[time-nuts] Question concerning failure and value of HP 5371A
Arnold.Tibus at gmx.de
Sun Oct 16 14:03:32 EDT 2005
I think we are not anymore in line with the subject line, but this device I want to
build is to replace the not existing interval counter for some measurement
tasks I want to solve.
On Sun, 16 Oct 2005 02:24:32 +0200 (CEST), Magnus Danielson wrote:
>From: "Arnold Tibus" <Arnold.Tibus at gmx.de>
>Subject: Re: [time-nuts] Question concerning failure and value of HP 5371A
>Date: Sun, 16 Oct 2005 01:36:51 +0200
>Message-ID: <E1EQvbQ-0005o2-J8 at febo.com>
>Are you aquanted with the term analog interpolators? Basically they try to
>measure the time-difference between the counting-clock edges and the start/stop
>However, I have yeat to understand what you are really after. There is many
>ways to skin a cat, but maybe if I knew what type of cat you are looking to
>skin I might give some more ideas on what to do...
Magnus, it is easy to tell what I want (but propably hard to realize):
I want to count, as precise as possible, a gated pulse train. The number of
the pulses will be the measure for the width of the gate and its variations,
which is opened and closed by separate start / stop signal edges.
Obviously, the shorter the pulses the better the resolution, the more precise
will be the absolute value.
This circuit I want to use shall be useful to determine the time difference
between two signal edges and to determine a possible drift and jitter between
both signals. The generation of the necessary pulses with 100 MHz to
500 MHz seem not to be that problem, I got good ideas from existing commercial
circuit. But for the gate and the counter I have not found stardard circuits or ICs.
The problem are the very short rise and fall times and high repetition/ clock rates.
More information about the time-nuts