From: Robert Baer
X-Mailer: Mozilla 4.75 [en] (Win98; U)
Subject: Re: easy fast rise time pulse generator?
Date: Tue, 15 Oct 2002 06:11:07 GMT
NNTP-Posting-Date: Mon, 14 Oct 2002 23:11:07 PDT
Organization: EarthLink Inc. -- http://www.EarthLink.net
Larry Gagnon wrote:
> I require a fast rise time pulse generator to calibrate the vertical
> amplifier pulse response of my HP1722B oscilloscope. Obviously the
> pulse marks do not need to be of high frequency train because the
> scope is calibrated for this at a time/div of only 10 microseconds.
> Could anyone suggest a simple circuit that would give me a fast rise
> pulse? Any help greatly appreciated.
> Larry Gagnon
> to send direct email remove "fake" from address
The old-fashioned way is to use a mercury reed relay inline with the
center conductor of a coax cable.
The preferred method is to use a General Radio "component insertion"
device, which had a 50 ohm GR connector on either end, and an open area
that could be shielded.
Pulse width can be set by connecting an open coax cable of one-half
the time required; the charging of the cable from a supply via a
One can generate pulses from millivolts to hundreds of volts, and if a
window is made to see the contacts, then a corresponding pulse of light
is also generated, providing the voltage is more than about 32 volts
(ionization potential of mercury). this latter form is used for
photomultiplier characterization, and is called a Huggins lamp (as i
Rise times are in the sub-nanosecond region; how good i cannot say
because in those daze when i was fiddling with such things, the fastest
scope i had available had a 1nSec rise time.
Of course, it will be a bitch to get a mercury relay of any type;
almost as hard as getting a radioactivity stabilized gas regulator.
An Esaki diode (AKA "tunnel diode") in parallel with an output winding
of a blocking oscillator can give decent rise times, but the amplitude
is obviously very limited and is nominally fixed.
Then again, you did not say what you considered to be fast.
The output of any TTL IC could be fast; use a 555...
Milliseconds, microseconds, nanoseconds, picoseconds, femtoseconds...
what is fast for you?