[sci.electronics] FAST pulses at several amps?

ch@well.UUCP (04/18/87)

[]
	Hello,

A communications system I'm working on requires the generation of
short, high-power, controlled pulses with a low repetition rate.
Can anyone please suggest ways to accomplish this?

Here are typical requirements:

Pulse  Duration : .5 through 5 ns, ~1-2ns is a practical compromise.
Repetition rate:  LOW duty cycle, ~ .1% probably.
Pulse Current  :  12-30 amperes.
Rise/Fall Time :  Preferably < or << the pulse duration.
Pulse Voltage  : > 5V to overcome R(load) (upper limit determined by R(G) )

The load is a LASER-diode, so I'm unsure in what manner parasitic /
non-linear capacitances / inductances will affect the pulse generator.

I know that similar requirements are often encountered in RADARs,
nuclear physics apparatus, and various other items, so I'm hoping that
this will prove to be relatively easy to accomplish.

I'll consider anything (MOSFETs, step recovery diodes, varactors, etc.)
Given the low utilization rate, power dissipation shouldn't be very
important, therefore low efficiency is tolerable, but transients should
be minimum, and OFF state current should be low.

Thanks,
-- 
	Chris Hayes
			UUCP: ucbvax!dual!well!ch
			 OR : {hplabs, ptsfa, lll-lcc}!well!ch

noise@eneevax.UUCP (Johnson Noise) (04/21/87)

For a fixed pulse duration, discharging a suitable length of
transmission line into the load is standard practice.  Hydrogen
thyratrons work well as fast, high current switches.  EG & G
manufactures their version (Krytrons) for this specific purpose.
Carefull construction and good impedence matching have yielded
sub-nanosecond risetimes.

Krytrons operate at relatively high voltages (2-5 kV), so an
attenuator may be required.  Krytrons exhibit finite lifetimes
(not more than 1E+7 discharges). Jitter not greater than 5ns
can be expected.

For more information, we refer you to EG & G.