to use atmega ICs but these go up to 20 MHz, if I am not mistaken. The idea is to use it for a voltage control in sample-and-hold circuit for short input pulses. Thanks!

100MHz pwm frequency or 100MHz time resolution?

[langwadt]

Frequency.

and how many bits of resolution?

[selcuk]

It seems that you will need fast/expensive ICs for that. How about implementing analog control by using comparators etc.? I'm not an expert in this but you can check datasheets of such controllers for block diagrams to have an idea.

[Zero999]

I still don't understand why you need 100MHz PWM.

The problem is, it needs to be able to generate much higher frequency harmonics. 10% duty cycle has a 1ns on time, so you need much sharper rise and fall times, say 100ps, so you need a slew-rate of 33GV/s for a 3.3V signal.

[coppice]

If you really want a PWM channel running at 100MHz with any kind of reasonable resolution for the pulse width trying looking at things like the TI C2000 MCUs. They have a pulse width resolution far beyond their clock rate, using calibrated delays.